From 9447082ae666fbf3adbe9c9152117daa31a8b737 Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Wed, 16 Nov 2016 12:21:19 -0200 Subject: [media] smiapp: Implement power-on and power-off sequences without runtime PM Power on the sensor when the module is loaded and power it off when it is removed. Signed-off-by: Sakari Ailus Signed-off-by: Mauro Carvalho Chehab --- drivers/media/i2c/smiapp/smiapp-core.c | 29 ++++++++++------------------- 1 file changed, 10 insertions(+), 19 deletions(-) diff --git a/drivers/media/i2c/smiapp/smiapp-core.c b/drivers/media/i2c/smiapp/smiapp-core.c index 59872b31f832..620f8ce8185e 100644 --- a/drivers/media/i2c/smiapp/smiapp-core.c +++ b/drivers/media/i2c/smiapp/smiapp-core.c @@ -2741,8 +2741,6 @@ static const struct v4l2_subdev_internal_ops smiapp_internal_ops = { * I2C Driver */ -#ifdef CONFIG_PM - static int smiapp_suspend(struct device *dev) { struct i2c_client *client = to_i2c_client(dev); @@ -2783,13 +2781,6 @@ static int smiapp_resume(struct device *dev) return rval; } -#else - -#define smiapp_suspend NULL -#define smiapp_resume NULL - -#endif /* CONFIG_PM */ - static struct smiapp_hwconfig *smiapp_get_hwconfig(struct device *dev) { struct smiapp_hwconfig *hwcfg; @@ -2913,13 +2904,9 @@ static int smiapp_probe(struct i2c_client *client, if (IS_ERR(sensor->xshutdown)) return PTR_ERR(sensor->xshutdown); - pm_runtime_enable(&client->dev); - - rval = pm_runtime_get_sync(&client->dev); - if (rval < 0) { - rval = -ENODEV; - goto out_power_off; - } + rval = smiapp_power_on(&client->dev); + if (rval < 0) + return rval; rval = smiapp_identify_module(sensor); if (rval) { @@ -3100,6 +3087,9 @@ static int smiapp_probe(struct i2c_client *client, if (rval < 0) goto out_media_entity_cleanup; + pm_runtime_set_active(&client->dev); + pm_runtime_get_noresume(&client->dev); + pm_runtime_enable(&client->dev); pm_runtime_set_autosuspend_delay(&client->dev, 1000); pm_runtime_use_autosuspend(&client->dev); pm_runtime_put_autosuspend(&client->dev); @@ -3113,8 +3103,7 @@ out_cleanup: smiapp_cleanup(sensor); out_power_off: - pm_runtime_put(&client->dev); - pm_runtime_disable(&client->dev); + smiapp_power_off(&client->dev); return rval; } @@ -3127,8 +3116,10 @@ static int smiapp_remove(struct i2c_client *client) v4l2_async_unregister_subdev(subdev); - pm_runtime_suspend(&client->dev); pm_runtime_disable(&client->dev); + if (!pm_runtime_status_suspended(&client->dev)) + smiapp_power_off(&client->dev); + pm_runtime_set_suspended(&client->dev); for (i = 0; i < sensor->ssds_used; i++) { v4l2_device_unregister_subdev(&sensor->ssds[i].sd); -- cgit v1.2.3 From 4bfb934b0067b7f6a24470682c5f7254fd4d8282 Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Sat, 19 Nov 2016 19:50:10 -0200 Subject: [media] smiapp: Make suspend and resume functions __maybe_unused The smiapp_suspend() and smiapp_resume() functions will end up being unused if CONFIG_PM is enabled but CONFIG_PM_SLEEP is disabled, causing a compiler warning from both of the function definitions. Fix this by marking the functions with __maybe_unused. Suggested-by: Arnd Bergmann Signed-off-by: Sakari Ailus Signed-off-by: Mauro Carvalho Chehab --- drivers/media/i2c/smiapp/smiapp-core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/media/i2c/smiapp/smiapp-core.c b/drivers/media/i2c/smiapp/smiapp-core.c index 620f8ce8185e..f4e92bdfe192 100644 --- a/drivers/media/i2c/smiapp/smiapp-core.c +++ b/drivers/media/i2c/smiapp/smiapp-core.c @@ -2741,7 +2741,7 @@ static const struct v4l2_subdev_internal_ops smiapp_internal_ops = { * I2C Driver */ -static int smiapp_suspend(struct device *dev) +static int __maybe_unused smiapp_suspend(struct device *dev) { struct i2c_client *client = to_i2c_client(dev); struct v4l2_subdev *subdev = i2c_get_clientdata(client); @@ -2766,7 +2766,7 @@ static int smiapp_suspend(struct device *dev) return 0; } -static int smiapp_resume(struct device *dev) +static int __maybe_unused smiapp_resume(struct device *dev) { struct i2c_client *client = to_i2c_client(dev); struct v4l2_subdev *subdev = i2c_get_clientdata(client); -- cgit v1.2.3 From f3854973f196baad5be6b62d8f5ea24b0346b63f Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Tue, 6 Dec 2016 11:17:12 -0200 Subject: [media] cec: fix report_current_latency In the (very) small print of the REPORT_CURRENT_LATENCY message there is a line that says that the last byte of the message (audio out delay) is only present if the 'audio out compensated' value is 3. I missed this, and so if this message was sent with a total length of 6 (i.e. without the audio out delay byte), then it was rejected by the framework since a minimum length of 7 was expected. Fix this minimum length check and update the wrappers in cec-funcs.h to do the right thing based on the message length. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/cec/cec-adap.c | 2 +- include/uapi/linux/cec-funcs.h | 10 +++++++--- 2 files changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/media/cec/cec-adap.c b/drivers/media/cec/cec-adap.c index 0ea4efb3de66..f15f6ffd75b2 100644 --- a/drivers/media/cec/cec-adap.c +++ b/drivers/media/cec/cec-adap.c @@ -851,7 +851,7 @@ static const u8 cec_msg_size[256] = { [CEC_MSG_REQUEST_ARC_TERMINATION] = 2 | DIRECTED, [CEC_MSG_TERMINATE_ARC] = 2 | DIRECTED, [CEC_MSG_REQUEST_CURRENT_LATENCY] = 4 | BCAST, - [CEC_MSG_REPORT_CURRENT_LATENCY] = 7 | BCAST, + [CEC_MSG_REPORT_CURRENT_LATENCY] = 6 | BCAST, [CEC_MSG_CDC_MESSAGE] = 2 | BCAST, }; diff --git a/include/uapi/linux/cec-funcs.h b/include/uapi/linux/cec-funcs.h index 3cbc327801d6..c451eec42a83 100644 --- a/include/uapi/linux/cec-funcs.h +++ b/include/uapi/linux/cec-funcs.h @@ -1665,14 +1665,15 @@ static inline void cec_msg_report_current_latency(struct cec_msg *msg, __u8 audio_out_compensated, __u8 audio_out_delay) { - msg->len = 7; + msg->len = 6; msg->msg[0] |= 0xf; /* broadcast */ msg->msg[1] = CEC_MSG_REPORT_CURRENT_LATENCY; msg->msg[2] = phys_addr >> 8; msg->msg[3] = phys_addr & 0xff; msg->msg[4] = video_latency; msg->msg[5] = (low_latency_mode << 2) | audio_out_compensated; - msg->msg[6] = audio_out_delay; + if (audio_out_compensated == 3) + msg->msg[msg->len++] = audio_out_delay; } static inline void cec_ops_report_current_latency(const struct cec_msg *msg, @@ -1686,7 +1687,10 @@ static inline void cec_ops_report_current_latency(const struct cec_msg *msg, *video_latency = msg->msg[4]; *low_latency_mode = (msg->msg[5] >> 2) & 1; *audio_out_compensated = msg->msg[5] & 3; - *audio_out_delay = msg->msg[6]; + if (*audio_out_compensated == 3 && msg->len >= 7) + *audio_out_delay = msg->msg[6]; + else + *audio_out_delay = 0; } static inline void cec_msg_request_current_latency(struct cec_msg *msg, -- cgit v1.2.3 From 120476123646ba3619c90db7bcbc6f8eea53c990 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Fri, 9 Dec 2016 11:14:32 -0200 Subject: [media] cec: when canceling a message, don't overwrite old status info When a pending message was canceled (e.g. due to a timeout), then the old tx_status info was overwritten instead of ORed. The same happened with the tx_error_cnt field. So just modify them instead of overwriting them. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/cec/cec-adap.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/media/cec/cec-adap.c b/drivers/media/cec/cec-adap.c index f15f6ffd75b2..3191c0c5eae1 100644 --- a/drivers/media/cec/cec-adap.c +++ b/drivers/media/cec/cec-adap.c @@ -288,10 +288,10 @@ static void cec_data_cancel(struct cec_data *data) /* Mark it as an error */ data->msg.tx_ts = ktime_get_ns(); - data->msg.tx_status = CEC_TX_STATUS_ERROR | - CEC_TX_STATUS_MAX_RETRIES; + data->msg.tx_status |= CEC_TX_STATUS_ERROR | + CEC_TX_STATUS_MAX_RETRIES; + data->msg.tx_error_cnt++; data->attempts = 0; - data->msg.tx_error_cnt = 1; /* Queue transmitted message for monitoring purposes */ cec_queue_msg_monitor(data->adap, &data->msg, 1); -- cgit v1.2.3 From a24f56d47930492c94ef6875bf45adf7607ca1a4 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Fri, 9 Dec 2016 11:28:19 -0200 Subject: [media] cec: CEC_MSG_GIVE_FEATURES should abort for CEC version < 2 This is a 2.0 only message, so it should return Feature Abort if the adapter is configured for CEC version 1.4. Right now it does nothing, which means that the sender will time out. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/cec/cec-adap.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/media/cec/cec-adap.c b/drivers/media/cec/cec-adap.c index 3191c0c5eae1..c05956fdd500 100644 --- a/drivers/media/cec/cec-adap.c +++ b/drivers/media/cec/cec-adap.c @@ -1777,9 +1777,9 @@ static int cec_receive_notify(struct cec_adapter *adap, struct cec_msg *msg, } case CEC_MSG_GIVE_FEATURES: - if (adap->log_addrs.cec_version >= CEC_OP_CEC_VERSION_2_0) - return cec_report_features(adap, la_idx); - return 0; + if (adap->log_addrs.cec_version < CEC_OP_CEC_VERSION_2_0) + return cec_feature_abort(adap, msg); + return cec_report_features(adap, la_idx); default: /* -- cgit v1.2.3 From 7af26f889eb67db272021a939f7d4a57e96dd961 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Fri, 9 Dec 2016 11:54:06 -0200 Subject: [media] cec: update log_addr[] before finishing configuration The loop that sets the unused logical addresses to INVALID should be done before 'configured' is set to true. This ensures that cec_log_addrs is consistent before it will be used. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/cec/cec-adap.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/media/cec/cec-adap.c b/drivers/media/cec/cec-adap.c index c05956fdd500..f3fef487f28d 100644 --- a/drivers/media/cec/cec-adap.c +++ b/drivers/media/cec/cec-adap.c @@ -1250,6 +1250,8 @@ configured: for (i = 1; i < las->num_log_addrs; i++) las->log_addr[i] = CEC_LOG_ADDR_INVALID; } + for (i = las->num_log_addrs; i < CEC_MAX_LOG_ADDRS; i++) + las->log_addr[i] = CEC_LOG_ADDR_INVALID; adap->is_configured = true; adap->is_configuring = false; cec_post_state_event(adap); @@ -1268,8 +1270,6 @@ configured: cec_report_features(adap, i); cec_report_phys_addr(adap, i); } - for (i = las->num_log_addrs; i < CEC_MAX_LOG_ADDRS; i++) - las->log_addr[i] = CEC_LOG_ADDR_INVALID; mutex_lock(&adap->lock); adap->kthread_config = NULL; mutex_unlock(&adap->lock); -- cgit v1.2.3 From 52bc30fda9622f492427d484bd4dd8ee42cc4667 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Fri, 9 Dec 2016 11:36:03 -0200 Subject: [media] cec: replace cec_report_features by cec_fill_msg_report_features The fill function just fills in the cec_msg struct, it doesn't transmit the message. This is now done explicitly. This makes it possible to switch to transmitting this message with adap->lock held. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/cec/cec-adap.c | 48 ++++++++++++++++++++++++-------------------- 1 file changed, 26 insertions(+), 22 deletions(-) diff --git a/drivers/media/cec/cec-adap.c b/drivers/media/cec/cec-adap.c index f3fef487f28d..2b668510ca36 100644 --- a/drivers/media/cec/cec-adap.c +++ b/drivers/media/cec/cec-adap.c @@ -30,8 +30,10 @@ #include "cec-priv.h" -static int cec_report_features(struct cec_adapter *adap, unsigned int la_idx); static int cec_report_phys_addr(struct cec_adapter *adap, unsigned int la_idx); +static void cec_fill_msg_report_features(struct cec_adapter *adap, + struct cec_msg *msg, + unsigned int la_idx); /* * 400 ms is the time it takes for one 16 byte message to be @@ -1258,16 +1260,21 @@ configured: mutex_unlock(&adap->lock); for (i = 0; i < las->num_log_addrs; i++) { + struct cec_msg msg = {}; + if (las->log_addr[i] == CEC_LOG_ADDR_INVALID || (las->flags & CEC_LOG_ADDRS_FL_CDC_ONLY)) continue; - /* - * Report Features must come first according - * to CEC 2.0 - */ - if (las->log_addr[i] != CEC_LOG_ADDR_UNREGISTERED) - cec_report_features(adap, i); + msg.msg[0] = (las->log_addr[i] << 4) | 0x0f; + + /* Report Features must come first according to CEC 2.0 */ + if (las->log_addr[i] != CEC_LOG_ADDR_UNREGISTERED && + adap->log_addrs.cec_version >= CEC_OP_CEC_VERSION_2_0) { + cec_fill_msg_report_features(adap, &msg, i); + cec_transmit_msg(adap, &msg, false); + } + cec_report_phys_addr(adap, i); } mutex_lock(&adap->lock); @@ -1526,36 +1533,32 @@ EXPORT_SYMBOL_GPL(cec_s_log_addrs); /* High-level core CEC message handling */ -/* Transmit the Report Features message */ -static int cec_report_features(struct cec_adapter *adap, unsigned int la_idx) +/* Fill in the Report Features message */ +static void cec_fill_msg_report_features(struct cec_adapter *adap, + struct cec_msg *msg, + unsigned int la_idx) { - struct cec_msg msg = { }; const struct cec_log_addrs *las = &adap->log_addrs; const u8 *features = las->features[la_idx]; bool op_is_dev_features = false; unsigned int idx; - /* This is 2.0 and up only */ - if (adap->log_addrs.cec_version < CEC_OP_CEC_VERSION_2_0) - return 0; - /* Report Features */ - msg.msg[0] = (las->log_addr[la_idx] << 4) | 0x0f; - msg.len = 4; - msg.msg[1] = CEC_MSG_REPORT_FEATURES; - msg.msg[2] = adap->log_addrs.cec_version; - msg.msg[3] = las->all_device_types[la_idx]; + msg->msg[0] = (las->log_addr[la_idx] << 4) | 0x0f; + msg->len = 4; + msg->msg[1] = CEC_MSG_REPORT_FEATURES; + msg->msg[2] = adap->log_addrs.cec_version; + msg->msg[3] = las->all_device_types[la_idx]; /* Write RC Profiles first, then Device Features */ for (idx = 0; idx < ARRAY_SIZE(las->features[0]); idx++) { - msg.msg[msg.len++] = features[idx]; + msg->msg[msg->len++] = features[idx]; if ((features[idx] & CEC_OP_FEAT_EXT) == 0) { if (op_is_dev_features) break; op_is_dev_features = true; } } - return cec_transmit_msg(adap, &msg, false); } /* Transmit the Report Physical Address message */ @@ -1779,7 +1782,8 @@ static int cec_receive_notify(struct cec_adapter *adap, struct cec_msg *msg, case CEC_MSG_GIVE_FEATURES: if (adap->log_addrs.cec_version < CEC_OP_CEC_VERSION_2_0) return cec_feature_abort(adap, msg); - return cec_report_features(adap, la_idx); + cec_fill_msg_report_features(adap, &tx_cec_msg, la_idx); + return cec_transmit_msg(adap, &tx_cec_msg, false); default: /* -- cgit v1.2.3 From d3d64bc7408f1ff0b0ff8354056e2a48eda5886d Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Fri, 9 Dec 2016 11:48:34 -0200 Subject: [media] cec: move cec_report_phys_addr into cec_config_thread_func It's only a small function and this makes it easier to switch to transmitting the message with adap->lock held in the next patch. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/cec/cec-adap.c | 25 +++++++------------------ 1 file changed, 7 insertions(+), 18 deletions(-) diff --git a/drivers/media/cec/cec-adap.c b/drivers/media/cec/cec-adap.c index 2b668510ca36..f3d495654a53 100644 --- a/drivers/media/cec/cec-adap.c +++ b/drivers/media/cec/cec-adap.c @@ -30,7 +30,6 @@ #include "cec-priv.h" -static int cec_report_phys_addr(struct cec_adapter *adap, unsigned int la_idx); static void cec_fill_msg_report_features(struct cec_adapter *adap, struct cec_msg *msg, unsigned int la_idx); @@ -1275,7 +1274,13 @@ configured: cec_transmit_msg(adap, &msg, false); } - cec_report_phys_addr(adap, i); + /* Report Physical Address */ + cec_msg_report_physical_addr(&msg, adap->phys_addr, + las->primary_device_type[i]); + dprintk(2, "config: la %d pa %x.%x.%x.%x\n", + las->log_addr[i], + cec_phys_addr_exp(adap->phys_addr)); + cec_transmit_msg(adap, &msg, false); } mutex_lock(&adap->lock); adap->kthread_config = NULL; @@ -1561,22 +1566,6 @@ static void cec_fill_msg_report_features(struct cec_adapter *adap, } } -/* Transmit the Report Physical Address message */ -static int cec_report_phys_addr(struct cec_adapter *adap, unsigned int la_idx) -{ - const struct cec_log_addrs *las = &adap->log_addrs; - struct cec_msg msg = { }; - - /* Report Physical Address */ - msg.msg[0] = (las->log_addr[la_idx] << 4) | 0x0f; - cec_msg_report_physical_addr(&msg, adap->phys_addr, - las->primary_device_type[la_idx]); - dprintk(2, "config: la %d pa %x.%x.%x.%x\n", - las->log_addr[la_idx], - cec_phys_addr_exp(adap->phys_addr)); - return cec_transmit_msg(adap, &msg, false); -} - /* Transmit the Feature Abort message */ static int cec_feature_abort_reason(struct cec_adapter *adap, struct cec_msg *msg, u8 reason) -- cgit v1.2.3 From f60f35609f89ef4fee73776bc1ef697923251995 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Fri, 9 Dec 2016 12:00:49 -0200 Subject: [media] cec: fix race between configuring and unconfiguring This race was discovered by running cec-compliance -A with the cec module debug parameter set to 2: suddenly the test would fail. It turns out that this happens when the test configures the adapter in non-blocking mode, then it waits for the CEC_EVENT_STATE_CHANGE event and once the event is received it unconfigures the adapter. What happened was that the unconfigure was executed while the configure was still transmitting the Report Features and Report Physical Address messages. This messed up the internal state of the cec_adapter. The fix is to transmit those messages with the adap->lock mutex held (this will just queue them up in the internal transmit queue, and not actually transmit anything yet). Only unlock the mutex once everything is done. The main thread will dequeue the messages from the internal transmit queue and transmit them one by one, unless an unconfigure was done, and in that case any messages are just dropped. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/cec/cec-adap.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff --git a/drivers/media/cec/cec-adap.c b/drivers/media/cec/cec-adap.c index f3d495654a53..ebb5e391b800 100644 --- a/drivers/media/cec/cec-adap.c +++ b/drivers/media/cec/cec-adap.c @@ -1256,8 +1256,17 @@ configured: adap->is_configured = true; adap->is_configuring = false; cec_post_state_event(adap); - mutex_unlock(&adap->lock); + /* + * Now post the Report Features and Report Physical Address broadcast + * messages. Note that these are non-blocking transmits, meaning that + * they are just queued up and once adap->lock is unlocked the main + * thread will kick in and start transmitting these. + * + * If after this function is done (but before one or more of these + * messages are actually transmitted) the CEC adapter is unconfigured, + * then any remaining messages will be dropped by the main thread. + */ for (i = 0; i < las->num_log_addrs; i++) { struct cec_msg msg = {}; @@ -1271,7 +1280,7 @@ configured: if (las->log_addr[i] != CEC_LOG_ADDR_UNREGISTERED && adap->log_addrs.cec_version >= CEC_OP_CEC_VERSION_2_0) { cec_fill_msg_report_features(adap, &msg, i); - cec_transmit_msg(adap, &msg, false); + cec_transmit_msg_fh(adap, &msg, NULL, false); } /* Report Physical Address */ @@ -1280,12 +1289,11 @@ configured: dprintk(2, "config: la %d pa %x.%x.%x.%x\n", las->log_addr[i], cec_phys_addr_exp(adap->phys_addr)); - cec_transmit_msg(adap, &msg, false); + cec_transmit_msg_fh(adap, &msg, NULL, false); } - mutex_lock(&adap->lock); adap->kthread_config = NULL; - mutex_unlock(&adap->lock); complete(&adap->config_completion); + mutex_unlock(&adap->lock); return 0; unconfigure: -- cgit v1.2.3 From 78ccbf9ff89bd7a20d36be039cb3eab71081648c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sun, 11 Sep 2016 10:31:28 -0300 Subject: [media] media/cobalt: use pci_irq_allocate_vectors Simply the interrupt setup by using the new PCI layer helpers. Despite using pci_enable_msi_range, this driver was only requesting a single MSI vector anyway. Signed-off-by: Christoph Hellwig Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/pci/cobalt/cobalt-driver.c | 8 ++------ drivers/media/pci/cobalt/cobalt-driver.h | 2 -- 2 files changed, 2 insertions(+), 8 deletions(-) diff --git a/drivers/media/pci/cobalt/cobalt-driver.c b/drivers/media/pci/cobalt/cobalt-driver.c index 979634000597..d5c911c09e2b 100644 --- a/drivers/media/pci/cobalt/cobalt-driver.c +++ b/drivers/media/pci/cobalt/cobalt-driver.c @@ -308,9 +308,7 @@ static void cobalt_pci_iounmap(struct cobalt *cobalt, struct pci_dev *pci_dev) static void cobalt_free_msi(struct cobalt *cobalt, struct pci_dev *pci_dev) { free_irq(pci_dev->irq, (void *)cobalt); - - if (cobalt->msi_enabled) - pci_disable_msi(pci_dev); + pci_free_irq_vectors(pci_dev); } static int cobalt_setup_pci(struct cobalt *cobalt, struct pci_dev *pci_dev, @@ -387,14 +385,12 @@ static int cobalt_setup_pci(struct cobalt *cobalt, struct pci_dev *pci_dev, from being generated. */ cobalt_set_interrupt(cobalt, false); - if (pci_enable_msi_range(pci_dev, 1, 1) < 1) { + if (pci_alloc_irq_vectors(pci_dev, 1, 1, PCI_IRQ_MSI) < 1) { cobalt_err("Could not enable MSI\n"); - cobalt->msi_enabled = false; ret = -EIO; goto err_release; } msi_config_show(cobalt, pci_dev); - cobalt->msi_enabled = true; /* Register IRQ */ if (request_irq(pci_dev->irq, cobalt_irq_handler, IRQF_SHARED, diff --git a/drivers/media/pci/cobalt/cobalt-driver.h b/drivers/media/pci/cobalt/cobalt-driver.h index ed00dc9d9399..00f773ec359a 100644 --- a/drivers/media/pci/cobalt/cobalt-driver.h +++ b/drivers/media/pci/cobalt/cobalt-driver.h @@ -287,8 +287,6 @@ struct cobalt { u32 irq_none; u32 irq_full_fifo; - bool msi_enabled; - /* omnitek dma */ int dma_channels; int first_fifo_channel; -- cgit v1.2.3 From 48775cb73c2e26b7ca9d679875a6e570c8b8e124 Mon Sep 17 00:00:00 2001 From: Max Kellermann Date: Thu, 15 Dec 2016 19:51:07 -0200 Subject: [media] pctv452e: move buffer to heap, no mutex commit 73d5c5c864f4 ("[media] pctv452e: don't do DMA on stack") caused a NULL pointer dereference which occurs when dvb_usb_init() calls dvb_usb_device_power_ctrl() for the first time, before the frontend has been attached. It also caused a recursive deadlock because tt3650_ci_msg_locked() has already locked the mutex. So, partially revert it, but move the buffer to the heap (DMA capable), not to the stack (may not be DMA capable). Instead of sharing one buffer which needs mutex protection, do a new heap allocation for each call. Fixes: commit 73d5c5c864f4 ("[media] pctv452e: don't do DMA on stack") Cc: stable@vger.kernel.org # For Kernel 4.9 Signed-off-by: Max Kellermann Signed-off-by: Mauro Carvalho Chehab --- drivers/media/usb/dvb-usb/pctv452e.c | 133 +++++++++++++++++++---------------- 1 file changed, 72 insertions(+), 61 deletions(-) diff --git a/drivers/media/usb/dvb-usb/pctv452e.c b/drivers/media/usb/dvb-usb/pctv452e.c index 07fa08be9e99..d54ebe7e0215 100644 --- a/drivers/media/usb/dvb-usb/pctv452e.c +++ b/drivers/media/usb/dvb-usb/pctv452e.c @@ -97,14 +97,13 @@ struct pctv452e_state { u8 c; /* transaction counter, wraps around... */ u8 initialized; /* set to 1 if 0x15 has been sent */ u16 last_rc_key; - - unsigned char data[80]; }; static int tt3650_ci_msg(struct dvb_usb_device *d, u8 cmd, u8 *data, unsigned int write_len, unsigned int read_len) { struct pctv452e_state *state = (struct pctv452e_state *)d->priv; + u8 *buf; u8 id; unsigned int rlen; int ret; @@ -114,36 +113,39 @@ static int tt3650_ci_msg(struct dvb_usb_device *d, u8 cmd, u8 *data, return -EIO; } - mutex_lock(&state->ca_mutex); + buf = kmalloc(64, GFP_KERNEL); + if (!buf) + return -ENOMEM; + id = state->c++; - state->data[0] = SYNC_BYTE_OUT; - state->data[1] = id; - state->data[2] = cmd; - state->data[3] = write_len; + buf[0] = SYNC_BYTE_OUT; + buf[1] = id; + buf[2] = cmd; + buf[3] = write_len; - memcpy(state->data + 4, data, write_len); + memcpy(buf + 4, data, write_len); rlen = (read_len > 0) ? 64 : 0; - ret = dvb_usb_generic_rw(d, state->data, 4 + write_len, - state->data, rlen, /* delay_ms */ 0); + ret = dvb_usb_generic_rw(d, buf, 4 + write_len, + buf, rlen, /* delay_ms */ 0); if (0 != ret) goto failed; ret = -EIO; - if (SYNC_BYTE_IN != state->data[0] || id != state->data[1]) + if (SYNC_BYTE_IN != buf[0] || id != buf[1]) goto failed; - memcpy(data, state->data + 4, read_len); + memcpy(data, buf + 4, read_len); - mutex_unlock(&state->ca_mutex); + kfree(buf); return 0; failed: err("CI error %d; %02X %02X %02X -> %*ph.", - ret, SYNC_BYTE_OUT, id, cmd, 3, state->data); + ret, SYNC_BYTE_OUT, id, cmd, 3, buf); - mutex_unlock(&state->ca_mutex); + kfree(buf); return ret; } @@ -410,53 +412,57 @@ static int pctv452e_i2c_msg(struct dvb_usb_device *d, u8 addr, u8 *rcv_buf, u8 rcv_len) { struct pctv452e_state *state = (struct pctv452e_state *)d->priv; + u8 *buf; u8 id; int ret; - mutex_lock(&state->ca_mutex); + buf = kmalloc(64, GFP_KERNEL); + if (!buf) + return -ENOMEM; + id = state->c++; ret = -EINVAL; if (snd_len > 64 - 7 || rcv_len > 64 - 7) goto failed; - state->data[0] = SYNC_BYTE_OUT; - state->data[1] = id; - state->data[2] = PCTV_CMD_I2C; - state->data[3] = snd_len + 3; - state->data[4] = addr << 1; - state->data[5] = snd_len; - state->data[6] = rcv_len; + buf[0] = SYNC_BYTE_OUT; + buf[1] = id; + buf[2] = PCTV_CMD_I2C; + buf[3] = snd_len + 3; + buf[4] = addr << 1; + buf[5] = snd_len; + buf[6] = rcv_len; - memcpy(state->data + 7, snd_buf, snd_len); + memcpy(buf + 7, snd_buf, snd_len); - ret = dvb_usb_generic_rw(d, state->data, 7 + snd_len, - state->data, /* rcv_len */ 64, + ret = dvb_usb_generic_rw(d, buf, 7 + snd_len, + buf, /* rcv_len */ 64, /* delay_ms */ 0); if (ret < 0) goto failed; /* TT USB protocol error. */ ret = -EIO; - if (SYNC_BYTE_IN != state->data[0] || id != state->data[1]) + if (SYNC_BYTE_IN != buf[0] || id != buf[1]) goto failed; /* I2C device didn't respond as expected. */ ret = -EREMOTEIO; - if (state->data[5] < snd_len || state->data[6] < rcv_len) + if (buf[5] < snd_len || buf[6] < rcv_len) goto failed; - memcpy(rcv_buf, state->data + 7, rcv_len); - mutex_unlock(&state->ca_mutex); + memcpy(rcv_buf, buf + 7, rcv_len); + kfree(buf); return rcv_len; failed: err("I2C error %d; %02X %02X %02X %02X %02X -> %*ph", ret, SYNC_BYTE_OUT, id, addr << 1, snd_len, rcv_len, - 7, state->data); + 7, buf); - mutex_unlock(&state->ca_mutex); + kfree(buf); return ret; } @@ -505,7 +511,7 @@ static u32 pctv452e_i2c_func(struct i2c_adapter *adapter) static int pctv452e_power_ctrl(struct dvb_usb_device *d, int i) { struct pctv452e_state *state = (struct pctv452e_state *)d->priv; - u8 *rx; + u8 *b0, *rx; int ret; info("%s: %d\n", __func__, i); @@ -516,11 +522,12 @@ static int pctv452e_power_ctrl(struct dvb_usb_device *d, int i) if (state->initialized) return 0; - rx = kmalloc(PCTV_ANSWER_LEN, GFP_KERNEL); - if (!rx) + b0 = kmalloc(5 + PCTV_ANSWER_LEN, GFP_KERNEL); + if (!b0) return -ENOMEM; - mutex_lock(&state->ca_mutex); + rx = b0 + 5; + /* hmm where shoud this should go? */ ret = usb_set_interface(d->udev, 0, ISOC_INTERFACE_ALTERNATIVE); if (ret != 0) @@ -528,66 +535,70 @@ static int pctv452e_power_ctrl(struct dvb_usb_device *d, int i) __func__, ret); /* this is a one-time initialization, dont know where to put */ - state->data[0] = 0xaa; - state->data[1] = state->c++; - state->data[2] = PCTV_CMD_RESET; - state->data[3] = 1; - state->data[4] = 0; + b0[0] = 0xaa; + b0[1] = state->c++; + b0[2] = PCTV_CMD_RESET; + b0[3] = 1; + b0[4] = 0; /* reset board */ - ret = dvb_usb_generic_rw(d, state->data, 5, rx, PCTV_ANSWER_LEN, 0); + ret = dvb_usb_generic_rw(d, b0, 5, rx, PCTV_ANSWER_LEN, 0); if (ret) goto ret; - state->data[1] = state->c++; - state->data[4] = 1; + b0[1] = state->c++; + b0[4] = 1; /* reset board (again?) */ - ret = dvb_usb_generic_rw(d, state->data, 5, rx, PCTV_ANSWER_LEN, 0); + ret = dvb_usb_generic_rw(d, b0, 5, rx, PCTV_ANSWER_LEN, 0); if (ret) goto ret; state->initialized = 1; ret: - mutex_unlock(&state->ca_mutex); - kfree(rx); + kfree(b0); return ret; } static int pctv452e_rc_query(struct dvb_usb_device *d) { struct pctv452e_state *state = (struct pctv452e_state *)d->priv; + u8 *b, *rx; int ret, i; u8 id; - mutex_lock(&state->ca_mutex); + b = kmalloc(CMD_BUFFER_SIZE + PCTV_ANSWER_LEN, GFP_KERNEL); + if (!b) + return -ENOMEM; + + rx = b + CMD_BUFFER_SIZE; + id = state->c++; /* prepare command header */ - state->data[0] = SYNC_BYTE_OUT; - state->data[1] = id; - state->data[2] = PCTV_CMD_IR; - state->data[3] = 0; + b[0] = SYNC_BYTE_OUT; + b[1] = id; + b[2] = PCTV_CMD_IR; + b[3] = 0; /* send ir request */ - ret = dvb_usb_generic_rw(d, state->data, 4, - state->data, PCTV_ANSWER_LEN, 0); + ret = dvb_usb_generic_rw(d, b, 4, rx, PCTV_ANSWER_LEN, 0); if (ret != 0) goto ret; if (debug > 3) { - info("%s: read: %2d: %*ph: ", __func__, ret, 3, state->data); - for (i = 0; (i < state->data[3]) && ((i + 3) < PCTV_ANSWER_LEN); i++) - info(" %02x", state->data[i + 3]); + info("%s: read: %2d: %*ph: ", __func__, ret, 3, rx); + for (i = 0; (i < rx[3]) && ((i+3) < PCTV_ANSWER_LEN); i++) + info(" %02x", rx[i+3]); info("\n"); } - if ((state->data[3] == 9) && (state->data[12] & 0x01)) { + if ((rx[3] == 9) && (rx[12] & 0x01)) { /* got a "press" event */ - state->last_rc_key = RC_SCANCODE_RC5(state->data[7], state->data[6]); + state->last_rc_key = RC_SCANCODE_RC5(rx[7], rx[6]); if (debug > 2) info("%s: cmd=0x%02x sys=0x%02x\n", - __func__, state->data[6], state->data[7]); + __func__, rx[6], rx[7]); rc_keydown(d->rc_dev, RC_TYPE_RC5, state->last_rc_key, 0); } else if (state->last_rc_key) { @@ -595,7 +606,7 @@ static int pctv452e_rc_query(struct dvb_usb_device *d) state->last_rc_key = 0; } ret: - mutex_unlock(&state->ca_mutex); + kfree(b); return ret; } -- cgit v1.2.3 From aff808e813fc2d311137754165cf53d4ee6ddcc2 Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Fri, 9 Dec 2016 09:47:17 -0200 Subject: [media] v4l: tvp5150: Reset device at probe time, not in get/set format handlers The tvp5150 doesn't support format setting through the subdev pad API and thus implements the set format handler as a get format operation. The single handler, tvp5150_fill_fmt(), resets the device by calling tvp5150_reset(). This causes malfunction as the device can be reset at will, possibly from userspace when the subdev userspace API is enabled. The reset call was added in commit ec2c4f3f93cb ("[media] media: tvp5150: Add mbus_fmt callbacks"), probably as an attempt to set the device to a known state before detecting the current TV standard. However, the get format handler doesn't access the hardware to get the TV standard since commit 963ddc63e20d ("[media] media: tvp5150: Add cropping support"). There is thus no need to reset the device when getting the format. However, removing the tvp5150_reset() from the get/set format handlers results in the function not being called at all if the bridge driver doesn't use the .reset() operation. The operation is nowadays abused and shouldn't be used, so shouldn't expect bridge drivers to call it. To make sure the device is properly initialize, move the reset call from the format handlers to the probe function. Cc: stable@vger.kernel.org # For Kernel 4.5 and upper Signed-off-by: Laurent Pinchart Signed-off-by: Mauro Carvalho Chehab --- drivers/media/i2c/tvp5150.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/media/i2c/tvp5150.c b/drivers/media/i2c/tvp5150.c index 3a0fe8cc64e9..a30bfcb4eec6 100644 --- a/drivers/media/i2c/tvp5150.c +++ b/drivers/media/i2c/tvp5150.c @@ -861,8 +861,6 @@ static int tvp5150_fill_fmt(struct v4l2_subdev *sd, f = &format->format; - tvp5150_reset(sd, 0); - f->width = decoder->rect.width; f->height = decoder->rect.height / 2; @@ -1524,7 +1522,6 @@ static int tvp5150_probe(struct i2c_client *c, res = core->hdl.error; goto err; } - v4l2_ctrl_handler_setup(&core->hdl); /* Default is no cropping */ core->rect.top = 0; @@ -1535,6 +1532,8 @@ static int tvp5150_probe(struct i2c_client *c, core->rect.left = 0; core->rect.width = TVP5150_H_MAX; + tvp5150_reset(sd, 0); /* Calls v4l2_ctrl_handler_setup() */ + res = v4l2_async_register_subdev(sd); if (res < 0) goto err; -- cgit v1.2.3 From b4b2de386bbb6589d81596999d4a924928dc119b Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Fri, 9 Dec 2016 09:47:18 -0200 Subject: [media] v4l: tvp5150: Fix comment regarding output pin muxing The FID/GLCO/VLK/HVLK and INTREQ/GPCL/VBLK pins are muxed differently depending on whether the input is an S-Video or composite signal. The comment that explains the logic doesn't reflect the code. It appears that the comment is incorrect, as disabling the output data bus in composite mode makes no sense. Update the comment to match the code. While at it define macros for the MISC_CTL register bits, the code is too confusing with numerical values. Cc: stable@vger.kernel.org # For Kernel 4.5 and upper Signed-off-by: Laurent Pinchart Signed-off-by: Mauro Carvalho Chehab --- drivers/media/i2c/tvp5150.c | 24 +++++++++++++++++------- drivers/media/i2c/tvp5150_reg.h | 9 +++++++++ 2 files changed, 26 insertions(+), 7 deletions(-) diff --git a/drivers/media/i2c/tvp5150.c b/drivers/media/i2c/tvp5150.c index a30bfcb4eec6..8852fa8c957b 100644 --- a/drivers/media/i2c/tvp5150.c +++ b/drivers/media/i2c/tvp5150.c @@ -291,8 +291,12 @@ static void tvp5150_selmux(struct v4l2_subdev *sd) tvp5150_write(sd, TVP5150_OP_MODE_CTL, opmode); tvp5150_write(sd, TVP5150_VD_IN_SRC_SEL_1, input); - /* Svideo should enable YCrCb output and disable GPCL output - * For Composite and TV, it should be the reverse + /* + * Setup the FID/GLCO/VLK/HVLK and INTREQ/GPCL/VBLK output signals. For + * S-Video we output the vertical lock (VLK) signal on FID/GLCO/VLK/HVLK + * and set INTREQ/GPCL/VBLK to logic 0. For composite we output the + * field indicator (FID) signal on FID/GLCO/VLK/HVLK and set + * INTREQ/GPCL/VBLK to logic 1. */ val = tvp5150_read(sd, TVP5150_MISC_CTL); if (val < 0) { @@ -301,9 +305,9 @@ static void tvp5150_selmux(struct v4l2_subdev *sd) } if (decoder->input == TVP5150_SVIDEO) - val = (val & ~0x40) | 0x10; + val = (val & ~TVP5150_MISC_CTL_GPCL) | TVP5150_MISC_CTL_HVLK; else - val = (val & ~0x10) | 0x40; + val = (val & ~TVP5150_MISC_CTL_HVLK) | TVP5150_MISC_CTL_GPCL; tvp5150_write(sd, TVP5150_MISC_CTL, val); }; @@ -455,7 +459,12 @@ static const struct i2c_reg_value tvp5150_init_enable[] = { },{ /* Automatic offset and AGC enabled */ TVP5150_ANAL_CHL_CTL, 0x15 },{ /* Activate YCrCb output 0x9 or 0xd ? */ - TVP5150_MISC_CTL, 0x6f + TVP5150_MISC_CTL, TVP5150_MISC_CTL_GPCL | + TVP5150_MISC_CTL_INTREQ_OE | + TVP5150_MISC_CTL_YCBCR_OE | + TVP5150_MISC_CTL_SYNC_OE | + TVP5150_MISC_CTL_VBLANK | + TVP5150_MISC_CTL_CLOCK_OE, },{ /* Activates video std autodetection for all standards */ TVP5150_AUTOSW_MSK, 0x0 },{ /* Default format: 0x47. For 4:2:2: 0x40 */ @@ -1050,11 +1059,12 @@ static int tvp5150_s_stream(struct v4l2_subdev *sd, int enable) { struct tvp5150 *decoder = to_tvp5150(sd); /* Output format: 8-bit ITU-R BT.656 with embedded syncs */ - int val = 0x09; + int val = TVP5150_MISC_CTL_YCBCR_OE | TVP5150_MISC_CTL_CLOCK_OE; /* Output format: 8-bit 4:2:2 YUV with discrete sync */ if (decoder->mbus_type == V4L2_MBUS_PARALLEL) - val = 0x0d; + val = TVP5150_MISC_CTL_YCBCR_OE | TVP5150_MISC_CTL_SYNC_OE + | TVP5150_MISC_CTL_CLOCK_OE; /* Initializes TVP5150 to its default values */ /* # set PCLK (27MHz) */ diff --git a/drivers/media/i2c/tvp5150_reg.h b/drivers/media/i2c/tvp5150_reg.h index 25a994944918..30a48c28d05a 100644 --- a/drivers/media/i2c/tvp5150_reg.h +++ b/drivers/media/i2c/tvp5150_reg.h @@ -9,6 +9,15 @@ #define TVP5150_ANAL_CHL_CTL 0x01 /* Analog channel controls */ #define TVP5150_OP_MODE_CTL 0x02 /* Operation mode controls */ #define TVP5150_MISC_CTL 0x03 /* Miscellaneous controls */ +#define TVP5150_MISC_CTL_VBLK_GPCL BIT(7) +#define TVP5150_MISC_CTL_GPCL BIT(6) +#define TVP5150_MISC_CTL_INTREQ_OE BIT(5) +#define TVP5150_MISC_CTL_HVLK BIT(4) +#define TVP5150_MISC_CTL_YCBCR_OE BIT(3) +#define TVP5150_MISC_CTL_SYNC_OE BIT(2) +#define TVP5150_MISC_CTL_VBLANK BIT(1) +#define TVP5150_MISC_CTL_CLOCK_OE BIT(0) + #define TVP5150_AUTOSW_MSK 0x04 /* Autoswitch mask: TVP5150A / TVP5150AM */ /* Reserved 05h */ -- cgit v1.2.3 From 79d6205a3f741c9fb89cfc47dfa0eddb1526726d Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Fri, 9 Dec 2016 09:47:19 -0200 Subject: [media] v4l: tvp5150: Don't override output pinmuxing at stream on/off time The s_stream() handler incorrectly writes the whole MISC_CTL register to enable or disable the outputs, overriding the output pinmuxing configuration. Fix it to only touch the output enable bits. The CONF_SHARED_PIN register is also written by the same function, resulting in muxing the INTREQ signal instead of the VBLK/GPCL signal on the INTREQ/GPCL/VBLK pin. As the driver doesn't support interrupts this is obviously incorrect, and breaks operation on other devices. Fix it by removing the write. Cc: stable@vger.kernel.org # For Kernel 4.5 and upper Signed-off-by: Laurent Pinchart Signed-off-by: Mauro Carvalho Chehab --- drivers/media/i2c/tvp5150.c | 33 +++++++++++++++++++-------------- 1 file changed, 19 insertions(+), 14 deletions(-) diff --git a/drivers/media/i2c/tvp5150.c b/drivers/media/i2c/tvp5150.c index 8852fa8c957b..48646a7f3fb0 100644 --- a/drivers/media/i2c/tvp5150.c +++ b/drivers/media/i2c/tvp5150.c @@ -1058,22 +1058,27 @@ static const struct media_entity_operations tvp5150_sd_media_ops = { static int tvp5150_s_stream(struct v4l2_subdev *sd, int enable) { struct tvp5150 *decoder = to_tvp5150(sd); - /* Output format: 8-bit ITU-R BT.656 with embedded syncs */ - int val = TVP5150_MISC_CTL_YCBCR_OE | TVP5150_MISC_CTL_CLOCK_OE; - - /* Output format: 8-bit 4:2:2 YUV with discrete sync */ - if (decoder->mbus_type == V4L2_MBUS_PARALLEL) - val = TVP5150_MISC_CTL_YCBCR_OE | TVP5150_MISC_CTL_SYNC_OE - | TVP5150_MISC_CTL_CLOCK_OE; + int val; - /* Initializes TVP5150 to its default values */ - /* # set PCLK (27MHz) */ - tvp5150_write(sd, TVP5150_CONF_SHARED_PIN, 0x00); + /* Enable or disable the video output signals. */ + val = tvp5150_read(sd, TVP5150_MISC_CTL); + if (val < 0) + return val; + + val &= ~(TVP5150_MISC_CTL_YCBCR_OE | TVP5150_MISC_CTL_SYNC_OE | + TVP5150_MISC_CTL_CLOCK_OE); + + if (enable) { + /* + * Enable the YCbCr and clock outputs. In discrete sync mode + * (non-BT.656) additionally enable the the sync outputs. + */ + val |= TVP5150_MISC_CTL_YCBCR_OE | TVP5150_MISC_CTL_CLOCK_OE; + if (decoder->mbus_type == V4L2_MBUS_PARALLEL) + val |= TVP5150_MISC_CTL_SYNC_OE; + } - if (enable) - tvp5150_write(sd, TVP5150_MISC_CTL, val); - else - tvp5150_write(sd, TVP5150_MISC_CTL, 0x00); + tvp5150_write(sd, TVP5150_MISC_CTL, val); return 0; } -- cgit v1.2.3 From 4dd19196c5539c377beaa9850fac30c18318c7a1 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 9 Dec 2016 09:36:29 -0200 Subject: [media] dvb: avoid warning in dvb_net With gcc-5 or higher on x86, we can get a bogus warning in the dvb-net code: drivers/media/dvb-core/dvb_net.c: In function 'dvb_net_ule': arch/x86/include/asm/string_32.h:78:22: error: '*((void *)&dest_addr+4)' may be used uninitialized in this function [-Werror=maybe-uninitialized] The problem here is that gcc doesn't track all of the conditions to prove it can't end up copying uninitialized data. This changes the logic around so we zero out the destination address earlier when we determine that it is not set here. This allows the compiler to figure it out. Signed-off-by: Arnd Bergmann Signed-off-by: Mauro Carvalho Chehab --- drivers/media/dvb-core/dvb_net.c | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/drivers/media/dvb-core/dvb_net.c b/drivers/media/dvb-core/dvb_net.c index bd833b0824c6..eb60cb1442f2 100644 --- a/drivers/media/dvb-core/dvb_net.c +++ b/drivers/media/dvb-core/dvb_net.c @@ -719,6 +719,9 @@ static void dvb_net_ule_check_crc(struct dvb_net_ule_handle *h, skb_copy_from_linear_data(h->priv->ule_skb, dest_addr, ETH_ALEN); skb_pull(h->priv->ule_skb, ETH_ALEN); + } else { + /* dest_addr buffer is only valid if h->priv->ule_dbit == 0 */ + eth_zero_addr(dest_addr); } /* Handle ULE Extension Headers. */ @@ -750,16 +753,8 @@ static void dvb_net_ule_check_crc(struct dvb_net_ule_handle *h, if (!h->priv->ule_bridged) { skb_push(h->priv->ule_skb, ETH_HLEN); h->ethh = (struct ethhdr *)h->priv->ule_skb->data; - if (!h->priv->ule_dbit) { - /* - * dest_addr buffer is only valid if - * h->priv->ule_dbit == 0 - */ - memcpy(h->ethh->h_dest, dest_addr, ETH_ALEN); - eth_zero_addr(h->ethh->h_source); - } else /* zeroize source and dest */ - memset(h->ethh, 0, ETH_ALEN * 2); - + memcpy(h->ethh->h_dest, dest_addr, ETH_ALEN); + eth_zero_addr(h->ethh->h_source); h->ethh->h_proto = htons(h->priv->ule_sndu_type); } /* else: skb is in correct state; nothing to do. */ -- cgit v1.2.3 From c739c0a7c3c2472d7562b8f802cdce44d2597c8b Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 9 Dec 2016 09:41:29 -0200 Subject: [media] s5k4ecgx: select CRC32 helper A rare randconfig build failure shows up in this driver when the CRC32 helper is not there: drivers/media/built-in.o: In function `s5k4ecgx_s_power': s5k4ecgx.c:(.text+0x9eb4): undefined reference to `crc32_le' This adds the 'select' that all other users of this function have. Fixes: 8b99312b7214 ("[media] Add v4l2 subdev driver for S5K4ECGX sensor") Signed-off-by: Arnd Bergmann Signed-off-by: Mauro Carvalho Chehab --- drivers/media/i2c/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/media/i2c/Kconfig b/drivers/media/i2c/Kconfig index b31fa6fae009..b979ea148251 100644 --- a/drivers/media/i2c/Kconfig +++ b/drivers/media/i2c/Kconfig @@ -655,6 +655,7 @@ config VIDEO_S5K6A3 config VIDEO_S5K4ECGX tristate "Samsung S5K4ECGX sensor support" depends on I2C && VIDEO_V4L2 && VIDEO_V4L2_SUBDEV_API + select CRC32 ---help--- This is a V4L2 sensor-level driver for Samsung S5K4ECGX 5M camera sensor with an embedded SoC image signal processor. -- cgit v1.2.3 From 4bbc84ffd137fe43d68aa633d317b0a96de8a828 Mon Sep 17 00:00:00 2001 From: Mike Kravetz Date: Mon, 19 Dec 2016 19:17:08 -0800 Subject: sparc: use symbolic names for tsb indexing Use symbolic names MM_TSB_BASE and MM_TSB_HUGE instead of numeric values 0 and 1 in __tsb_context_switch. Code cleanup only, no functional change. Suggested-by: Sam Ravnborg Signed-off-by: Mike Kravetz Acked-by: Sam Ravnborg Signed-off-by: David S. Miller --- arch/sparc/include/asm/mmu_context_64.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/sparc/include/asm/mmu_context_64.h b/arch/sparc/include/asm/mmu_context_64.h index b84be675e507..d0317993e947 100644 --- a/arch/sparc/include/asm/mmu_context_64.h +++ b/arch/sparc/include/asm/mmu_context_64.h @@ -35,15 +35,15 @@ void __tsb_context_switch(unsigned long pgd_pa, static inline void tsb_context_switch(struct mm_struct *mm) { __tsb_context_switch(__pa(mm->pgd), - &mm->context.tsb_block[0], + &mm->context.tsb_block[MM_TSB_BASE], #if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) - (mm->context.tsb_block[1].tsb ? - &mm->context.tsb_block[1] : + (mm->context.tsb_block[MM_TSB_HUGE].tsb ? + &mm->context.tsb_block[MM_TSB_HUGE] : NULL) #else NULL #endif - , __pa(&mm->context.tsb_descr[0])); + , __pa(&mm->context.tsb_descr[MM_TSB_BASE])); } void tsb_grow(struct mm_struct *mm, -- cgit v1.2.3 From 4ea33ef0f9e95b69db9131d7afd98563713e81b0 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Tue, 27 Dec 2016 08:51:17 +0100 Subject: batman-adv: Decrease hardif refcnt on fragmentation send error An error before the hardif is found has to free the skb. But every error after that has to free the skb + put the hard interface. Fixes: 8def0be82dd1 ("batman-adv: Consume skb in batadv_frag_send_packet") Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/fragmentation.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c index 9c561e683f4b..0854ebd8613e 100644 --- a/net/batman-adv/fragmentation.c +++ b/net/batman-adv/fragmentation.c @@ -474,7 +474,7 @@ int batadv_frag_send_packet(struct sk_buff *skb, primary_if = batadv_primary_if_get_selected(bat_priv); if (!primary_if) { ret = -EINVAL; - goto put_primary_if; + goto free_skb; } /* Create one header to be copied to all fragments */ @@ -502,7 +502,7 @@ int batadv_frag_send_packet(struct sk_buff *skb, skb_fragment = batadv_frag_create(skb, &frag_header, mtu); if (!skb_fragment) { ret = -ENOMEM; - goto free_skb; + goto put_primary_if; } batadv_inc_counter(bat_priv, BATADV_CNT_FRAG_TX); @@ -511,7 +511,7 @@ int batadv_frag_send_packet(struct sk_buff *skb, ret = batadv_send_unicast_skb(skb_fragment, neigh_node); if (ret != NET_XMIT_SUCCESS) { ret = NET_XMIT_DROP; - goto free_skb; + goto put_primary_if; } frag_header.no++; @@ -519,7 +519,7 @@ int batadv_frag_send_packet(struct sk_buff *skb, /* The initial check in this function should cover this case */ if (frag_header.no == BATADV_FRAG_MAX_FRAGMENTS - 1) { ret = -EINVAL; - goto free_skb; + goto put_primary_if; } } @@ -527,7 +527,7 @@ int batadv_frag_send_packet(struct sk_buff *skb, if (batadv_skb_head_push(skb, header_size) < 0 || pskb_expand_head(skb, header_size + ETH_HLEN, 0, GFP_ATOMIC) < 0) { ret = -ENOMEM; - goto free_skb; + goto put_primary_if; } memcpy(skb->data, &frag_header, header_size); -- cgit v1.2.3 From e0edc8c546463f268d41d064d855bcff994c52fa Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 6 Jan 2017 11:48:50 -0500 Subject: libata: apply MAX_SEC_1024 to all CX1-JB*-HP devices Marko reports that CX1-JB512-HP shows the same timeout issues as CX1-JB256-HP. Let's apply MAX_SEC_128 to all devices in the series. Signed-off-by: Tejun Heo Reported-by: Marko Koski-Vähälä Cc: stable@vger.kernel.org # v3.19+ --- drivers/ata/libata-core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 9cd0a2d41816..a2bd5acc8a89 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4356,10 +4356,10 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { { "ST380013AS", "3.20", ATA_HORKAGE_MAX_SEC_1024 }, /* - * Device times out with higher max sects. + * These devices time out with higher max sects. * https://bugzilla.kernel.org/show_bug.cgi?id=121671 */ - { "LITEON CX1-JB256-HP", NULL, ATA_HORKAGE_MAX_SEC_1024 }, + { "LITEON CX1-JB*-HP", NULL, ATA_HORKAGE_MAX_SEC_1024 }, /* Devices we expect to fail diagnostics */ -- cgit v1.2.3 From 2dae99558e86894e9e5dbf097477baaa5eb70134 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Mon, 19 Dec 2016 10:17:40 +0900 Subject: libata: Fix ATA request sense For an ATA device supporting the sense data reporting feature set, a failed command will trigger the execution of ata_eh_request_sense if the result task file of the failed command has the ATA_SENSE bit set (sense data available bit). ata_eh_request_sense executes the REQUEST SENSE DATA EXT command to retrieve the sense data of the failed command. On success of REQUEST SENSE DATA EXT, the ATA_SENSE bit will NOT be set (the command succeeded) but ata_eh_request_sense nevertheless tests the availability of sense data by testing that bit presence in the result tf of the REQUEST SENSE DATA EXT command. This leads us to falsely assume that request sense data failed and to the warning message: atax.xx: request sense failed stat 50 emask 0 Upon success of REQUEST SENSE DATA EXT, set the ATA_SENSE bit in the result task file command so that sense data can be returned by ata_eh_request_sense. Signed-off-by: Damien Le Moal Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org --- drivers/ata/libata-core.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index a2bd5acc8a89..c2d3785ec227 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -1702,6 +1702,8 @@ unsigned ata_exec_internal_sg(struct ata_device *dev, if (qc->err_mask & ~AC_ERR_OTHER) qc->err_mask &= ~AC_ERR_OTHER; + } else if (qc->tf.command == ATA_CMD_REQ_SENSE_DATA) { + qc->result_tf.command |= ATA_SENSE; } /* finish up */ -- cgit v1.2.3 From 064c3db9c564cc5be514ac21fb4aa26cc33db746 Mon Sep 17 00:00:00 2001 From: Arvind Yadav Date: Mon, 12 Dec 2016 23:13:27 +0530 Subject: ata: sata_mv:- Handle return value of devm_ioremap. Here, If devm_ioremap will fail. It will return NULL. Then hpriv->base = NULL - 0x20000; Kernel can run into a NULL-pointer dereference. This error check will avoid NULL pointer dereference. Signed-off-by: Arvind Yadav Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org --- drivers/ata/sata_mv.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c index 823e938c9a78..2f32782cea6d 100644 --- a/drivers/ata/sata_mv.c +++ b/drivers/ata/sata_mv.c @@ -4132,6 +4132,9 @@ static int mv_platform_probe(struct platform_device *pdev) host->iomap = NULL; hpriv->base = devm_ioremap(&pdev->dev, res->start, resource_size(res)); + if (!hpriv->base) + return -ENOMEM; + hpriv->base -= SATAHC0_REG_BASE; hpriv->clk = clk_get(&pdev->dev, NULL); -- cgit v1.2.3 From cc31d43b4154ad5a7d8aa5543255a93b7e89edc2 Mon Sep 17 00:00:00 2001 From: Pau Espin Pedrol Date: Fri, 6 Jan 2017 20:33:27 +0100 Subject: netfilter: use fwmark_reflect in nf_send_reset Otherwise, RST packets generated by ipt_REJECT always have mark 0 when the routing is checked later in the same code path. Fixes: e110861f8609 ("net: add a sysctl to reflect the fwmark on replies") Cc: Lorenzo Colitti Signed-off-by: Pau Espin Pedrol Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/nf_reject_ipv4.c | 2 ++ net/ipv6/netfilter/nf_reject_ipv6.c | 3 +++ 2 files changed, 5 insertions(+) diff --git a/net/ipv4/netfilter/nf_reject_ipv4.c b/net/ipv4/netfilter/nf_reject_ipv4.c index fd8220213afc..146d86105183 100644 --- a/net/ipv4/netfilter/nf_reject_ipv4.c +++ b/net/ipv4/netfilter/nf_reject_ipv4.c @@ -126,6 +126,8 @@ void nf_send_reset(struct net *net, struct sk_buff *oldskb, int hook) /* ip_route_me_harder expects skb->dst to be set */ skb_dst_set_noref(nskb, skb_dst(oldskb)); + nskb->mark = IP4_REPLY_MARK(net, oldskb->mark); + skb_reserve(nskb, LL_MAX_HEADER); niph = nf_reject_iphdr_put(nskb, oldskb, IPPROTO_TCP, ip4_dst_hoplimit(skb_dst(nskb))); diff --git a/net/ipv6/netfilter/nf_reject_ipv6.c b/net/ipv6/netfilter/nf_reject_ipv6.c index 10090400c72f..eedee5d108d9 100644 --- a/net/ipv6/netfilter/nf_reject_ipv6.c +++ b/net/ipv6/netfilter/nf_reject_ipv6.c @@ -157,6 +157,7 @@ void nf_send_reset6(struct net *net, struct sk_buff *oldskb, int hook) fl6.fl6_sport = otcph->dest; fl6.fl6_dport = otcph->source; fl6.flowi6_oif = l3mdev_master_ifindex(skb_dst(oldskb)->dev); + fl6.flowi6_mark = IP6_REPLY_MARK(net, oldskb->mark); security_skb_classify_flow(oldskb, flowi6_to_flowi(&fl6)); dst = ip6_route_output(net, NULL, &fl6); if (dst->error) { @@ -180,6 +181,8 @@ void nf_send_reset6(struct net *net, struct sk_buff *oldskb, int hook) skb_dst_set(nskb, dst); + nskb->mark = fl6.flowi6_mark; + skb_reserve(nskb, hh_len + dst->header_len); ip6h = nf_reject_ip6hdr_put(nskb, oldskb, IPPROTO_TCP, ip6_dst_hoplimit(dst)); -- cgit v1.2.3 From bf99b4ded5f8a4767dbb9d180626f06c51f9881f Mon Sep 17 00:00:00 2001 From: Pau Espin Pedrol Date: Fri, 6 Jan 2017 20:33:28 +0100 Subject: tcp: fix mark propagation with fwmark_reflect enabled Otherwise, RST packets generated by the TCP stack for non-existing sockets always have mark 0. The mark from the original packet is assigned to the netns_ipv4/6 socket used to send the response so that it can get copied into the response skb when the socket sends it. Fixes: e110861f8609 ("net: add a sysctl to reflect the fwmark on replies") Cc: Lorenzo Colitti Signed-off-by: Pau Espin Pedrol Signed-off-by: Pablo Neira Ayuso --- net/ipv4/ip_output.c | 1 + net/ipv6/tcp_ipv6.c | 1 + 2 files changed, 2 insertions(+) diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index fac275c48108..b67719f45953 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -1629,6 +1629,7 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb, sk->sk_protocol = ip_hdr(skb)->protocol; sk->sk_bound_dev_if = arg->bound_dev_if; sk->sk_sndbuf = sysctl_wmem_default; + sk->sk_mark = fl4.flowi4_mark; err = ip_append_data(sk, &fl4, ip_reply_glue_bits, arg->iov->iov_base, len, 0, &ipc, &rt, MSG_DONTWAIT); if (unlikely(err)) { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 73bc8fc68acd..2b20622a5824 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -840,6 +840,7 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32 dst = ip6_dst_lookup_flow(ctl_sk, &fl6, NULL); if (!IS_ERR(dst)) { skb_dst_set(buff, dst); + ctl_sk->sk_mark = fl6.flowi6_mark; ip6_xmit(ctl_sk, buff, &fl6, NULL, tclass); TCP_INC_STATS(net, TCP_MIB_OUTSEGS); if (rst) -- cgit v1.2.3 From 4fe7c2962e110dfd58e61888514726aac419562f Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Thu, 22 Dec 2016 07:04:59 -0800 Subject: iw_cxgb4: refactor sq/rq drain logic With the addition of the IB/Core drain API, iw_cxgb4 supported drain by watching the CQs when the QP was out of RTS and signalling "drain complete" when the last CQE is polled. This, however, doesn't fully support the drain semantics. Namely, the drain logic is supposed to signal "drain complete" only when the application has _processed_ the last CQE, not just removed them from the CQ. Thus a small timing hole exists that can cause touch after free type bugs in applications using the drain API (nvmf, iSER, for example). So iw_cxgb4 needs a better solution. The iWARP Verbs spec mandates that "_at some point_ after the QP is moved to ERROR", the iWARP driver MUST synchronously fail post_send and post_recv calls. iw_cxgb4 was currently not allowing any posts once the QP is in ERROR. This was in part due to the fact that the HW queues for the QP in ERROR state are disabled at this point, so there wasn't much else to do but fail the post operation synchronously. This restriction is what drove the first drain implementation in iw_cxgb4 that has the above mentioned flaw. This patch changes iw_cxgb4 to allow post_send and post_recv WRs after the QP is moved to ERROR state for kernel mode users, thus still adhering to the Verbs spec for user mode users, but allowing flush WRs for kernel users. Since the HW queues are disabled, we just synthesize a CQE for this post, queue it to the SW CQ, and then call the CQ event handler. This enables proper drain operations for the various storage applications. Signed-off-by: Steve Wise Signed-off-by: Doug Ledford --- drivers/infiniband/hw/cxgb4/cq.c | 21 ++++--- drivers/infiniband/hw/cxgb4/iw_cxgb4.h | 6 +- drivers/infiniband/hw/cxgb4/provider.c | 2 - drivers/infiniband/hw/cxgb4/qp.c | 112 ++++++++++++++++++++------------- drivers/infiniband/hw/cxgb4/t4.h | 2 + 5 files changed, 85 insertions(+), 58 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index 19c6477af19f..bec82a600d77 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -504,6 +504,15 @@ static int poll_cq(struct t4_wq *wq, struct t4_cq *cq, struct t4_cqe *cqe, goto skip_cqe; } + /* + * Special cqe for drain WR completions... + */ + if (CQE_OPCODE(hw_cqe) == C4IW_DRAIN_OPCODE) { + *cookie = CQE_DRAIN_COOKIE(hw_cqe); + *cqe = *hw_cqe; + goto skip_cqe; + } + /* * Gotta tweak READ completions: * 1) the cqe doesn't contain the sq_wptr from the wr. @@ -753,6 +762,9 @@ static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc) c4iw_invalidate_mr(qhp->rhp, CQE_WRID_FR_STAG(&cqe)); break; + case C4IW_DRAIN_OPCODE: + wc->opcode = IB_WC_SEND; + break; default: printk(KERN_ERR MOD "Unexpected opcode %d " "in the CQE received for QPID=0x%0x\n", @@ -817,15 +829,8 @@ static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc) } } out: - if (wq) { - if (unlikely(qhp->attr.state != C4IW_QP_STATE_RTS)) { - if (t4_sq_empty(wq)) - complete(&qhp->sq_drained); - if (t4_rq_empty(wq)) - complete(&qhp->rq_drained); - } + if (wq) spin_unlock(&qhp->lock); - } return ret; } diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h index 4788e1a46fde..7b1e465b2a5e 100644 --- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h +++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h @@ -480,8 +480,6 @@ struct c4iw_qp { wait_queue_head_t wait; struct timer_list timer; int sq_sig_all; - struct completion rq_drained; - struct completion sq_drained; }; static inline struct c4iw_qp *to_c4iw_qp(struct ib_qp *ibqp) @@ -615,6 +613,8 @@ static inline int to_ib_qp_state(int c4iw_qp_state) return IB_QPS_ERR; } +#define C4IW_DRAIN_OPCODE FW_RI_SGE_EC_CR_RETURN + static inline u32 c4iw_ib_to_tpt_access(int a) { return (a & IB_ACCESS_REMOTE_WRITE ? FW_RI_MEM_ACCESS_REM_WRITE : 0) | @@ -997,8 +997,6 @@ extern int c4iw_wr_log; extern int db_fc_threshold; extern int db_coalescing_threshold; extern int use_dsgl; -void c4iw_drain_rq(struct ib_qp *qp); -void c4iw_drain_sq(struct ib_qp *qp); void c4iw_invalidate_mr(struct c4iw_dev *rhp, u32 rkey); #endif diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c index 49b51b7e0fd7..c156413515b1 100644 --- a/drivers/infiniband/hw/cxgb4/provider.c +++ b/drivers/infiniband/hw/cxgb4/provider.c @@ -607,8 +607,6 @@ int c4iw_register_device(struct c4iw_dev *dev) dev->ibdev.uverbs_abi_ver = C4IW_UVERBS_ABI_VERSION; dev->ibdev.get_port_immutable = c4iw_port_immutable; dev->ibdev.get_dev_fw_str = get_dev_fw_str; - dev->ibdev.drain_sq = c4iw_drain_sq; - dev->ibdev.drain_rq = c4iw_drain_rq; dev->ibdev.iwcm = kmalloc(sizeof(struct iw_cm_verbs), GFP_KERNEL); if (!dev->ibdev.iwcm) diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index cda5542e13a2..31ab4512f827 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -776,6 +776,64 @@ static int ring_kernel_rq_db(struct c4iw_qp *qhp, u16 inc) return 0; } +static void complete_sq_drain_wr(struct c4iw_qp *qhp, struct ib_send_wr *wr) +{ + struct t4_cqe cqe = {}; + struct c4iw_cq *schp; + unsigned long flag; + struct t4_cq *cq; + + schp = to_c4iw_cq(qhp->ibqp.send_cq); + cq = &schp->cq; + + cqe.u.drain_cookie = wr->wr_id; + cqe.header = cpu_to_be32(CQE_STATUS_V(T4_ERR_SWFLUSH) | + CQE_OPCODE_V(C4IW_DRAIN_OPCODE) | + CQE_TYPE_V(1) | + CQE_SWCQE_V(1) | + CQE_QPID_V(qhp->wq.sq.qid)); + + spin_lock_irqsave(&schp->lock, flag); + cqe.bits_type_ts = cpu_to_be64(CQE_GENBIT_V((u64)cq->gen)); + cq->sw_queue[cq->sw_pidx] = cqe; + t4_swcq_produce(cq); + spin_unlock_irqrestore(&schp->lock, flag); + + spin_lock_irqsave(&schp->comp_handler_lock, flag); + (*schp->ibcq.comp_handler)(&schp->ibcq, + schp->ibcq.cq_context); + spin_unlock_irqrestore(&schp->comp_handler_lock, flag); +} + +static void complete_rq_drain_wr(struct c4iw_qp *qhp, struct ib_recv_wr *wr) +{ + struct t4_cqe cqe = {}; + struct c4iw_cq *rchp; + unsigned long flag; + struct t4_cq *cq; + + rchp = to_c4iw_cq(qhp->ibqp.recv_cq); + cq = &rchp->cq; + + cqe.u.drain_cookie = wr->wr_id; + cqe.header = cpu_to_be32(CQE_STATUS_V(T4_ERR_SWFLUSH) | + CQE_OPCODE_V(C4IW_DRAIN_OPCODE) | + CQE_TYPE_V(0) | + CQE_SWCQE_V(1) | + CQE_QPID_V(qhp->wq.sq.qid)); + + spin_lock_irqsave(&rchp->lock, flag); + cqe.bits_type_ts = cpu_to_be64(CQE_GENBIT_V((u64)cq->gen)); + cq->sw_queue[cq->sw_pidx] = cqe; + t4_swcq_produce(cq); + spin_unlock_irqrestore(&rchp->lock, flag); + + spin_lock_irqsave(&rchp->comp_handler_lock, flag); + (*rchp->ibcq.comp_handler)(&rchp->ibcq, + rchp->ibcq.cq_context); + spin_unlock_irqrestore(&rchp->comp_handler_lock, flag); +} + int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, struct ib_send_wr **bad_wr) { @@ -794,8 +852,8 @@ int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, spin_lock_irqsave(&qhp->lock, flag); if (t4_wq_in_error(&qhp->wq)) { spin_unlock_irqrestore(&qhp->lock, flag); - *bad_wr = wr; - return -EINVAL; + complete_sq_drain_wr(qhp, wr); + return err; } num_wrs = t4_sq_avail(&qhp->wq); if (num_wrs == 0) { @@ -937,8 +995,8 @@ int c4iw_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, spin_lock_irqsave(&qhp->lock, flag); if (t4_wq_in_error(&qhp->wq)) { spin_unlock_irqrestore(&qhp->lock, flag); - *bad_wr = wr; - return -EINVAL; + complete_rq_drain_wr(qhp, wr); + return err; } num_wrs = t4_rq_avail(&qhp->wq); if (num_wrs == 0) { @@ -1550,7 +1608,12 @@ int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp, } break; case C4IW_QP_STATE_CLOSING: - if (!internal) { + + /* + * Allow kernel users to move to ERROR for qp draining. + */ + if (!internal && (qhp->ibqp.uobject || attrs->next_state != + C4IW_QP_STATE_ERROR)) { ret = -EINVAL; goto out; } @@ -1763,8 +1826,6 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, qhp->attr.max_ird = 0; qhp->sq_sig_all = attrs->sq_sig_type == IB_SIGNAL_ALL_WR; spin_lock_init(&qhp->lock); - init_completion(&qhp->sq_drained); - init_completion(&qhp->rq_drained); mutex_init(&qhp->mutex); init_waitqueue_head(&qhp->wait); kref_init(&qhp->kref); @@ -1958,40 +2019,3 @@ int c4iw_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, init_attr->sq_sig_type = qhp->sq_sig_all ? IB_SIGNAL_ALL_WR : 0; return 0; } - -static void move_qp_to_err(struct c4iw_qp *qp) -{ - struct c4iw_qp_attributes attrs = { .next_state = C4IW_QP_STATE_ERROR }; - - (void)c4iw_modify_qp(qp->rhp, qp, C4IW_QP_ATTR_NEXT_STATE, &attrs, 1); -} - -void c4iw_drain_sq(struct ib_qp *ibqp) -{ - struct c4iw_qp *qp = to_c4iw_qp(ibqp); - unsigned long flag; - bool need_to_wait; - - move_qp_to_err(qp); - spin_lock_irqsave(&qp->lock, flag); - need_to_wait = !t4_sq_empty(&qp->wq); - spin_unlock_irqrestore(&qp->lock, flag); - - if (need_to_wait) - wait_for_completion(&qp->sq_drained); -} - -void c4iw_drain_rq(struct ib_qp *ibqp) -{ - struct c4iw_qp *qp = to_c4iw_qp(ibqp); - unsigned long flag; - bool need_to_wait; - - move_qp_to_err(qp); - spin_lock_irqsave(&qp->lock, flag); - need_to_wait = !t4_rq_empty(&qp->wq); - spin_unlock_irqrestore(&qp->lock, flag); - - if (need_to_wait) - wait_for_completion(&qp->rq_drained); -} diff --git a/drivers/infiniband/hw/cxgb4/t4.h b/drivers/infiniband/hw/cxgb4/t4.h index 862381aa83c8..640d22148a3e 100644 --- a/drivers/infiniband/hw/cxgb4/t4.h +++ b/drivers/infiniband/hw/cxgb4/t4.h @@ -179,6 +179,7 @@ struct t4_cqe { __be32 wrid_hi; __be32 wrid_low; } gen; + u64 drain_cookie; } u; __be64 reserved; __be64 bits_type_ts; @@ -238,6 +239,7 @@ struct t4_cqe { /* generic accessor macros */ #define CQE_WRID_HI(x) (be32_to_cpu((x)->u.gen.wrid_hi)) #define CQE_WRID_LOW(x) (be32_to_cpu((x)->u.gen.wrid_low)) +#define CQE_DRAIN_COOKIE(x) ((x)->u.drain_cookie) /* macros for flit 3 of the cqe */ #define CQE_GENBIT_S 63 -- cgit v1.2.3 From c12a67fec8d99bb554e8d4e99120d418f1a39c87 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Thu, 22 Dec 2016 07:40:36 -0800 Subject: iw_cxgb4: free EQ queue memory on last deref Commit ad61a4c7a9b7 ("iw_cxgb4: don't block in destroy_qp awaiting the last deref") introduced a bug where the RDMA QP EQ queue memory (and QIDs) are possibly freed before the underlying connection has been fully shutdown. The result being a possible DMA read issued by HW after the queue memory has been unmapped and freed. This results in possible WR corruption in the worst case, system bus errors if an IOMMU is in use, and SGE "bad WR" errors reported in the very least. The fix is to defer unmap/free of queue memory and QID resources until the QP struct has been fully dereferenced. To do this, the c4iw_ucontext must also be kept around until the last QP that references it is fully freed. In addition, since the last QP deref can happen in an IRQ disabled context, we need a new workqueue thread to do the final unmap/free of the EQ queue memory. Fixes: ad61a4c7a9b7 ("iw_cxgb4: don't block in destroy_qp awaiting the last deref") Cc: stable@vger.kernel.org Signed-off-by: Steve Wise Signed-off-by: Doug Ledford --- drivers/infiniband/hw/cxgb4/device.c | 9 +++++++++ drivers/infiniband/hw/cxgb4/iw_cxgb4.h | 18 +++++++++++++++++ drivers/infiniband/hw/cxgb4/provider.c | 20 +++++++++++++++---- drivers/infiniband/hw/cxgb4/qp.c | 35 +++++++++++++++++++++++++--------- 4 files changed, 69 insertions(+), 13 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index 516b0ae6dc3f..40c0e7b9fc6e 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -846,9 +846,17 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) } } + rdev->free_workq = create_singlethread_workqueue("iw_cxgb4_free"); + if (!rdev->free_workq) { + err = -ENOMEM; + goto err_free_status_page; + } + rdev->status_page->db_off = 0; return 0; +err_free_status_page: + free_page((unsigned long)rdev->status_page); destroy_ocqp_pool: c4iw_ocqp_pool_destroy(rdev); destroy_rqtpool: @@ -862,6 +870,7 @@ destroy_resource: static void c4iw_rdev_close(struct c4iw_rdev *rdev) { + destroy_workqueue(rdev->free_workq); kfree(rdev->wr_log); free_page((unsigned long)rdev->status_page); c4iw_pblpool_destroy(rdev); diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h index 7b1e465b2a5e..8cd4d054a87e 100644 --- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h +++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h @@ -45,6 +45,7 @@ #include #include #include +#include #include @@ -107,6 +108,7 @@ struct c4iw_dev_ucontext { struct list_head qpids; struct list_head cqids; struct mutex lock; + struct kref kref; }; enum c4iw_rdev_flags { @@ -183,6 +185,7 @@ struct c4iw_rdev { atomic_t wr_log_idx; struct wr_log_entry *wr_log; int wr_log_size; + struct workqueue_struct *free_workq; }; static inline int c4iw_fatal_error(struct c4iw_rdev *rdev) @@ -480,6 +483,8 @@ struct c4iw_qp { wait_queue_head_t wait; struct timer_list timer; int sq_sig_all; + struct work_struct free_work; + struct c4iw_ucontext *ucontext; }; static inline struct c4iw_qp *to_c4iw_qp(struct ib_qp *ibqp) @@ -493,6 +498,7 @@ struct c4iw_ucontext { u32 key; spinlock_t mmap_lock; struct list_head mmaps; + struct kref kref; }; static inline struct c4iw_ucontext *to_c4iw_ucontext(struct ib_ucontext *c) @@ -500,6 +506,18 @@ static inline struct c4iw_ucontext *to_c4iw_ucontext(struct ib_ucontext *c) return container_of(c, struct c4iw_ucontext, ibucontext); } +void _c4iw_free_ucontext(struct kref *kref); + +static inline void c4iw_put_ucontext(struct c4iw_ucontext *ucontext) +{ + kref_put(&ucontext->kref, _c4iw_free_ucontext); +} + +static inline void c4iw_get_ucontext(struct c4iw_ucontext *ucontext) +{ + kref_get(&ucontext->kref); +} + struct c4iw_mm_entry { struct list_head entry; u64 addr; diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c index c156413515b1..fa64f5d93b11 100644 --- a/drivers/infiniband/hw/cxgb4/provider.c +++ b/drivers/infiniband/hw/cxgb4/provider.c @@ -93,17 +93,28 @@ static int c4iw_process_mad(struct ib_device *ibdev, int mad_flags, return -ENOSYS; } -static int c4iw_dealloc_ucontext(struct ib_ucontext *context) +void _c4iw_free_ucontext(struct kref *kref) { - struct c4iw_dev *rhp = to_c4iw_dev(context->device); - struct c4iw_ucontext *ucontext = to_c4iw_ucontext(context); + struct c4iw_ucontext *ucontext; + struct c4iw_dev *rhp; struct c4iw_mm_entry *mm, *tmp; - PDBG("%s context %p\n", __func__, context); + ucontext = container_of(kref, struct c4iw_ucontext, kref); + rhp = to_c4iw_dev(ucontext->ibucontext.device); + + PDBG("%s ucontext %p\n", __func__, ucontext); list_for_each_entry_safe(mm, tmp, &ucontext->mmaps, entry) kfree(mm); c4iw_release_dev_ucontext(&rhp->rdev, &ucontext->uctx); kfree(ucontext); +} + +static int c4iw_dealloc_ucontext(struct ib_ucontext *context) +{ + struct c4iw_ucontext *ucontext = to_c4iw_ucontext(context); + + PDBG("%s context %p\n", __func__, context); + c4iw_put_ucontext(ucontext); return 0; } @@ -127,6 +138,7 @@ static struct ib_ucontext *c4iw_alloc_ucontext(struct ib_device *ibdev, c4iw_init_dev_ucontext(&rhp->rdev, &context->uctx); INIT_LIST_HEAD(&context->mmaps); spin_lock_init(&context->mmap_lock); + kref_init(&context->kref); if (udata->outlen < sizeof(uresp) - sizeof(uresp.reserved)) { if (!warned++) diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index 31ab4512f827..04c1c382dedb 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -715,13 +715,32 @@ static int build_inv_stag(union t4_wr *wqe, struct ib_send_wr *wr, u8 *len16) return 0; } -static void _free_qp(struct kref *kref) +static void free_qp_work(struct work_struct *work) +{ + struct c4iw_ucontext *ucontext; + struct c4iw_qp *qhp; + struct c4iw_dev *rhp; + + qhp = container_of(work, struct c4iw_qp, free_work); + ucontext = qhp->ucontext; + rhp = qhp->rhp; + + PDBG("%s qhp %p ucontext %p\n", __func__, qhp, ucontext); + destroy_qp(&rhp->rdev, &qhp->wq, + ucontext ? &ucontext->uctx : &rhp->rdev.uctx); + + if (ucontext) + c4iw_put_ucontext(ucontext); + kfree(qhp); +} + +static void queue_qp_free(struct kref *kref) { struct c4iw_qp *qhp; qhp = container_of(kref, struct c4iw_qp, kref); PDBG("%s qhp %p\n", __func__, qhp); - kfree(qhp); + queue_work(qhp->rhp->rdev.free_workq, &qhp->free_work); } void c4iw_qp_add_ref(struct ib_qp *qp) @@ -733,7 +752,7 @@ void c4iw_qp_add_ref(struct ib_qp *qp) void c4iw_qp_rem_ref(struct ib_qp *qp) { PDBG("%s ib_qp %p\n", __func__, qp); - kref_put(&to_c4iw_qp(qp)->kref, _free_qp); + kref_put(&to_c4iw_qp(qp)->kref, queue_qp_free); } static void add_to_fc_list(struct list_head *head, struct list_head *entry) @@ -1706,7 +1725,6 @@ int c4iw_destroy_qp(struct ib_qp *ib_qp) struct c4iw_dev *rhp; struct c4iw_qp *qhp; struct c4iw_qp_attributes attrs; - struct c4iw_ucontext *ucontext; qhp = to_c4iw_qp(ib_qp); rhp = qhp->rhp; @@ -1726,11 +1744,6 @@ int c4iw_destroy_qp(struct ib_qp *ib_qp) spin_unlock_irq(&rhp->lock); free_ird(rhp, qhp->attr.max_ird); - ucontext = ib_qp->uobject ? - to_c4iw_ucontext(ib_qp->uobject->context) : NULL; - destroy_qp(&rhp->rdev, &qhp->wq, - ucontext ? &ucontext->uctx : &rhp->rdev.uctx); - c4iw_qp_rem_ref(ib_qp); PDBG("%s ib_qp %p qpid 0x%0x\n", __func__, ib_qp, qhp->wq.sq.qid); @@ -1829,6 +1842,7 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, mutex_init(&qhp->mutex); init_waitqueue_head(&qhp->wait); kref_init(&qhp->kref); + INIT_WORK(&qhp->free_work, free_qp_work); ret = insert_handle(rhp, &rhp->qpidr, qhp, qhp->wq.sq.qid); if (ret) @@ -1915,6 +1929,9 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, ma_sync_key_mm->len = PAGE_SIZE; insert_mmap(ucontext, ma_sync_key_mm); } + + c4iw_get_ucontext(ucontext); + qhp->ucontext = ucontext; } qhp->ibqp.qp_num = qhp->wq.sq.qid; init_timer(&(qhp->timer)); -- cgit v1.2.3 From 3bcf96e0183f5c863657cb6ae9adad307a0f6071 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Thu, 22 Dec 2016 07:40:37 -0800 Subject: iw_cxgb4: do not send RX_DATA_ACK CPLs after close/abort Function rx_data(), which handles ingress CPL_RX_DATA messages, was always sending an RX_DATA_ACK with the goal of updating the credits. However, if the RDMA connection is moved out of FPDU mode abruptly, then it is possible for iw_cxgb4 to process queued RX_DATA CPLs after HW has aborted the connection. These CPLs should not trigger RX_DATA_ACKS. If they do, HW can see a READ after DELETE of the DB_LE hash entry for the tid and post a LE_DB HashTblMemCrcError. Signed-off-by: Steve Wise Signed-off-by: Doug Ledford --- drivers/infiniband/hw/cxgb4/cm.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index f1510cc76d2d..9398143d7c5e 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -1804,20 +1804,21 @@ static int rx_data(struct c4iw_dev *dev, struct sk_buff *skb) skb_trim(skb, dlen); mutex_lock(&ep->com.mutex); - /* update RX credits */ - update_rx_credits(ep, dlen); - switch (ep->com.state) { case MPA_REQ_SENT: + update_rx_credits(ep, dlen); ep->rcv_seq += dlen; disconnect = process_mpa_reply(ep, skb); break; case MPA_REQ_WAIT: + update_rx_credits(ep, dlen); ep->rcv_seq += dlen; disconnect = process_mpa_request(ep, skb); break; case FPDU_MODE: { struct c4iw_qp_attributes attrs; + + update_rx_credits(ep, dlen); BUG_ON(!ep->com.qp); if (status) pr_err("%s Unexpected streaming data." \ -- cgit v1.2.3 From ecc8995363ee6231b32dad61c955b371b79cc4cf Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Tue, 10 Jan 2017 17:31:56 +0300 Subject: pinctrl: broxton: Use correct PADCFGLOCK offset PADCFGLOCK (and PADCFGLOCK_TX) offset in Broxton actually starts at 0x060 and not 0x090 as used in the driver. Fix it to use the correct offset. Signed-off-by: Mika Westerberg Reviewed-by: Andy Shevchenko Signed-off-by: Linus Walleij --- drivers/pinctrl/intel/pinctrl-broxton.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pinctrl/intel/pinctrl-broxton.c b/drivers/pinctrl/intel/pinctrl-broxton.c index 59cb7a6fc5be..901b356b09d7 100644 --- a/drivers/pinctrl/intel/pinctrl-broxton.c +++ b/drivers/pinctrl/intel/pinctrl-broxton.c @@ -19,7 +19,7 @@ #define BXT_PAD_OWN 0x020 #define BXT_HOSTSW_OWN 0x080 -#define BXT_PADCFGLOCK 0x090 +#define BXT_PADCFGLOCK 0x060 #define BXT_GPI_IE 0x110 #define BXT_COMMUNITY(s, e) \ -- cgit v1.2.3 From 17fab473693e8357a9aa6fee4fbed6c13a34bd81 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 2 Jan 2017 14:07:22 +0200 Subject: pinctrl: intel: Set pin direction properly There are two bits in the PADCFG0 register to configure direction, one per TX/RX buffers. For now we wrongly assume that the GPIO is always requested before it is being used, which is not true when the GPIO is used through irqchip. In this case the GPIO is never requested and we never enable RX buffer for it. Fix this by setting both bits accordingly. Reported-by: Jarkko Nikula Signed-off-by: Andy Shevchenko Signed-off-by: Linus Walleij --- drivers/pinctrl/intel/pinctrl-intel.c | 30 +++++++++++++++++++----------- 1 file changed, 19 insertions(+), 11 deletions(-) diff --git a/drivers/pinctrl/intel/pinctrl-intel.c b/drivers/pinctrl/intel/pinctrl-intel.c index 1e139672f1af..6df35dcb29ae 100644 --- a/drivers/pinctrl/intel/pinctrl-intel.c +++ b/drivers/pinctrl/intel/pinctrl-intel.c @@ -353,6 +353,21 @@ static int intel_pinmux_set_mux(struct pinctrl_dev *pctldev, unsigned function, return 0; } +static void __intel_gpio_set_direction(void __iomem *padcfg0, bool input) +{ + u32 value; + + value = readl(padcfg0); + if (input) { + value &= ~PADCFG0_GPIORXDIS; + value |= PADCFG0_GPIOTXDIS; + } else { + value &= ~PADCFG0_GPIOTXDIS; + value |= PADCFG0_GPIORXDIS; + } + writel(value, padcfg0); +} + static int intel_gpio_request_enable(struct pinctrl_dev *pctldev, struct pinctrl_gpio_range *range, unsigned pin) @@ -375,11 +390,11 @@ static int intel_gpio_request_enable(struct pinctrl_dev *pctldev, /* Disable SCI/SMI/NMI generation */ value &= ~(PADCFG0_GPIROUTIOXAPIC | PADCFG0_GPIROUTSCI); value &= ~(PADCFG0_GPIROUTSMI | PADCFG0_GPIROUTNMI); - /* Disable TX buffer and enable RX (this will be input) */ - value &= ~PADCFG0_GPIORXDIS; - value |= PADCFG0_GPIOTXDIS; writel(value, padcfg0); + /* Disable TX buffer and enable RX (this will be input) */ + __intel_gpio_set_direction(padcfg0, true); + raw_spin_unlock_irqrestore(&pctrl->lock, flags); return 0; @@ -392,18 +407,11 @@ static int intel_gpio_set_direction(struct pinctrl_dev *pctldev, struct intel_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev); void __iomem *padcfg0; unsigned long flags; - u32 value; raw_spin_lock_irqsave(&pctrl->lock, flags); padcfg0 = intel_get_padcfg(pctrl, pin, PADCFG0); - - value = readl(padcfg0); - if (input) - value |= PADCFG0_GPIOTXDIS; - else - value &= ~PADCFG0_GPIOTXDIS; - writel(value, padcfg0); + __intel_gpio_set_direction(padcfg0, input); raw_spin_unlock_irqrestore(&pctrl->lock, flags); -- cgit v1.2.3 From 04ff5a095d662e0879f0eb04b9247e092210aeff Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 10 Jan 2017 16:38:52 +0200 Subject: pinctrl: baytrail: Rectify debounce support The commit 658b476c742f ("pinctrl: baytrail: Add debounce configuration") implements debounce for Baytrail pin control, but seems wasn't tested properly. The register which keeps debounce value is separated from the configuration one. Writing wrong values to the latter will guarantee wrong behaviour of the driver and even might break something physically. Besides above there is missed case how to disable it, which is actually done through the bit in configuration register. Rectify implementation here by using proper register for debounce value. Fixes: 658b476c742f ("pinctrl: baytrail: Add debounce configuration") Cc: Cristina Ciocan Signed-off-by: Andy Shevchenko Acked-by: Mika Westerberg Signed-off-by: Linus Walleij --- drivers/pinctrl/intel/pinctrl-baytrail.c | 28 +++++++++++++++++----------- 1 file changed, 17 insertions(+), 11 deletions(-) diff --git a/drivers/pinctrl/intel/pinctrl-baytrail.c b/drivers/pinctrl/intel/pinctrl-baytrail.c index 37300634b7d2..67e92699b84e 100644 --- a/drivers/pinctrl/intel/pinctrl-baytrail.c +++ b/drivers/pinctrl/intel/pinctrl-baytrail.c @@ -1092,6 +1092,7 @@ static int byt_pin_config_get(struct pinctrl_dev *pctl_dev, unsigned int offset, enum pin_config_param param = pinconf_to_config_param(*config); void __iomem *conf_reg = byt_gpio_reg(vg, offset, BYT_CONF0_REG); void __iomem *val_reg = byt_gpio_reg(vg, offset, BYT_VAL_REG); + void __iomem *db_reg = byt_gpio_reg(vg, offset, BYT_DEBOUNCE_REG); unsigned long flags; u32 conf, pull, val, debounce; u16 arg = 0; @@ -1128,7 +1129,7 @@ static int byt_pin_config_get(struct pinctrl_dev *pctl_dev, unsigned int offset, return -EINVAL; raw_spin_lock_irqsave(&vg->lock, flags); - debounce = readl(byt_gpio_reg(vg, offset, BYT_DEBOUNCE_REG)); + debounce = readl(db_reg); raw_spin_unlock_irqrestore(&vg->lock, flags); switch (debounce & BYT_DEBOUNCE_PULSE_MASK) { @@ -1176,6 +1177,7 @@ static int byt_pin_config_set(struct pinctrl_dev *pctl_dev, unsigned int param, arg; void __iomem *conf_reg = byt_gpio_reg(vg, offset, BYT_CONF0_REG); void __iomem *val_reg = byt_gpio_reg(vg, offset, BYT_VAL_REG); + void __iomem *db_reg = byt_gpio_reg(vg, offset, BYT_DEBOUNCE_REG); unsigned long flags; u32 conf, val, debounce; int i, ret = 0; @@ -1238,36 +1240,40 @@ static int byt_pin_config_set(struct pinctrl_dev *pctl_dev, break; case PIN_CONFIG_INPUT_DEBOUNCE: - debounce = readl(byt_gpio_reg(vg, offset, - BYT_DEBOUNCE_REG)); - conf &= ~BYT_DEBOUNCE_PULSE_MASK; + debounce = readl(db_reg); + debounce &= ~BYT_DEBOUNCE_PULSE_MASK; switch (arg) { + case 0: + conf &= BYT_DEBOUNCE_EN; + break; case 375: - conf |= BYT_DEBOUNCE_PULSE_375US; + debounce |= BYT_DEBOUNCE_PULSE_375US; break; case 750: - conf |= BYT_DEBOUNCE_PULSE_750US; + debounce |= BYT_DEBOUNCE_PULSE_750US; break; case 1500: - conf |= BYT_DEBOUNCE_PULSE_1500US; + debounce |= BYT_DEBOUNCE_PULSE_1500US; break; case 3000: - conf |= BYT_DEBOUNCE_PULSE_3MS; + debounce |= BYT_DEBOUNCE_PULSE_3MS; break; case 6000: - conf |= BYT_DEBOUNCE_PULSE_6MS; + debounce |= BYT_DEBOUNCE_PULSE_6MS; break; case 12000: - conf |= BYT_DEBOUNCE_PULSE_12MS; + debounce |= BYT_DEBOUNCE_PULSE_12MS; break; case 24000: - conf |= BYT_DEBOUNCE_PULSE_24MS; + debounce |= BYT_DEBOUNCE_PULSE_24MS; break; default: ret = -EINVAL; } + if (!ret) + writel(debounce, db_reg); break; default: ret = -ENOTSUPP; -- cgit v1.2.3 From 49c03096263871a68c9dea3e86b7d1e163d2fba8 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Wed, 11 Jan 2017 15:49:00 +0200 Subject: pinctrl: baytrail: Do not add all GPIOs to IRQ domain When DIRECT_IRQ_EN is set, the pin is routed directly to the IO-APIC bypassing the GPIO driver completely. However, the mask register is still used to determine if the pin is supposed to generate IRQ or not. So with commit 3ae02c14d964 the IRQ core masks all IRQs (because of handle_bad_irq()) the pin connected to the touchscreen gets masked as well and hence no interrupts. To make this all work as expected we do not add those GPIOs to the IRQ domain that can actually propagate interrupts. Fixes: 3ae02c14d964 ("pinctrl: intel: set default handler to be handle_bad_irq()") Reported-by: Robert R. Howell Suggested-by: Mika Westerberg Signed-off-by: Andy Shevchenko Acked-by: Mika Westerberg Signed-off-by: Linus Walleij --- drivers/pinctrl/intel/pinctrl-baytrail.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/pinctrl/intel/pinctrl-baytrail.c b/drivers/pinctrl/intel/pinctrl-baytrail.c index 67e92699b84e..c123488266ce 100644 --- a/drivers/pinctrl/intel/pinctrl-baytrail.c +++ b/drivers/pinctrl/intel/pinctrl-baytrail.c @@ -1623,6 +1623,8 @@ static void byt_gpio_irq_handler(struct irq_desc *desc) static void byt_gpio_irq_init_hw(struct byt_gpio *vg) { + struct gpio_chip *gc = &vg->chip; + struct device *dev = &vg->pdev->dev; void __iomem *reg; u32 base, value; int i; @@ -1644,10 +1646,12 @@ static void byt_gpio_irq_init_hw(struct byt_gpio *vg) } value = readl(reg); - if ((value & BYT_PIN_MUX) == byt_get_gpio_mux(vg, i) && - !(value & BYT_DIRECT_IRQ_EN)) { + if (value & BYT_DIRECT_IRQ_EN) { + clear_bit(i, gc->irq_valid_mask); + dev_dbg(dev, "excluding GPIO %d from IRQ domain\n", i); + } else if ((value & BYT_PIN_MUX) == byt_get_gpio_mux(vg, i)) { byt_gpio_clear_triggering(vg, i); - dev_dbg(&vg->pdev->dev, "disabling GPIO %d\n", i); + dev_dbg(dev, "disabling GPIO %d\n", i); } } @@ -1686,6 +1690,7 @@ static int byt_gpio_probe(struct byt_gpio *vg) gc->can_sleep = false; gc->parent = &vg->pdev->dev; gc->ngpio = vg->soc_data->npins; + gc->irq_need_valid_mask = true; #ifdef CONFIG_PM_SLEEP vg->saved_context = devm_kcalloc(&vg->pdev->dev, gc->ngpio, -- cgit v1.2.3 From 581d3c2025632f838fb08e5160dab752b3a1f527 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 11 Jan 2017 15:36:56 +0100 Subject: pinctrl: amd: avoid maybe-uninitalized warning Since gpio_dev->hwbank_num is now a variable, the compiler cannot figure out if pin_num is initialized at all: drivers/pinctrl/pinctrl-amd.c: In function 'amd_gpio_dbg_show': drivers/pinctrl/pinctrl-amd.c:210:3: warning: 'pin_num' may be used uninitialized in this function [-Wmaybe-uninitialized] for (; i < pin_num; i++) { ^~~ drivers/pinctrl/pinctrl-amd.c:172:21: warning: 'i' may be used uninitialized in this function [-Wmaybe-uninitialized] This adds a 'default' statement to make that case well-defined. Fixes: 3bfd44306c65 ("pinctrl: amd: Add support for additional GPIO") Signed-off-by: Arnd Bergmann Signed-off-by: Linus Walleij --- drivers/pinctrl/pinctrl-amd.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/pinctrl/pinctrl-amd.c b/drivers/pinctrl/pinctrl-amd.c index c9a146948192..537b52055756 100644 --- a/drivers/pinctrl/pinctrl-amd.c +++ b/drivers/pinctrl/pinctrl-amd.c @@ -202,6 +202,8 @@ static void amd_gpio_dbg_show(struct seq_file *s, struct gpio_chip *gc) i = 128; pin_num = AMD_GPIO_PINS_BANK2 + i; break; + default: + return; } for (; i < pin_num; i++) { -- cgit v1.2.3 From 81ddd8c0c5e1cb41184d66567140cb48c53eb3d1 Mon Sep 17 00:00:00 2001 From: Rabin Vincent Date: Fri, 13 Jan 2017 15:00:16 +0100 Subject: cifs: initialize file_info_lock Reviewed-by: Jeff Layton CC: Stable file_info_lock is not initalized in initiate_cifs_search(), leading to the following splat after a simple "mount.cifs ... dir && ls dir/": BUG: spinlock bad magic on CPU#0, ls/486 lock: 0xffff880009301110, .magic: 00000000, .owner: /-1, .owner_cpu: 0 CPU: 0 PID: 486 Comm: ls Not tainted 4.9.0 #27 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) ffffc900042f3db0 ffffffff81327533 0000000000000000 ffff880009301110 ffffc900042f3dd0 ffffffff810baf75 ffff880009301110 ffffffff817ae077 ffffc900042f3df0 ffffffff810baff6 ffff880009301110 ffff880008d69900 Call Trace: [] dump_stack+0x65/0x92 [] spin_dump+0x85/0xe0 [] spin_bug+0x26/0x30 [] do_raw_spin_lock+0xe9/0x130 [] _raw_spin_lock+0x1f/0x30 [] cifs_closedir+0x4d/0x100 [] __fput+0x5d/0x160 [] ____fput+0xe/0x10 [] task_work_run+0x7e/0xa0 [] exit_to_usermode_loop+0x92/0xa0 [] syscall_return_slowpath+0x49/0x50 [] entry_SYSCALL_64_fastpath+0xa7/0xa9 Fixes: 3afca265b5f53a0 ("Clarify locking of cifs file and tcon structures and make more granular") Signed-off-by: Rabin Vincent Signed-off-by: Steve French --- fs/cifs/readdir.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 8f6a2a5863b9..a27fc8791551 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -285,6 +285,7 @@ initiate_cifs_search(const unsigned int xid, struct file *file) rc = -ENOMEM; goto error_exit; } + spin_lock_init(&cifsFile->file_info_lock); file->private_data = cifsFile; cifsFile->tlink = cifs_get_tlink(tlink); tcon = tlink_tcon(tlink); -- cgit v1.2.3 From d7f5762c5e532dfe8247ce1bc60d97af27ff8d00 Mon Sep 17 00:00:00 2001 From: Alexander Alemayhu Date: Wed, 4 Jan 2017 23:16:41 +0100 Subject: netfilter: nf_tables: fix spelling mistakes o s/numerice/numeric o s/opertaor/operator Signed-off-by: Alexander Alemayhu Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/nf_tables.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 881d49e94569..e3f27e09eb2b 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -235,7 +235,7 @@ enum nft_rule_compat_flags { /** * enum nft_rule_compat_attributes - nf_tables rule compat attributes * - * @NFTA_RULE_COMPAT_PROTO: numerice value of handled protocol (NLA_U32) + * @NFTA_RULE_COMPAT_PROTO: numeric value of handled protocol (NLA_U32) * @NFTA_RULE_COMPAT_FLAGS: bitmask of enum nft_rule_compat_flags (NLA_U32) */ enum nft_rule_compat_attributes { @@ -499,7 +499,7 @@ enum nft_bitwise_attributes { * enum nft_byteorder_ops - nf_tables byteorder operators * * @NFT_BYTEORDER_NTOH: network to host operator - * @NFT_BYTEORDER_HTON: host to network opertaor + * @NFT_BYTEORDER_HTON: host to network operator */ enum nft_byteorder_ops { NFT_BYTEORDER_NTOH, -- cgit v1.2.3 From 6443ebc3fdd6f3c766d9442c18be274b3d736050 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sat, 7 Jan 2017 20:49:18 +0800 Subject: netfilter: rpfilter: fix incorrect loopback packet judgment Currently, we check the existing rtable in PREROUTING hook, if RTCF_LOCAL is set, we assume that the packet is loopback. But this assumption is incorrect, for example, a packet encapsulated in ipsec transport mode was received and routed to local, after decapsulation, it would be delivered to local again, and the rtable was not dropped, so RTCF_LOCAL check would trigger. But actually, the packet was not loopback. So for these normal loopback packets, we can check whether the in device is IFF_LOOPBACK or not. For these locally generated broadcast/multicast, we can check whether the skb->pkt_type is PACKET_LOOPBACK or not. Finally, there's a subtle difference between nft fib expr and xtables rpfilter extension, user can add the following nft rule to do strict rpfilter check: # nft add rule x y meta iif eth0 fib saddr . iif oif != eth0 drop So when the packet is loopback, it's better to store the in device instead of the LOOPBACK_IFINDEX, otherwise, after adding the above nft rule, locally generated broad/multicast packets will be dropped incorrectly. Fixes: f83a7ea2075c ("netfilter: xt_rpfilter: skip locally generated broadcast/multicast, too") Fixes: f6d0cbcf09c5 ("netfilter: nf_tables: add fib expression") Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nft_fib.h | 6 ++++++ net/ipv4/netfilter/ipt_rpfilter.c | 8 ++++---- net/ipv4/netfilter/nft_fib_ipv4.c | 15 +++++---------- net/ipv6/netfilter/ip6t_rpfilter.c | 8 ++++---- net/ipv6/netfilter/nft_fib_ipv6.c | 13 ++++--------- 5 files changed, 23 insertions(+), 27 deletions(-) diff --git a/include/net/netfilter/nft_fib.h b/include/net/netfilter/nft_fib.h index cbedda077db2..5ceb2205e4e3 100644 --- a/include/net/netfilter/nft_fib.h +++ b/include/net/netfilter/nft_fib.h @@ -9,6 +9,12 @@ struct nft_fib { extern const struct nla_policy nft_fib_policy[]; +static inline bool +nft_fib_is_loopback(const struct sk_buff *skb, const struct net_device *in) +{ + return skb->pkt_type == PACKET_LOOPBACK || in->flags & IFF_LOOPBACK; +} + int nft_fib_dump(struct sk_buff *skb, const struct nft_expr *expr); int nft_fib_init(const struct nft_ctx *ctx, const struct nft_expr *expr, const struct nlattr * const tb[]); diff --git a/net/ipv4/netfilter/ipt_rpfilter.c b/net/ipv4/netfilter/ipt_rpfilter.c index f273098e48fd..37fb9552e858 100644 --- a/net/ipv4/netfilter/ipt_rpfilter.c +++ b/net/ipv4/netfilter/ipt_rpfilter.c @@ -63,10 +63,10 @@ static bool rpfilter_lookup_reverse(struct net *net, struct flowi4 *fl4, return dev_match || flags & XT_RPFILTER_LOOSE; } -static bool rpfilter_is_local(const struct sk_buff *skb) +static bool +rpfilter_is_loopback(const struct sk_buff *skb, const struct net_device *in) { - const struct rtable *rt = skb_rtable(skb); - return rt && (rt->rt_flags & RTCF_LOCAL); + return skb->pkt_type == PACKET_LOOPBACK || in->flags & IFF_LOOPBACK; } static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par) @@ -79,7 +79,7 @@ static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par) info = par->matchinfo; invert = info->flags & XT_RPFILTER_INVERT; - if (rpfilter_is_local(skb)) + if (rpfilter_is_loopback(skb, xt_in(par))) return true ^ invert; iph = ip_hdr(skb); diff --git a/net/ipv4/netfilter/nft_fib_ipv4.c b/net/ipv4/netfilter/nft_fib_ipv4.c index 965b1a161369..2981291910dd 100644 --- a/net/ipv4/netfilter/nft_fib_ipv4.c +++ b/net/ipv4/netfilter/nft_fib_ipv4.c @@ -26,13 +26,6 @@ static __be32 get_saddr(__be32 addr) return addr; } -static bool fib4_is_local(const struct sk_buff *skb) -{ - const struct rtable *rt = skb_rtable(skb); - - return rt && (rt->rt_flags & RTCF_LOCAL); -} - #define DSCP_BITS 0xfc void nft_fib4_eval_type(const struct nft_expr *expr, struct nft_regs *regs, @@ -95,8 +88,10 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs, else oif = NULL; - if (nft_hook(pkt) == NF_INET_PRE_ROUTING && fib4_is_local(pkt->skb)) { - nft_fib_store_result(dest, priv->result, pkt, LOOPBACK_IFINDEX); + if (nft_hook(pkt) == NF_INET_PRE_ROUTING && + nft_fib_is_loopback(pkt->skb, nft_in(pkt))) { + nft_fib_store_result(dest, priv->result, pkt, + nft_in(pkt)->ifindex); return; } @@ -131,7 +126,7 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs, switch (res.type) { case RTN_UNICAST: break; - case RTN_LOCAL: /* should not appear here, see fib4_is_local() above */ + case RTN_LOCAL: /* Should not see RTN_LOCAL here */ return; default: break; diff --git a/net/ipv6/netfilter/ip6t_rpfilter.c b/net/ipv6/netfilter/ip6t_rpfilter.c index d5263dc364a9..b12e61b7b16c 100644 --- a/net/ipv6/netfilter/ip6t_rpfilter.c +++ b/net/ipv6/netfilter/ip6t_rpfilter.c @@ -72,10 +72,10 @@ static bool rpfilter_lookup_reverse6(struct net *net, const struct sk_buff *skb, return ret; } -static bool rpfilter_is_local(const struct sk_buff *skb) +static bool +rpfilter_is_loopback(const struct sk_buff *skb, const struct net_device *in) { - const struct rt6_info *rt = (const void *) skb_dst(skb); - return rt && (rt->rt6i_flags & RTF_LOCAL); + return skb->pkt_type == PACKET_LOOPBACK || in->flags & IFF_LOOPBACK; } static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par) @@ -85,7 +85,7 @@ static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par) struct ipv6hdr *iph; bool invert = info->flags & XT_RPFILTER_INVERT; - if (rpfilter_is_local(skb)) + if (rpfilter_is_loopback(skb, xt_in(par))) return true ^ invert; iph = ipv6_hdr(skb); diff --git a/net/ipv6/netfilter/nft_fib_ipv6.c b/net/ipv6/netfilter/nft_fib_ipv6.c index c947aad8bcc6..765facf03d45 100644 --- a/net/ipv6/netfilter/nft_fib_ipv6.c +++ b/net/ipv6/netfilter/nft_fib_ipv6.c @@ -18,13 +18,6 @@ #include #include -static bool fib6_is_local(const struct sk_buff *skb) -{ - const struct rt6_info *rt = (const void *)skb_dst(skb); - - return rt && (rt->rt6i_flags & RTF_LOCAL); -} - static int get_ifindex(const struct net_device *dev) { return dev ? dev->ifindex : 0; @@ -164,8 +157,10 @@ void nft_fib6_eval(const struct nft_expr *expr, struct nft_regs *regs, lookup_flags = nft_fib6_flowi_init(&fl6, priv, pkt, oif); - if (nft_hook(pkt) == NF_INET_PRE_ROUTING && fib6_is_local(pkt->skb)) { - nft_fib_store_result(dest, priv->result, pkt, LOOPBACK_IFINDEX); + if (nft_hook(pkt) == NF_INET_PRE_ROUTING && + nft_fib_is_loopback(pkt->skb, nft_in(pkt))) { + nft_fib_store_result(dest, priv->result, pkt, + nft_in(pkt)->ifindex); return; } -- cgit v1.2.3 From d21e540b4dd74a26df7a66ebab75c693a4a6a861 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sat, 7 Jan 2017 20:51:50 +0800 Subject: netfilter: nf_tables: fix possible oops when dumping stateful objects When dumping nft stateful objects, if NFTA_OBJ_TABLE and NFTA_OBJ_TYPE attributes are not specified either, filter will become NULL, so oops will happen(actually nft utility will always set NFTA_OBJ_TABLE attr, so I write a test program to make this happen): BUG: unable to handle kernel NULL pointer dereference at (null) IP: nf_tables_dump_obj+0x17c/0x330 [nf_tables] [...] Call Trace: ? nf_tables_dump_obj+0x5/0x330 [nf_tables] ? __kmalloc_reserve.isra.35+0x31/0x90 ? __alloc_skb+0x5b/0x1e0 netlink_dump+0x124/0x2a0 __netlink_dump_start+0x161/0x190 nf_tables_getobj+0xe8/0x280 [nf_tables] Fixes: a9fea2a3c3cf ("netfilter: nf_tables: allow to filter stateful object dumps by type") Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 0db5f9782265..091d2dcc63b2 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -4262,10 +4262,11 @@ static int nf_tables_dump_obj(struct sk_buff *skb, struct netlink_callback *cb) if (idx > s_idx) memset(&cb->args[1], 0, sizeof(cb->args) - sizeof(cb->args[0])); - if (filter->table[0] && + if (filter && filter->table[0] && strcmp(filter->table, table->name)) goto cont; - if (filter->type != NFT_OBJECT_UNSPEC && + if (filter && + filter->type != NFT_OBJECT_UNSPEC && obj->type->type != filter->type) goto cont; -- cgit v1.2.3 From e4670b058af64639ec1aef4db845c39bfdfff7c4 Mon Sep 17 00:00:00 2001 From: William Breathitt Gray Date: Mon, 9 Jan 2017 17:24:18 -0500 Subject: netfilter: Fix typo in NF_CONNTRACK Kconfig option description The NF_CONNTRACK Kconfig option description makes an incorrect reference to the "meta" expression where the "ct" expression would be correct.This patch fixes the respective typographical error. Fixes: d497c6352736 ("netfilter: add help information to new nf_tables Kconfig options") Signed-off-by: William Breathitt Gray Signed-off-by: Pablo Neira Ayuso --- net/netfilter/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 63729b489c2c..bbc45f8a7b2d 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -494,7 +494,7 @@ config NFT_CT depends on NF_CONNTRACK tristate "Netfilter nf_tables conntrack module" help - This option adds the "meta" expression that you can use to match + This option adds the "ct" expression that you can use to match connection tracking information such as the flow state. config NFT_SET_RBTREE -- cgit v1.2.3 From 7e9081c5aac73b8a0bc22e0b3e7a12c3e9cf5256 Mon Sep 17 00:00:00 2001 From: Gustavo Padovan Date: Fri, 13 Jan 2017 12:22:09 -0200 Subject: drm/fence: fix memory overwrite when setting out_fence fd Currently if the userspace declares a int variable to store the out_fence fd and pass it to OUT_FENCE_PTR the kernel will overwrite the 32 bits above the int variable on 64 bits systems. Fix this by making the internal storage of out_fence in the kernel a s32 pointer. Reported-by: Chad Versace Signed-off-by: Gustavo Padovan Fixes: beaf5af48034 ("drm/fence: add out-fences support") Cc: Daniel Vetter Cc: Rafael Antognolli Cc: Laurent Pinchart Acked-by: Laurent Pinchart Reviewed-and-Tested-by: Chad Versace Link: http://patchwork.freedesktop.org/patch/msgid/1484317329-9293-1-git-send-email-gustavo@padovan.org --- drivers/gpu/drm/drm_atomic.c | 12 ++++++------ include/drm/drm_atomic.h | 2 +- include/drm/drm_mode_config.h | 2 +- 3 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index 60697482b94c..50f5cf7b69d1 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -291,15 +291,15 @@ drm_atomic_get_crtc_state(struct drm_atomic_state *state, EXPORT_SYMBOL(drm_atomic_get_crtc_state); static void set_out_fence_for_crtc(struct drm_atomic_state *state, - struct drm_crtc *crtc, s64 __user *fence_ptr) + struct drm_crtc *crtc, s32 __user *fence_ptr) { state->crtcs[drm_crtc_index(crtc)].out_fence_ptr = fence_ptr; } -static s64 __user *get_out_fence_for_crtc(struct drm_atomic_state *state, +static s32 __user *get_out_fence_for_crtc(struct drm_atomic_state *state, struct drm_crtc *crtc) { - s64 __user *fence_ptr; + s32 __user *fence_ptr; fence_ptr = state->crtcs[drm_crtc_index(crtc)].out_fence_ptr; state->crtcs[drm_crtc_index(crtc)].out_fence_ptr = NULL; @@ -512,7 +512,7 @@ int drm_atomic_crtc_set_property(struct drm_crtc *crtc, state->color_mgmt_changed |= replaced; return ret; } else if (property == config->prop_out_fence_ptr) { - s64 __user *fence_ptr = u64_to_user_ptr(val); + s32 __user *fence_ptr = u64_to_user_ptr(val); if (!fence_ptr) return 0; @@ -1915,7 +1915,7 @@ EXPORT_SYMBOL(drm_atomic_clean_old_fb); */ struct drm_out_fence_state { - s64 __user *out_fence_ptr; + s32 __user *out_fence_ptr; struct sync_file *sync_file; int fd; }; @@ -1952,7 +1952,7 @@ static int prepare_crtc_signaling(struct drm_device *dev, return 0; for_each_crtc_in_state(state, crtc, crtc_state, i) { - u64 __user *fence_ptr; + s32 __user *fence_ptr; fence_ptr = get_out_fence_for_crtc(crtc_state->state, crtc); diff --git a/include/drm/drm_atomic.h b/include/drm/drm_atomic.h index d6d241f63b9f..56814e8ae7ea 100644 --- a/include/drm/drm_atomic.h +++ b/include/drm/drm_atomic.h @@ -144,7 +144,7 @@ struct __drm_crtcs_state { struct drm_crtc *ptr; struct drm_crtc_state *state; struct drm_crtc_commit *commit; - s64 __user *out_fence_ptr; + s32 __user *out_fence_ptr; }; struct __drm_connnectors_state { diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h index bf9991b20611..137432386310 100644 --- a/include/drm/drm_mode_config.h +++ b/include/drm/drm_mode_config.h @@ -488,7 +488,7 @@ struct drm_mode_config { /** * @prop_out_fence_ptr: Sync File fd pointer representing the * outgoing fences for a CRTC. Userspace should provide a pointer to a - * value of type s64, and then cast that pointer to u64. + * value of type s32, and then cast that pointer to u64. */ struct drm_property *prop_out_fence_ptr; /** -- cgit v1.2.3 From 5d0e7705774dd412a465896d08d59a81a345c1e4 Mon Sep 17 00:00:00 2001 From: Tom Hromatka Date: Tue, 10 Jan 2017 10:57:56 -0700 Subject: sparc: Fixed typo in sstate.c. Replaced panicing with panicking Signed-off-by: Tom Hromatka Signed-off-by: David S. Miller --- arch/sparc/kernel/sstate.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/sparc/kernel/sstate.c b/arch/sparc/kernel/sstate.c index c59af546f522..3caed4023589 100644 --- a/arch/sparc/kernel/sstate.c +++ b/arch/sparc/kernel/sstate.c @@ -43,8 +43,8 @@ static const char poweroff_msg[32] __attribute__((aligned(32))) = "Linux powering off"; static const char rebooting_msg[32] __attribute__((aligned(32))) = "Linux rebooting"; -static const char panicing_msg[32] __attribute__((aligned(32))) = - "Linux panicing"; +static const char panicking_msg[32] __attribute__((aligned(32))) = + "Linux panicking"; static int sstate_reboot_call(struct notifier_block *np, unsigned long type, void *_unused) { @@ -76,7 +76,7 @@ static struct notifier_block sstate_reboot_notifier = { static int sstate_panic_event(struct notifier_block *n, unsigned long event, void *ptr) { - do_set_sstate(HV_SOFT_STATE_TRANSITION, panicing_msg); + do_set_sstate(HV_SOFT_STATE_TRANSITION, panicking_msg); return NOTIFY_DONE; } -- cgit v1.2.3 From 58c744da9dcc82a4b55a18e05149ae0e32624d11 Mon Sep 17 00:00:00 2001 From: Tina Zhang Date: Tue, 17 Jan 2017 14:33:16 +0800 Subject: drm/i915/gvt: Enable the shadow batch buffer Shadow batch buffer is used to shadow the privileged batch buffer which is submitted by vGPU's workload. This patch is used to unmark this functionality. Signed-off-by: Tina Zhang Signed-off-by: Zhenyu Wang --- drivers/gpu/drm/i915/gvt/cmd_parser.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c index d26a092c70e8..e4563984cb1e 100644 --- a/drivers/gpu/drm/i915/gvt/cmd_parser.c +++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c @@ -481,7 +481,6 @@ struct parser_exec_state { (s->vgpu->gvt->device_info.gmadr_bytes_in_cmd >> 2) static unsigned long bypass_scan_mask = 0; -static bool bypass_batch_buffer_scan = true; /* ring ALL, type = 0 */ static struct sub_op_bits sub_op_mi[] = { @@ -1525,9 +1524,6 @@ static int batch_buffer_needs_scan(struct parser_exec_state *s) { struct intel_gvt *gvt = s->vgpu->gvt; - if (bypass_batch_buffer_scan) - return 0; - if (IS_BROADWELL(gvt->dev_priv) || IS_SKYLAKE(gvt->dev_priv)) { /* BDW decides privilege based on address space */ if (cmd_val(s, 0) & (1 << 8)) -- cgit v1.2.3 From 62f0a11e2339e1ba154600d1f49ef5d5d84eaae4 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 6 Jan 2017 19:58:16 +0000 Subject: drm/i915/gvt: Fix relocation of shadow bb set_gma_to_bb_cmd() is completely bogus - it is (incorrectly) applying the rules to read a GTT offset from a command as opposed to writing the GTT offset. And to cap it all set_gma_to_bb_cmd() is called within a list iterator of the most strange construction. Fixes: be1da7070aea ("drm/i915/gvt: vGPU command scanner") Signed-off-by: Chris Wilson Cc: Zhenyu Wang Cc: Zhi Wang Cc: Yulei Zhang Cc: # v4.10-rc1+ Tested-by: Tina Zhang Signed-off-by: Zhenyu Wang --- drivers/gpu/drm/i915/gvt/execlist.c | 64 ++++++++++-------------------------- drivers/gpu/drm/i915/gvt/scheduler.h | 2 +- 2 files changed, 19 insertions(+), 47 deletions(-) diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c index f32bb6f6495c..fb852c51d00e 100644 --- a/drivers/gpu/drm/i915/gvt/execlist.c +++ b/drivers/gpu/drm/i915/gvt/execlist.c @@ -364,58 +364,30 @@ static void free_workload(struct intel_vgpu_workload *workload) #define get_desc_from_elsp_dwords(ed, i) \ ((struct execlist_ctx_descriptor_format *)&((ed)->data[i * 2])) - -#define BATCH_BUFFER_ADDR_MASK ((1UL << 32) - (1U << 2)) -#define BATCH_BUFFER_ADDR_HIGH_MASK ((1UL << 16) - (1U)) -static int set_gma_to_bb_cmd(struct intel_shadow_bb_entry *entry_obj, - unsigned long add, int gmadr_bytes) -{ - if (WARN_ON(gmadr_bytes != 4 && gmadr_bytes != 8)) - return -1; - - *((u32 *)(entry_obj->bb_start_cmd_va + (1 << 2))) = add & - BATCH_BUFFER_ADDR_MASK; - if (gmadr_bytes == 8) { - *((u32 *)(entry_obj->bb_start_cmd_va + (2 << 2))) = - add & BATCH_BUFFER_ADDR_HIGH_MASK; - } - - return 0; -} - static void prepare_shadow_batch_buffer(struct intel_vgpu_workload *workload) { - int gmadr_bytes = workload->vgpu->gvt->device_info.gmadr_bytes_in_cmd; + const int gmadr_bytes = workload->vgpu->gvt->device_info.gmadr_bytes_in_cmd; + struct intel_shadow_bb_entry *entry_obj; /* pin the gem object to ggtt */ - if (!list_empty(&workload->shadow_bb)) { - struct intel_shadow_bb_entry *entry_obj = - list_first_entry(&workload->shadow_bb, - struct intel_shadow_bb_entry, - list); - struct intel_shadow_bb_entry *temp; + list_for_each_entry(entry_obj, &workload->shadow_bb, list) { + struct i915_vma *vma; - list_for_each_entry_safe(entry_obj, temp, &workload->shadow_bb, - list) { - struct i915_vma *vma; - - vma = i915_gem_object_ggtt_pin(entry_obj->obj, NULL, 0, - 4, 0); - if (IS_ERR(vma)) { - gvt_err("Cannot pin\n"); - return; - } - - /* FIXME: we are not tracking our pinned VMA leaving it - * up to the core to fix up the stray pin_count upon - * free. - */ - - /* update the relocate gma with shadow batch buffer*/ - set_gma_to_bb_cmd(entry_obj, - i915_ggtt_offset(vma), - gmadr_bytes); + vma = i915_gem_object_ggtt_pin(entry_obj->obj, NULL, 0, 4, 0); + if (IS_ERR(vma)) { + gvt_err("Cannot pin\n"); + return; } + + /* FIXME: we are not tracking our pinned VMA leaving it + * up to the core to fix up the stray pin_count upon + * free. + */ + + /* update the relocate gma with shadow batch buffer*/ + entry_obj->bb_start_cmd_va[1] = i915_ggtt_offset(vma); + if (gmadr_bytes == 8) + entry_obj->bb_start_cmd_va[2] = 0; } } diff --git a/drivers/gpu/drm/i915/gvt/scheduler.h b/drivers/gpu/drm/i915/gvt/scheduler.h index 3b30c28bff51..2833dfa8c9ae 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.h +++ b/drivers/gpu/drm/i915/gvt/scheduler.h @@ -113,7 +113,7 @@ struct intel_shadow_bb_entry { struct drm_i915_gem_object *obj; void *va; unsigned long len; - void *bb_start_cmd_va; + u32 *bb_start_cmd_va; }; #define workload_q_head(vgpu, ring_id) \ -- cgit v1.2.3 From 7622b25543665567d8830a63210385b7d705924b Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Mon, 10 Oct 2016 09:44:06 -0700 Subject: drm/vc4: Fix memory leak of the CRTC state. The underscores variant frees the pointers inside, while the no-underscores variant calls underscores and then frees the struct. Signed-off-by: Eric Anholt Fixes: d8dbf44f13b9 ("drm/vc4: Make the CRTCs cooperate on allocating display lists.") Cc: stable@vger.kernel.org --- drivers/gpu/drm/vc4/vc4_crtc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index a0fd3e66bc4b..7aadce1f7e7a 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -839,7 +839,7 @@ static void vc4_crtc_destroy_state(struct drm_crtc *crtc, } - __drm_atomic_helper_crtc_destroy_state(state); + drm_atomic_helper_crtc_destroy_state(crtc, state); } static const struct drm_crtc_funcs vc4_crtc_funcs = { -- cgit v1.2.3 From 21ccc32496b2f63228f5232b3ac0e426e8fb3c31 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 13 Jan 2017 10:49:00 +0300 Subject: drm/vc4: fix a bounds check We accidentally return success even if vc4_full_res_bounds_check() fails. Fixes: d5b1a78a772f ("drm/vc4: Add support for drawing 3D frames.") Signed-off-by: Dan Carpenter Reviewed-by: Eric Engestrom Reviewed-by: Eric Anholt --- drivers/gpu/drm/vc4/vc4_render_cl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vc4/vc4_render_cl.c b/drivers/gpu/drm/vc4/vc4_render_cl.c index 08886a309757..5cdd003605f5 100644 --- a/drivers/gpu/drm/vc4/vc4_render_cl.c +++ b/drivers/gpu/drm/vc4/vc4_render_cl.c @@ -461,7 +461,7 @@ static int vc4_rcl_surface_setup(struct vc4_exec_info *exec, } ret = vc4_full_res_bounds_check(exec, *obj, surf); - if (!ret) + if (ret) return ret; return 0; -- cgit v1.2.3 From 0f2ff82e11c86c05d051cae32b58226392d33bbf Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Tue, 17 Jan 2017 21:42:53 +1100 Subject: drm/vc4: Fix an integer overflow in temporary allocation layout. We copy the unvalidated ioctl arguments from the user into kernel temporary memory to run the validation from, to avoid a race where the user updates the unvalidate contents in between validating them and copying them into the validated BO. However, in setting up the layout of the kernel side, we failed to check one of the additions (the roundup() for shader_rec_offset) against integer overflow, allowing a nearly MAX_UINT value of bin_cl_size to cause us to under-allocate the temporary space that we then copy_from_user into. Reported-by: Murray McAllister Signed-off-by: Eric Anholt Fixes: d5b1a78a772f ("drm/vc4: Add support for drawing 3D frames.") --- drivers/gpu/drm/vc4/vc4_gem.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c index db920771bfb5..c5fe3554858e 100644 --- a/drivers/gpu/drm/vc4/vc4_gem.c +++ b/drivers/gpu/drm/vc4/vc4_gem.c @@ -594,7 +594,8 @@ vc4_get_bcl(struct drm_device *dev, struct vc4_exec_info *exec) args->shader_rec_count); struct vc4_bo *bo; - if (uniforms_offset < shader_rec_offset || + if (shader_rec_offset < args->bin_cl_size || + uniforms_offset < shader_rec_offset || exec_size < uniforms_offset || args->shader_rec_count >= (UINT_MAX / sizeof(struct vc4_shader_state)) || -- cgit v1.2.3 From 6b8ac63847bc2f958dd93c09edc941a0118992d9 Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Tue, 17 Jan 2017 21:58:06 +1100 Subject: drm/vc4: Return -EINVAL on the overflow checks failing. By failing to set the errno, we'd continue on to trying to set up the RCL, and then oops on trying to dereference the tile_bo that binning validation should have set up. Reported-by: Ingo Molnar Signed-off-by: Eric Anholt Fixes: d5b1a78a772f ("drm/vc4: Add support for drawing 3D frames.") --- drivers/gpu/drm/vc4/vc4_gem.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c index c5fe3554858e..ab3016982466 100644 --- a/drivers/gpu/drm/vc4/vc4_gem.c +++ b/drivers/gpu/drm/vc4/vc4_gem.c @@ -601,6 +601,7 @@ vc4_get_bcl(struct drm_device *dev, struct vc4_exec_info *exec) sizeof(struct vc4_shader_state)) || temp_size < exec_size) { DRM_ERROR("overflow in exec arguments\n"); + ret = -EINVAL; goto fail; } -- cgit v1.2.3 From 7195439d1d71bc4a6c33cfb57bc669a7cd041041 Mon Sep 17 00:00:00 2001 From: Rafał Miłecki Date: Fri, 13 Jan 2017 12:23:35 +0100 Subject: Revert "bcma: init serial console directly from ChipCommon code" This reverts commit 4c81acab3816 ("bcma: init serial console directly from ChipCommon code") as it broke IRQ assignment. Getting IRQ with bcma_core_irq helper on SoC requires MIPS core to be set. It happens *after* ChipCommon initialization so we can't do this so early. This fixes a user reported regression. It wasn't critical as serial was still somehow working but lack of IRQs was making in unreliable. Fixes: 4c81acab3816 ("bcma: init serial console directly from ChipCommon code") Reported-by: Felix Fietkau Cc: stable@vger.kernel.org # 4.6+ Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo --- drivers/bcma/bcma_private.h | 3 +++ drivers/bcma/driver_chipcommon.c | 11 +++-------- drivers/bcma/driver_mips.c | 3 +++ 3 files changed, 9 insertions(+), 8 deletions(-) diff --git a/drivers/bcma/bcma_private.h b/drivers/bcma/bcma_private.h index f642c4264c27..168fa175d65a 100644 --- a/drivers/bcma/bcma_private.h +++ b/drivers/bcma/bcma_private.h @@ -45,6 +45,9 @@ int bcma_sprom_get(struct bcma_bus *bus); void bcma_core_chipcommon_early_init(struct bcma_drv_cc *cc); void bcma_core_chipcommon_init(struct bcma_drv_cc *cc); void bcma_chipco_bcm4331_ext_pa_lines_ctl(struct bcma_drv_cc *cc, bool enable); +#ifdef CONFIG_BCMA_DRIVER_MIPS +void bcma_chipco_serial_init(struct bcma_drv_cc *cc); +#endif /* CONFIG_BCMA_DRIVER_MIPS */ /* driver_chipcommon_b.c */ int bcma_core_chipcommon_b_init(struct bcma_drv_cc_b *ccb); diff --git a/drivers/bcma/driver_chipcommon.c b/drivers/bcma/driver_chipcommon.c index b4f6520e74f0..62f5bfa5065d 100644 --- a/drivers/bcma/driver_chipcommon.c +++ b/drivers/bcma/driver_chipcommon.c @@ -15,8 +15,6 @@ #include #include -static void bcma_chipco_serial_init(struct bcma_drv_cc *cc); - static inline u32 bcma_cc_write32_masked(struct bcma_drv_cc *cc, u16 offset, u32 mask, u32 value) { @@ -186,9 +184,6 @@ void bcma_core_chipcommon_early_init(struct bcma_drv_cc *cc) if (cc->capabilities & BCMA_CC_CAP_PMU) bcma_pmu_early_init(cc); - if (IS_BUILTIN(CONFIG_BCM47XX) && bus->hosttype == BCMA_HOSTTYPE_SOC) - bcma_chipco_serial_init(cc); - if (bus->hosttype == BCMA_HOSTTYPE_SOC) bcma_core_chipcommon_flash_detect(cc); @@ -378,9 +373,9 @@ u32 bcma_chipco_gpio_pulldown(struct bcma_drv_cc *cc, u32 mask, u32 value) return res; } -static void bcma_chipco_serial_init(struct bcma_drv_cc *cc) +#ifdef CONFIG_BCMA_DRIVER_MIPS +void bcma_chipco_serial_init(struct bcma_drv_cc *cc) { -#if IS_BUILTIN(CONFIG_BCM47XX) unsigned int irq; u32 baud_base; u32 i; @@ -422,5 +417,5 @@ static void bcma_chipco_serial_init(struct bcma_drv_cc *cc) ports[i].baud_base = baud_base; ports[i].reg_shift = 0; } -#endif /* CONFIG_BCM47XX */ } +#endif /* CONFIG_BCMA_DRIVER_MIPS */ diff --git a/drivers/bcma/driver_mips.c b/drivers/bcma/driver_mips.c index 96f171328200..89af807cf29c 100644 --- a/drivers/bcma/driver_mips.c +++ b/drivers/bcma/driver_mips.c @@ -278,9 +278,12 @@ static void bcma_core_mips_nvram_init(struct bcma_drv_mips *mcore) void bcma_core_mips_early_init(struct bcma_drv_mips *mcore) { + struct bcma_bus *bus = mcore->core->bus; + if (mcore->early_setup_done) return; + bcma_chipco_serial_init(&bus->drv_cc); bcma_core_mips_nvram_init(mcore); mcore->early_setup_done = true; -- cgit v1.2.3 From 501db511397fd6efff3aa5b4e8de415b55559550 Mon Sep 17 00:00:00 2001 From: Rolf Neugebauer Date: Tue, 17 Jan 2017 18:13:51 +0000 Subject: virtio: don't set VIRTIO_NET_HDR_F_DATA_VALID on xmit This patch part reverts fd2a0437dc33 and e858fae2b0b8 which introduced a subtle change in how the virtio_net flags are derived from the SKBs ip_summed field. With the above commits, the flags are set to VIRTIO_NET_HDR_F_DATA_VALID when ip_summed == CHECKSUM_UNNECESSARY, thus treating it differently to ip_summed == CHECKSUM_NONE, which should be the same. Further, the virtio spec 1.0 / CS04 explicitly says that VIRTIO_NET_HDR_F_DATA_VALID must not be set by the driver. Fixes: fd2a0437dc33 ("virtio_net: introduce virtio_net_hdr_{from,to}_skb") Fixes: e858fae2b0b8 (" virtio_net: use common code for virtio_net_hdr and skb GSO conversion") Signed-off-by: Rolf Neugebauer Acked-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- include/linux/virtio_net.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/include/linux/virtio_net.h b/include/linux/virtio_net.h index 66204007d7ac..56436472ccc7 100644 --- a/include/linux/virtio_net.h +++ b/include/linux/virtio_net.h @@ -91,8 +91,6 @@ static inline int virtio_net_hdr_from_skb(const struct sk_buff *skb, skb_checksum_start_offset(skb)); hdr->csum_offset = __cpu_to_virtio16(little_endian, skb->csum_offset); - } else if (skb->ip_summed == CHECKSUM_UNNECESSARY) { - hdr->flags = VIRTIO_NET_HDR_F_DATA_VALID; } /* else everything is zero */ return 0; -- cgit v1.2.3 From a249708bc2aa1fe3ddf15dfac22bee519d15996b Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Tue, 17 Jan 2017 12:23:21 +0100 Subject: stmmac: add missing of_node_put The function stmmac_dt_phy provides several possibilities for initializing plat->mdio_node, all of which have the effect of increasing the reference count of the assigned value. This field is not updated elsewhere, so the value is live until the end of the lifetime of plat (devm_allocated), just after the end of stmmac_remove_config_dt. Thus, add an of_node_put on plat->mdio_node in stmmac_remove_config_dt. It is possible that the field mdio_node is never initialized, but of_node_put is NULL-safe, so it is also safe to call of_node_put in that case. Signed-off-by: Julia Lawall Acked-by: Alexandre TORGUE Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 082cd48db6a7..36942f5a6a53 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -351,6 +351,7 @@ void stmmac_remove_config_dt(struct platform_device *pdev, if (of_phy_is_fixed_link(np)) of_phy_deregister_fixed_link(np); of_node_put(plat->phy_node); + of_node_put(plat->mdio_node); } #else struct plat_stmmacenet_data * -- cgit v1.2.3 From d5ff72d9af73bc3cbaa3edb541333a851f8c7295 Mon Sep 17 00:00:00 2001 From: Lance Richardson Date: Mon, 16 Jan 2017 18:37:58 -0500 Subject: vxlan: fix byte order of vxlan-gpe port number vxlan->cfg.dst_port is in network byte order, so an htons() is needed here. Also reduced comment length to stay closer to 80 column width (still slightly over, however). Fixes: e1e5314de08b ("vxlan: implement GPE") Signed-off-by: Lance Richardson Acked-by: Jiri Benc Signed-off-by: David S. Miller --- drivers/net/vxlan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index ca7196c40060..8a79cfcac889 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2890,7 +2890,7 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev, memcpy(&vxlan->cfg, conf, sizeof(*conf)); if (!vxlan->cfg.dst_port) { if (conf->flags & VXLAN_F_GPE) - vxlan->cfg.dst_port = 4790; /* IANA assigned VXLAN-GPE port */ + vxlan->cfg.dst_port = htons(4790); /* IANA VXLAN-GPE port */ else vxlan->cfg.dst_port = default_port; } -- cgit v1.2.3 From 3fbfadce6012e7bb384b2e9ad47869d5177f7209 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Mon, 16 Jan 2017 22:17:29 -0800 Subject: bpf: Fix test_lru_sanity5() in test_lru_map.c test_lru_sanity5() fails when the number of online cpus is fewer than the number of possible cpus. It can be reproduced with qemu by using cmd args "--smp cpus=2,maxcpus=8". The problem is the loop in test_lru_sanity5() is testing 'i' which is incorrect. This patch: 1. Make sched_next_online() always return -1 if it cannot find a next cpu to schedule the process. 2. In test_lru_sanity5(), the parent process does sched_setaffinity() first (through sched_next_online()) and the forked process will inherit it according to the 'man sched_setaffinity'. Fixes: 5db58faf989f ("bpf: Add tests for the LRU bpf_htab") Reported-by: Daniel Borkmann Signed-off-by: Martin KaFai Lau Acked-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- tools/testing/selftests/bpf/test_lru_map.c | 53 +++++++++++++++--------------- 1 file changed, 27 insertions(+), 26 deletions(-) diff --git a/tools/testing/selftests/bpf/test_lru_map.c b/tools/testing/selftests/bpf/test_lru_map.c index b13fed534d76..9f7bd1915c21 100644 --- a/tools/testing/selftests/bpf/test_lru_map.c +++ b/tools/testing/selftests/bpf/test_lru_map.c @@ -67,21 +67,23 @@ static int map_equal(int lru_map, int expected) return map_subset(lru_map, expected) && map_subset(expected, lru_map); } -static int sched_next_online(int pid, int next_to_try) +static int sched_next_online(int pid, int *next_to_try) { cpu_set_t cpuset; + int next = *next_to_try; + int ret = -1; - if (next_to_try == nr_cpus) - return -1; - - while (next_to_try < nr_cpus) { + while (next < nr_cpus) { CPU_ZERO(&cpuset); - CPU_SET(next_to_try++, &cpuset); - if (!sched_setaffinity(pid, sizeof(cpuset), &cpuset)) + CPU_SET(next++, &cpuset); + if (!sched_setaffinity(pid, sizeof(cpuset), &cpuset)) { + ret = 0; break; + } } - return next_to_try; + *next_to_try = next; + return ret; } /* Size of the LRU amp is 2 @@ -96,11 +98,12 @@ static void test_lru_sanity0(int map_type, int map_flags) { unsigned long long key, value[nr_cpus]; int lru_map_fd, expected_map_fd; + int next_cpu = 0; printf("%s (map_type:%d map_flags:0x%X): ", __func__, map_type, map_flags); - assert(sched_next_online(0, 0) != -1); + assert(sched_next_online(0, &next_cpu) != -1); if (map_flags & BPF_F_NO_COMMON_LRU) lru_map_fd = create_map(map_type, map_flags, 2 * nr_cpus); @@ -183,6 +186,7 @@ static void test_lru_sanity1(int map_type, int map_flags, unsigned int tgt_free) int lru_map_fd, expected_map_fd; unsigned int batch_size; unsigned int map_size; + int next_cpu = 0; if (map_flags & BPF_F_NO_COMMON_LRU) /* Ther percpu lru list (i.e each cpu has its own LRU @@ -196,7 +200,7 @@ static void test_lru_sanity1(int map_type, int map_flags, unsigned int tgt_free) printf("%s (map_type:%d map_flags:0x%X): ", __func__, map_type, map_flags); - assert(sched_next_online(0, 0) != -1); + assert(sched_next_online(0, &next_cpu) != -1); batch_size = tgt_free / 2; assert(batch_size * 2 == tgt_free); @@ -262,6 +266,7 @@ static void test_lru_sanity2(int map_type, int map_flags, unsigned int tgt_free) int lru_map_fd, expected_map_fd; unsigned int batch_size; unsigned int map_size; + int next_cpu = 0; if (map_flags & BPF_F_NO_COMMON_LRU) /* Ther percpu lru list (i.e each cpu has its own LRU @@ -275,7 +280,7 @@ static void test_lru_sanity2(int map_type, int map_flags, unsigned int tgt_free) printf("%s (map_type:%d map_flags:0x%X): ", __func__, map_type, map_flags); - assert(sched_next_online(0, 0) != -1); + assert(sched_next_online(0, &next_cpu) != -1); batch_size = tgt_free / 2; assert(batch_size * 2 == tgt_free); @@ -370,11 +375,12 @@ static void test_lru_sanity3(int map_type, int map_flags, unsigned int tgt_free) int lru_map_fd, expected_map_fd; unsigned int batch_size; unsigned int map_size; + int next_cpu = 0; printf("%s (map_type:%d map_flags:0x%X): ", __func__, map_type, map_flags); - assert(sched_next_online(0, 0) != -1); + assert(sched_next_online(0, &next_cpu) != -1); batch_size = tgt_free / 2; assert(batch_size * 2 == tgt_free); @@ -430,11 +436,12 @@ static void test_lru_sanity4(int map_type, int map_flags, unsigned int tgt_free) int lru_map_fd, expected_map_fd; unsigned long long key, value[nr_cpus]; unsigned long long end_key; + int next_cpu = 0; printf("%s (map_type:%d map_flags:0x%X): ", __func__, map_type, map_flags); - assert(sched_next_online(0, 0) != -1); + assert(sched_next_online(0, &next_cpu) != -1); if (map_flags & BPF_F_NO_COMMON_LRU) lru_map_fd = create_map(map_type, map_flags, @@ -502,9 +509,8 @@ static void do_test_lru_sanity5(unsigned long long last_key, int map_fd) static void test_lru_sanity5(int map_type, int map_flags) { unsigned long long key, value[nr_cpus]; - int next_sched_cpu = 0; + int next_cpu = 0; int map_fd; - int i; if (map_flags & BPF_F_NO_COMMON_LRU) return; @@ -519,27 +525,20 @@ static void test_lru_sanity5(int map_type, int map_flags) key = 0; assert(!bpf_map_update(map_fd, &key, value, BPF_NOEXIST)); - for (i = 0; i < nr_cpus; i++) { + while (sched_next_online(0, &next_cpu) != -1) { pid_t pid; pid = fork(); if (pid == 0) { - next_sched_cpu = sched_next_online(0, next_sched_cpu); - if (next_sched_cpu != -1) - do_test_lru_sanity5(key, map_fd); + do_test_lru_sanity5(key, map_fd); exit(0); } else if (pid == -1) { - printf("couldn't spawn #%d process\n", i); + printf("couldn't spawn process to test key:%llu\n", + key); exit(1); } else { int status; - /* It is mostly redundant and just allow the parent - * process to update next_shced_cpu for the next child - * process - */ - next_sched_cpu = sched_next_online(pid, next_sched_cpu); - assert(waitpid(pid, &status, 0) == pid); assert(status == 0); key++; @@ -547,6 +546,8 @@ static void test_lru_sanity5(int map_type, int map_flags) } close(map_fd); + /* At least one key should be tested */ + assert(key > 0); printf("Pass\n"); } -- cgit v1.2.3 From 93b43fd137cd8865adf9978ab9870a344365d3af Mon Sep 17 00:00:00 2001 From: Alvaro G. M Date: Tue, 17 Jan 2017 09:08:16 +0100 Subject: net: phy: dp83848: add DP83620 PHY support This PHY with fiber support is register compatible with DP83848, so add support for it. Signed-off-by: Alvaro Gamez Machado Signed-off-by: David S. Miller --- drivers/net/phy/dp83848.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/phy/dp83848.c b/drivers/net/phy/dp83848.c index 800b39f06279..a10d0e7fc5f7 100644 --- a/drivers/net/phy/dp83848.c +++ b/drivers/net/phy/dp83848.c @@ -17,6 +17,7 @@ #include #define TI_DP83848C_PHY_ID 0x20005ca0 +#define TI_DP83620_PHY_ID 0x20005ce0 #define NS_DP83848C_PHY_ID 0x20005c90 #define TLK10X_PHY_ID 0x2000a210 #define TI_DP83822_PHY_ID 0x2000a240 @@ -77,6 +78,7 @@ static int dp83848_config_intr(struct phy_device *phydev) static struct mdio_device_id __maybe_unused dp83848_tbl[] = { { TI_DP83848C_PHY_ID, 0xfffffff0 }, { NS_DP83848C_PHY_ID, 0xfffffff0 }, + { TI_DP83620_PHY_ID, 0xfffffff0 }, { TLK10X_PHY_ID, 0xfffffff0 }, { TI_DP83822_PHY_ID, 0xfffffff0 }, { } @@ -106,6 +108,7 @@ MODULE_DEVICE_TABLE(mdio, dp83848_tbl); static struct phy_driver dp83848_driver[] = { DP83848_PHY_DRIVER(TI_DP83848C_PHY_ID, "TI DP83848C 10/100 Mbps PHY"), DP83848_PHY_DRIVER(NS_DP83848C_PHY_ID, "NS DP83848C 10/100 Mbps PHY"), + DP83848_PHY_DRIVER(TI_DP83620_PHY_ID, "TI DP83620 10/100 Mbps PHY"), DP83848_PHY_DRIVER(TLK10X_PHY_ID, "TI TLK10X 10/100 Mbps PHY"), DP83848_PHY_DRIVER(TI_DP83822_PHY_ID, "TI DP83822 10/100 Mbps PHY"), }; -- cgit v1.2.3 From c97c52be78b8463ac5407f1cf1f22f8f6cf93a37 Mon Sep 17 00:00:00 2001 From: Einar Jón Date: Fri, 12 Aug 2016 13:50:41 +0200 Subject: can: c_can_pci: fix null-pointer-deref in c_can_start() - set device pointer The priv->device pointer for c_can_pci is never set, but it is used without a NULL check in c_can_start(). Setting it in c_can_pci_probe() like c_can_plat_probe() prevents c_can_pci.ko from crashing, with and without CONFIG_PM. This might also cause the pm_runtime_*() functions in c_can.c to actually be executed for c_can_pci devices - they are the only other place where priv->device is used, but they all contain a null check. Signed-off-by: Einar Jón Cc: linux-stable Signed-off-by: Marc Kleine-Budde --- drivers/net/can/c_can/c_can_pci.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/can/c_can/c_can_pci.c b/drivers/net/can/c_can/c_can_pci.c index 7be393c96b1a..cf7c18947189 100644 --- a/drivers/net/can/c_can/c_can_pci.c +++ b/drivers/net/can/c_can/c_can_pci.c @@ -161,6 +161,7 @@ static int c_can_pci_probe(struct pci_dev *pdev, dev->irq = pdev->irq; priv->base = addr; + priv->device = &pdev->dev; if (!c_can_pci_data->freq) { dev_err(&pdev->dev, "no clock frequency defined\n"); -- cgit v1.2.3 From befa60113ce7ea270cb51eada28443ca2756f480 Mon Sep 17 00:00:00 2001 From: Yegor Yefremov Date: Wed, 18 Jan 2017 11:35:57 +0100 Subject: can: ti_hecc: add missing prepare and unprepare of the clock In order to make the driver work with the common clock framework, this patch converts the clk_enable()/clk_disable() to clk_prepare_enable()/clk_disable_unprepare(). Also add error checking for clk_prepare_enable(). Signed-off-by: Yegor Yefremov Cc: linux-stable Signed-off-by: Marc Kleine-Budde --- drivers/net/can/ti_hecc.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/drivers/net/can/ti_hecc.c b/drivers/net/can/ti_hecc.c index 680d1ff07a55..6749b1829469 100644 --- a/drivers/net/can/ti_hecc.c +++ b/drivers/net/can/ti_hecc.c @@ -948,7 +948,12 @@ static int ti_hecc_probe(struct platform_device *pdev) netif_napi_add(ndev, &priv->napi, ti_hecc_rx_poll, HECC_DEF_NAPI_WEIGHT); - clk_enable(priv->clk); + err = clk_prepare_enable(priv->clk); + if (err) { + dev_err(&pdev->dev, "clk_prepare_enable() failed\n"); + goto probe_exit_clk; + } + err = register_candev(ndev); if (err) { dev_err(&pdev->dev, "register_candev() failed\n"); @@ -981,7 +986,7 @@ static int ti_hecc_remove(struct platform_device *pdev) struct ti_hecc_priv *priv = netdev_priv(ndev); unregister_candev(ndev); - clk_disable(priv->clk); + clk_disable_unprepare(priv->clk); clk_put(priv->clk); res = platform_get_resource(pdev, IORESOURCE_MEM, 0); iounmap(priv->base); @@ -1006,7 +1011,7 @@ static int ti_hecc_suspend(struct platform_device *pdev, pm_message_t state) hecc_set_bit(priv, HECC_CANMC, HECC_CANMC_PDR); priv->can.state = CAN_STATE_SLEEPING; - clk_disable(priv->clk); + clk_disable_unprepare(priv->clk); return 0; } @@ -1015,8 +1020,11 @@ static int ti_hecc_resume(struct platform_device *pdev) { struct net_device *dev = platform_get_drvdata(pdev); struct ti_hecc_priv *priv = netdev_priv(dev); + int err; - clk_enable(priv->clk); + err = clk_prepare_enable(priv->clk); + if (err) + return err; hecc_clear_bit(priv, HECC_CANMC, HECC_CANMC_PDR); priv->can.state = CAN_STATE_ERROR_ACTIVE; -- cgit v1.2.3 From b98acbff9a05b371c5f0ca6e44a3af8ce9274379 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Wed, 11 Jan 2017 15:36:20 +0000 Subject: regulator: twl6030: fix range comparison, allowing vsel = 59 The range min_uV > 1350000 && min_uV <= 150000 is never reachable because of a typo in the previous range check and hence vsel = 59 is never reached. Fix the previous range check to enable the vsel = 59 setting. Fixes CoverityScan CID#728454 ("Logially dead code") Signed-off-by: Colin Ian King Signed-off-by: Mark Brown --- drivers/regulator/twl6030-regulator.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/regulator/twl6030-regulator.c b/drivers/regulator/twl6030-regulator.c index 4864b9d742c0..716191046a70 100644 --- a/drivers/regulator/twl6030-regulator.c +++ b/drivers/regulator/twl6030-regulator.c @@ -452,7 +452,7 @@ static int twl6030smps_map_voltage(struct regulator_dev *rdev, int min_uV, vsel = 62; else if ((min_uV > 1800000) && (min_uV <= 1900000)) vsel = 61; - else if ((min_uV > 1350000) && (min_uV <= 1800000)) + else if ((min_uV > 1500000) && (min_uV <= 1800000)) vsel = 60; else if ((min_uV > 1350000) && (min_uV <= 1500000)) vsel = 59; -- cgit v1.2.3 From 31a86d137219373c3222ca5f4f912e9a4d8065bb Mon Sep 17 00:00:00 2001 From: Eran Ben Elisha Date: Tue, 17 Jan 2017 19:19:17 +0200 Subject: net: ethtool: Initialize buffer when querying device channel settings Ethtool channels respond struct was uninitialized when querying device channel boundaries settings. As a result, unreported fields by the driver hold garbage. This may cause sending unsupported params to driver. Fixes: 8bf368620486 ('ethtool: ensure channel counts are within bounds ...') Signed-off-by: Eran Ben Elisha Signed-off-by: Tariq Toukan CC: John W. Linville Signed-off-by: David S. Miller --- net/core/ethtool.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/ethtool.c b/net/core/ethtool.c index e23766c7e3ba..236a21e3c878 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -1712,7 +1712,7 @@ static noinline_for_stack int ethtool_get_channels(struct net_device *dev, static noinline_for_stack int ethtool_set_channels(struct net_device *dev, void __user *useraddr) { - struct ethtool_channels channels, max; + struct ethtool_channels channels, max = { .cmd = ETHTOOL_GCHANNELS }; u32 max_rx_in_use = 0; if (!dev->ethtool_ops->set_channels || !dev->ethtool_ops->get_channels) -- cgit v1.2.3 From e91ef71dfe834e11b57411f1715cd2e2bb4401f1 Mon Sep 17 00:00:00 2001 From: Eran Ben Elisha Date: Tue, 17 Jan 2017 19:19:18 +0200 Subject: net/mlx4_en: Remove unnecessary checks when setting num channels Boundaries checks for the number of RX, TX, other and combined channels should be checked by the caller and not in the driver. In addition, remove wrong memset on get channels as it overrides the cmd field in the requester struct. Signed-off-by: Eran Ben Elisha Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index d9c9f86a30df..d5a9372ed84d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -1732,8 +1732,6 @@ static void mlx4_en_get_channels(struct net_device *dev, { struct mlx4_en_priv *priv = netdev_priv(dev); - memset(channel, 0, sizeof(*channel)); - channel->max_rx = MAX_RX_RINGS; channel->max_tx = MLX4_EN_MAX_TX_RING_P_UP; @@ -1752,10 +1750,7 @@ static int mlx4_en_set_channels(struct net_device *dev, int xdp_count; int err = 0; - if (channel->other_count || channel->combined_count || - channel->tx_count > MLX4_EN_MAX_TX_RING_P_UP || - channel->rx_count > MAX_RX_RINGS || - !channel->tx_count || !channel->rx_count) + if (!channel->tx_count || !channel->rx_count) return -EINVAL; tmp = kzalloc(sizeof(*tmp), GFP_KERNEL); -- cgit v1.2.3 From 639e9e94160e59469305fc2c5e6f9c2733744958 Mon Sep 17 00:00:00 2001 From: Eran Ben Elisha Date: Tue, 17 Jan 2017 19:19:19 +0200 Subject: net/mlx5e: Remove unnecessary checks when setting num channels Boundaries checks for the number of RX and TX should be checked by the caller and not in the driver. Signed-off-by: Eran Ben Elisha Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 33a399a8b5d5..b1b9eb6ee135 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -554,16 +554,6 @@ static int mlx5e_set_channels(struct net_device *dev, __func__); return -EINVAL; } - if (ch->rx_count || ch->tx_count) { - netdev_info(dev, "%s: separate rx/tx count not supported\n", - __func__); - return -EINVAL; - } - if (count > ncv) { - netdev_info(dev, "%s: count (%d) > max (%d)\n", - __func__, count, ncv); - return -EINVAL; - } if (priv->params.num_channels == count) return 0; -- cgit v1.2.3 From 3fd0b634de7d6b9a85f34a4cf9d8afc1df465cc9 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 13 Jan 2017 16:41:03 +0100 Subject: netfilter: ipt_CLUSTERIP: fix build error without procfs We can't access c->pde if CONFIG_PROC_FS is disabled: net/ipv4/netfilter/ipt_CLUSTERIP.c: In function 'clusterip_config_find_get': net/ipv4/netfilter/ipt_CLUSTERIP.c:147:9: error: 'struct clusterip_config' has no member named 'pde' This moves the check inside of another #ifdef. Fixes: 6c5d5cfbe3c5 ("netfilter: ipt_CLUSTERIP: check duplicate config when initializing") Signed-off-by: Arnd Bergmann Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/ipt_CLUSTERIP.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index a6b8c1a4102b..0a783cd73faf 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -144,7 +144,12 @@ clusterip_config_find_get(struct net *net, __be32 clusterip, int entry) rcu_read_lock_bh(); c = __clusterip_config_find(net, clusterip); if (c) { - if (!c->pde || unlikely(!atomic_inc_not_zero(&c->refcount))) +#ifdef CONFIG_PROC_FS + if (!c->pde) + c = NULL; + else +#endif + if (unlikely(!atomic_inc_not_zero(&c->refcount))) c = NULL; else if (entry) atomic_inc(&c->entries); -- cgit v1.2.3 From 9a6cdf52b85ea5fb21d2bb31e4a7bc61b79923a7 Mon Sep 17 00:00:00 2001 From: Igor Druzhinin Date: Tue, 17 Jan 2017 20:49:37 +0000 Subject: xen-netback: fix memory leaks on XenBus disconnect Eliminate memory leaks introduced several years ago by cleaning the queue resources which are allocated on XenBus connection event. Namely, queue structure array and pages used for IO rings. Signed-off-by: Igor Druzhinin Reviewed-by: Paul Durrant Acked-by: Wei Liu Signed-off-by: David S. Miller --- drivers/net/xen-netback/xenbus.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c index 3124eaec9427..34e4af9272ea 100644 --- a/drivers/net/xen-netback/xenbus.c +++ b/drivers/net/xen-netback/xenbus.c @@ -493,11 +493,20 @@ static int backend_create_xenvif(struct backend_info *be) static void backend_disconnect(struct backend_info *be) { if (be->vif) { + unsigned int queue_index; + xen_unregister_watchers(be->vif); #ifdef CONFIG_DEBUG_FS xenvif_debugfs_delif(be->vif); #endif /* CONFIG_DEBUG_FS */ xenvif_disconnect_data(be->vif); + for (queue_index = 0; queue_index < be->vif->num_queues; ++queue_index) + xenvif_deinit_queue(&be->vif->queues[queue_index]); + + vfree(be->vif->queues); + be->vif->num_queues = 0; + be->vif->queues = NULL; + xenvif_disconnect_ctrl(be->vif); } } @@ -1034,6 +1043,8 @@ static void connect(struct backend_info *be) err: if (be->vif->num_queues > 0) xenvif_disconnect_data(be->vif); /* Clean up existing queues */ + for (queue_index = 0; queue_index < be->vif->num_queues; ++queue_index) + xenvif_deinit_queue(&be->vif->queues[queue_index]); vfree(be->vif->queues); be->vif->queues = NULL; be->vif->num_queues = 0; -- cgit v1.2.3 From f16f1df65f1cf139ff9e9f84661e6573d6bb27fc Mon Sep 17 00:00:00 2001 From: Igor Druzhinin Date: Tue, 17 Jan 2017 20:49:38 +0000 Subject: xen-netback: protect resource cleaning on XenBus disconnect vif->lock is used to protect statistics gathering agents from using the queue structure during cleaning. Signed-off-by: Igor Druzhinin Acked-by: Wei Liu Reviewed-by: Paul Durrant Signed-off-by: David S. Miller --- drivers/net/xen-netback/interface.c | 6 ++++-- drivers/net/xen-netback/xenbus.c | 2 ++ 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c index e30ffd29b7e9..579521327b03 100644 --- a/drivers/net/xen-netback/interface.c +++ b/drivers/net/xen-netback/interface.c @@ -221,18 +221,18 @@ static struct net_device_stats *xenvif_get_stats(struct net_device *dev) { struct xenvif *vif = netdev_priv(dev); struct xenvif_queue *queue = NULL; - unsigned int num_queues = vif->num_queues; unsigned long rx_bytes = 0; unsigned long rx_packets = 0; unsigned long tx_bytes = 0; unsigned long tx_packets = 0; unsigned int index; + spin_lock(&vif->lock); if (vif->queues == NULL) goto out; /* Aggregate tx and rx stats from each queue */ - for (index = 0; index < num_queues; ++index) { + for (index = 0; index < vif->num_queues; ++index) { queue = &vif->queues[index]; rx_bytes += queue->stats.rx_bytes; rx_packets += queue->stats.rx_packets; @@ -241,6 +241,8 @@ static struct net_device_stats *xenvif_get_stats(struct net_device *dev) } out: + spin_unlock(&vif->lock); + vif->dev->stats.rx_bytes = rx_bytes; vif->dev->stats.rx_packets = rx_packets; vif->dev->stats.tx_bytes = tx_bytes; diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c index 34e4af9272ea..85b742e1c42f 100644 --- a/drivers/net/xen-netback/xenbus.c +++ b/drivers/net/xen-netback/xenbus.c @@ -503,9 +503,11 @@ static void backend_disconnect(struct backend_info *be) for (queue_index = 0; queue_index < be->vif->num_queues; ++queue_index) xenvif_deinit_queue(&be->vif->queues[queue_index]); + spin_lock(&be->vif->lock); vfree(be->vif->queues); be->vif->num_queues = 0; be->vif->queues = NULL; + spin_unlock(&be->vif->lock); xenvif_disconnect_ctrl(be->vif); } -- cgit v1.2.3 From 7be2c82cfd5d28d7adb66821a992604eb6dd112e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 18 Jan 2017 12:12:17 -0800 Subject: net: fix harmonize_features() vs NETIF_F_HIGHDMA Ashizuka reported a highmem oddity and sent a patch for freescale fec driver. But the problem root cause is that core networking stack must ensure no skb with highmem fragment is ever sent through a device that does not assert NETIF_F_HIGHDMA in its features. We need to call illegal_highdma() from harmonize_features() regardless of CSUM checks. Fixes: ec5f06156423 ("net: Kill link between CSUM and SG features.") Signed-off-by: Eric Dumazet Cc: Pravin Shelar Reported-by: "Ashizuka, Yuusuke" Signed-off-by: David S. Miller --- net/core/dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index 07b307b0b414..7f218e095361 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2795,9 +2795,9 @@ static netdev_features_t harmonize_features(struct sk_buff *skb, if (skb->ip_summed != CHECKSUM_NONE && !can_checksum_protocol(features, type)) { features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); - } else if (illegal_highdma(skb->dev, skb)) { - features &= ~NETIF_F_SG; } + if (illegal_highdma(skb->dev, skb)) + features &= ~NETIF_F_SG; return features; } -- cgit v1.2.3 From cd33b3e0da43522ff8e8f2b2b71d3d08298512b0 Mon Sep 17 00:00:00 2001 From: Daniel Gonzalez Cabanelas Date: Tue, 17 Jan 2017 16:26:55 -0800 Subject: net: phy: bcm63xx: Utilize correct config_intr function Commit a1cba5613edf ("net: phy: Add Broadcom phy library for common interfaces") make the BCM63xx PHY driver utilize bcm_phy_config_intr() which would appear to do the right thing, except that it does not write to the MII_BCM63XX_IR register but to MII_BCM54XX_ECR which is different. This would be causing invalid link parameters and events from being generated by the PHY interrupt. Fixes: a1cba5613edf ("net: phy: Add Broadcom phy library for common interfaces") Signed-off-by: Daniel Gonzalez Cabanelas Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/bcm63xx.c | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/bcm63xx.c b/drivers/net/phy/bcm63xx.c index e741bf614c4e..b0492ef2cdaa 100644 --- a/drivers/net/phy/bcm63xx.c +++ b/drivers/net/phy/bcm63xx.c @@ -21,6 +21,23 @@ MODULE_DESCRIPTION("Broadcom 63xx internal PHY driver"); MODULE_AUTHOR("Maxime Bizon "); MODULE_LICENSE("GPL"); +static int bcm63xx_config_intr(struct phy_device *phydev) +{ + int reg, err; + + reg = phy_read(phydev, MII_BCM63XX_IR); + if (reg < 0) + return reg; + + if (phydev->interrupts == PHY_INTERRUPT_ENABLED) + reg &= ~MII_BCM63XX_IR_GMASK; + else + reg |= MII_BCM63XX_IR_GMASK; + + err = phy_write(phydev, MII_BCM63XX_IR, reg); + return err; +} + static int bcm63xx_config_init(struct phy_device *phydev) { int reg, err; @@ -55,7 +72,7 @@ static struct phy_driver bcm63xx_driver[] = { .config_aneg = genphy_config_aneg, .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, - .config_intr = bcm_phy_config_intr, + .config_intr = bcm63xx_config_intr, }, { /* same phy as above, with just a different OUI */ .phy_id = 0x002bdc00, @@ -67,7 +84,7 @@ static struct phy_driver bcm63xx_driver[] = { .config_aneg = genphy_config_aneg, .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, - .config_intr = bcm_phy_config_intr, + .config_intr = bcm63xx_config_intr, } }; module_phy_driver(bcm63xx_driver); -- cgit v1.2.3 From 719ca8111402aa6157bd83a3c966d184db0d8956 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Tue, 17 Jan 2017 22:07:19 -0500 Subject: bnxt_en: Fix "uninitialized variable" bug in TPA code path. In the TPA GRO code path, initialize the tcp_opt_len variable to 0 so that it will be correct for packets without TCP timestamps. The bug caused the SKB fields to be incorrectly set up for packets without TCP timestamps, leading to these packets being rejected by the stack. Reported-by: Andy Gospodarek Acked-by: Andy Gospodarek Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 9608cb49a11c..53e686fdf2ce 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -1099,7 +1099,7 @@ static struct sk_buff *bnxt_gro_func_5730x(struct bnxt_tpa_info *tpa_info, { #ifdef CONFIG_INET struct tcphdr *th; - int len, nw_off, tcp_opt_len; + int len, nw_off, tcp_opt_len = 0; if (tcp_ts) tcp_opt_len = 12; -- cgit v1.2.3 From 9ed59592e3e379b2e9557dc1d9e9ec8fcbb33f16 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Tue, 17 Jan 2017 14:57:36 -0800 Subject: lwtunnel: fix autoload of lwt modules Trying to add an mpls encap route when the MPLS modules are not loaded hangs. For example: CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m CONFIG_MPLS_IPTUNNEL=m $ ip route add 10.10.10.10/32 encap mpls 100 via inet 10.100.1.2 The ip command hangs: root 880 826 0 21:25 pts/0 00:00:00 ip route add 10.10.10.10/32 encap mpls 100 via inet 10.100.1.2 $ cat /proc/880/stack [] call_usermodehelper_exec+0xd6/0x134 [] __request_module+0x27b/0x30a [] lwtunnel_build_state+0xe4/0x178 [] fib_create_info+0x47f/0xdd4 [] fib_table_insert+0x90/0x41f [] inet_rtm_newroute+0x4b/0x52 ... modprobe is trying to load rtnl-lwt-MPLS: root 881 5 0 21:25 ? 00:00:00 /sbin/modprobe -q -- rtnl-lwt-MPLS and it hangs after loading mpls_router: $ cat /proc/881/stack [] rtnl_lock+0x12/0x14 [] register_netdevice_notifier+0x16/0x179 [] mpls_init+0x25/0x1000 [mpls_router] [] do_one_initcall+0x8e/0x13f [] do_init_module+0x5a/0x1e5 [] load_module+0x13bd/0x17d6 ... The problem is that lwtunnel_build_state is called with rtnl lock held preventing mpls_init from registering. Given the potential references held by the time lwtunnel_build_state it can not drop the rtnl lock to the load module. So, extract the module loading code from lwtunnel_build_state into a new function to validate the encap type. The new function is called while converting the user request into a fib_config which is well before any table, device or fib entries are examined. Fixes: 745041e2aaf1 ("lwtunnel: autoload of lwt modules") Signed-off-by: David Ahern Signed-off-by: David S. Miller --- include/net/lwtunnel.h | 11 +++++++++ net/core/lwtunnel.c | 62 ++++++++++++++++++++++++++++++++++++++++++++----- net/ipv4/fib_frontend.c | 8 +++++++ net/ipv6/route.c | 12 +++++++++- 4 files changed, 86 insertions(+), 7 deletions(-) diff --git a/include/net/lwtunnel.h b/include/net/lwtunnel.h index d4c1c75b8862..0b585f1fd340 100644 --- a/include/net/lwtunnel.h +++ b/include/net/lwtunnel.h @@ -105,6 +105,8 @@ int lwtunnel_encap_add_ops(const struct lwtunnel_encap_ops *op, unsigned int num); int lwtunnel_encap_del_ops(const struct lwtunnel_encap_ops *op, unsigned int num); +int lwtunnel_valid_encap_type(u16 encap_type); +int lwtunnel_valid_encap_type_attr(struct nlattr *attr, int len); int lwtunnel_build_state(struct net_device *dev, u16 encap_type, struct nlattr *encap, unsigned int family, const void *cfg, @@ -168,6 +170,15 @@ static inline int lwtunnel_encap_del_ops(const struct lwtunnel_encap_ops *op, return -EOPNOTSUPP; } +static inline int lwtunnel_valid_encap_type(u16 encap_type) +{ + return -EOPNOTSUPP; +} +static inline int lwtunnel_valid_encap_type_attr(struct nlattr *attr, int len) +{ + return -EOPNOTSUPP; +} + static inline int lwtunnel_build_state(struct net_device *dev, u16 encap_type, struct nlattr *encap, unsigned int family, const void *cfg, diff --git a/net/core/lwtunnel.c b/net/core/lwtunnel.c index a5d4e866ce88..47b1dd65947b 100644 --- a/net/core/lwtunnel.c +++ b/net/core/lwtunnel.c @@ -26,6 +26,7 @@ #include #include #include +#include #ifdef CONFIG_MODULES @@ -114,25 +115,74 @@ int lwtunnel_build_state(struct net_device *dev, u16 encap_type, ret = -EOPNOTSUPP; rcu_read_lock(); ops = rcu_dereference(lwtun_encaps[encap_type]); + if (likely(ops && ops->build_state)) + ret = ops->build_state(dev, encap, family, cfg, lws); + rcu_read_unlock(); + + return ret; +} +EXPORT_SYMBOL(lwtunnel_build_state); + +int lwtunnel_valid_encap_type(u16 encap_type) +{ + const struct lwtunnel_encap_ops *ops; + int ret = -EINVAL; + + if (encap_type == LWTUNNEL_ENCAP_NONE || + encap_type > LWTUNNEL_ENCAP_MAX) + return ret; + + rcu_read_lock(); + ops = rcu_dereference(lwtun_encaps[encap_type]); + rcu_read_unlock(); #ifdef CONFIG_MODULES if (!ops) { const char *encap_type_str = lwtunnel_encap_str(encap_type); if (encap_type_str) { - rcu_read_unlock(); + __rtnl_unlock(); request_module("rtnl-lwt-%s", encap_type_str); + rtnl_lock(); + rcu_read_lock(); ops = rcu_dereference(lwtun_encaps[encap_type]); + rcu_read_unlock(); } } #endif - if (likely(ops && ops->build_state)) - ret = ops->build_state(dev, encap, family, cfg, lws); - rcu_read_unlock(); + return ops ? 0 : -EOPNOTSUPP; +} +EXPORT_SYMBOL(lwtunnel_valid_encap_type); - return ret; +int lwtunnel_valid_encap_type_attr(struct nlattr *attr, int remaining) +{ + struct rtnexthop *rtnh = (struct rtnexthop *)attr; + struct nlattr *nla_entype; + struct nlattr *attrs; + struct nlattr *nla; + u16 encap_type; + int attrlen; + + while (rtnh_ok(rtnh, remaining)) { + attrlen = rtnh_attrlen(rtnh); + if (attrlen > 0) { + attrs = rtnh_attrs(rtnh); + nla = nla_find(attrs, attrlen, RTA_ENCAP); + nla_entype = nla_find(attrs, attrlen, RTA_ENCAP_TYPE); + + if (nla_entype) { + encap_type = nla_get_u16(nla_entype); + + if (lwtunnel_valid_encap_type(encap_type) != 0) + return -EOPNOTSUPP; + } + } + rtnh = rtnh_next(rtnh, &remaining); + } + + return 0; } -EXPORT_SYMBOL(lwtunnel_build_state); +EXPORT_SYMBOL(lwtunnel_valid_encap_type_attr); void lwtstate_free(struct lwtunnel_state *lws) { diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index eae0332b0e8c..7db2ad2e82d3 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -46,6 +46,7 @@ #include #include #include +#include #include #ifndef CONFIG_IP_MULTIPLE_TABLES @@ -677,6 +678,10 @@ static int rtm_to_fib_config(struct net *net, struct sk_buff *skb, cfg->fc_mx_len = nla_len(attr); break; case RTA_MULTIPATH: + err = lwtunnel_valid_encap_type_attr(nla_data(attr), + nla_len(attr)); + if (err < 0) + goto errout; cfg->fc_mp = nla_data(attr); cfg->fc_mp_len = nla_len(attr); break; @@ -691,6 +696,9 @@ static int rtm_to_fib_config(struct net *net, struct sk_buff *skb, break; case RTA_ENCAP_TYPE: cfg->fc_encap_type = nla_get_u16(attr); + err = lwtunnel_valid_encap_type(cfg->fc_encap_type); + if (err < 0) + goto errout; break; } } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 4f6b067c8753..7ea85370c11c 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2896,6 +2896,11 @@ static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh, if (tb[RTA_MULTIPATH]) { cfg->fc_mp = nla_data(tb[RTA_MULTIPATH]); cfg->fc_mp_len = nla_len(tb[RTA_MULTIPATH]); + + err = lwtunnel_valid_encap_type_attr(cfg->fc_mp, + cfg->fc_mp_len); + if (err < 0) + goto errout; } if (tb[RTA_PREF]) { @@ -2909,9 +2914,14 @@ static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh, if (tb[RTA_ENCAP]) cfg->fc_encap = tb[RTA_ENCAP]; - if (tb[RTA_ENCAP_TYPE]) + if (tb[RTA_ENCAP_TYPE]) { cfg->fc_encap_type = nla_get_u16(tb[RTA_ENCAP_TYPE]); + err = lwtunnel_valid_encap_type(cfg->fc_encap_type); + if (err < 0) + goto errout; + } + if (tb[RTA_EXPIRES]) { unsigned long timeout = addrconf_timeout_fixup(nla_get_u32(tb[RTA_EXPIRES]), HZ); -- cgit v1.2.3 From d407bd25a204bd66b7346dde24bd3d37ef0e0b05 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 18 Jan 2017 15:14:17 +0100 Subject: bpf: don't trigger OOM killer under pressure with map alloc This patch adds two helpers, bpf_map_area_alloc() and bpf_map_area_free(), that are to be used for map allocations. Using kmalloc() for very large allocations can cause excessive work within the page allocator, so i) fall back earlier to vmalloc() when the attempt is considered costly anyway, and even more importantly ii) don't trigger OOM killer with any of the allocators. Since this is based on a user space request, for example, when creating maps with element pre-allocation, we really want such requests to fail instead of killing other user space processes. Also, don't spam the kernel log with warnings should any of the allocations fail under pressure. Given that, we can make backend selection in bpf_map_area_alloc() generic, and convert all maps over to use this API for spots with potentially large allocation requests. Note, replacing the one kmalloc_array() is fine as overflow checks happen earlier in htab_map_alloc(), since it must also protect the multiplication for vmalloc() should kmalloc_array() fail. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 2 ++ kernel/bpf/arraymap.c | 18 +++++++----------- kernel/bpf/hashtab.c | 22 +++++++++------------- kernel/bpf/stackmap.c | 20 ++++++++------------ kernel/bpf/syscall.c | 26 ++++++++++++++++++++++++++ 5 files changed, 52 insertions(+), 36 deletions(-) diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 05cf951df3fe..3ed1f3b1d594 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -247,6 +247,8 @@ struct bpf_map * __must_check bpf_map_inc(struct bpf_map *map, bool uref); void bpf_map_put_with_uref(struct bpf_map *map); void bpf_map_put(struct bpf_map *map); int bpf_map_precharge_memlock(u32 pages); +void *bpf_map_area_alloc(size_t size); +void bpf_map_area_free(void *base); extern int sysctl_unprivileged_bpf_disabled; diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 229a5d5df977..3d55d95dcf49 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -11,7 +11,6 @@ */ #include #include -#include #include #include #include @@ -74,14 +73,10 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) if (array_size >= U32_MAX - PAGE_SIZE) return ERR_PTR(-ENOMEM); - /* allocate all map elements and zero-initialize them */ - array = kzalloc(array_size, GFP_USER | __GFP_NOWARN); - if (!array) { - array = vzalloc(array_size); - if (!array) - return ERR_PTR(-ENOMEM); - } + array = bpf_map_area_alloc(array_size); + if (!array) + return ERR_PTR(-ENOMEM); /* copy mandatory map attributes */ array->map.map_type = attr->map_type; @@ -97,7 +92,7 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) if (array_size >= U32_MAX - PAGE_SIZE || elem_size > PCPU_MIN_UNIT_SIZE || bpf_array_alloc_percpu(array)) { - kvfree(array); + bpf_map_area_free(array); return ERR_PTR(-ENOMEM); } out: @@ -262,7 +257,7 @@ static void array_map_free(struct bpf_map *map) if (array->map.map_type == BPF_MAP_TYPE_PERCPU_ARRAY) bpf_array_free_percpu(array); - kvfree(array); + bpf_map_area_free(array); } static const struct bpf_map_ops array_ops = { @@ -319,7 +314,8 @@ static void fd_array_map_free(struct bpf_map *map) /* make sure it's empty */ for (i = 0; i < array->map.max_entries; i++) BUG_ON(array->ptrs[i] != NULL); - kvfree(array); + + bpf_map_area_free(array); } static void *fd_array_map_lookup_elem(struct bpf_map *map, void *key) diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c index 3f2bb58952d8..a753bbe7df0a 100644 --- a/kernel/bpf/hashtab.c +++ b/kernel/bpf/hashtab.c @@ -13,7 +13,6 @@ #include #include #include -#include #include "percpu_freelist.h" #include "bpf_lru_list.h" @@ -103,7 +102,7 @@ static void htab_free_elems(struct bpf_htab *htab) free_percpu(pptr); } free_elems: - vfree(htab->elems); + bpf_map_area_free(htab->elems); } static struct htab_elem *prealloc_lru_pop(struct bpf_htab *htab, void *key, @@ -125,7 +124,8 @@ static int prealloc_init(struct bpf_htab *htab) { int err = -ENOMEM, i; - htab->elems = vzalloc(htab->elem_size * htab->map.max_entries); + htab->elems = bpf_map_area_alloc(htab->elem_size * + htab->map.max_entries); if (!htab->elems) return -ENOMEM; @@ -320,14 +320,10 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr) goto free_htab; err = -ENOMEM; - htab->buckets = kmalloc_array(htab->n_buckets, sizeof(struct bucket), - GFP_USER | __GFP_NOWARN); - - if (!htab->buckets) { - htab->buckets = vmalloc(htab->n_buckets * sizeof(struct bucket)); - if (!htab->buckets) - goto free_htab; - } + htab->buckets = bpf_map_area_alloc(htab->n_buckets * + sizeof(struct bucket)); + if (!htab->buckets) + goto free_htab; for (i = 0; i < htab->n_buckets; i++) { INIT_HLIST_HEAD(&htab->buckets[i].head); @@ -354,7 +350,7 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr) free_extra_elems: free_percpu(htab->extra_elems); free_buckets: - kvfree(htab->buckets); + bpf_map_area_free(htab->buckets); free_htab: kfree(htab); return ERR_PTR(err); @@ -1014,7 +1010,7 @@ static void htab_map_free(struct bpf_map *map) prealloc_destroy(htab); free_percpu(htab->extra_elems); - kvfree(htab->buckets); + bpf_map_area_free(htab->buckets); kfree(htab); } diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c index 732ae16d12b7..be8519148c25 100644 --- a/kernel/bpf/stackmap.c +++ b/kernel/bpf/stackmap.c @@ -7,7 +7,6 @@ #include #include #include -#include #include #include #include "percpu_freelist.h" @@ -32,7 +31,7 @@ static int prealloc_elems_and_freelist(struct bpf_stack_map *smap) u32 elem_size = sizeof(struct stack_map_bucket) + smap->map.value_size; int err; - smap->elems = vzalloc(elem_size * smap->map.max_entries); + smap->elems = bpf_map_area_alloc(elem_size * smap->map.max_entries); if (!smap->elems) return -ENOMEM; @@ -45,7 +44,7 @@ static int prealloc_elems_and_freelist(struct bpf_stack_map *smap) return 0; free_elems: - vfree(smap->elems); + bpf_map_area_free(smap->elems); return err; } @@ -76,12 +75,9 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr) if (cost >= U32_MAX - PAGE_SIZE) return ERR_PTR(-E2BIG); - smap = kzalloc(cost, GFP_USER | __GFP_NOWARN); - if (!smap) { - smap = vzalloc(cost); - if (!smap) - return ERR_PTR(-ENOMEM); - } + smap = bpf_map_area_alloc(cost); + if (!smap) + return ERR_PTR(-ENOMEM); err = -E2BIG; cost += n_buckets * (value_size + sizeof(struct stack_map_bucket)); @@ -112,7 +108,7 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr) put_buffers: put_callchain_buffers(); free_smap: - kvfree(smap); + bpf_map_area_free(smap); return ERR_PTR(err); } @@ -262,9 +258,9 @@ static void stack_map_free(struct bpf_map *map) /* wait for bpf programs to complete before freeing stack map */ synchronize_rcu(); - vfree(smap->elems); + bpf_map_area_free(smap->elems); pcpu_freelist_destroy(&smap->freelist); - kvfree(smap); + bpf_map_area_free(smap); put_callchain_buffers(); } diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 1d6b29e4e2c3..19b6129eab23 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -12,6 +12,8 @@ #include #include #include +#include +#include #include #include #include @@ -49,6 +51,30 @@ void bpf_register_map_type(struct bpf_map_type_list *tl) list_add(&tl->list_node, &bpf_map_types); } +void *bpf_map_area_alloc(size_t size) +{ + /* We definitely need __GFP_NORETRY, so OOM killer doesn't + * trigger under memory pressure as we really just want to + * fail instead. + */ + const gfp_t flags = __GFP_NOWARN | __GFP_NORETRY | __GFP_ZERO; + void *area; + + if (size <= (PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER)) { + area = kmalloc(size, GFP_USER | flags); + if (area != NULL) + return area; + } + + return __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM | flags, + PAGE_KERNEL); +} + +void bpf_map_area_free(void *area) +{ + kvfree(area); +} + int bpf_map_precharge_memlock(u32 pages) { struct user_struct *user = get_current_user(); -- cgit v1.2.3 From b5b46c4740aed1538544f0fa849c5b76c7823469 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Wed, 18 Jan 2017 14:29:21 +0100 Subject: objtool: Fix IRET's opcode The IRET opcode is 0xcf according to the Intel manual and also to objdump of my vmlinux: 1ea8: 48 cf iretq Fix the opcode in arch_decode_instruction(). The previous value (0xc5) seems to correspond to LDS. Signed-off-by: Jiri Slaby Acked-by: Josh Poimboeuf Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20170118132921.19319-1-jslaby@suse.cz Signed-off-by: Ingo Molnar --- tools/objtool/arch/x86/decode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/objtool/arch/x86/decode.c b/tools/objtool/arch/x86/decode.c index 5e0dea2cdc01..039636ffb6c8 100644 --- a/tools/objtool/arch/x86/decode.c +++ b/tools/objtool/arch/x86/decode.c @@ -150,9 +150,9 @@ int arch_decode_instruction(struct elf *elf, struct section *sec, *type = INSN_RETURN; break; - case 0xc5: /* iret */ case 0xca: /* retf */ case 0xcb: /* retf */ + case 0xcf: /* iret */ *type = INSN_CONTEXT_SWITCH; break; -- cgit v1.2.3 From 24d615a694d649aa2e167c3f97f62bdad07e3f84 Mon Sep 17 00:00:00 2001 From: Aleksander Morgado Date: Wed, 18 Jan 2017 21:31:31 +0100 Subject: USB: serial: qcserial: add Dell DW5570 QDL The Dell DW5570 is a re-branded Sierra Wireless MC8805 which will by default boot with vid 0x413c and pid 0x81a3. When triggered QDL download mode, the device switches to pid 0x81a6 and provides the standard TTY used for firmware upgrade. Cc: Signed-off-by: Aleksander Morgado Signed-off-by: Johan Hovold --- drivers/usb/serial/qcserial.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/serial/qcserial.c b/drivers/usb/serial/qcserial.c index 1bc6089b9008..696458db7e3c 100644 --- a/drivers/usb/serial/qcserial.c +++ b/drivers/usb/serial/qcserial.c @@ -124,6 +124,7 @@ static const struct usb_device_id id_table[] = { {USB_DEVICE(0x1410, 0xa021)}, /* Novatel Gobi 3000 Composite */ {USB_DEVICE(0x413c, 0x8193)}, /* Dell Gobi 3000 QDL */ {USB_DEVICE(0x413c, 0x8194)}, /* Dell Gobi 3000 Composite */ + {USB_DEVICE(0x413c, 0x81a6)}, /* Dell DW5570 QDL (MC8805) */ {USB_DEVICE(0x1199, 0x68a4)}, /* Sierra Wireless QDL */ {USB_DEVICE(0x1199, 0x68a5)}, /* Sierra Wireless Modem */ {USB_DEVICE(0x1199, 0x68a8)}, /* Sierra Wireless QDL */ -- cgit v1.2.3 From b27e36482c02a94194fec71fb29696f4c8e9241c Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Sun, 15 Jan 2017 23:20:28 +0100 Subject: pinctrl: meson: fix uart_ao_b for GXBB and GXL/GXM The GXBB and GXL/GXM pinctrl drivers had a configuration which conflicts with uart_ao_a. According to the GXBB ("S905") datasheet the AO UART functions are: - GPIOAO_0: Func1 = UART_TX_AO_A (bit 12), Func2 = UART_TX_AO_B (bit 26) - GPIOAO_1: Func1 = UART_RX_AO_A (bit 11), Func2 = UART_RX_AO_B (bit 25) - GPIOAO_4: Func2 = UART_TX_AO_B (bit 24) - GPIOAO_5: Func2 = UART_RX_AO_B (bit 25) The existing definition for uart_AO_A already uses GPIOAO_0 and GPIOAO_1. The old definition of uart_AO_B however was broken, as it used GPIOAO_0 for TX (which would be fine) and two pins (GPIOAO_1 and GPIOAO_5) for RX (which does not make any sense). This fixes the uart_AO_B configuration by moving it to GPIOAO_4 and GPIOAO_5 (it would be possible to use GPIOAO_0 and GPIOAO_1 in theory, but all existing hardware uses uart_AO_A there). The fix for GXBB and GXL/GXM is identical since it seems that these specific pins are identical on both SoC variants. Signed-off-by: Martin Blumenstingl Reviewed-by: Kevin Hilman Signed-off-by: Linus Walleij --- drivers/pinctrl/meson/pinctrl-meson-gxbb.c | 7 +++---- drivers/pinctrl/meson/pinctrl-meson-gxl.c | 7 +++---- 2 files changed, 6 insertions(+), 8 deletions(-) diff --git a/drivers/pinctrl/meson/pinctrl-meson-gxbb.c b/drivers/pinctrl/meson/pinctrl-meson-gxbb.c index c3928aa3fefa..e0bca4df2a2f 100644 --- a/drivers/pinctrl/meson/pinctrl-meson-gxbb.c +++ b/drivers/pinctrl/meson/pinctrl-meson-gxbb.c @@ -253,9 +253,8 @@ static const unsigned int uart_tx_ao_a_pins[] = { PIN(GPIOAO_0, 0) }; static const unsigned int uart_rx_ao_a_pins[] = { PIN(GPIOAO_1, 0) }; static const unsigned int uart_cts_ao_a_pins[] = { PIN(GPIOAO_2, 0) }; static const unsigned int uart_rts_ao_a_pins[] = { PIN(GPIOAO_3, 0) }; -static const unsigned int uart_tx_ao_b_pins[] = { PIN(GPIOAO_0, 0) }; -static const unsigned int uart_rx_ao_b_pins[] = { PIN(GPIOAO_1, 0), - PIN(GPIOAO_5, 0) }; +static const unsigned int uart_tx_ao_b_pins[] = { PIN(GPIOAO_4, 0) }; +static const unsigned int uart_rx_ao_b_pins[] = { PIN(GPIOAO_5, 0) }; static const unsigned int uart_cts_ao_b_pins[] = { PIN(GPIOAO_2, 0) }; static const unsigned int uart_rts_ao_b_pins[] = { PIN(GPIOAO_3, 0) }; @@ -498,7 +497,7 @@ static struct meson_pmx_group meson_gxbb_aobus_groups[] = { GPIO_GROUP(GPIOAO_13, 0), /* bank AO */ - GROUP(uart_tx_ao_b, 0, 26), + GROUP(uart_tx_ao_b, 0, 24), GROUP(uart_rx_ao_b, 0, 25), GROUP(uart_tx_ao_a, 0, 12), GROUP(uart_rx_ao_a, 0, 11), diff --git a/drivers/pinctrl/meson/pinctrl-meson-gxl.c b/drivers/pinctrl/meson/pinctrl-meson-gxl.c index 25694f7094c7..b69743b07a1d 100644 --- a/drivers/pinctrl/meson/pinctrl-meson-gxl.c +++ b/drivers/pinctrl/meson/pinctrl-meson-gxl.c @@ -214,9 +214,8 @@ static const unsigned int uart_tx_ao_a_pins[] = { PIN(GPIOAO_0, 0) }; static const unsigned int uart_rx_ao_a_pins[] = { PIN(GPIOAO_1, 0) }; static const unsigned int uart_cts_ao_a_pins[] = { PIN(GPIOAO_2, 0) }; static const unsigned int uart_rts_ao_a_pins[] = { PIN(GPIOAO_3, 0) }; -static const unsigned int uart_tx_ao_b_pins[] = { PIN(GPIOAO_0, 0) }; -static const unsigned int uart_rx_ao_b_pins[] = { PIN(GPIOAO_1, 0), - PIN(GPIOAO_5, 0) }; +static const unsigned int uart_tx_ao_b_pins[] = { PIN(GPIOAO_4, 0) }; +static const unsigned int uart_rx_ao_b_pins[] = { PIN(GPIOAO_5, 0) }; static const unsigned int uart_cts_ao_b_pins[] = { PIN(GPIOAO_2, 0) }; static const unsigned int uart_rts_ao_b_pins[] = { PIN(GPIOAO_3, 0) }; @@ -409,7 +408,7 @@ static struct meson_pmx_group meson_gxl_aobus_groups[] = { GPIO_GROUP(GPIOAO_9, 0), /* bank AO */ - GROUP(uart_tx_ao_b, 0, 26), + GROUP(uart_tx_ao_b, 0, 24), GROUP(uart_rx_ao_b, 0, 25), GROUP(uart_tx_ao_a, 0, 12), GROUP(uart_rx_ao_a, 0, 11), -- cgit v1.2.3 From df1539c25cce98e2ac69881958850c6535240707 Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Tue, 17 Jan 2017 19:52:54 +0900 Subject: pinctrl: uniphier: fix Ethernet (RMII) pin-mux setting for LD20 Fix the pin-mux values for the MDC, MDIO, MDIO_INTL, PHYRSTL pins. Fixes: 1e359ab1285e ("pinctrl: uniphier: add Ethernet pin-mux settings") Signed-off-by: Masahiro Yamada Signed-off-by: Linus Walleij --- drivers/pinctrl/uniphier/pinctrl-uniphier-ld20.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pinctrl/uniphier/pinctrl-uniphier-ld20.c b/drivers/pinctrl/uniphier/pinctrl-uniphier-ld20.c index aa8bd9794683..96686336e3a3 100644 --- a/drivers/pinctrl/uniphier/pinctrl-uniphier-ld20.c +++ b/drivers/pinctrl/uniphier/pinctrl-uniphier-ld20.c @@ -561,7 +561,7 @@ static const int ether_rgmii_muxvals[] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; static const unsigned ether_rmii_pins[] = {30, 31, 32, 33, 34, 35, 36, 37, 39, 41, 42, 45}; -static const int ether_rmii_muxvals[] = {1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1}; +static const int ether_rmii_muxvals[] = {0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1}; static const unsigned i2c0_pins[] = {63, 64}; static const int i2c0_muxvals[] = {0, 0}; static const unsigned i2c1_pins[] = {65, 66}; -- cgit v1.2.3 From 524b698db06b9b6da7192e749f637904e2f62d7b Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 16 Jan 2017 18:24:56 +0100 Subject: netfilter: conntrack: remove GC_MAX_EVICTS break Instead of breaking loop and instant resched, don't bother checking this in first place (the loop calls cond_resched for every bucket anyway). Suggested-by: Nicolas Dichtel Signed-off-by: Florian Westphal Acked-by: Nicolas Dichtel Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_core.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 3a073cd9fcf4..6feb5d370319 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -88,8 +88,6 @@ static __read_mostly bool nf_conntrack_locks_all; #define GC_MAX_BUCKETS_DIV 64u /* upper bound of scan intervals */ #define GC_INTERVAL_MAX (2 * HZ) -/* maximum conntracks to evict per gc run */ -#define GC_MAX_EVICTS 256u static struct conntrack_gc_work conntrack_gc_work; @@ -979,8 +977,7 @@ static void gc_worker(struct work_struct *work) */ rcu_read_unlock(); cond_resched_rcu_qs(); - } while (++buckets < goal && - expired_count < GC_MAX_EVICTS); + } while (++buckets < goal); if (gc_work->exiting) return; @@ -1005,7 +1002,7 @@ static void gc_worker(struct work_struct *work) * In case we have lots of evictions next scan is done immediately. */ ratio = scanned ? expired_count * 100 / scanned : 0; - if (ratio >= 90 || expired_count == GC_MAX_EVICTS) { + if (ratio >= 90) { gc_work->next_gc_run = 0; next_run = 0; } else if (expired_count) { -- cgit v1.2.3 From e5072053b09642b8ff417d47da05b84720aea3ee Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 18 Jan 2017 02:01:22 +0100 Subject: netfilter: conntrack: refine gc worker heuristics, redux This further refines the changes made to conntrack gc_worker in commit e0df8cae6c16 ("netfilter: conntrack: refine gc worker heuristics"). The main idea of that change was to reduce the scan interval when evictions take place. However, on the reporters' setup, there are 1-2 million conntrack entries in total and roughly 8k new (and closing) connections per second. In this case we'll always evict at least one entry per gc cycle and scan interval is always at 1 jiffy because of this test: } else if (expired_count) { gc_work->next_gc_run /= 2U; next_run = msecs_to_jiffies(1); being true almost all the time. Given we scan ~10k entries per run its clearly wrong to reduce interval based on nonzero eviction count, it will only waste cpu cycles since a vast majorities of conntracks are not timed out. Thus only look at the ratio (scanned entries vs. evicted entries) to make a decision on whether to reduce or not. Because evictor is supposed to only kick in when system turns idle after a busy period, pick a high ratio -- this makes it 50%. We thus keep the idea of increasing scan rate when its likely that table contains many expired entries. In order to not let timed-out entries hang around for too long (important when using event logging, in which case we want to timely destroy events), we now scan the full table within at most GC_MAX_SCAN_JIFFIES (16 seconds) even in worst-case scenario where all timed-out entries sit in same slot. I tested this with a vm under synflood (with sysctl net.netfilter.nf_conntrack_tcp_timeout_syn_recv=3). While flood is ongoing, interval now stays at its max rate (GC_MAX_SCAN_JIFFIES / GC_MAX_BUCKETS_DIV -> 125ms). With feedback from Nicolas Dichtel. Reported-by: Denys Fedoryshchenko Cc: Nicolas Dichtel Fixes: b87a2f9199ea82eaadc ("netfilter: conntrack: add gc worker to remove timed-out entries") Signed-off-by: Florian Westphal Tested-by: Nicolas Dichtel Acked-by: Nicolas Dichtel Tested-by: Denys Fedoryshchenko Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_core.c | 39 ++++++++++++++++++++------------------- 1 file changed, 20 insertions(+), 19 deletions(-) diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 6feb5d370319..4e8083c5e01d 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -85,9 +85,11 @@ static __read_mostly DEFINE_SPINLOCK(nf_conntrack_locks_all_lock); static __read_mostly bool nf_conntrack_locks_all; /* every gc cycle scans at most 1/GC_MAX_BUCKETS_DIV part of table */ -#define GC_MAX_BUCKETS_DIV 64u -/* upper bound of scan intervals */ -#define GC_INTERVAL_MAX (2 * HZ) +#define GC_MAX_BUCKETS_DIV 128u +/* upper bound of full table scan */ +#define GC_MAX_SCAN_JIFFIES (16u * HZ) +/* desired ratio of entries found to be expired */ +#define GC_EVICT_RATIO 50u static struct conntrack_gc_work conntrack_gc_work; @@ -936,6 +938,7 @@ static noinline int early_drop(struct net *net, unsigned int _hash) static void gc_worker(struct work_struct *work) { + unsigned int min_interval = max(HZ / GC_MAX_BUCKETS_DIV, 1u); unsigned int i, goal, buckets = 0, expired_count = 0; struct conntrack_gc_work *gc_work; unsigned int ratio, scanned = 0; @@ -994,27 +997,25 @@ static void gc_worker(struct work_struct *work) * 1. Minimize time until we notice a stale entry * 2. Maximize scan intervals to not waste cycles * - * Normally, expired_count will be 0, this increases the next_run time - * to priorize 2) above. + * Normally, expire ratio will be close to 0. * - * As soon as a timed-out entry is found, move towards 1) and increase - * the scan frequency. - * In case we have lots of evictions next scan is done immediately. + * As soon as a sizeable fraction of the entries have expired + * increase scan frequency. */ ratio = scanned ? expired_count * 100 / scanned : 0; - if (ratio >= 90) { - gc_work->next_gc_run = 0; - next_run = 0; - } else if (expired_count) { - gc_work->next_gc_run /= 2U; - next_run = msecs_to_jiffies(1); + if (ratio > GC_EVICT_RATIO) { + gc_work->next_gc_run = min_interval; } else { - if (gc_work->next_gc_run < GC_INTERVAL_MAX) - gc_work->next_gc_run += msecs_to_jiffies(1); + unsigned int max = GC_MAX_SCAN_JIFFIES / GC_MAX_BUCKETS_DIV; - next_run = gc_work->next_gc_run; + BUILD_BUG_ON((GC_MAX_SCAN_JIFFIES / GC_MAX_BUCKETS_DIV) == 0); + + gc_work->next_gc_run += min_interval; + if (gc_work->next_gc_run > max) + gc_work->next_gc_run = max; } + next_run = gc_work->next_gc_run; gc_work->last_bucket = i; queue_delayed_work(system_long_wq, &gc_work->dwork, next_run); } @@ -1022,7 +1023,7 @@ static void gc_worker(struct work_struct *work) static void conntrack_gc_work_init(struct conntrack_gc_work *gc_work) { INIT_DELAYED_WORK(&gc_work->dwork, gc_worker); - gc_work->next_gc_run = GC_INTERVAL_MAX; + gc_work->next_gc_run = HZ; gc_work->exiting = false; } @@ -1914,7 +1915,7 @@ int nf_conntrack_init_start(void) nf_ct_untracked_status_or(IPS_CONFIRMED | IPS_UNTRACKED); conntrack_gc_work_init(&conntrack_gc_work); - queue_delayed_work(system_long_wq, &conntrack_gc_work.dwork, GC_INTERVAL_MAX); + queue_delayed_work(system_long_wq, &conntrack_gc_work.dwork, HZ); return 0; -- cgit v1.2.3 From 59cfa789d04c35b6c647aacf4cc89b3d4d430cfe Mon Sep 17 00:00:00 2001 From: Hariprasad Shenai Date: Thu, 19 Jan 2017 11:50:32 +0530 Subject: MAINTAINERS: update cxgb4 maintainer Ganesg will be taking over as maintainer from now Signed-off-by: Hariprasad Shenai Signed-off-by: David S. Miller --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index c36976d3bd1a..af3456d33a18 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3564,7 +3564,7 @@ F: drivers/infiniband/hw/cxgb3/ F: include/uapi/rdma/cxgb3-abi.h CXGB4 ETHERNET DRIVER (CXGB4) -M: Hariprasad S +M: Ganesh Goudar L: netdev@vger.kernel.org W: http://www.chelsio.com S: Supported -- cgit v1.2.3 From 03e4deff4987f79c34112c5ba4eb195d4f9382b0 Mon Sep 17 00:00:00 2001 From: Kefeng Wang Date: Thu, 19 Jan 2017 16:26:21 +0800 Subject: ipv6: addrconf: Avoid addrconf_disable_change() using RCU read-side lock Just like commit 4acd4945cd1e ("ipv6: addrconf: Avoid calling netdevice notifiers with RCU read-side lock"), it is unnecessary to make addrconf_disable_change() use RCU iteration over the netdev list, since it already holds the RTNL lock, or we may meet Illegal context switch in RCU read-side critical section. Signed-off-by: Kefeng Wang Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index c1e124bc8e1e..f60e88e56255 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -5540,8 +5540,7 @@ static void addrconf_disable_change(struct net *net, __s32 newf) struct net_device *dev; struct inet6_dev *idev; - rcu_read_lock(); - for_each_netdev_rcu(net, dev) { + for_each_netdev(net, dev) { idev = __in6_dev_get(dev); if (idev) { int changed = (!idev->cnf.disable_ipv6) ^ (!newf); @@ -5550,7 +5549,6 @@ static void addrconf_disable_change(struct net *net, __s32 newf) dev_disable_change(idev); } } - rcu_read_unlock(); } static int addrconf_disable_ipv6(struct ctl_table *table, int *p, int newf) -- cgit v1.2.3 From ad05df399f3343b10664827a3860669a8a80782d Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 19 Jan 2017 10:33:29 +0100 Subject: net/mlx5e: Remove unused variable A cleanup removed the only user of this variable mlx5/core/en_ethtool.c: In function 'mlx5e_set_channels': mlx5/core/en_ethtool.c:546:6: error: unused variable 'ncv' [-Werror=unused-variable] Let's remove the declaration as well. Fixes: 639e9e94160e ("net/mlx5e: Remove unnecessary checks when setting num channels") Signed-off-by: Arnd Bergmann Reported-by: Stephen Rothwell Reviewed-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index b1b9eb6ee135..5197817e4b2f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -543,7 +543,6 @@ static int mlx5e_set_channels(struct net_device *dev, struct ethtool_channels *ch) { struct mlx5e_priv *priv = netdev_priv(dev); - int ncv = mlx5e_get_max_num_channels(priv->mdev); unsigned int count = ch->combined_count; bool arfs_enabled; bool was_opened; -- cgit v1.2.3 From 0dbd7ff3ac5017a46033a9d0a87a8267d69119d9 Mon Sep 17 00:00:00 2001 From: Alexey Kodanev Date: Thu, 19 Jan 2017 16:36:39 +0300 Subject: tcp: initialize max window for a new fastopen socket Found that if we run LTP netstress test with large MSS (65K), the first attempt from server to send data comparable to this MSS on fastopen connection will be delayed by the probe timer. Here is an example: < S seq 0:0 win 43690 options [mss 65495 wscale 7 tfo cookie] length 32 > S. seq 0:0 ack 1 win 43690 options [mss 65495 wscale 7] length 0 < . ack 1 win 342 length 0 Inside tcp_sendmsg(), tcp_send_mss() returns max MSS in 'mss_now', as well as in 'size_goal'. This results the segment not queued for transmition until all the data copied from user buffer. Then, inside __tcp_push_pending_frames(), it breaks on send window test and continues with the check probe timer. Fragmentation occurs in tcp_write_wakeup()... +0.2 > P. seq 1:43777 ack 1 win 342 length 43776 < . ack 43777, win 1365 length 0 > P. seq 43777:65001 ack 1 win 342 options [...] length 21224 ... This also contradicts with the fact that we should bound to the half of the window if it is large. Fix this flaw by correctly initializing max_window. Before that, it could have large values that affect further calculations of 'size_goal'. Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path") Signed-off-by: Alexey Kodanev Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index f51919535ca7..dd2560c83a85 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -205,6 +205,7 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk, * scaled. So correct it appropriately. */ tp->snd_wnd = ntohs(tcp_hdr(skb)->window); + tp->max_window = tp->snd_wnd; /* Activate the retrans timer so that SYNACK can be retransmitted. * The request socket is not added to the ehash -- cgit v1.2.3 From 69fed99baac186013840ced3524562841296034f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 18 Jan 2017 19:44:42 -0800 Subject: gianfar: Do not reuse pages from emergency reserve A driver using dev_alloc_page() must not reuse a page that had to use emergency memory reserve. Otherwise all packets using this page will be immediately dropped, unless for very specific sockets having SOCK_MEMALLOC bit set. This issue might be hard to debug, because only a fraction of the RX ring buffer would suffer from drops. Fixes: 75354148ce69 ("gianfar: Add paged allocation and Rx S/G") Signed-off-by: Eric Dumazet Cc: Claudiu Manoil Acked-by: Claudiu Manoil Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/gianfar.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index a6e7afa878be..c1b671667920 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -2948,7 +2948,7 @@ static bool gfar_add_rx_frag(struct gfar_rx_buff *rxb, u32 lstatus, } /* try reuse page */ - if (unlikely(page_count(page) != 1)) + if (unlikely(page_count(page) != 1 || page_is_pfmemalloc(page))) return false; /* change offset to the other half */ -- cgit v1.2.3 From 47b5d64691350e116093c9b47b55ca6b9433bc50 Mon Sep 17 00:00:00 2001 From: Wang Xiaoguang Date: Wed, 7 Sep 2016 20:17:38 +0800 Subject: btrfs: fix false enospc error when truncating heavily reflinked file Below test script can reveal this bug: dd if=/dev/zero of=fs.img bs=$((1024*1024)) count=100 dev=$(losetup --show -f fs.img) mkdir -p /mnt/mntpoint mkfs.btrfs -f $dev mount $dev /mnt/mntpoint cd /mnt/mntpoint echo "workdir is: /mnt/mntpoint" blocksize=$((128 * 1024)) dd if=/dev/zero of=testfile bs=$blocksize count=1 sync count=$((17*1024*1024*1024/blocksize)) echo "file size is:" $((count*blocksize)) for ((i = 1; i <= $count; i++)); do dst_offset=$((blocksize * i)) xfs_io -f -c "reflink testfile 0 $dst_offset $blocksize"\ testfile > /dev/null done sync truncate --size 0 testfile The last truncate operation will fail for ENOSPC reason, but indeed it should not fail. In btrfs_truncate(), we use a temporary block_rsv to do truncate operation. With every btrfs_truncate_inode_items() call, we migrate space to this block_rsv, but forget to cleanup previous reservation, which will make this block_rsv's reserved bytes keep growing, and this reserved space will only be released in the end of btrfs_truncate(), this metadata leak will impact other's metadata reservation. In this case, it's "btrfs_start_transaction(root, 2);" fails for enospc error, which make this truncate operation fail. Call btrfs_block_rsv_release() to fix this bug. Signed-off-by: Wang Xiaoguang Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 128e52489f81..415a6ad5ba22 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -9215,6 +9215,7 @@ static int btrfs_truncate(struct inode *inode) break; } + btrfs_block_rsv_release(fs_info, rsv, -1); ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, rsv, min_size, 0); BUG_ON(ret); /* shouldn't happen */ -- cgit v1.2.3 From 97dcdea076ecef41ea4aaa23d4397c2f622e4265 Mon Sep 17 00:00:00 2001 From: Chandan Rajendra Date: Fri, 23 Dec 2016 15:00:18 +0530 Subject: Btrfs: Fix deadlock between direct IO and fast fsync The following deadlock is seen when executing generic/113 test, ---------------------------------------------------------+---------------------------------------------------- Direct I/O task Fast fsync task ---------------------------------------------------------+---------------------------------------------------- btrfs_direct_IO __blockdev_direct_IO do_blockdev_direct_IO do_direct_IO btrfs_get_blocks_direct while (blocks needs to written) get_more_blocks (first iteration) btrfs_get_blocks_direct btrfs_create_dio_extent down_read(&BTRFS_I(inode) >dio_sem) Create and add extent map and ordered extent up_read(&BTRFS_I(inode) >dio_sem) btrfs_sync_file btrfs_log_dentry_safe btrfs_log_inode_parent btrfs_log_inode btrfs_log_changed_extents down_write(&BTRFS_I(inode) >dio_sem) Collect new extent maps and ordered extents wait for ordered extent completion get_more_blocks (second iteration) btrfs_get_blocks_direct btrfs_create_dio_extent down_read(&BTRFS_I(inode) >dio_sem) -------------------------------------------------------------------------------------------------------------- In the above description, Btrfs direct I/O code path has not yet started submitting bios for file range covered by the initial ordered extent. Meanwhile, The fast fsync task obtains the write semaphore and waits for I/O on the ordered extent to get completed. However, the Direct I/O task is now blocked on obtaining the read semaphore. To resolve the deadlock, this commit modifies the Direct I/O code path to obtain the read semaphore before invoking __blockdev_direct_IO(). The semaphore is then given up after __blockdev_direct_IO() returns. This allows the Direct I/O code to complete I/O on all the ordered extents it creates. Signed-off-by: Chandan Rajendra Reviewed-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/inode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 415a6ad5ba22..bed6cf5f67a4 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7215,7 +7215,6 @@ static struct extent_map *btrfs_create_dio_extent(struct inode *inode, struct extent_map *em = NULL; int ret; - down_read(&BTRFS_I(inode)->dio_sem); if (type != BTRFS_ORDERED_NOCOW) { em = create_pinned_em(inode, start, len, orig_start, block_start, block_len, orig_block_len, @@ -7234,7 +7233,6 @@ static struct extent_map *btrfs_create_dio_extent(struct inode *inode, em = ERR_PTR(ret); } out: - up_read(&BTRFS_I(inode)->dio_sem); return em; } @@ -8695,6 +8693,7 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) dio_data.unsubmitted_oe_range_start = (u64)offset; dio_data.unsubmitted_oe_range_end = (u64)offset; current->journal_info = &dio_data; + down_read(&BTRFS_I(inode)->dio_sem); } else if (test_bit(BTRFS_INODE_READDIO_NEED_LOCK, &BTRFS_I(inode)->runtime_flags)) { inode_dio_end(inode); @@ -8707,6 +8706,7 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) iter, btrfs_get_blocks_direct, NULL, btrfs_submit_direct, flags); if (iov_iter_rw(iter) == WRITE) { + up_read(&BTRFS_I(inode)->dio_sem); current->journal_info = NULL; if (ret < 0 && ret != -EIOCBQUEUED) { if (dio_data.reserve) -- cgit v1.2.3 From 91298eec05cd8d4e828cf7ee5d4a6334f70cf69a Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Thu, 1 Dec 2016 13:43:31 -0800 Subject: Btrfs: fix truncate down when no_holes feature is enabled For such a file mapping, [0-4k][hole][8k-12k] In NO_HOLES mode, we don't have the [hole] extent any more. Commit c1aa45759e90 ("Btrfs: fix shrinking truncate when the no_holes feature is enabled") fixed disk isize not being updated in NO_HOLES mode when data is not flushed. However, even if data has been flushed, we can still have trouble in updating disk isize since we updated disk isize to 'start' of the last evicted extent. Reviewed-by: Chris Mason Signed-off-by: Liu Bo Signed-off-by: David Sterba --- fs/btrfs/inode.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index bed6cf5f67a4..177e7284909c 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4505,8 +4505,19 @@ search_again: if (found_type > min_type) { del_item = 1; } else { - if (item_end < new_size) + if (item_end < new_size) { + /* + * With NO_HOLES mode, for the following mapping + * + * [0-4k][hole][8k-12k] + * + * if truncating isize down to 6k, it ends up + * isize being 8k. + */ + if (btrfs_fs_incompat(root->fs_info, NO_HOLES)) + last_size = new_size; break; + } if (found_key.offset >= new_size) del_item = 1; else -- cgit v1.2.3 From e326ce013a8e851193eb337aafb1aa396c533a61 Mon Sep 17 00:00:00 2001 From: Rafael J. Wysocki Date: Fri, 20 Jan 2017 03:25:34 +0100 Subject: Revert "PM / sleep / ACPI: Use the ACPI_FADT_LOW_POWER_S0 flag" Revert commit 08b98d329165 (PM / sleep / ACPI: Use the ACPI_FADT_LOW_POWER_S0 flag) as it caused system suspend (in the default configuration) to fail on Dell XPS13 (9360) with the Kaby Lake processor. Fixes: 08b98d329165 (PM / sleep / ACPI: Use the ACPI_FADT_LOW_POWER_S0 flag) Reported-by: Paul Menzel Signed-off-by: Rafael J. Wysocki --- Documentation/power/states.txt | 4 +--- drivers/acpi/sleep.c | 8 -------- include/linux/suspend.h | 2 -- kernel/power/suspend.c | 4 ++-- 4 files changed, 3 insertions(+), 15 deletions(-) diff --git a/Documentation/power/states.txt b/Documentation/power/states.txt index 8a39ce45d8a0..008ecb588317 100644 --- a/Documentation/power/states.txt +++ b/Documentation/power/states.txt @@ -35,9 +35,7 @@ only one way to cause the system to go into the Suspend-To-RAM state (write The default suspend mode (ie. the one to be used without writing anything into /sys/power/mem_sleep) is either "deep" (if Suspend-To-RAM is supported) or "s2idle", but it can be overridden by the value of the "mem_sleep_default" -parameter in the kernel command line. On some ACPI-based systems, depending on -the information in the FADT, the default may be "s2idle" even if Suspend-To-RAM -is supported. +parameter in the kernel command line. The properties of all of the sleep states are described below. diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c index ce1855fd584b..deb0ff78eba8 100644 --- a/drivers/acpi/sleep.c +++ b/drivers/acpi/sleep.c @@ -691,14 +691,6 @@ static void acpi_sleep_suspend_setup(void) if (acpi_sleep_state_supported(i)) sleep_states[i] = 1; - /* - * Use suspend-to-idle by default if ACPI_FADT_LOW_POWER_S0 is set and - * the default suspend mode was not selected from the command line. - */ - if (acpi_gbl_FADT.flags & ACPI_FADT_LOW_POWER_S0 && - mem_sleep_default > PM_SUSPEND_MEM) - mem_sleep_default = PM_SUSPEND_FREEZE; - suspend_set_ops(old_suspend_ordering ? &acpi_suspend_ops_old : &acpi_suspend_ops); freeze_set_ops(&acpi_freeze_ops); diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 0c729c3c8549..d9718378a8be 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -194,8 +194,6 @@ struct platform_freeze_ops { }; #ifdef CONFIG_SUSPEND -extern suspend_state_t mem_sleep_default; - /** * suspend_set_ops - set platform dependent suspend operations * @ops: The new suspend operations to set. diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index f67ceb7768b8..15e6baef5c73 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -46,7 +46,7 @@ static const char * const mem_sleep_labels[] = { const char *mem_sleep_states[PM_SUSPEND_MAX]; suspend_state_t mem_sleep_current = PM_SUSPEND_FREEZE; -suspend_state_t mem_sleep_default = PM_SUSPEND_MAX; +static suspend_state_t mem_sleep_default = PM_SUSPEND_MEM; unsigned int pm_suspend_global_flags; EXPORT_SYMBOL_GPL(pm_suspend_global_flags); @@ -168,7 +168,7 @@ void suspend_set_ops(const struct platform_suspend_ops *ops) } if (valid_state(PM_SUSPEND_MEM)) { mem_sleep_states[PM_SUSPEND_MEM] = mem_sleep_labels[PM_SUSPEND_MEM]; - if (mem_sleep_default >= PM_SUSPEND_MEM) + if (mem_sleep_default == PM_SUSPEND_MEM) mem_sleep_current = PM_SUSPEND_MEM; } -- cgit v1.2.3 From 1443ebbacfd7f8d84fbbbf629ef66a12dc8a4b4e Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Wed, 18 Jan 2017 10:48:22 -0800 Subject: cpufreq: intel_pstate: Fix sysfs limits enforcement for performance policy A side effect of keeping intel_pstate sysfs limits in sync with cpufreq is that the now sysfs limits can't enforced under performance policy. For example, if the max_perf_pct is changed from 100 to 80, this will call intel_pstate_set_policy(), which will change the max_perf to 100 again for performance policy. Same issue happens, when no_turbo is set. This change calculates max and min frequency using sysfs performance limits in intel_pstate_verify_policy() and adjusts policy limits by calling cpufreq_verify_within_limits(). Also, it causes the setting of performance limits to be skipped if no_turbo is set. Fixes: 111b8b3fe4fa (cpufreq: intel_pstate: Always keep all limits settings in sync) Signed-off-by: Srinivas Pandruvada Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/intel_pstate.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index e261438cd690..ed215c5592f6 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -1951,7 +1951,8 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy) limits = &performance_limits; perf_limits = limits; } - if (policy->max >= policy->cpuinfo.max_freq) { + if (policy->max >= policy->cpuinfo.max_freq && + !limits->no_turbo) { pr_debug("set performance\n"); intel_pstate_set_performance_limits(perf_limits); goto out; @@ -1993,6 +1994,17 @@ static int intel_pstate_verify_policy(struct cpufreq_policy *policy) policy->policy != CPUFREQ_POLICY_PERFORMANCE) return -EINVAL; + /* When per-CPU limits are used, sysfs limits are not used */ + if (!per_cpu_limits) { + unsigned int max_freq, min_freq; + + max_freq = policy->cpuinfo.max_freq * + limits->max_sysfs_pct / 100; + min_freq = policy->cpuinfo.max_freq * + limits->min_sysfs_pct / 100; + cpufreq_verify_within_limits(policy, min_freq, max_freq); + } + return 0; } -- cgit v1.2.3 From 7a37052adb5e843bcfff6c98aee9b60bb087b910 Mon Sep 17 00:00:00 2001 From: Lv Zheng Date: Thu, 19 Jan 2017 15:21:34 +0800 Subject: ACPICA: Tables: Fix hidden logic related to acpi_tb_install_standard_table() There is a hidden logic for acpi_tb_install_standard_table() as it can be invoked from the boot stage and during runtime. 1. When it is invoked from the OS boot stage, the ACPICA mutex may not have been initialized yet and so acpi_ut_acquire_mutex()/acpi_ut_release_mutex() are not invoked in these code paths: acpi_initialize_tables acpi_tb_parse_root_table acpi_tb_install_standard_table (4 invocations) acpi_install_table acpi_tb_install_standard_table 2. When it is invoked during the runtime, ACPICA mutex is used as appropriate: acpi_ex_load_op acpi_tb_install_and_load_table acpi_tb_install_standard_table acpi_load_table acpi_tb_install_and_load_table acpi_tb_install_standard_table The mutex is now used in acpi_tb_install_and_load_table(), while it actually should be in acpi_tb_install_standard_table(). This introduces another problem in acpi_tb_install_standard_table() where acpi_gbl_table_handler is invoked from and the lock contexts are thus not consistent for the table handlers. This triggers a regression when acpi_get_table()/acpi_put_table() start to hold table mutex during runtime. The regression is noticed by LKP as new errors reported by ACPICA mutex debugging facility. [ 2.043693] ACPI Error: Mutex [ACPI_MTX_Tables] already acquired by this thread [497483776] (20160930/utmutex-254) [ 2.054084] ACPI Error: Mutex [0x2] is not acquired, cannot release (20160930/utmutex-326) And it triggers a deadlock: [ 247.066214] INFO: task swapper/0:1 blocked for more than 120 seconds. ... [ 247.091271] Call Trace: ... [ 247.121523] down_timeout+0x47/0x50 [ 247.125065] acpi_os_wait_semaphore+0x47/0x62 [ 247.129475] acpi_ut_acquire_mutex+0x43/0x81 [ 247.133798] acpi_get_table+0x2d/0x84 [ 247.137513] acpi_table_attr_init+0xcd/0x100 [ 247.146590] acpi_sysfs_table_handler+0x5d/0xb8 [ 247.151174] acpi_bus_table_handler+0x23/0x2a [ 247.155583] acpi_tb_install_standard_table+0xe0/0x213 [ 247.164489] acpi_tb_install_and_load_table+0x3a/0x82 [ 247.169592] acpi_ex_load_op+0x194/0x201 ... [ 247.200108] acpi_ns_evaluate+0x1bb/0x247 [ 247.204170] acpi_evaluate_object+0x178/0x274 [ 247.213249] acpi_processor_set_pdc+0x154/0x17b ... The table mutex is held in acpi_tb_install_and_load_table() and is re-visited by acpi_get_table(). Noticing that the early mutex requirement actually belongs to the OSL layer and has already been handled in acpi_os_wait_semaphore()/acpi_os_signal_semaphore(), the regression canbe fixed by removing this hidden logic from the ACPICA core to the OS-specific code. Fixes: 174cc7187e6f ("ACPICA: Tables: Back port acpi_get_table_with_size() and early_acpi_os_unmap_memory() from Linux kernel") Reported-and-tested-by: Tomi Sarvela Reported-by: Ye Xiaolong Signed-off-by: Lv Zheng Signed-off-by: Rafael J. Wysocki --- drivers/acpi/acpica/tbdata.c | 9 ++------- drivers/acpi/acpica/tbinstal.c | 17 +++++++++++++++-- 2 files changed, 17 insertions(+), 9 deletions(-) diff --git a/drivers/acpi/acpica/tbdata.c b/drivers/acpi/acpica/tbdata.c index 82b0b5710979..b0399e8f6d27 100644 --- a/drivers/acpi/acpica/tbdata.c +++ b/drivers/acpi/acpica/tbdata.c @@ -852,23 +852,18 @@ acpi_tb_install_and_load_table(acpi_physical_address address, ACPI_FUNCTION_TRACE(tb_install_and_load_table); - (void)acpi_ut_acquire_mutex(ACPI_MTX_TABLES); - /* Install the table and load it into the namespace */ status = acpi_tb_install_standard_table(address, flags, TRUE, override, &i); if (ACPI_FAILURE(status)) { - goto unlock_and_exit; + goto exit; } - (void)acpi_ut_release_mutex(ACPI_MTX_TABLES); status = acpi_tb_load_table(i, acpi_gbl_root_node); - (void)acpi_ut_acquire_mutex(ACPI_MTX_TABLES); -unlock_and_exit: +exit: *table_index = i; - (void)acpi_ut_release_mutex(ACPI_MTX_TABLES); return_ACPI_STATUS(status); } diff --git a/drivers/acpi/acpica/tbinstal.c b/drivers/acpi/acpica/tbinstal.c index 5fdf251a9f97..01e1b3d63fc0 100644 --- a/drivers/acpi/acpica/tbinstal.c +++ b/drivers/acpi/acpica/tbinstal.c @@ -217,6 +217,10 @@ acpi_tb_install_standard_table(acpi_physical_address address, goto release_and_exit; } + /* Acquire the table lock */ + + (void)acpi_ut_acquire_mutex(ACPI_MTX_TABLES); + if (reload) { /* * Validate the incoming table signature. @@ -244,7 +248,7 @@ acpi_tb_install_standard_table(acpi_physical_address address, new_table_desc.signature.integer)); status = AE_BAD_SIGNATURE; - goto release_and_exit; + goto unlock_and_exit; } /* Check if table is already registered */ @@ -279,7 +283,7 @@ acpi_tb_install_standard_table(acpi_physical_address address, /* Table is still loaded, this is an error */ status = AE_ALREADY_EXISTS; - goto release_and_exit; + goto unlock_and_exit; } else { /* * Table was unloaded, allow it to be reloaded. @@ -290,6 +294,7 @@ acpi_tb_install_standard_table(acpi_physical_address address, * indicate the re-installation. */ acpi_tb_uninstall_table(&new_table_desc); + (void)acpi_ut_release_mutex(ACPI_MTX_TABLES); *table_index = i; return_ACPI_STATUS(AE_OK); } @@ -303,11 +308,19 @@ acpi_tb_install_standard_table(acpi_physical_address address, /* Invoke table handler if present */ + (void)acpi_ut_release_mutex(ACPI_MTX_TABLES); if (acpi_gbl_table_handler) { (void)acpi_gbl_table_handler(ACPI_TABLE_EVENT_INSTALL, new_table_desc.pointer, acpi_gbl_table_handler_context); } + (void)acpi_ut_acquire_mutex(ACPI_MTX_TABLES); + +unlock_and_exit: + + /* Release the table lock */ + + (void)acpi_ut_release_mutex(ACPI_MTX_TABLES); release_and_exit: -- cgit v1.2.3 From f1225ee4c8fcf09afaa199b8b1f0450f38b8cd11 Mon Sep 17 00:00:00 2001 From: Stefano Stabellini Date: Thu, 19 Jan 2017 10:39:09 -0800 Subject: swiotlb-xen: update dev_addr after swapping pages In xen_swiotlb_map_page and xen_swiotlb_map_sg_attrs, if the original page is not suitable, we swap it for another page from the swiotlb pool. In these cases, we don't update the previously calculated dma address for the page before calling xen_dma_map_page. Thus, we end up calling xen_dma_map_page passing the wrong dev_addr, resulting in xen_dma_map_page mistakenly assuming that the page is foreign when it is local. Fix the bug by updating dev_addr appropriately. This change has no effect on x86, because xen_dma_map_page is a stub there. Signed-off-by: Stefano Stabellini Signed-off-by: Pooya Keshavarzi Tested-by: Pooya Keshavarzi Reviewed-by: Boris Ostrovsky Signed-off-by: Konrad Rzeszutek Wilk --- drivers/xen/swiotlb-xen.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/xen/swiotlb-xen.c b/drivers/xen/swiotlb-xen.c index f905d6eeb048..f8afc6dcc29f 100644 --- a/drivers/xen/swiotlb-xen.c +++ b/drivers/xen/swiotlb-xen.c @@ -414,9 +414,9 @@ dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page, if (map == SWIOTLB_MAP_ERROR) return DMA_ERROR_CODE; + dev_addr = xen_phys_to_bus(map); xen_dma_map_page(dev, pfn_to_page(map >> PAGE_SHIFT), dev_addr, map & ~PAGE_MASK, size, dir, attrs); - dev_addr = xen_phys_to_bus(map); /* * Ensure that the address returned is DMA'ble @@ -575,13 +575,14 @@ xen_swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, sg_dma_len(sgl) = 0; return 0; } + dev_addr = xen_phys_to_bus(map); xen_dma_map_page(hwdev, pfn_to_page(map >> PAGE_SHIFT), dev_addr, map & ~PAGE_MASK, sg->length, dir, attrs); - sg->dma_address = xen_phys_to_bus(map); + sg->dma_address = dev_addr; } else { /* we are not interested in the dma_addr returned by * xen_dma_map_page, only in the potential cache flushes executed -- cgit v1.2.3 From 6391a4481ba0796805d6581e42f9f0418c099e34 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Fri, 20 Jan 2017 14:32:42 +0800 Subject: virtio-net: restore VIRTIO_HDR_F_DATA_VALID on receiving Commit 501db511397f ("virtio: don't set VIRTIO_NET_HDR_F_DATA_VALID on xmit") in fact disables VIRTIO_HDR_F_DATA_VALID on receiving path too, fixing this by adding a hint (has_data_valid) and set it only on the receiving path. Cc: Rolf Neugebauer Signed-off-by: Jason Wang Acked-by: Rolf Neugebauer Signed-off-by: David S. Miller --- drivers/net/macvtap.c | 2 +- drivers/net/tun.c | 2 +- drivers/net/virtio_net.c | 2 +- include/linux/virtio_net.h | 6 +++++- net/packet/af_packet.c | 4 ++-- 5 files changed, 10 insertions(+), 6 deletions(-) diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index 5c26653eceb5..402618565838 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -825,7 +825,7 @@ static ssize_t macvtap_put_user(struct macvtap_queue *q, return -EINVAL; if (virtio_net_hdr_from_skb(skb, &vnet_hdr, - macvtap_is_little_endian(q))) + macvtap_is_little_endian(q), true)) BUG(); if (copy_to_iter(&vnet_hdr, sizeof(vnet_hdr), iter) != diff --git a/drivers/net/tun.c b/drivers/net/tun.c index cd8e02c94be0..2cd10b26b650 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1360,7 +1360,7 @@ static ssize_t tun_put_user(struct tun_struct *tun, return -EINVAL; if (virtio_net_hdr_from_skb(skb, &gso, - tun_is_little_endian(tun))) { + tun_is_little_endian(tun), true)) { struct skb_shared_info *sinfo = skb_shinfo(skb); pr_err("unexpected GSO type: " "0x%x, gso_size %d, hdr_len %d\n", diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 4a105006ca63..347424351ade 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -1104,7 +1104,7 @@ static int xmit_skb(struct send_queue *sq, struct sk_buff *skb) hdr = skb_vnet_hdr(skb); if (virtio_net_hdr_from_skb(skb, &hdr->hdr, - virtio_is_little_endian(vi->vdev))) + virtio_is_little_endian(vi->vdev), false)) BUG(); if (vi->mergeable_rx_bufs) diff --git a/include/linux/virtio_net.h b/include/linux/virtio_net.h index 56436472ccc7..5209b5ed2a64 100644 --- a/include/linux/virtio_net.h +++ b/include/linux/virtio_net.h @@ -56,7 +56,8 @@ static inline int virtio_net_hdr_to_skb(struct sk_buff *skb, static inline int virtio_net_hdr_from_skb(const struct sk_buff *skb, struct virtio_net_hdr *hdr, - bool little_endian) + bool little_endian, + bool has_data_valid) { memset(hdr, 0, sizeof(*hdr)); /* no info leak */ @@ -91,6 +92,9 @@ static inline int virtio_net_hdr_from_skb(const struct sk_buff *skb, skb_checksum_start_offset(skb)); hdr->csum_offset = __cpu_to_virtio16(little_endian, skb->csum_offset); + } else if (has_data_valid && + skb->ip_summed == CHECKSUM_UNNECESSARY) { + hdr->flags = VIRTIO_NET_HDR_F_DATA_VALID; } /* else everything is zero */ return 0; diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index b9e1a13b4ba3..3d555c79a7b5 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1976,7 +1976,7 @@ static int packet_rcv_vnet(struct msghdr *msg, const struct sk_buff *skb, return -EINVAL; *len -= sizeof(vnet_hdr); - if (virtio_net_hdr_from_skb(skb, &vnet_hdr, vio_le())) + if (virtio_net_hdr_from_skb(skb, &vnet_hdr, vio_le(), true)) return -EINVAL; return memcpy_to_msg(msg, (void *)&vnet_hdr, sizeof(vnet_hdr)); @@ -2237,7 +2237,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, if (po->has_vnet_hdr) { if (virtio_net_hdr_from_skb(skb, h.raw + macoff - sizeof(struct virtio_net_hdr), - vio_le())) { + vio_le(), true)) { spin_lock(&sk->sk_receive_queue.lock); goto drop_n_account; } -- cgit v1.2.3 From 2c561b2b728ca4013e76d6439bde2c137503745e Mon Sep 17 00:00:00 2001 From: hayeswang Date: Fri, 20 Jan 2017 14:33:55 +0800 Subject: r8152: fix rtl8152_post_reset function The rtl8152_post_reset() should sumbit rx urb and interrupt transfer, otherwise the rx wouldn't work and the linking change couldn't be detected. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller --- drivers/net/usb/r8152.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index f3b48ad90865..0e99af090734 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -3545,12 +3545,14 @@ static int rtl8152_post_reset(struct usb_interface *intf) if (netif_carrier_ok(netdev)) { mutex_lock(&tp->control); tp->rtl_ops.enable(tp); + rtl_start_rx(tp); rtl8152_set_rx_mode(netdev); mutex_unlock(&tp->control); netif_wake_queue(netdev); } napi_enable(&tp->napi); + usb_submit_urb(tp->intr_urb, GFP_KERNEL); return 0; } -- cgit v1.2.3 From e363116b90906f326c9cde5473b4b9a99ba476df Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 20 Jan 2017 07:57:42 -0800 Subject: ipv6: seg6_genl_set_tunsrc() must check kmemdup() return value seg6_genl_get_tunsrc() and set_tun_src() do not handle tun_src being possibly NULL, so we must check kmemdup() return value and abort if it is NULL Fixes: 915d7e5e5930 ("ipv6: sr: add code base for control plane support of SR-IPv6") Signed-off-by: Eric Dumazet Cc: David Lebrun Acked-by: David Lebrun Signed-off-by: David S. Miller --- net/ipv6/seg6.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c index b172d85c650a..a855eb325b03 100644 --- a/net/ipv6/seg6.c +++ b/net/ipv6/seg6.c @@ -176,6 +176,8 @@ static int seg6_genl_set_tunsrc(struct sk_buff *skb, struct genl_info *info) val = nla_data(info->attrs[SEG6_ATTR_DST]); t_new = kmemdup(val, sizeof(*val), GFP_KERNEL); + if (!t_new) + return -ENOMEM; mutex_lock(&sdata->lock); -- cgit v1.2.3 From 0e73fc9a56f22f2eec4d2b2910c649f7af67b74d Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Fri, 20 Jan 2017 13:01:57 +0000 Subject: net: sctp: fix array overrun read on sctp_timer_tbl The comparison on the timeout can lead to an array overrun read on sctp_timer_tbl because of an off-by-one error. Fix this by using < instead of <= and also compare to the array size rather than SCTP_EVENT_TIMEOUT_MAX. Fixes CoverityScan CID#1397639 ("Out-of-bounds read") Signed-off-by: Colin Ian King Signed-off-by: David S. Miller --- net/sctp/debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sctp/debug.c b/net/sctp/debug.c index 95d7b15dad21..e371a0d90068 100644 --- a/net/sctp/debug.c +++ b/net/sctp/debug.c @@ -166,7 +166,7 @@ static const char *const sctp_timer_tbl[] = { /* Lookup timer debug name. */ const char *sctp_tname(const sctp_subtype_t id) { - if (id.timeout <= SCTP_EVENT_TIMEOUT_MAX) + if (id.timeout < ARRAY_SIZE(sctp_timer_tbl)) return sctp_timer_tbl[id.timeout]; return "unknown_timer"; } -- cgit v1.2.3 From 91e744653cb80554f3fdfd1d31c5ddf7b6169f37 Mon Sep 17 00:00:00 2001 From: David S. Miller Date: Fri, 20 Jan 2017 11:29:43 -0500 Subject: Revert "net: sctp: fix array overrun read on sctp_timer_tbl" This reverts commit 0e73fc9a56f22f2eec4d2b2910c649f7af67b74d. This fix wasn't correct, a better one is coming right up. Signed-off-by: David S. Miller --- net/sctp/debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sctp/debug.c b/net/sctp/debug.c index e371a0d90068..95d7b15dad21 100644 --- a/net/sctp/debug.c +++ b/net/sctp/debug.c @@ -166,7 +166,7 @@ static const char *const sctp_timer_tbl[] = { /* Lookup timer debug name. */ const char *sctp_tname(const sctp_subtype_t id) { - if (id.timeout < ARRAY_SIZE(sctp_timer_tbl)) + if (id.timeout <= SCTP_EVENT_TIMEOUT_MAX) return sctp_timer_tbl[id.timeout]; return "unknown_timer"; } -- cgit v1.2.3 From 0629a330cf55454962168dd3ee46fad53a39323e Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 18 Jan 2017 15:52:52 +0100 Subject: qed: avoid possible stack overflow in qed_ll2_acquire_connection struct qed_ll2_info is rather large, so putting it on the stack can cause an overflow, as this warning tries to tell us: drivers/net/ethernet/qlogic/qed/qed_ll2.c: In function 'qed_ll2_start': drivers/net/ethernet/qlogic/qed/qed_ll2.c:2159:1: error: the frame size of 1056 bytes is larger than 1024 bytes [-Werror=frame-larger-than=] qed_ll2_start_ooo() already uses a dynamic allocation for the structure to work around that problem, and we could do the same in qed_ll2_start() as well as qed_roce_ll2_start(), but since the structure is only used to pass a couple of initialization values here, it seems nicer to replace it with a different structure. Lacking any idea for better naming, I'm adding 'struct qed_ll2_conn', which now contains all the initialization data, and this now simply gets copied into struct qed_ll2_info rather than assigning all members one by one. Signed-off-by: Arnd Bergmann Acked-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_ll2.c | 88 +++++++++++++----------------- drivers/net/ethernet/qlogic/qed/qed_ll2.h | 24 ++++---- drivers/net/ethernet/qlogic/qed/qed_roce.c | 2 +- 3 files changed, 53 insertions(+), 61 deletions(-) diff --git a/drivers/net/ethernet/qlogic/qed/qed_ll2.c b/drivers/net/ethernet/qlogic/qed/qed_ll2.c index 8e5cb7605b0f..873ce2cd76ba 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_ll2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_ll2.c @@ -297,7 +297,7 @@ static void qed_ll2_txq_flush(struct qed_hwfn *p_hwfn, u8 connection_handle) list_del(&p_pkt->list_entry); b_last_packet = list_empty(&p_tx->active_descq); list_add_tail(&p_pkt->list_entry, &p_tx->free_descq); - if (p_ll2_conn->conn_type == QED_LL2_TYPE_ISCSI_OOO) { + if (p_ll2_conn->conn.conn_type == QED_LL2_TYPE_ISCSI_OOO) { struct qed_ooo_buffer *p_buffer; p_buffer = (struct qed_ooo_buffer *)p_pkt->cookie; @@ -309,7 +309,7 @@ static void qed_ll2_txq_flush(struct qed_hwfn *p_hwfn, u8 connection_handle) b_last_frag = p_tx->cur_completing_bd_idx == p_pkt->bd_used; tx_frag = p_pkt->bds_set[0].tx_frag; - if (p_ll2_conn->gsi_enable) + if (p_ll2_conn->conn.gsi_enable) qed_ll2b_release_tx_gsi_packet(p_hwfn, p_ll2_conn-> my_id, @@ -378,7 +378,7 @@ static int qed_ll2_txq_completion(struct qed_hwfn *p_hwfn, void *p_cookie) spin_unlock_irqrestore(&p_tx->lock, flags); tx_frag = p_pkt->bds_set[0].tx_frag; - if (p_ll2_conn->gsi_enable) + if (p_ll2_conn->conn.gsi_enable) qed_ll2b_complete_tx_gsi_packet(p_hwfn, p_ll2_conn->my_id, p_pkt->cookie, @@ -550,7 +550,7 @@ static void qed_ll2_rxq_flush(struct qed_hwfn *p_hwfn, u8 connection_handle) list_move_tail(&p_pkt->list_entry, &p_rx->free_descq); - if (p_ll2_conn->conn_type == QED_LL2_TYPE_ISCSI_OOO) { + if (p_ll2_conn->conn.conn_type == QED_LL2_TYPE_ISCSI_OOO) { struct qed_ooo_buffer *p_buffer; p_buffer = (struct qed_ooo_buffer *)p_pkt->cookie; @@ -738,7 +738,7 @@ qed_ooo_submit_tx_buffers(struct qed_hwfn *p_hwfn, rc = qed_ll2_prepare_tx_packet(p_hwfn, p_ll2_conn->my_id, 1, p_buffer->vlan, bd_flags, l4_hdr_offset_w, - p_ll2_conn->tx_dest, 0, + p_ll2_conn->conn.tx_dest, 0, first_frag, p_buffer->packet_length, p_buffer, true); @@ -858,7 +858,7 @@ qed_ll2_acquire_connection_ooo(struct qed_hwfn *p_hwfn, u16 buf_idx; int rc = 0; - if (p_ll2_info->conn_type != QED_LL2_TYPE_ISCSI_OOO) + if (p_ll2_info->conn.conn_type != QED_LL2_TYPE_ISCSI_OOO) return rc; if (!rx_num_ooo_buffers) @@ -901,7 +901,7 @@ static void qed_ll2_establish_connection_ooo(struct qed_hwfn *p_hwfn, struct qed_ll2_info *p_ll2_conn) { - if (p_ll2_conn->conn_type != QED_LL2_TYPE_ISCSI_OOO) + if (p_ll2_conn->conn.conn_type != QED_LL2_TYPE_ISCSI_OOO) return; qed_ooo_release_all_isles(p_hwfn, p_hwfn->p_ooo_info); @@ -913,7 +913,7 @@ static void qed_ll2_release_connection_ooo(struct qed_hwfn *p_hwfn, { struct qed_ooo_buffer *p_buffer; - if (p_ll2_conn->conn_type != QED_LL2_TYPE_ISCSI_OOO) + if (p_ll2_conn->conn.conn_type != QED_LL2_TYPE_ISCSI_OOO) return; qed_ooo_release_all_isles(p_hwfn, p_hwfn->p_ooo_info); @@ -945,23 +945,19 @@ static int qed_ll2_start_ooo(struct qed_dev *cdev, { struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); u8 *handle = &hwfn->pf_params.iscsi_pf_params.ll2_ooo_queue_id; - struct qed_ll2_info *ll2_info; + struct qed_ll2_conn ll2_info; int rc; - ll2_info = kzalloc(sizeof(*ll2_info), GFP_KERNEL); - if (!ll2_info) - return -ENOMEM; - ll2_info->conn_type = QED_LL2_TYPE_ISCSI_OOO; - ll2_info->mtu = params->mtu; - ll2_info->rx_drop_ttl0_flg = params->drop_ttl0_packets; - ll2_info->rx_vlan_removal_en = params->rx_vlan_stripping; - ll2_info->tx_tc = OOO_LB_TC; - ll2_info->tx_dest = CORE_TX_DEST_LB; - - rc = qed_ll2_acquire_connection(hwfn, ll2_info, + ll2_info.conn_type = QED_LL2_TYPE_ISCSI_OOO; + ll2_info.mtu = params->mtu; + ll2_info.rx_drop_ttl0_flg = params->drop_ttl0_packets; + ll2_info.rx_vlan_removal_en = params->rx_vlan_stripping; + ll2_info.tx_tc = OOO_LB_TC; + ll2_info.tx_dest = CORE_TX_DEST_LB; + + rc = qed_ll2_acquire_connection(hwfn, &ll2_info, QED_LL2_RX_SIZE, QED_LL2_TX_SIZE, handle); - kfree(ll2_info); if (rc) { DP_INFO(cdev, "Failed to acquire LL2 OOO connection\n"); goto out; @@ -1006,7 +1002,7 @@ static int qed_sp_ll2_rx_queue_start(struct qed_hwfn *p_hwfn, struct qed_ll2_info *p_ll2_conn, u8 action_on_error) { - enum qed_ll2_conn_type conn_type = p_ll2_conn->conn_type; + enum qed_ll2_conn_type conn_type = p_ll2_conn->conn.conn_type; struct qed_ll2_rx_queue *p_rx = &p_ll2_conn->rx_queue; struct core_rx_start_ramrod_data *p_ramrod = NULL; struct qed_spq_entry *p_ent = NULL; @@ -1032,7 +1028,7 @@ static int qed_sp_ll2_rx_queue_start(struct qed_hwfn *p_hwfn, p_ramrod->sb_index = p_rx->rx_sb_index; p_ramrod->complete_event_flg = 1; - p_ramrod->mtu = cpu_to_le16(p_ll2_conn->mtu); + p_ramrod->mtu = cpu_to_le16(p_ll2_conn->conn.mtu); DMA_REGPAIR_LE(p_ramrod->bd_base, p_rx->rxq_chain.p_phys_addr); cqe_pbl_size = (u16)qed_chain_get_page_cnt(&p_rx->rcq_chain); @@ -1040,8 +1036,8 @@ static int qed_sp_ll2_rx_queue_start(struct qed_hwfn *p_hwfn, DMA_REGPAIR_LE(p_ramrod->cqe_pbl_addr, qed_chain_get_pbl_phys(&p_rx->rcq_chain)); - p_ramrod->drop_ttl0_flg = p_ll2_conn->rx_drop_ttl0_flg; - p_ramrod->inner_vlan_removal_en = p_ll2_conn->rx_vlan_removal_en; + p_ramrod->drop_ttl0_flg = p_ll2_conn->conn.rx_drop_ttl0_flg; + p_ramrod->inner_vlan_removal_en = p_ll2_conn->conn.rx_vlan_removal_en; p_ramrod->queue_id = p_ll2_conn->queue_id; p_ramrod->main_func_queue = (conn_type == QED_LL2_TYPE_ISCSI_OOO) ? 0 : 1; @@ -1056,14 +1052,14 @@ static int qed_sp_ll2_rx_queue_start(struct qed_hwfn *p_hwfn, } p_ramrod->action_on_error.error_type = action_on_error; - p_ramrod->gsi_offload_flag = p_ll2_conn->gsi_enable; + p_ramrod->gsi_offload_flag = p_ll2_conn->conn.gsi_enable; return qed_spq_post(p_hwfn, p_ent, NULL); } static int qed_sp_ll2_tx_queue_start(struct qed_hwfn *p_hwfn, struct qed_ll2_info *p_ll2_conn) { - enum qed_ll2_conn_type conn_type = p_ll2_conn->conn_type; + enum qed_ll2_conn_type conn_type = p_ll2_conn->conn.conn_type; struct qed_ll2_tx_queue *p_tx = &p_ll2_conn->tx_queue; struct core_tx_start_ramrod_data *p_ramrod = NULL; struct qed_spq_entry *p_ent = NULL; @@ -1075,7 +1071,7 @@ static int qed_sp_ll2_tx_queue_start(struct qed_hwfn *p_hwfn, if (!QED_LL2_TX_REGISTERED(p_ll2_conn)) return 0; - if (p_ll2_conn->conn_type == QED_LL2_TYPE_ISCSI_OOO) + if (p_ll2_conn->conn.conn_type == QED_LL2_TYPE_ISCSI_OOO) p_ll2_conn->tx_stats_en = 0; else p_ll2_conn->tx_stats_en = 1; @@ -1096,7 +1092,7 @@ static int qed_sp_ll2_tx_queue_start(struct qed_hwfn *p_hwfn, p_ramrod->sb_id = cpu_to_le16(qed_int_get_sp_sb_id(p_hwfn)); p_ramrod->sb_index = p_tx->tx_sb_index; - p_ramrod->mtu = cpu_to_le16(p_ll2_conn->mtu); + p_ramrod->mtu = cpu_to_le16(p_ll2_conn->conn.mtu); p_ramrod->stats_en = p_ll2_conn->tx_stats_en; p_ramrod->stats_id = p_ll2_conn->tx_stats_id; @@ -1106,7 +1102,7 @@ static int qed_sp_ll2_tx_queue_start(struct qed_hwfn *p_hwfn, p_ramrod->pbl_size = cpu_to_le16(pbl_size); memset(&pq_params, 0, sizeof(pq_params)); - pq_params.core.tc = p_ll2_conn->tx_tc; + pq_params.core.tc = p_ll2_conn->conn.tx_tc; pq_id = qed_get_qm_pq(p_hwfn, PROTOCOLID_CORE, &pq_params); p_ramrod->qm_pq_id = cpu_to_le16(pq_id); @@ -1123,7 +1119,7 @@ static int qed_sp_ll2_tx_queue_start(struct qed_hwfn *p_hwfn, DP_NOTICE(p_hwfn, "Unknown connection type: %d\n", conn_type); } - p_ramrod->gsi_offload_flag = p_ll2_conn->gsi_enable; + p_ramrod->gsi_offload_flag = p_ll2_conn->conn.gsi_enable; return qed_spq_post(p_hwfn, p_ent, NULL); } @@ -1224,7 +1220,7 @@ qed_ll2_acquire_connection_rx(struct qed_hwfn *p_hwfn, DP_VERBOSE(p_hwfn, QED_MSG_LL2, "Allocated LL2 Rxq [Type %08x] with 0x%08x buffers\n", - p_ll2_info->conn_type, rx_num_desc); + p_ll2_info->conn.conn_type, rx_num_desc); out: return rc; @@ -1262,7 +1258,7 @@ static int qed_ll2_acquire_connection_tx(struct qed_hwfn *p_hwfn, DP_VERBOSE(p_hwfn, QED_MSG_LL2, "Allocated LL2 Txq [Type %08x] with 0x%08x buffers\n", - p_ll2_info->conn_type, tx_num_desc); + p_ll2_info->conn.conn_type, tx_num_desc); out: if (rc) @@ -1273,7 +1269,7 @@ out: } int qed_ll2_acquire_connection(struct qed_hwfn *p_hwfn, - struct qed_ll2_info *p_params, + struct qed_ll2_conn *p_params, u16 rx_num_desc, u16 tx_num_desc, u8 *p_connection_handle) @@ -1302,15 +1298,7 @@ int qed_ll2_acquire_connection(struct qed_hwfn *p_hwfn, if (!p_ll2_info) return -EBUSY; - p_ll2_info->conn_type = p_params->conn_type; - p_ll2_info->mtu = p_params->mtu; - p_ll2_info->rx_drop_ttl0_flg = p_params->rx_drop_ttl0_flg; - p_ll2_info->rx_vlan_removal_en = p_params->rx_vlan_removal_en; - p_ll2_info->tx_tc = p_params->tx_tc; - p_ll2_info->tx_dest = p_params->tx_dest; - p_ll2_info->ai_err_packet_too_big = p_params->ai_err_packet_too_big; - p_ll2_info->ai_err_no_buf = p_params->ai_err_no_buf; - p_ll2_info->gsi_enable = p_params->gsi_enable; + p_ll2_info->conn = *p_params; rc = qed_ll2_acquire_connection_rx(p_hwfn, p_ll2_info, rx_num_desc); if (rc) @@ -1371,9 +1359,9 @@ static int qed_ll2_establish_connection_rx(struct qed_hwfn *p_hwfn, SET_FIELD(action_on_error, CORE_RX_ACTION_ON_ERROR_PACKET_TOO_BIG, - p_ll2_conn->ai_err_packet_too_big); + p_ll2_conn->conn.ai_err_packet_too_big); SET_FIELD(action_on_error, - CORE_RX_ACTION_ON_ERROR_NO_BUFF, p_ll2_conn->ai_err_no_buf); + CORE_RX_ACTION_ON_ERROR_NO_BUFF, p_ll2_conn->conn.ai_err_no_buf); return qed_sp_ll2_rx_queue_start(p_hwfn, p_ll2_conn, action_on_error); } @@ -1600,7 +1588,7 @@ static void qed_ll2_prepare_tx_packet_set_bd(struct qed_hwfn *p_hwfn, "LL2 [q 0x%02x cid 0x%08x type 0x%08x] Tx Producer at [0x%04x] - set with a %04x bytes %02x BDs buffer at %08x:%08x\n", p_ll2->queue_id, p_ll2->cid, - p_ll2->conn_type, + p_ll2->conn.conn_type, prod_idx, first_frag_len, num_of_bds, @@ -1676,7 +1664,7 @@ static void qed_ll2_tx_packet_notify(struct qed_hwfn *p_hwfn, (NETIF_MSG_TX_QUEUED | QED_MSG_LL2), "LL2 [q 0x%02x cid 0x%08x type 0x%08x] Doorbelled [producer 0x%04x]\n", p_ll2_conn->queue_id, - p_ll2_conn->cid, p_ll2_conn->conn_type, db_msg.spq_prod); + p_ll2_conn->cid, p_ll2_conn->conn.conn_type, db_msg.spq_prod); } int qed_ll2_prepare_tx_packet(struct qed_hwfn *p_hwfn, @@ -1817,7 +1805,7 @@ int qed_ll2_terminate_connection(struct qed_hwfn *p_hwfn, u8 connection_handle) qed_ll2_rxq_flush(p_hwfn, connection_handle); } - if (p_ll2_conn->conn_type == QED_LL2_TYPE_ISCSI_OOO) + if (p_ll2_conn->conn.conn_type == QED_LL2_TYPE_ISCSI_OOO) qed_ooo_release_all_isles(p_hwfn, p_hwfn->p_ooo_info); return rc; @@ -1993,7 +1981,7 @@ static void qed_ll2_register_cb_ops(struct qed_dev *cdev, static int qed_ll2_start(struct qed_dev *cdev, struct qed_ll2_params *params) { - struct qed_ll2_info ll2_info; + struct qed_ll2_conn ll2_info; struct qed_ll2_buffer *buffer, *tmp_buffer; enum qed_ll2_conn_type conn_type; struct qed_ptt *p_ptt; @@ -2041,6 +2029,7 @@ static int qed_ll2_start(struct qed_dev *cdev, struct qed_ll2_params *params) /* Prepare the temporary ll2 information */ memset(&ll2_info, 0, sizeof(ll2_info)); + ll2_info.conn_type = conn_type; ll2_info.mtu = params->mtu; ll2_info.rx_drop_ttl0_flg = params->drop_ttl0_packets; @@ -2120,7 +2109,6 @@ static int qed_ll2_start(struct qed_dev *cdev, struct qed_ll2_params *params) } ether_addr_copy(cdev->ll2_mac_address, params->ll2_mac_address); - return 0; release_terminate_all: diff --git a/drivers/net/ethernet/qlogic/qed/qed_ll2.h b/drivers/net/ethernet/qlogic/qed/qed_ll2.h index 6625a3ae5a33..31417928b635 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_ll2.h +++ b/drivers/net/ethernet/qlogic/qed/qed_ll2.h @@ -112,15 +112,8 @@ struct qed_ll2_tx_queue { bool b_completing_packet; }; -struct qed_ll2_info { - /* Lock protecting the state of LL2 */ - struct mutex mutex; +struct qed_ll2_conn { enum qed_ll2_conn_type conn_type; - u32 cid; - u8 my_id; - u8 queue_id; - u8 tx_stats_id; - bool b_active; u16 mtu; u8 rx_drop_ttl0_flg; u8 rx_vlan_removal_en; @@ -128,10 +121,21 @@ struct qed_ll2_info { enum core_tx_dest tx_dest; enum core_error_handle ai_err_packet_too_big; enum core_error_handle ai_err_no_buf; + u8 gsi_enable; +}; + +struct qed_ll2_info { + /* Lock protecting the state of LL2 */ + struct mutex mutex; + struct qed_ll2_conn conn; + u32 cid; + u8 my_id; + u8 queue_id; + u8 tx_stats_id; + bool b_active; u8 tx_stats_en; struct qed_ll2_rx_queue rx_queue; struct qed_ll2_tx_queue tx_queue; - u8 gsi_enable; }; /** @@ -149,7 +153,7 @@ struct qed_ll2_info { * @return 0 on success, failure otherwise */ int qed_ll2_acquire_connection(struct qed_hwfn *p_hwfn, - struct qed_ll2_info *p_params, + struct qed_ll2_conn *p_params, u16 rx_num_desc, u16 tx_num_desc, u8 *p_connection_handle); diff --git a/drivers/net/ethernet/qlogic/qed/qed_roce.c b/drivers/net/ethernet/qlogic/qed/qed_roce.c index 2a16547c8966..2dbdb3298991 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_roce.c +++ b/drivers/net/ethernet/qlogic/qed/qed_roce.c @@ -2632,7 +2632,7 @@ static int qed_roce_ll2_start(struct qed_dev *cdev, { struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); struct qed_roce_ll2_info *roce_ll2; - struct qed_ll2_info ll2_params; + struct qed_ll2_conn ll2_params; int rc; if (!params) { -- cgit v1.2.3 From df384d435a5c034c735df3d9ea87a03172c59b56 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 18 Jan 2017 15:52:53 +0100 Subject: bcm63xx_enet: avoid uninitialized variable warning gcc-7 and probably earlier versions get confused by this function and print a harmless warning: drivers/net/ethernet/broadcom/bcm63xx_enet.c: In function 'bcm_enet_open': drivers/net/ethernet/broadcom/bcm63xx_enet.c:1130:3: error: 'phydev' may be used uninitialized in this function [-Werror=maybe-uninitialized] This adds an initialization for the 'phydev' variable when it is unused and changes the check to test for that NULL pointer to make it clear that we always pass a valid pointer here. Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bcm63xx_enet.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c index 3b14d5144228..c483618b57bd 100644 --- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c +++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c @@ -913,6 +913,8 @@ static int bcm_enet_open(struct net_device *dev) priv->old_link = 0; priv->old_duplex = -1; priv->old_pause = -1; + } else { + phydev = NULL; } /* mask all interrupts and request them */ @@ -1083,7 +1085,7 @@ static int bcm_enet_open(struct net_device *dev) enet_dmac_writel(priv, priv->dma_chan_int_mask, ENETDMAC_IRMASK, priv->tx_chan); - if (priv->has_phy) + if (phydev) phy_start(phydev); else bcm_enet_adjust_link(dev); @@ -1126,7 +1128,7 @@ out_freeirq: free_irq(dev->irq, dev); out_phy_disconnect: - if (priv->has_phy) + if (phydev) phy_disconnect(phydev); return ret; -- cgit v1.2.3 From cdb749cef16bceb74950fc8668f2632ff7cac9aa Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Wed, 18 Jan 2017 17:19:00 +0100 Subject: bpf: fix samples xdp_tx_iptunnel and tc_l2_redirect with fake KBUILD_MODNAME Fix build errors for samples/bpf xdp_tx_iptunnel and tc_l2_redirect, when dynamic debugging is enabled (CONFIG_DYNAMIC_DEBUG) by defining a fake KBUILD_MODNAME. Just like Daniel Borkmann fixed other samples/bpf in commit 96a8eb1eeed2 ("bpf: fix samples to add fake KBUILD_MODNAME"). Fixes: 12d8bb64e3f6 ("bpf: xdp: Add XDP example for head adjustment") Fixes: 90e02896f1a4 ("bpf: Add test for bpf_redirect to ipip/ip6tnl") Signed-off-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller --- samples/bpf/tc_l2_redirect_kern.c | 1 + samples/bpf/xdp_tx_iptunnel_kern.c | 1 + 2 files changed, 2 insertions(+) diff --git a/samples/bpf/tc_l2_redirect_kern.c b/samples/bpf/tc_l2_redirect_kern.c index 92a44729dbe4..7ef2a12b25b2 100644 --- a/samples/bpf/tc_l2_redirect_kern.c +++ b/samples/bpf/tc_l2_redirect_kern.c @@ -4,6 +4,7 @@ * modify it under the terms of version 2 of the GNU General Public * License as published by the Free Software Foundation. */ +#define KBUILD_MODNAME "foo" #include #include #include diff --git a/samples/bpf/xdp_tx_iptunnel_kern.c b/samples/bpf/xdp_tx_iptunnel_kern.c index 85c38ecd3a2d..0f4f6e8c8611 100644 --- a/samples/bpf/xdp_tx_iptunnel_kern.c +++ b/samples/bpf/xdp_tx_iptunnel_kern.c @@ -8,6 +8,7 @@ * encapsulating the incoming packet in an IPv4/v6 header * and then XDP_TX it out. */ +#define KBUILD_MODNAME "foo" #include #include #include -- cgit v1.2.3 From e048fc50d7bde23136e098e04a324d7e3404408d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 18 Jan 2017 23:03:08 -0800 Subject: net/mlx5e: Do not recycle pages from emergency reserve A driver using dev_alloc_page() must not reuse a page allocated from emergency memory reserve. Otherwise all packets using this page will be immediately dropped, unless for very specific sockets having SOCK_MEMALLOC bit set. This issue might be hard to debug, because only a fraction of received packets would be dropped. Fixes: 4415a0319f92 ("net/mlx5e: Implement RX mapped page cache for page recycle") Signed-off-by: Eric Dumazet Cc: Tariq Toukan Cc: Saeed Mahameed Acked-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 0e2fb3ed1790..06d5e6fecb0a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -193,6 +193,9 @@ static inline bool mlx5e_rx_cache_put(struct mlx5e_rq *rq, return false; } + if (unlikely(page_is_pfmemalloc(dma_info->page))) + return false; + cache->page_cache[cache->tail] = *dma_info; cache->tail = tail_next; return true; -- cgit v1.2.3 From 90c311b0eeead647b708a723dbdde1eda3dcad05 Mon Sep 17 00:00:00 2001 From: Vineeth Remanan Pillai Date: Thu, 19 Jan 2017 08:35:39 -0800 Subject: xen-netfront: Fix Rx stall during network stress and OOM During an OOM scenario, request slots could not be created as skb allocation fails. So the netback cannot pass in packets and netfront wrongly assumes that there is no more work to be done and it disables polling. This causes Rx to stall. The issue is with the retry logic which schedules the timer if the created slots are less than NET_RX_SLOTS_MIN. The count of new request slots to be pushed are calculated as a difference between new req_prod and rsp_cons which could be more than the actual slots, if there are unconsumed responses. The fix is to calculate the count of newly created slots as the difference between new req_prod and old req_prod. Signed-off-by: Vineeth Remanan Pillai Reviewed-by: Juergen Gross Signed-off-by: David S. Miller --- drivers/net/xen-netfront.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index a479cd99911d..8315fe73ecd0 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -321,7 +321,7 @@ static void xennet_alloc_rx_buffers(struct netfront_queue *queue) queue->rx.req_prod_pvt = req_prod; /* Not enough requests? Try again later. */ - if (req_prod - queue->rx.rsp_cons < NET_RX_SLOTS_MIN) { + if (req_prod - queue->rx.sring->req_prod < NET_RX_SLOTS_MIN) { mod_timer(&queue->rx_refill_timer, jiffies + (HZ/10)); return; } -- cgit v1.2.3 From b6677449dff674cf5b81429b11d5c7f358852ef9 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Fri, 20 Jan 2017 18:12:17 +0100 Subject: bridge: netlink: call br_changelink() during br_dev_newlink() Any bridge options specified during link creation (e.g. ip link add) are ignored as br_dev_newlink() does not process them. Use br_changelink() to do it. Fixes: 133235161721 ("bridge: implement rtnl_link_ops->changelink") Signed-off-by: Ivan Vecera Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- net/bridge/br_netlink.c | 33 +++++++++++++++++++-------------- 1 file changed, 19 insertions(+), 14 deletions(-) diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 71c7453268c1..7109b389ea58 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -781,20 +781,6 @@ static int br_validate(struct nlattr *tb[], struct nlattr *data[]) return 0; } -static int br_dev_newlink(struct net *src_net, struct net_device *dev, - struct nlattr *tb[], struct nlattr *data[]) -{ - struct net_bridge *br = netdev_priv(dev); - - if (tb[IFLA_ADDRESS]) { - spin_lock_bh(&br->lock); - br_stp_change_bridge_id(br, nla_data(tb[IFLA_ADDRESS])); - spin_unlock_bh(&br->lock); - } - - return register_netdevice(dev); -} - static int br_port_slave_changelink(struct net_device *brdev, struct net_device *dev, struct nlattr *tb[], @@ -1115,6 +1101,25 @@ static int br_changelink(struct net_device *brdev, struct nlattr *tb[], return 0; } +static int br_dev_newlink(struct net *src_net, struct net_device *dev, + struct nlattr *tb[], struct nlattr *data[]) +{ + struct net_bridge *br = netdev_priv(dev); + int err; + + if (tb[IFLA_ADDRESS]) { + spin_lock_bh(&br->lock); + br_stp_change_bridge_id(br, nla_data(tb[IFLA_ADDRESS])); + spin_unlock_bh(&br->lock); + } + + err = br_changelink(dev, tb, data); + if (err) + return err; + + return register_netdevice(dev); +} + static size_t br_get_size(const struct net_device *brdev) { return nla_total_size(sizeof(u32)) + /* IFLA_BR_FORWARD_DELAY */ -- cgit v1.2.3 From 773c7220e22d193e5667c352fcbf8d47eefc817f Mon Sep 17 00:00:00 2001 From: Eric Farman Date: Fri, 13 Jan 2017 12:48:06 -0500 Subject: scsi: virtio_scsi: Reject commands when virtqueue is broken In the case of a graceful set of detaches, where the virtio-scsi-ccw disk is removed from the guest prior to the controller, the guest behaves quite normally. Specifically, the detach gets us into sd_sync_cache to issue a Synchronize Cache(10) command, which immediately fails (and is retried a couple of times) because the device has been removed. Later, the removal of the controller sees two CRWs presented, but there's no further indication of the removal from the guest viewpoint. [ 17.217458] sd 0:0:0:0: [sda] Synchronizing SCSI cache [ 17.219257] sd 0:0:0:0: [sda] Synchronize Cache(10) failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK [ 21.449400] crw_info : CRW reports slct=0, oflw=0, chn=1, rsc=3, anc=0, erc=4, rsid=2 [ 21.449406] crw_info : CRW reports slct=0, oflw=0, chn=0, rsc=3, anc=0, erc=4, rsid=0 However, on s390, the SCSI disks can be removed "by surprise" when an entire controller (host) is removed and all associated disks are removed via the loop in scsi_forget_host. The same call to sd_sync_cache is made, but because the controller has already been removed, the Synchronize Cache(10) command is neither issued (and then failed) nor rejected. That the I/O isn't returned means the guest cannot have other devices added nor removed, and other tasks (such as shutdown or reboot) issued by the guest will not complete either. The virtio ring has already been marked as broken (via virtio_break_device in virtio_ccw_remove), but we still attempt to queue the command only to have it remain there. The calling sequence provides a bit of distinction for us: virtscsi_queuecommand() -> virtscsi_kick_cmd() -> virtscsi_add_cmd() -> virtqueue_add_sgs() -> virtqueue_add() if success return 0 elseif vq->broken or vring_mapping_error() return -EIO else return -ENOSPC A return of ENOSPC is generally a temporary condition, so returning "host busy" from virtscsi_queuecommand makes sense here, to have it redriven in a moment or two. But the EIO return code is more of a permanent error and so it would be wise to return the I/O itself and allow the calling thread to finish gracefully. The result is these four kernel messages in the guest (the fourth one does not occur prior to this patch): [ 22.921562] crw_info : CRW reports slct=0, oflw=0, chn=1, rsc=3, anc=0, erc=4, rsid=2 [ 22.921580] crw_info : CRW reports slct=0, oflw=0, chn=0, rsc=3, anc=0, erc=4, rsid=0 [ 22.921978] sd 0:0:0:0: [sda] Synchronizing SCSI cache [ 22.921993] sd 0:0:0:0: [sda] Synchronize Cache(10) failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK I opted to fill in the same response data that is returned from the more graceful device detach, where the disk device is removed prior to the controller device. Signed-off-by: Eric Farman Reviewed-by: Fam Zheng Signed-off-by: Martin K. Petersen --- drivers/scsi/virtio_scsi.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c index ec91bd07f00a..c680d7641311 100644 --- a/drivers/scsi/virtio_scsi.c +++ b/drivers/scsi/virtio_scsi.c @@ -534,7 +534,9 @@ static int virtscsi_queuecommand(struct virtio_scsi *vscsi, { struct Scsi_Host *shost = virtio_scsi_host(vscsi->vdev); struct virtio_scsi_cmd *cmd = scsi_cmd_priv(sc); + unsigned long flags; int req_size; + int ret; BUG_ON(scsi_sg_count(sc) > shost->sg_tablesize); @@ -562,8 +564,15 @@ static int virtscsi_queuecommand(struct virtio_scsi *vscsi, req_size = sizeof(cmd->req.cmd); } - if (virtscsi_kick_cmd(req_vq, cmd, req_size, sizeof(cmd->resp.cmd)) != 0) + ret = virtscsi_kick_cmd(req_vq, cmd, req_size, sizeof(cmd->resp.cmd)); + if (ret == -EIO) { + cmd->resp.cmd.response = VIRTIO_SCSI_S_BAD_TARGET; + spin_lock_irqsave(&req_vq->vq_lock, flags); + virtscsi_complete_cmd(vscsi, cmd); + spin_unlock_irqrestore(&req_vq->vq_lock, flags); + } else if (ret != 0) { return SCSI_MLQUEUE_HOST_BUSY; + } return 0; } -- cgit v1.2.3 From 92549cdc288f47f3a98cf80ac5890c91f5876a06 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 17 Jan 2017 14:22:24 -0800 Subject: iwlwifi: fix kernel crash when unregistering thermal zone A recent firmware change seems to have enabled thermal zones on the iwlwifi driver. Unfortunately, my device fails when registering the thermal zone. This doesn't stop the driver from attempting to unregister the thermal zone at unload time, triggering a NULL pointer deference in strlen() off the thermal_zone_device_unregister() path. Don't unregister if name is NULL, for that case we failed registering. Do the same for the cooling zone. Signed-off-by: Jens Axboe Signed-off-by: Kalle Valo --- drivers/net/wireless/intel/iwlwifi/mvm/tt.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c index 63a051be832e..bec7d9c46087 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c @@ -843,8 +843,10 @@ static void iwl_mvm_thermal_zone_unregister(struct iwl_mvm *mvm) return; IWL_DEBUG_TEMP(mvm, "Thermal zone device unregister\n"); - thermal_zone_device_unregister(mvm->tz_device.tzone); - mvm->tz_device.tzone = NULL; + if (mvm->tz_device.tzone) { + thermal_zone_device_unregister(mvm->tz_device.tzone); + mvm->tz_device.tzone = NULL; + } } static void iwl_mvm_cooling_device_unregister(struct iwl_mvm *mvm) @@ -853,8 +855,10 @@ static void iwl_mvm_cooling_device_unregister(struct iwl_mvm *mvm) return; IWL_DEBUG_TEMP(mvm, "Cooling device unregister\n"); - thermal_cooling_device_unregister(mvm->cooling_dev.cdev); - mvm->cooling_dev.cdev = NULL; + if (mvm->cooling_dev.cdev) { + thermal_cooling_device_unregister(mvm->cooling_dev.cdev); + mvm->cooling_dev.cdev = NULL; + } } #endif /* CONFIG_THERMAL */ -- cgit v1.2.3 From a5badd1e97e6caeca78ad74191f12fc923c403a8 Mon Sep 17 00:00:00 2001 From: Alison Schofield Date: Sat, 14 Jan 2017 19:51:52 -0800 Subject: iio: health: afe4403: retrieve a valid iio_dev in suspend/resume The suspend/resume functions were using dev_to_iio_dev() to get the iio_dev. That only works on IIO dev's. Replace it with spi functions to get the correct iio_dev. Signed-off-by: Alison Schofield Acked-by: Andrew F. Davis Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/health/afe4403.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/iio/health/afe4403.c b/drivers/iio/health/afe4403.c index 9a081465c42f..6bb23a49e81e 100644 --- a/drivers/iio/health/afe4403.c +++ b/drivers/iio/health/afe4403.c @@ -422,7 +422,7 @@ MODULE_DEVICE_TABLE(of, afe4403_of_match); static int __maybe_unused afe4403_suspend(struct device *dev) { - struct iio_dev *indio_dev = dev_to_iio_dev(dev); + struct iio_dev *indio_dev = spi_get_drvdata(to_spi_device(dev)); struct afe4403_data *afe = iio_priv(indio_dev); int ret; @@ -443,7 +443,7 @@ static int __maybe_unused afe4403_suspend(struct device *dev) static int __maybe_unused afe4403_resume(struct device *dev) { - struct iio_dev *indio_dev = dev_to_iio_dev(dev); + struct iio_dev *indio_dev = spi_get_drvdata(to_spi_device(dev)); struct afe4403_data *afe = iio_priv(indio_dev); int ret; -- cgit v1.2.3 From 802ecfc113df1e15af1d028427cbbe785ae9cc4a Mon Sep 17 00:00:00 2001 From: Alison Schofield Date: Sat, 14 Jan 2017 19:52:50 -0800 Subject: iio: health: afe4404: retrieve a valid iio_dev in suspend/resume The suspend/resume functions were using dev_to_iio_dev() to get the iio_dev. That only works on IIO dev's. Replace it with i2c functions to get the correct iio_dev. Signed-off-by: Alison Schofield Acked-by: Andrew F. Davis Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/health/afe4404.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/iio/health/afe4404.c b/drivers/iio/health/afe4404.c index 45266404f7e3..964f5231a831 100644 --- a/drivers/iio/health/afe4404.c +++ b/drivers/iio/health/afe4404.c @@ -428,7 +428,7 @@ MODULE_DEVICE_TABLE(of, afe4404_of_match); static int __maybe_unused afe4404_suspend(struct device *dev) { - struct iio_dev *indio_dev = dev_to_iio_dev(dev); + struct iio_dev *indio_dev = i2c_get_clientdata(to_i2c_client(dev)); struct afe4404_data *afe = iio_priv(indio_dev); int ret; @@ -449,7 +449,7 @@ static int __maybe_unused afe4404_suspend(struct device *dev) static int __maybe_unused afe4404_resume(struct device *dev) { - struct iio_dev *indio_dev = dev_to_iio_dev(dev); + struct iio_dev *indio_dev = i2c_get_clientdata(to_i2c_client(dev)); struct afe4404_data *afe = iio_priv(indio_dev); int ret; -- cgit v1.2.3 From 828f84ee8f84710ea1818b3565add268bcb824c8 Mon Sep 17 00:00:00 2001 From: Matt Ranostay Date: Mon, 16 Jan 2017 18:04:18 -0800 Subject: iio: health: max30100: fixed parenthesis around FIFO count check FIFO was being read every sample after the "almost full" state was reached. This was due to an incorrect placement of the parenthesis in the while condition check. Note - the fixes tag is not actually correct, but the fix in this patch would also be needed for it to function correctly so we'll go with that one. Backports should pick up both. Signed-off-by: Matt Ranostay Fixes: b74fccad7 ("iio: health: max30100: correct FIFO check condition") Cc: Stable@vger.kernel.org> Signed-off-by: Jonathan Cameron --- drivers/iio/health/max30100.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/health/max30100.c b/drivers/iio/health/max30100.c index 90ab8a2d2846..183c14329d6e 100644 --- a/drivers/iio/health/max30100.c +++ b/drivers/iio/health/max30100.c @@ -238,7 +238,7 @@ static irqreturn_t max30100_interrupt_handler(int irq, void *private) mutex_lock(&data->lock); - while (cnt || (cnt = max30100_fifo_count(data) > 0)) { + while (cnt || (cnt = max30100_fifo_count(data)) > 0) { ret = max30100_read_measurement(data); if (ret) break; -- cgit v1.2.3 From d1aaf20ee655888c227d5137b7a63551f8d15416 Mon Sep 17 00:00:00 2001 From: Alison Schofield Date: Mon, 16 Jan 2017 11:27:52 -0800 Subject: iio: adc: palmas_gpadc: retrieve a valid iio_dev in suspend/resume The suspend/resume functions were using dev_to_iio_dev() to get the iio_dev. That only works on IIO dev's. Use dev_get_drvdata() for a platform device to get the correct iio_dev. Signed-off-by: Alison Schofield Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/palmas_gpadc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/iio/adc/palmas_gpadc.c b/drivers/iio/adc/palmas_gpadc.c index 2bbf0c521beb..7d61b566e148 100644 --- a/drivers/iio/adc/palmas_gpadc.c +++ b/drivers/iio/adc/palmas_gpadc.c @@ -775,7 +775,7 @@ static int palmas_adc_wakeup_reset(struct palmas_gpadc *adc) static int palmas_gpadc_suspend(struct device *dev) { - struct iio_dev *indio_dev = dev_to_iio_dev(dev); + struct iio_dev *indio_dev = dev_get_drvdata(dev); struct palmas_gpadc *adc = iio_priv(indio_dev); int wakeup = adc->wakeup1_enable || adc->wakeup2_enable; int ret; @@ -798,7 +798,7 @@ static int palmas_gpadc_suspend(struct device *dev) static int palmas_gpadc_resume(struct device *dev) { - struct iio_dev *indio_dev = dev_to_iio_dev(dev); + struct iio_dev *indio_dev = dev_get_drvdata(dev); struct palmas_gpadc *adc = iio_priv(indio_dev); int wakeup = adc->wakeup1_enable || adc->wakeup2_enable; int ret; -- cgit v1.2.3 From 5c113b5e0082e90d2e1c7b12e96a7b8cf0623e27 Mon Sep 17 00:00:00 2001 From: John Brooks Date: Wed, 18 Jan 2017 21:50:39 +0000 Subject: iio: dht11: Use usleep_range instead of msleep for start signal The DHT22 (AM2302) datasheet specifies that the LOW start pulse should not exceed 20ms. However, observations with an oscilloscope of an RPi Model 2B (rev 1.1) communicating with a DHT22 sensor showed that the driver was consistently sending start pulses longer than 20ms: Kernel 4.7.10-v7+ (n=132): Minimum pulse length: 20.20ms Maximum: 29.84ms Mean: 24.96ms StDev: 2.82ms Sensor response rate: 100% Read success rate: 76% On kernel 4.8, the start pulse was so long that the sensor would not even respond 97% of the time: Kernel 4.8.16-v7+ (n=100): Minimum pulse length: 30.4ms Maximum: 74.4ms Mean: 39.3ms StDev: 10.2ms Sensor response rate: 3% Read success rate: 3% The driver would return ETIMEDOUT and write log messages like this: [ 51.430987] dht11 dht11@0: Only 1 signal edges detected [ 66.311019] dht11 dht11@0: Only 0 signal edges detected Replacing msleep(18) with usleep_range(18000, 20000) made the pulse length sane again and restored responsiveness: Kernel 4.8.16-v7+ with usleep_range (n=123): Minimum pulse length: 18.16ms Maximum: 20.20ms Mean: 19.85ms StDev: 0.51ms Sensor response rate: 100% Read success rate: 84% Cc: stable@vger.kernel.org Signed-off-by: John Brooks Reviewed-by: Harald Geyer Signed-off-by: Jonathan Cameron --- drivers/iio/humidity/dht11.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/iio/humidity/dht11.c b/drivers/iio/humidity/dht11.c index 9c47bc98f3ac..2a22ad920333 100644 --- a/drivers/iio/humidity/dht11.c +++ b/drivers/iio/humidity/dht11.c @@ -71,7 +71,8 @@ * a) select an implementation using busy loop polling on those systems * b) use the checksum to do some probabilistic decoding */ -#define DHT11_START_TRANSMISSION 18 /* ms */ +#define DHT11_START_TRANSMISSION_MIN 18000 /* us */ +#define DHT11_START_TRANSMISSION_MAX 20000 /* us */ #define DHT11_MIN_TIMERES 34000 /* ns */ #define DHT11_THRESHOLD 49000 /* ns */ #define DHT11_AMBIG_LOW 23000 /* ns */ @@ -228,7 +229,8 @@ static int dht11_read_raw(struct iio_dev *iio_dev, ret = gpio_direction_output(dht11->gpio, 0); if (ret) goto err; - msleep(DHT11_START_TRANSMISSION); + usleep_range(DHT11_START_TRANSMISSION_MIN, + DHT11_START_TRANSMISSION_MAX); ret = gpio_direction_input(dht11->gpio); if (ret) goto err; -- cgit v1.2.3 From 4eccbfc36186926b570310bfbd44f4216cd05c63 Mon Sep 17 00:00:00 2001 From: Lendacky, Thomas Date: Fri, 20 Jan 2017 12:14:03 -0600 Subject: amd-xgbe: Add a hardware quirk for register definitions A newer version of the hardware is using the same PCI ids for the network device but has altered register definitions for determining the window settings for the indirect PCS access. Add support to check for this hardware and if found use the new register values. Signed-off-by: Tom Lendacky Signed-off-by: David S. Miller --- drivers/net/ethernet/amd/xgbe/xgbe-common.h | 2 ++ drivers/net/ethernet/amd/xgbe/xgbe-dev.c | 4 ++-- drivers/net/ethernet/amd/xgbe/xgbe-pci.c | 15 ++++++++++++++- drivers/net/ethernet/amd/xgbe/xgbe.h | 2 ++ 4 files changed, 20 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-common.h b/drivers/net/ethernet/amd/xgbe/xgbe-common.h index 5b7ba25e0065..8a280e7d66bd 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-common.h +++ b/drivers/net/ethernet/amd/xgbe/xgbe-common.h @@ -891,6 +891,8 @@ #define PCS_V1_WINDOW_SELECT 0x03fc #define PCS_V2_WINDOW_DEF 0x9060 #define PCS_V2_WINDOW_SELECT 0x9064 +#define PCS_V2_RV_WINDOW_DEF 0x1060 +#define PCS_V2_RV_WINDOW_SELECT 0x1064 /* PCS register entry bit positions and sizes */ #define PCS_V2_WINDOW_DEF_OFFSET_INDEX 6 diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c index aaf0350076a9..c8e8a4a271ec 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c @@ -1151,7 +1151,7 @@ static int xgbe_read_mmd_regs_v2(struct xgbe_prv_data *pdata, int prtad, offset = pdata->xpcs_window + (mmd_address & pdata->xpcs_window_mask); spin_lock_irqsave(&pdata->xpcs_lock, flags); - XPCS32_IOWRITE(pdata, PCS_V2_WINDOW_SELECT, index); + XPCS32_IOWRITE(pdata, pdata->xpcs_window_sel_reg, index); mmd_data = XPCS16_IOREAD(pdata, offset); spin_unlock_irqrestore(&pdata->xpcs_lock, flags); @@ -1183,7 +1183,7 @@ static void xgbe_write_mmd_regs_v2(struct xgbe_prv_data *pdata, int prtad, offset = pdata->xpcs_window + (mmd_address & pdata->xpcs_window_mask); spin_lock_irqsave(&pdata->xpcs_lock, flags); - XPCS32_IOWRITE(pdata, PCS_V2_WINDOW_SELECT, index); + XPCS32_IOWRITE(pdata, pdata->xpcs_window_sel_reg, index); XPCS16_IOWRITE(pdata, offset, mmd_data); spin_unlock_irqrestore(&pdata->xpcs_lock, flags); } diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-pci.c b/drivers/net/ethernet/amd/xgbe/xgbe-pci.c index e76b7f65b805..c2730f15bd8b 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-pci.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-pci.c @@ -265,6 +265,7 @@ static int xgbe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) struct xgbe_prv_data *pdata; struct device *dev = &pdev->dev; void __iomem * const *iomap_table; + struct pci_dev *rdev; unsigned int ma_lo, ma_hi; unsigned int reg; int bar_mask; @@ -326,8 +327,20 @@ static int xgbe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) if (netif_msg_probe(pdata)) dev_dbg(dev, "xpcs_regs = %p\n", pdata->xpcs_regs); + /* Set the PCS indirect addressing definition registers */ + rdev = pci_get_domain_bus_and_slot(0, 0, PCI_DEVFN(0, 0)); + if (rdev && + (rdev->vendor == PCI_VENDOR_ID_AMD) && (rdev->device == 0x15d0)) { + pdata->xpcs_window_def_reg = PCS_V2_RV_WINDOW_DEF; + pdata->xpcs_window_sel_reg = PCS_V2_RV_WINDOW_SELECT; + } else { + pdata->xpcs_window_def_reg = PCS_V2_WINDOW_DEF; + pdata->xpcs_window_sel_reg = PCS_V2_WINDOW_SELECT; + } + pci_dev_put(rdev); + /* Configure the PCS indirect addressing support */ - reg = XPCS32_IOREAD(pdata, PCS_V2_WINDOW_DEF); + reg = XPCS32_IOREAD(pdata, pdata->xpcs_window_def_reg); pdata->xpcs_window = XPCS_GET_BITS(reg, PCS_V2_WINDOW_DEF, OFFSET); pdata->xpcs_window <<= 6; pdata->xpcs_window_size = XPCS_GET_BITS(reg, PCS_V2_WINDOW_DEF, SIZE); diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h index f52a9bd05bac..00108815b55e 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe.h +++ b/drivers/net/ethernet/amd/xgbe/xgbe.h @@ -955,6 +955,8 @@ struct xgbe_prv_data { /* XPCS indirect addressing lock */ spinlock_t xpcs_lock; + unsigned int xpcs_window_def_reg; + unsigned int xpcs_window_sel_reg; unsigned int xpcs_window; unsigned int xpcs_window_size; unsigned int xpcs_window_mask; -- cgit v1.2.3 From 738f7f647371ff4cfc9646c99dba5b58ad142db3 Mon Sep 17 00:00:00 2001 From: Lendacky, Thomas Date: Fri, 20 Jan 2017 12:14:13 -0600 Subject: amd-xgbe: Check xgbe_init() return code The xgbe_init() routine returns a return code indicating success or failure, but the return code is not checked. Add code to xgbe_init() to issue a message when failures are seen and add code to check the xgbe_init() return code. Signed-off-by: Tom Lendacky Signed-off-by: David S. Miller --- drivers/net/ethernet/amd/xgbe/xgbe-dev.c | 4 +++- drivers/net/ethernet/amd/xgbe/xgbe-drv.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c index c8e8a4a271ec..a7d16db5c4b2 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c @@ -3407,8 +3407,10 @@ static int xgbe_init(struct xgbe_prv_data *pdata) /* Flush Tx queues */ ret = xgbe_flush_tx_queues(pdata); - if (ret) + if (ret) { + netdev_err(pdata->netdev, "error flushing TX queues\n"); return ret; + } /* * Initialize DMA related features diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c index 9943629fcbf9..1c87cc204075 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c @@ -1070,7 +1070,9 @@ static int xgbe_start(struct xgbe_prv_data *pdata) DBGPR("-->xgbe_start\n"); - hw_if->init(pdata); + ret = hw_if->init(pdata); + if (ret) + return ret; xgbe_napi_enable(pdata, 1); -- cgit v1.2.3 From c26665ab5c49ad3e142e0f054ca3204f259ba09c Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Fri, 20 Jan 2017 21:29:40 +0100 Subject: x86/microcode/intel: Drop stashed AP patch pointer optimization This was meant to save us the scanning of the microcode containter in the initrd since the first AP had already done that but it can also hurt us: Imagine a single hyperthreaded CPU (Intel(R) Atom(TM) CPU N270, for example) which updates the microcode on the BSP but since the microcode engine is shared between the two threads, the update on CPU1 doesn't happen because it has already happened on CPU0 and we don't find a newer microcode revision on CPU1. Which doesn't set the intel_ucode_patch pointer and at initrd jettisoning time we don't save the microcode patch for later application. Now, when we suspend to RAM, the loaded microcode gets cleared so we need to reload but there's no patch saved in the cache. Removing the optimization fixes this issue and all is fine and dandy. Fixes: 06b8534cb728 ("x86/microcode: Rework microcode loading") Signed-off-by: Borislav Petkov Reviewed-by: Thomas Gleixner Link: http://lkml.kernel.org/r/20170120202955.4091-2-bp@alien8.de Signed-off-by: Thomas Gleixner --- arch/x86/kernel/cpu/microcode/intel.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) diff --git a/arch/x86/kernel/cpu/microcode/intel.c b/arch/x86/kernel/cpu/microcode/intel.c index 3f329b74e040..8325d8a09ab0 100644 --- a/arch/x86/kernel/cpu/microcode/intel.c +++ b/arch/x86/kernel/cpu/microcode/intel.c @@ -41,7 +41,7 @@ static const char ucode_path[] = "kernel/x86/microcode/GenuineIntel.bin"; -/* Current microcode patch used in early patching */ +/* Current microcode patch used in early patching on the APs. */ struct microcode_intel *intel_ucode_patch; static inline bool cpu_signatures_match(unsigned int s1, unsigned int p1, @@ -607,12 +607,6 @@ int __init save_microcode_in_initrd_intel(void) struct ucode_cpu_info uci; struct cpio_data cp; - /* - * AP loading didn't find any microcode patch, no need to save anything. - */ - if (!intel_ucode_patch || IS_ERR(intel_ucode_patch)) - return 0; - if (!load_builtin_intel_microcode(&cp)) cp = find_microcode_in_initrd(ucode_path, false); @@ -628,7 +622,6 @@ int __init save_microcode_in_initrd_intel(void) return 0; } - /* * @res_patch, output: a pointer to the patch we found. */ -- cgit v1.2.3 From e9748e0364fe82dc037d22900ff13a62d04518bf Mon Sep 17 00:00:00 2001 From: Ziyuan Xu Date: Tue, 17 Jan 2017 09:22:56 +0800 Subject: mmc: dw_mmc: force setup bus if active slots exist It's necessary to setup bus if any slots are present. - update clock after ctrl reset - if the host has genpd node, we can guarantee the clock is available before starting request. Otherwies, the clock register is reset once power off the pd, and host can't output the active clock during communication. Fixes: e9ed8835e990 ("mmc: dw_mmc: add runtime PM callback") Fixes: df9bcc2bc0a1 ("mmc: dw_mmc: add missing codes for runtime resume") cc: Reported-by: Randy Li Reported-by: S. Gilles Signed-off-by: Ziyuan Xu Signed-off-by: Shawn Lin Signed-off-by: Ulf Hansson --- drivers/mmc/host/dw_mmc.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c index b44306b886cb..73db08558e4d 100644 --- a/drivers/mmc/host/dw_mmc.c +++ b/drivers/mmc/host/dw_mmc.c @@ -3354,10 +3354,11 @@ int dw_mci_runtime_resume(struct device *dev) if (!slot) continue; - if (slot->mmc->pm_flags & MMC_PM_KEEP_POWER) { + if (slot->mmc->pm_flags & MMC_PM_KEEP_POWER) dw_mci_set_ios(slot->mmc, &slot->mmc->ios); - dw_mci_setup_bus(slot, true); - } + + /* Force setup bus to guarantee available clock output */ + dw_mci_setup_bus(slot, true); } /* Now that slots are all setup, we can enable card detect */ -- cgit v1.2.3 From 4c833368f0bf748d4147bf301b1f95bc8eccb3c0 Mon Sep 17 00:00:00 2001 From: Kevin Hao Date: Sun, 22 Jan 2017 16:50:23 +0800 Subject: x86/fpu: Set the xcomp_bv when we fake up a XSAVES area I got the following calltrace on a Apollo Lake SoC with 32-bit kernel: WARNING: CPU: 2 PID: 261 at arch/x86/include/asm/fpu/internal.h:363 fpu__restore+0x1f5/0x260 [...] Hardware name: Intel Corp. Broxton P/NOTEBOOK, BIOS APLIRVPA.X64.0138.B35.1608091058 08/09/2016 Call Trace: dump_stack() __warn() ? fpu__restore() warn_slowpath_null() fpu__restore() __fpu__restore_sig() fpu__restore_sig() restore_sigcontext.isra.9() sys_sigreturn() do_int80_syscall_32() entry_INT80_32() The reason is that a #GP occurs when executing XRSTORS. The root cause is that we forget to set the xcomp_bv when we fake up the XSAVES area in the copyin_to_xsaves() function. Signed-off-by: Kevin Hao Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Dave Hansen Cc: Denys Vlasenko Cc: Fenghua Yu Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Oleg Nesterov Cc: Peter Zijlstra Cc: Quentin Casasnovas Cc: Rik van Riel Cc: Thomas Gleixner Cc: Yu-cheng Yu Link: http://lkml.kernel.org/r/1485075023-30161-1-git-send-email-haokexin@gmail.com Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- arch/x86/kernel/fpu/xstate.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c index 1d7770447b3e..e287b9075527 100644 --- a/arch/x86/kernel/fpu/xstate.c +++ b/arch/x86/kernel/fpu/xstate.c @@ -1070,6 +1070,7 @@ int copyin_to_xsaves(const void *kbuf, const void __user *ubuf, * Add back in the features that came in from userspace: */ xsave->header.xfeatures |= xfeatures; + xsave->header.xcomp_bv = XCOMP_BV_COMPACTED_FORMAT | xsave->header.xfeatures; return 0; } -- cgit v1.2.3 From 7941c59e45f3b6d30e07375e9b6713427e0a9f98 Mon Sep 17 00:00:00 2001 From: Jürg Billeter Date: Mon, 10 Oct 2016 18:30:01 +0200 Subject: iwlwifi: fix double hyphen in MODULE_FIRMWARE for 8000 Mistakenly, the driver is trying to load the 8000C firmware with an incorrect name (i.e. with two hyphens where there should be only one) and that fails. Fix that by removing the hyphen from the format macro. Fixes: e1ba684f762b ("iwlwifi: 8000: fix MODULE_FIRMWARE input") Signed-off-by: Jürg Billeter Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/iwl-8000.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-8000.c b/drivers/net/wireless/intel/iwlwifi/iwl-8000.c index d02ca1491d16..8d3e53fac1da 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-8000.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-8000.c @@ -91,7 +91,7 @@ #define IWL8000_FW_PRE "iwlwifi-8000C-" #define IWL8000_MODULE_FIRMWARE(api) \ - IWL8000_FW_PRE "-" __stringify(api) ".ucode" + IWL8000_FW_PRE __stringify(api) ".ucode" #define IWL8265_FW_PRE "iwlwifi-8265-" #define IWL8265_MODULE_FIRMWARE(api) \ -- cgit v1.2.3 From 03c902bff524e0cf664737a33f2365f7837040bf Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 2 Dec 2016 12:03:36 +0100 Subject: iwlwifi: mvm: avoid crash on restart w/o reserved queues When the firmware restarts in a situation in which any station has no queue reserved anymore because that queue was used, the code will crash trying to access the queue_info array at the offset 255, which is far too big. Fix this by checking that a queue is actually reserved before writing its status. Fixes: 8d98ae6eb0d5 ("iwlwifi: mvm: re-assign old queues after hw restart in dqa mode") Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index 636c8b03e318..09e9e2e3ed04 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -1164,9 +1164,10 @@ static void iwl_mvm_realloc_queues_after_restart(struct iwl_mvm *mvm, .frame_limit = IWL_FRAME_LIMIT, }; - /* Make sure reserved queue is still marked as such (or allocated) */ - mvm->queue_info[mvm_sta->reserved_queue].status = - IWL_MVM_QUEUE_RESERVED; + /* Make sure reserved queue is still marked as such (if allocated) */ + if (mvm_sta->reserved_queue != IEEE80211_INVAL_HW_QUEUE) + mvm->queue_info[mvm_sta->reserved_queue].status = + IWL_MVM_QUEUE_RESERVED; for (i = 0; i <= IWL_MAX_TID_COUNT; i++) { struct iwl_mvm_tid_data *tid_data = &mvm_sta->tid_data[i]; -- cgit v1.2.3 From d6040764adcb5cb6de1489422411d701c158bb69 Mon Sep 17 00:00:00 2001 From: Salvatore Benedetto Date: Fri, 13 Jan 2017 11:54:08 +0000 Subject: crypto: api - Clear CRYPTO_ALG_DEAD bit before registering an alg Make sure CRYPTO_ALG_DEAD bit is cleared before proceeding with the algorithm registration. This fixes qat-dh registration when driver is restarted Cc: Signed-off-by: Salvatore Benedetto Signed-off-by: Herbert Xu --- crypto/algapi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/crypto/algapi.c b/crypto/algapi.c index df939b54b09f..1fad2a6b3bbb 100644 --- a/crypto/algapi.c +++ b/crypto/algapi.c @@ -356,6 +356,7 @@ int crypto_register_alg(struct crypto_alg *alg) struct crypto_larval *larval; int err; + alg->cra_flags &= ~CRYPTO_ALG_DEAD; err = crypto_check_alg(alg); if (err) return err; -- cgit v1.2.3 From 11e3b725cfc282efe9d4a354153e99d86a16af08 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Tue, 17 Jan 2017 13:46:29 +0000 Subject: crypto: arm64/aes-blk - honour iv_out requirement in CBC and CTR modes Update the ARMv8 Crypto Extensions and the plain NEON AES implementations in CBC and CTR modes to return the next IV back to the skcipher API client. This is necessary for chaining to work correctly. Note that for CTR, this is only done if the request is a round multiple of the block size, since otherwise, chaining is impossible anyway. Cc: # v3.16+ Signed-off-by: Ard Biesheuvel Signed-off-by: Herbert Xu --- arch/arm64/crypto/aes-modes.S | 88 +++++++++++++++++++++---------------------- 1 file changed, 42 insertions(+), 46 deletions(-) diff --git a/arch/arm64/crypto/aes-modes.S b/arch/arm64/crypto/aes-modes.S index c53dbeae79f2..838dad5c209f 100644 --- a/arch/arm64/crypto/aes-modes.S +++ b/arch/arm64/crypto/aes-modes.S @@ -193,15 +193,16 @@ AES_ENTRY(aes_cbc_encrypt) cbz w6, .Lcbcencloop ld1 {v0.16b}, [x5] /* get iv */ - enc_prepare w3, x2, x5 + enc_prepare w3, x2, x6 .Lcbcencloop: ld1 {v1.16b}, [x1], #16 /* get next pt block */ eor v0.16b, v0.16b, v1.16b /* ..and xor with iv */ - encrypt_block v0, w3, x2, x5, w6 + encrypt_block v0, w3, x2, x6, w7 st1 {v0.16b}, [x0], #16 subs w4, w4, #1 bne .Lcbcencloop + st1 {v0.16b}, [x5] /* return iv */ ret AES_ENDPROC(aes_cbc_encrypt) @@ -211,7 +212,7 @@ AES_ENTRY(aes_cbc_decrypt) cbz w6, .LcbcdecloopNx ld1 {v7.16b}, [x5] /* get iv */ - dec_prepare w3, x2, x5 + dec_prepare w3, x2, x6 .LcbcdecloopNx: #if INTERLEAVE >= 2 @@ -248,7 +249,7 @@ AES_ENTRY(aes_cbc_decrypt) .Lcbcdecloop: ld1 {v1.16b}, [x1], #16 /* get next ct block */ mov v0.16b, v1.16b /* ...and copy to v0 */ - decrypt_block v0, w3, x2, x5, w6 + decrypt_block v0, w3, x2, x6, w7 eor v0.16b, v0.16b, v7.16b /* xor with iv => pt */ mov v7.16b, v1.16b /* ct is next iv */ st1 {v0.16b}, [x0], #16 @@ -256,6 +257,7 @@ AES_ENTRY(aes_cbc_decrypt) bne .Lcbcdecloop .Lcbcdecout: FRAME_POP + st1 {v7.16b}, [x5] /* return iv */ ret AES_ENDPROC(aes_cbc_decrypt) @@ -267,24 +269,15 @@ AES_ENDPROC(aes_cbc_decrypt) AES_ENTRY(aes_ctr_encrypt) FRAME_PUSH - cbnz w6, .Lctrfirst /* 1st time around? */ - umov x5, v4.d[1] /* keep swabbed ctr in reg */ - rev x5, x5 -#if INTERLEAVE >= 2 - cmn w5, w4 /* 32 bit overflow? */ - bcs .Lctrinc - add x5, x5, #1 /* increment BE ctr */ - b .LctrincNx -#else - b .Lctrinc -#endif -.Lctrfirst: + cbz w6, .Lctrnotfirst /* 1st time around? */ enc_prepare w3, x2, x6 ld1 {v4.16b}, [x5] - umov x5, v4.d[1] /* keep swabbed ctr in reg */ - rev x5, x5 + +.Lctrnotfirst: + umov x8, v4.d[1] /* keep swabbed ctr in reg */ + rev x8, x8 #if INTERLEAVE >= 2 - cmn w5, w4 /* 32 bit overflow? */ + cmn w8, w4 /* 32 bit overflow? */ bcs .Lctrloop .LctrloopNx: subs w4, w4, #INTERLEAVE @@ -292,11 +285,11 @@ AES_ENTRY(aes_ctr_encrypt) #if INTERLEAVE == 2 mov v0.8b, v4.8b mov v1.8b, v4.8b - rev x7, x5 - add x5, x5, #1 + rev x7, x8 + add x8, x8, #1 ins v0.d[1], x7 - rev x7, x5 - add x5, x5, #1 + rev x7, x8 + add x8, x8, #1 ins v1.d[1], x7 ld1 {v2.16b-v3.16b}, [x1], #32 /* get 2 input blocks */ do_encrypt_block2x @@ -305,7 +298,7 @@ AES_ENTRY(aes_ctr_encrypt) st1 {v0.16b-v1.16b}, [x0], #32 #else ldr q8, =0x30000000200000001 /* addends 1,2,3[,0] */ - dup v7.4s, w5 + dup v7.4s, w8 mov v0.16b, v4.16b add v7.4s, v7.4s, v8.4s mov v1.16b, v4.16b @@ -323,18 +316,12 @@ AES_ENTRY(aes_ctr_encrypt) eor v2.16b, v7.16b, v2.16b eor v3.16b, v5.16b, v3.16b st1 {v0.16b-v3.16b}, [x0], #64 - add x5, x5, #INTERLEAVE + add x8, x8, #INTERLEAVE #endif - cbz w4, .LctroutNx -.LctrincNx: - rev x7, x5 + rev x7, x8 ins v4.d[1], x7 + cbz w4, .Lctrout b .LctrloopNx -.LctroutNx: - sub x5, x5, #1 - rev x7, x5 - ins v4.d[1], x7 - b .Lctrout .Lctr1x: adds w4, w4, #INTERLEAVE beq .Lctrout @@ -342,30 +329,39 @@ AES_ENTRY(aes_ctr_encrypt) .Lctrloop: mov v0.16b, v4.16b encrypt_block v0, w3, x2, x6, w7 + + adds x8, x8, #1 /* increment BE ctr */ + rev x7, x8 + ins v4.d[1], x7 + bcs .Lctrcarry /* overflow? */ + +.Lctrcarrydone: subs w4, w4, #1 bmi .Lctrhalfblock /* blocks < 0 means 1/2 block */ ld1 {v3.16b}, [x1], #16 eor v3.16b, v0.16b, v3.16b st1 {v3.16b}, [x0], #16 - beq .Lctrout -.Lctrinc: - adds x5, x5, #1 /* increment BE ctr */ - rev x7, x5 - ins v4.d[1], x7 - bcc .Lctrloop /* no overflow? */ - umov x7, v4.d[0] /* load upper word of ctr */ - rev x7, x7 /* ... to handle the carry */ - add x7, x7, #1 - rev x7, x7 - ins v4.d[0], x7 - b .Lctrloop + bne .Lctrloop + +.Lctrout: + st1 {v4.16b}, [x5] /* return next CTR value */ + FRAME_POP + ret + .Lctrhalfblock: ld1 {v3.8b}, [x1] eor v3.8b, v0.8b, v3.8b st1 {v3.8b}, [x0] -.Lctrout: FRAME_POP ret + +.Lctrcarry: + umov x7, v4.d[0] /* load upper word of ctr */ + rev x7, x7 /* ... to handle the carry */ + add x7, x7, #1 + rev x7, x7 + ins v4.d[0], x7 + b .Lctrcarrydone AES_ENDPROC(aes_ctr_encrypt) .ltorg -- cgit v1.2.3 From 9f427a0e474a67b454420c131709600d44850486 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Fri, 20 Jan 2017 12:58:34 -0800 Subject: net: mpls: Fix multipath selection for LSR use case MPLS multipath for LSR is broken -- always selecting the first nexthop in the one label case. For example: $ ip -f mpls ro ls 100 nexthop as to 200 via inet 172.16.2.2 dev virt12 nexthop as to 300 via inet 172.16.3.2 dev virt13 101 nexthop as to 201 via inet6 2000:2::2 dev virt12 nexthop as to 301 via inet6 2000:3::2 dev virt13 In this example incoming packets have a single MPLS labels which means BOS bit is set. The BOS bit is passed from mpls_forward down to mpls_multipath_hash which never processes the hash loop because BOS is 1. Update mpls_multipath_hash to process the entire label stack. mpls_hdr_len tracks the total mpls header length on each pass (on pass N mpls_hdr_len is N * sizeof(mpls_shim_hdr)). When the label is found with the BOS set it verifies the skb has sufficient header for ipv4 or ipv6, and find the IPv4 and IPv6 header by using the last mpls_hdr pointer and adding 1 to advance past it. With these changes I have verified the code correctly sees the label, BOS, IPv4 and IPv6 addresses in the network header and icmp/tcp/udp traffic for ipv4 and ipv6 are distributed across the nexthops. Fixes: 1c78efa8319ca ("mpls: flow-based multipath selection") Acked-by: Robert Shearman Signed-off-by: David Ahern Signed-off-by: David S. Miller --- net/mpls/af_mpls.c | 48 +++++++++++++++++++++++++----------------------- 1 file changed, 25 insertions(+), 23 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 15fe97644ffe..5b77377e5a15 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -98,18 +98,19 @@ bool mpls_pkt_too_big(const struct sk_buff *skb, unsigned int mtu) } EXPORT_SYMBOL_GPL(mpls_pkt_too_big); -static u32 mpls_multipath_hash(struct mpls_route *rt, - struct sk_buff *skb, bool bos) +static u32 mpls_multipath_hash(struct mpls_route *rt, struct sk_buff *skb) { struct mpls_entry_decoded dec; + unsigned int mpls_hdr_len = 0; struct mpls_shim_hdr *hdr; bool eli_seen = false; int label_index; u32 hash = 0; - for (label_index = 0; label_index < MAX_MP_SELECT_LABELS && !bos; + for (label_index = 0; label_index < MAX_MP_SELECT_LABELS; label_index++) { - if (!pskb_may_pull(skb, sizeof(*hdr) * label_index)) + mpls_hdr_len += sizeof(*hdr); + if (!pskb_may_pull(skb, mpls_hdr_len)) break; /* Read and decode the current label */ @@ -134,37 +135,38 @@ static u32 mpls_multipath_hash(struct mpls_route *rt, eli_seen = true; } - bos = dec.bos; - if (bos && pskb_may_pull(skb, sizeof(*hdr) * label_index + - sizeof(struct iphdr))) { + if (!dec.bos) + continue; + + /* found bottom label; does skb have room for a header? */ + if (pskb_may_pull(skb, mpls_hdr_len + sizeof(struct iphdr))) { const struct iphdr *v4hdr; - v4hdr = (const struct iphdr *)(mpls_hdr(skb) + - label_index); + v4hdr = (const struct iphdr *)(hdr + 1); if (v4hdr->version == 4) { hash = jhash_3words(ntohl(v4hdr->saddr), ntohl(v4hdr->daddr), v4hdr->protocol, hash); } else if (v4hdr->version == 6 && - pskb_may_pull(skb, sizeof(*hdr) * label_index + - sizeof(struct ipv6hdr))) { + pskb_may_pull(skb, mpls_hdr_len + + sizeof(struct ipv6hdr))) { const struct ipv6hdr *v6hdr; - v6hdr = (const struct ipv6hdr *)(mpls_hdr(skb) + - label_index); - + v6hdr = (const struct ipv6hdr *)(hdr + 1); hash = __ipv6_addr_jhash(&v6hdr->saddr, hash); hash = __ipv6_addr_jhash(&v6hdr->daddr, hash); hash = jhash_1word(v6hdr->nexthdr, hash); } } + + break; } return hash; } static struct mpls_nh *mpls_select_multipath(struct mpls_route *rt, - struct sk_buff *skb, bool bos) + struct sk_buff *skb) { int alive = ACCESS_ONCE(rt->rt_nhn_alive); u32 hash = 0; @@ -180,7 +182,7 @@ static struct mpls_nh *mpls_select_multipath(struct mpls_route *rt, if (alive <= 0) return NULL; - hash = mpls_multipath_hash(rt, skb, bos); + hash = mpls_multipath_hash(rt, skb); nh_index = hash % alive; if (alive == rt->rt_nhn) goto out; @@ -278,17 +280,11 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev, hdr = mpls_hdr(skb); dec = mpls_entry_decode(hdr); - /* Pop the label */ - skb_pull(skb, sizeof(*hdr)); - skb_reset_network_header(skb); - - skb_orphan(skb); - rt = mpls_route_input_rcu(net, dec.label); if (!rt) goto drop; - nh = mpls_select_multipath(rt, skb, dec.bos); + nh = mpls_select_multipath(rt, skb); if (!nh) goto drop; @@ -297,6 +293,12 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev, if (!mpls_output_possible(out_dev)) goto drop; + /* Pop the label */ + skb_pull(skb, sizeof(*hdr)); + skb_reset_network_header(skb); + + skb_orphan(skb); + if (skb_warn_if_lro(skb)) goto drop; -- cgit v1.2.3 From d0e287a401d9acf67b75180b26e2d62b7d482652 Mon Sep 17 00:00:00 2001 From: Rask Ingemann Lambertsen Date: Sat, 21 Jan 2017 17:11:43 +0100 Subject: regulator: axp20x: AXP806: Fix dcdcb being set instead of dcdce A typo or copy-paste bug means that the register access intended for regulator dcdce goes to dcdcb instead. This patch corrects it. Fixes: 2ca342d391e3 (regulator: axp20x: Support AXP806 variant) Signed-off-by: Rask Ingemann Lambertsen Acked-by: Chen-Yu Tsai Signed-off-by: Mark Brown Cc: stable@vger.kernel.org --- drivers/regulator/axp20x-regulator.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/regulator/axp20x-regulator.c b/drivers/regulator/axp20x-regulator.c index e6a512ebeae2..a3ade9e4ef47 100644 --- a/drivers/regulator/axp20x-regulator.c +++ b/drivers/regulator/axp20x-regulator.c @@ -272,7 +272,7 @@ static const struct regulator_desc axp806_regulators[] = { 64, AXP806_DCDCD_V_CTRL, 0x3f, AXP806_PWR_OUT_CTRL1, BIT(3)), AXP_DESC(AXP806, DCDCE, "dcdce", "vine", 1100, 3400, 100, - AXP806_DCDCB_V_CTRL, 0x1f, AXP806_PWR_OUT_CTRL1, BIT(4)), + AXP806_DCDCE_V_CTRL, 0x1f, AXP806_PWR_OUT_CTRL1, BIT(4)), AXP_DESC(AXP806, ALDO1, "aldo1", "aldoin", 700, 3300, 100, AXP806_ALDO1_V_CTRL, 0x1f, AXP806_PWR_OUT_CTRL1, BIT(5)), AXP_DESC(AXP806, ALDO2, "aldo2", "aldoin", 700, 3400, 100, -- cgit v1.2.3 From b32728ffef7f233dbdabb3f11814bdf692aaf501 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 23 Jan 2017 08:12:19 -0700 Subject: xen-blkfront: feature flags handling adjustments Don't truncate the "feature-persistent" value read from xenstore: Any non-zero value is supposed to enable the feature, just like is already being done for feature_secdiscard. Just like the other feature_* fields, feature_flush and feature_fua are boolean flags, and hence fit well into a single bit. Keep all bit fields together to limit gaps. Signed-off-by: Jan Beulich Signed-off-by: Konrad Rzeszutek Wilk --- drivers/block/xen-blkfront.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index b2bdfa81f929..386b17208605 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -197,13 +197,13 @@ struct blkfront_info /* Number of pages per ring buffer. */ unsigned int nr_ring_pages; struct request_queue *rq; - unsigned int feature_flush; - unsigned int feature_fua; + unsigned int feature_flush:1; + unsigned int feature_fua:1; unsigned int feature_discard:1; unsigned int feature_secdiscard:1; + unsigned int feature_persistent:1; unsigned int discard_granularity; unsigned int discard_alignment; - unsigned int feature_persistent:1; /* Number of 4KB segments handled */ unsigned int max_indirect_segments; int is_ready; @@ -2323,8 +2323,8 @@ static void blkfront_gather_backend_features(struct blkfront_info *info) blkfront_setup_discard(info); info->feature_persistent = - xenbus_read_unsigned(info->xbdev->otherend, - "feature-persistent", 0); + !!xenbus_read_unsigned(info->xbdev->otherend, + "feature-persistent", 0); indirect_segments = xenbus_read_unsigned(info->xbdev->otherend, "feature-max-indirect-segments", 0); -- cgit v1.2.3 From 3b4f18843e511193e7eb616710e838f5852e661d Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 23 Jan 2017 08:11:37 -0700 Subject: xen-blkfront: correct maximum segment accounting Making use of "max_indirect_segments=" has issues: - blkfront_setup_indirect() may end up with zero psegs when PAGE_SIZE is sufficiently much larger than XEN_PAGE_SIZE - the variable driven by the command line option (xen_blkif_max_segments) has a somewhat different purpose, and hence should namely never end up being zero - as long as the specified value is lower than the legacy default, we better don't use indirect segments at all (or we'd in fact lower throughput) Signed-off-by: Jan Beulich Signed-off-by: Konrad Rzeszutek Wilk --- drivers/block/xen-blkfront.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 386b17208605..265f1a7072e9 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -2223,7 +2223,7 @@ static int blkfront_setup_indirect(struct blkfront_ring_info *rinfo) } else grants = info->max_indirect_segments; - psegs = grants / GRANTS_PER_PSEG; + psegs = DIV_ROUND_UP(grants, GRANTS_PER_PSEG); err = fill_grant_buffer(rinfo, (grants + INDIRECT_GREFS(grants)) * BLK_RING_SIZE(info)); @@ -2328,8 +2328,11 @@ static void blkfront_gather_backend_features(struct blkfront_info *info) indirect_segments = xenbus_read_unsigned(info->xbdev->otherend, "feature-max-indirect-segments", 0); - info->max_indirect_segments = min(indirect_segments, - xen_blkif_max_segments); + if (indirect_segments > xen_blkif_max_segments) + indirect_segments = xen_blkif_max_segments; + if (indirect_segments <= BLKIF_MAX_SEGMENTS_PER_REQUEST) + indirect_segments = 0; + info->max_indirect_segments = indirect_segments; } /* @@ -2652,6 +2655,9 @@ static int __init xlblk_init(void) if (!xen_domain()) return -ENODEV; + if (xen_blkif_max_segments < BLKIF_MAX_SEGMENTS_PER_REQUEST) + xen_blkif_max_segments = BLKIF_MAX_SEGMENTS_PER_REQUEST; + if (xen_blkif_max_ring_order > XENBUS_MAX_RING_GRANT_ORDER) { pr_info("Invalid max_ring_order (%d), will use default max: %d.\n", xen_blkif_max_ring_order, XENBUS_MAX_RING_GRANT_ORDER); -- cgit v1.2.3 From d2b3964a0780d2d2994eba57f950d6c9fe489ed8 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 20 Jan 2017 09:31:54 -0800 Subject: xfs: fix COW writeback race Due to the way how xfs_iomap_write_allocate tries to convert the whole found extents from delalloc to real space we can run into a race condition with multiple threads doing writes to this same extent. For the non-COW case that is harmless as the only thing that can happen is that we call xfs_bmapi_write on an extent that has already been converted to a real allocation. For COW writes where we move the extent from the COW to the data fork after I/O completion the race is, however, not quite as harmless. In the worst case we are now calling xfs_bmapi_write on a region that contains hole in the COW work, which will trip up an assert in debug builds or lead to file system corruption in non-debug builds. This seems to be reproducible with workloads of small O_DSYNC write, although so far I've not managed to come up with a with an isolated reproducer. The fix for the issue is relatively simple: tell xfs_bmapi_write that we are only asked to convert delayed allocations and skip holes in that case. Signed-off-by: Christoph Hellwig Reviewed-by: Brian Foster Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_bmap.c | 44 ++++++++++++++++++++++++++++++++------------ fs/xfs/libxfs/xfs_bmap.h | 6 +++++- fs/xfs/xfs_iomap.c | 2 +- 3 files changed, 38 insertions(+), 14 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 44773c9eb957..ab82dd4a4980 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4514,8 +4514,6 @@ xfs_bmapi_write( int n; /* current extent index */ xfs_fileoff_t obno; /* old block number (offset) */ int whichfork; /* data or attr fork */ - char inhole; /* current location is hole in file */ - char wasdelay; /* old extent was delayed */ #ifdef DEBUG xfs_fileoff_t orig_bno; /* original block number value */ @@ -4603,22 +4601,44 @@ xfs_bmapi_write( bma.firstblock = firstblock; while (bno < end && n < *nmap) { - inhole = eof || bma.got.br_startoff > bno; - wasdelay = !inhole && isnullstartblock(bma.got.br_startblock); + bool need_alloc = false, wasdelay = false; - /* - * Make sure we only reflink into a hole. - */ - if (flags & XFS_BMAPI_REMAP) - ASSERT(inhole); - if (flags & XFS_BMAPI_COWFORK) - ASSERT(!inhole); + /* in hole or beyoned EOF? */ + if (eof || bma.got.br_startoff > bno) { + if (flags & XFS_BMAPI_DELALLOC) { + /* + * For the COW fork we can reasonably get a + * request for converting an extent that races + * with other threads already having converted + * part of it, as there converting COW to + * regular blocks is not protected using the + * IOLOCK. + */ + ASSERT(flags & XFS_BMAPI_COWFORK); + if (!(flags & XFS_BMAPI_COWFORK)) { + error = -EIO; + goto error0; + } + + if (eof || bno >= end) + break; + } else { + need_alloc = true; + } + } else { + /* + * Make sure we only reflink into a hole. + */ + ASSERT(!(flags & XFS_BMAPI_REMAP)); + if (isnullstartblock(bma.got.br_startblock)) + wasdelay = true; + } /* * First, deal with the hole before the allocated space * that we found, if any. */ - if (inhole || wasdelay) { + if (need_alloc || wasdelay) { bma.eof = eof; bma.conv = !!(flags & XFS_BMAPI_CONVERT); bma.wasdel = wasdelay; diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index cecd094404cc..cdef87db5262 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -110,6 +110,9 @@ struct xfs_extent_free_item /* Map something in the CoW fork. */ #define XFS_BMAPI_COWFORK 0x200 +/* Only convert delalloc space, don't allocate entirely new extents */ +#define XFS_BMAPI_DELALLOC 0x400 + #define XFS_BMAPI_FLAGS \ { XFS_BMAPI_ENTIRE, "ENTIRE" }, \ { XFS_BMAPI_METADATA, "METADATA" }, \ @@ -120,7 +123,8 @@ struct xfs_extent_free_item { XFS_BMAPI_CONVERT, "CONVERT" }, \ { XFS_BMAPI_ZERO, "ZERO" }, \ { XFS_BMAPI_REMAP, "REMAP" }, \ - { XFS_BMAPI_COWFORK, "COWFORK" } + { XFS_BMAPI_COWFORK, "COWFORK" }, \ + { XFS_BMAPI_DELALLOC, "DELALLOC" } static inline int xfs_bmapi_aflag(int w) diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 0d147428971e..1aa3abd67b36 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -681,7 +681,7 @@ xfs_iomap_write_allocate( xfs_trans_t *tp; int nimaps; int error = 0; - int flags = 0; + int flags = XFS_BMAPI_DELALLOC; int nres; if (whichfork == XFS_COW_FORK) -- cgit v1.2.3 From eab127717a6af54401ba534790c793ec143cd1fc Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 20 Jan 2017 15:31:52 -0800 Subject: net: phy: Avoid deadlock during phy_error() phy_error() is called in the PHY state machine workqueue context, and calls phy_trigger_machine() which does a cancel_delayed_work_sync() of the workqueue we execute from, causing a deadlock situation. Augment phy_trigger_machine() machine with a sync boolean indicating whether we should use cancel_*_sync() or just cancel_*_work(). Fixes: 3c293f4e08b5 ("net: phy: Trigger state machine on state change and not polling.") Reported-by: Russell King Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/phy.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 48da6e93c3f7..e687a9cb4a37 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -649,14 +649,18 @@ void phy_start_machine(struct phy_device *phydev) * phy_trigger_machine - trigger the state machine to run * * @phydev: the phy_device struct + * @sync: indicate whether we should wait for the workqueue cancelation * * Description: There has been a change in state which requires that the * state machine runs. */ -static void phy_trigger_machine(struct phy_device *phydev) +static void phy_trigger_machine(struct phy_device *phydev, bool sync) { - cancel_delayed_work_sync(&phydev->state_queue); + if (sync) + cancel_delayed_work_sync(&phydev->state_queue); + else + cancel_delayed_work(&phydev->state_queue); queue_delayed_work(system_power_efficient_wq, &phydev->state_queue, 0); } @@ -693,7 +697,7 @@ static void phy_error(struct phy_device *phydev) phydev->state = PHY_HALTED; mutex_unlock(&phydev->lock); - phy_trigger_machine(phydev); + phy_trigger_machine(phydev, false); } /** @@ -840,7 +844,7 @@ void phy_change(struct phy_device *phydev) } /* reschedule state queue work to run as soon as possible */ - phy_trigger_machine(phydev); + phy_trigger_machine(phydev, true); return; ignore: @@ -942,7 +946,7 @@ void phy_start(struct phy_device *phydev) if (do_resume) phy_resume(phydev); - phy_trigger_machine(phydev); + phy_trigger_machine(phydev, true); } EXPORT_SYMBOL(phy_start); -- cgit v1.2.3 From 4078b76cac68e50ccf1f76a74e7d3d5788aec3fe Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 20 Jan 2017 16:05:05 -0800 Subject: net: dsa: Check return value of phy_connect_direct() We need to check the return value of phy_connect_direct() in dsa_slave_phy_connect() otherwise we may be continuing the initialization of a slave network device with a PHY that already attached somewhere else and which will soon be in error because the PHY device is in error. The conditions for such an error to occur are that we have a port of our switch that is not disabled, and has the same port number as a PHY address (say both 5) that can be probed using the DSA slave MII bus. We end-up having this slave network device find a PHY at the same address as our port number, and we try to attach to it. A slave network (e.g: port 0) has already attached to our PHY device, and we try to re-attach it with a different network device, but since we ignore the error we would end-up initializating incorrect device references by the time the slave network interface is opened. The code has been (re)organized several times, making it hard to provide an exact Fixes tag, this is a bugfix nonetheless. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/slave.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 68c9eea00518..ba1b6b9630d2 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -1105,10 +1105,8 @@ static int dsa_slave_phy_connect(struct dsa_slave_priv *p, /* Use already configured phy mode */ if (p->phy_interface == PHY_INTERFACE_MODE_NA) p->phy_interface = p->phy->interface; - phy_connect_direct(slave_dev, p->phy, dsa_slave_adjust_link, - p->phy_interface); - - return 0; + return phy_connect_direct(slave_dev, p->phy, dsa_slave_adjust_link, + p->phy_interface); } static int dsa_slave_phy_setup(struct dsa_slave_priv *p, -- cgit v1.2.3 From c5f21c9f878b8dcd54d0b9739c025ca73cb4c091 Mon Sep 17 00:00:00 2001 From: Ding Pixel Date: Wed, 18 Jan 2017 17:26:38 +0800 Subject: drm/amdgpu: check ring being ready before using Return success when the ring is properly initialized, otherwise return failure. Tonga SRIOV VF doesn't have UVD and VCE engines, the initialization of these IPs is bypassed. The system crashes if application submit IB to their rings which are not ready to use. It could be a common issue if IP having ring buffer is disabled for some reason on specific ASIC, so it should check the ring being ready to use. Bug: amdgpu_test crashes system on Tonga VF. Signed-off-by: Ding Pixel Reviewed-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 29d6d84d1c28..41e41f90265d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -83,6 +83,13 @@ int amdgpu_cs_get_ring(struct amdgpu_device *adev, u32 ip_type, } break; } + + if (!(*out_ring && (*out_ring)->adev)) { + DRM_ERROR("Ring %d is not initialized on IP %d\n", + ring, ip_type); + return -EINVAL; + } + return 0; } -- cgit v1.2.3 From 3a1d19a29670aa7eb58576a31883d0aa9fb77549 Mon Sep 17 00:00:00 2001 From: Xiangliang Yu Date: Thu, 19 Jan 2017 09:57:41 +0800 Subject: drm/amdgpu: fix unload driver issue for virtual display Virtual display doesn't allocate amdgpu_encoder when initializing, so will get invaild pointer if try to free amdgpu_encoder when unloading driver. Signed-off-by: Xiangliang Yu Reviewed-by: Alex Deucher Acked-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/dce_virtual.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c index 762f8e82ceb7..e9a176891e13 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c @@ -627,11 +627,8 @@ static const struct drm_encoder_helper_funcs dce_virtual_encoder_helper_funcs = static void dce_virtual_encoder_destroy(struct drm_encoder *encoder) { - struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder); - - kfree(amdgpu_encoder->enc_priv); drm_encoder_cleanup(encoder); - kfree(amdgpu_encoder); + kfree(encoder); } static const struct drm_encoder_funcs dce_virtual_encoder_funcs = { -- cgit v1.2.3 From b2fbd04498789def80ceba3d5bbc5af7f2f70a5f Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Fri, 20 Jan 2017 21:03:03 +0800 Subject: netfilter: nf_tables: validate the name size when possible Currently, if the user add a stateful object with the name size exceed NFT_OBJ_MAXNAMELEN - 1 (i.e. 31), we truncate it down to 31 silently. This is not friendly, furthermore, this will cause duplicated stateful objects when the first 31 characters of the name is same. So limit the stateful object's name size to NFT_OBJ_MAXNAMELEN - 1. After apply this patch, error message will be printed out like this: # name_32=$(printf "%0.sQ" {1..32}) # nft add counter filter $name_32 :1:1-52: Error: Could not process rule: Numerical result out of range add counter filter QQQQQQQQQQQQQQQQQQQQQQQQQQQQQQQQ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Also this patch cleans up the codes which missing the name size limit validation in nftables. Fixes: e50092404c1b ("netfilter: nf_tables: add stateful objects") Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 21 ++++++++++++++------- net/netfilter/nft_dynset.c | 3 ++- net/netfilter/nft_lookup.c | 3 ++- net/netfilter/nft_objref.c | 6 ++++-- 4 files changed, 22 insertions(+), 11 deletions(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 091d2dcc63b2..b84c7b25219b 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -928,7 +928,8 @@ static struct nft_chain *nf_tables_chain_lookup(const struct nft_table *table, } static const struct nla_policy nft_chain_policy[NFTA_CHAIN_MAX + 1] = { - [NFTA_CHAIN_TABLE] = { .type = NLA_STRING }, + [NFTA_CHAIN_TABLE] = { .type = NLA_STRING, + .len = NFT_TABLE_MAXNAMELEN - 1 }, [NFTA_CHAIN_HANDLE] = { .type = NLA_U64 }, [NFTA_CHAIN_NAME] = { .type = NLA_STRING, .len = NFT_CHAIN_MAXNAMELEN - 1 }, @@ -1854,7 +1855,8 @@ static struct nft_rule *nf_tables_rule_lookup(const struct nft_chain *chain, } static const struct nla_policy nft_rule_policy[NFTA_RULE_MAX + 1] = { - [NFTA_RULE_TABLE] = { .type = NLA_STRING }, + [NFTA_RULE_TABLE] = { .type = NLA_STRING, + .len = NFT_TABLE_MAXNAMELEN - 1 }, [NFTA_RULE_CHAIN] = { .type = NLA_STRING, .len = NFT_CHAIN_MAXNAMELEN - 1 }, [NFTA_RULE_HANDLE] = { .type = NLA_U64 }, @@ -2443,7 +2445,8 @@ nft_select_set_ops(const struct nlattr * const nla[], } static const struct nla_policy nft_set_policy[NFTA_SET_MAX + 1] = { - [NFTA_SET_TABLE] = { .type = NLA_STRING }, + [NFTA_SET_TABLE] = { .type = NLA_STRING, + .len = NFT_TABLE_MAXNAMELEN - 1 }, [NFTA_SET_NAME] = { .type = NLA_STRING, .len = NFT_SET_MAXNAMELEN - 1 }, [NFTA_SET_FLAGS] = { .type = NLA_U32 }, @@ -3192,8 +3195,10 @@ static const struct nla_policy nft_set_elem_policy[NFTA_SET_ELEM_MAX + 1] = { }; static const struct nla_policy nft_set_elem_list_policy[NFTA_SET_ELEM_LIST_MAX + 1] = { - [NFTA_SET_ELEM_LIST_TABLE] = { .type = NLA_STRING }, - [NFTA_SET_ELEM_LIST_SET] = { .type = NLA_STRING }, + [NFTA_SET_ELEM_LIST_TABLE] = { .type = NLA_STRING, + .len = NFT_TABLE_MAXNAMELEN - 1 }, + [NFTA_SET_ELEM_LIST_SET] = { .type = NLA_STRING, + .len = NFT_SET_MAXNAMELEN - 1 }, [NFTA_SET_ELEM_LIST_ELEMENTS] = { .type = NLA_NESTED }, [NFTA_SET_ELEM_LIST_SET_ID] = { .type = NLA_U32 }, }; @@ -4032,8 +4037,10 @@ struct nft_object *nf_tables_obj_lookup(const struct nft_table *table, EXPORT_SYMBOL_GPL(nf_tables_obj_lookup); static const struct nla_policy nft_obj_policy[NFTA_OBJ_MAX + 1] = { - [NFTA_OBJ_TABLE] = { .type = NLA_STRING }, - [NFTA_OBJ_NAME] = { .type = NLA_STRING }, + [NFTA_OBJ_TABLE] = { .type = NLA_STRING, + .len = NFT_TABLE_MAXNAMELEN - 1 }, + [NFTA_OBJ_NAME] = { .type = NLA_STRING, + .len = NFT_OBJ_MAXNAMELEN - 1 }, [NFTA_OBJ_TYPE] = { .type = NLA_U32 }, [NFTA_OBJ_DATA] = { .type = NLA_NESTED }, }; diff --git a/net/netfilter/nft_dynset.c b/net/netfilter/nft_dynset.c index 7de2f46734a4..049ad2d9ee66 100644 --- a/net/netfilter/nft_dynset.c +++ b/net/netfilter/nft_dynset.c @@ -98,7 +98,8 @@ out: } static const struct nla_policy nft_dynset_policy[NFTA_DYNSET_MAX + 1] = { - [NFTA_DYNSET_SET_NAME] = { .type = NLA_STRING }, + [NFTA_DYNSET_SET_NAME] = { .type = NLA_STRING, + .len = NFT_SET_MAXNAMELEN - 1 }, [NFTA_DYNSET_SET_ID] = { .type = NLA_U32 }, [NFTA_DYNSET_OP] = { .type = NLA_U32 }, [NFTA_DYNSET_SREG_KEY] = { .type = NLA_U32 }, diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index d4f97fa7e21d..e21aea7e5ec8 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -49,7 +49,8 @@ static void nft_lookup_eval(const struct nft_expr *expr, } static const struct nla_policy nft_lookup_policy[NFTA_LOOKUP_MAX + 1] = { - [NFTA_LOOKUP_SET] = { .type = NLA_STRING }, + [NFTA_LOOKUP_SET] = { .type = NLA_STRING, + .len = NFT_SET_MAXNAMELEN - 1 }, [NFTA_LOOKUP_SET_ID] = { .type = NLA_U32 }, [NFTA_LOOKUP_SREG] = { .type = NLA_U32 }, [NFTA_LOOKUP_DREG] = { .type = NLA_U32 }, diff --git a/net/netfilter/nft_objref.c b/net/netfilter/nft_objref.c index 415a65ba2b85..1ae8c49ca4a1 100644 --- a/net/netfilter/nft_objref.c +++ b/net/netfilter/nft_objref.c @@ -193,10 +193,12 @@ nft_objref_select_ops(const struct nft_ctx *ctx, } static const struct nla_policy nft_objref_policy[NFTA_OBJREF_MAX + 1] = { - [NFTA_OBJREF_IMM_NAME] = { .type = NLA_STRING }, + [NFTA_OBJREF_IMM_NAME] = { .type = NLA_STRING, + .len = NFT_OBJ_MAXNAMELEN - 1 }, [NFTA_OBJREF_IMM_TYPE] = { .type = NLA_U32 }, [NFTA_OBJREF_SET_SREG] = { .type = NLA_U32 }, - [NFTA_OBJREF_SET_NAME] = { .type = NLA_STRING }, + [NFTA_OBJREF_SET_NAME] = { .type = NLA_STRING, + .len = NFT_SET_MAXNAMELEN - 1 }, [NFTA_OBJREF_SET_ID] = { .type = NLA_U32 }, }; -- cgit v1.2.3 From 8ac092519ad91931c96d306c4bfae2c6587c325f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 23 Jan 2017 22:44:12 -0500 Subject: NFSv4.1: Fix a deadlock in layoutget We cannot call nfs4_handle_exception() without first ensuring that the slot has been freed. If not, we end up deadlocking with the process waiting for recovery to complete, and recovery waiting for the slot table to drain. Fixes: 2e80dbe7ac51 ("NFSv4.1: Close callback races for OPEN, LAYOUTGET...") Cc: stable@vger.kernel.org # v4.8+ Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index ecc151697fd4..59bb574d7d7c 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -8490,6 +8490,7 @@ nfs4_layoutget_handle_exception(struct rpc_task *task, goto out; } + nfs4_sequence_free_slot(&lgp->res.seq_res); err = nfs4_handle_exception(server, nfs4err, exception); if (!status) { if (exception->retry) -- cgit v1.2.3 From af2b7fa17eb92e52b65f96604448ff7a2a89ee99 Mon Sep 17 00:00:00 2001 From: Darren Stevens Date: Mon, 23 Jan 2017 19:42:54 +0000 Subject: powerpc: Add missing error check to prom_find_boot_cpu() prom_init.c calls 'instance-to-package' twice, but the return is not checked during prom_find_boot_cpu(). The result is then passed to prom_getprop(), which could be PROM_ERROR. Add a return check to prevent this. This was found on a pasemi system, where CFE doesn't have a working 'instance-to package' prom call. Before Commit 5c0484e25ec0 ('powerpc: Endian safe trampoline') the area around addr 0 was mostly 0's and this doesn't cause a problem. Once the macro 'FIXUP_ENDIAN' has been added to head_64.S, the low memory area now has non-zero values, which cause the prom_getprop() call to hang. mpe: Also confirmed that under SLOF if 'instance-to-package' did fail with PROM_ERROR we would crash in SLOF. So the bug is not specific to CFE, it's just that other open firmwares don't trigger it because they have a working 'instance-to-package'. Fixes: 5c0484e25ec0 ("powerpc: Endian safe trampoline") Cc: stable@vger.kernel.org # v3.13+ Signed-off-by: Darren Stevens Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/prom_init.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index ec47a939cbdd..ac83eb04a8b8 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -2834,6 +2834,9 @@ static void __init prom_find_boot_cpu(void) cpu_pkg = call_prom("instance-to-package", 1, 1, prom_cpu); + if (!PHANDLE_VALID(cpu_pkg)) + return; + prom_getprop(cpu_pkg, "reg", &rval, sizeof(rval)); prom.cpu = be32_to_cpu(rval); -- cgit v1.2.3 From 9dce990d2cf57b5ed4e71a9cdbd7eae4335111ff Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Tue, 24 Jan 2017 08:05:52 +0100 Subject: s390/ptrace: Preserve previous registers for short regset write Ensure that if userspace supplies insufficient data to PTRACE_SETREGSET to fill all the registers, the thread's old registers are preserved. convert_vx_to_fp() is adapted to handle only a specified number of registers rather than unconditionally handling all of them: other callers of this function are adapted appropriately. Based on an initial patch by Dave Martin. Cc: stable@vger.kernel.org Reported-by: Dave Martin Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/ptrace.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 7447ba509c30..12020b55887b 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -963,6 +963,11 @@ static int s390_fpregs_set(struct task_struct *target, if (target == current) save_fpu_regs(); + if (MACHINE_HAS_VX) + convert_vx_to_fp(fprs, target->thread.fpu.vxrs); + else + memcpy(&fprs, target->thread.fpu.fprs, sizeof(fprs)); + /* If setting FPC, must validate it first. */ if (count > 0 && pos < offsetof(s390_fp_regs, fprs)) { u32 ufpc[2] = { target->thread.fpu.fpc, 0 }; @@ -1067,6 +1072,9 @@ static int s390_vxrs_low_set(struct task_struct *target, if (target == current) save_fpu_regs(); + for (i = 0; i < __NUM_VXRS_LOW; i++) + vxrs[i] = *((__u64 *)(target->thread.fpu.vxrs + i) + 1); + rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1); if (rc == 0) for (i = 0; i < __NUM_VXRS_LOW; i++) -- cgit v1.2.3 From dffba9a31c7769be3231c420d4b364c92ba3f1ac Mon Sep 17 00:00:00 2001 From: Yu-cheng Yu Date: Mon, 23 Jan 2017 14:54:44 -0800 Subject: x86/fpu/xstate: Fix xcomp_bv in XSAVES header The compacted-format XSAVES area is determined at boot time and never changed after. The field xsave.header.xcomp_bv indicates which components are in the fixed XSAVES format. In fpstate_init() we did not set xcomp_bv to reflect the XSAVES format since at the time there is no valid data. However, after we do copy_init_fpstate_to_fpregs() in fpu__clear(), as in commit: b22cbe404a9c x86/fpu: Fix invalid FPU ptrace state after execve() and when __fpu_restore_sig() does fpu__restore() for a COMPAT-mode app, a #GP occurs. This can be easily triggered by doing valgrind on a COMPAT-mode "Hello World," as reported by Joakim Tjernlund and others: https://bugzilla.kernel.org/show_bug.cgi?id=190061 Fix it by setting xcomp_bv correctly. This patch also moves the xcomp_bv initialization to the proper place, which was in copyin_to_xsaves() as of: 4c833368f0bf x86/fpu: Set the xcomp_bv when we fake up a XSAVES area which fixed the bug too, but it's more efficient and cleaner to initialize things once per boot, not for every signal handling operation. Reported-by: Kevin Hao Reported-by: Joakim Tjernlund Signed-off-by: Yu-cheng Yu Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Dave Hansen Cc: Fenghua Yu Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Ravi V. Shankar Cc: Thomas Gleixner Cc: haokexin@gmail.com Link: http://lkml.kernel.org/r/1485212084-4418-1-git-send-email-yu-cheng.yu@intel.com [ Combined it with 4c833368f0bf. ] Signed-off-by: Ingo Molnar --- arch/x86/kernel/fpu/core.c | 4 +++- arch/x86/kernel/fpu/xstate.c | 1 - 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/fpu/core.c b/arch/x86/kernel/fpu/core.c index e4e97a5355ce..de7234401275 100644 --- a/arch/x86/kernel/fpu/core.c +++ b/arch/x86/kernel/fpu/core.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include @@ -183,7 +184,8 @@ void fpstate_init(union fpregs_state *state) * it will #GP. Make sure it is replaced after the memset(). */ if (static_cpu_has(X86_FEATURE_XSAVES)) - state->xsave.header.xcomp_bv = XCOMP_BV_COMPACTED_FORMAT; + state->xsave.header.xcomp_bv = XCOMP_BV_COMPACTED_FORMAT | + xfeatures_mask; if (static_cpu_has(X86_FEATURE_FXSR)) fpstate_init_fxstate(&state->fxsave); diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c index e287b9075527..1d7770447b3e 100644 --- a/arch/x86/kernel/fpu/xstate.c +++ b/arch/x86/kernel/fpu/xstate.c @@ -1070,7 +1070,6 @@ int copyin_to_xsaves(const void *kbuf, const void __user *ubuf, * Add back in the features that came in from userspace: */ xsave->header.xfeatures |= xfeatures; - xsave->header.xcomp_bv = XCOMP_BV_COMPACTED_FORMAT | xsave->header.xfeatures; return 0; } -- cgit v1.2.3 From f05fea5b3574a5926c53865eea27139bb40b2f2b Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Thu, 19 Jan 2017 10:10:16 +1100 Subject: powerpc/eeh: Fix wrong flag passed to eeh_unfreeze_pe() In __eeh_clear_pe_frozen_state(), we should pass the flag's value instead of its address to eeh_unfreeze_pe(). The isolated flag is cleared if no error returned from __eeh_clear_pe_frozen_state(). We never observed the error from the function. So the isolated flag should have been always cleared, no real issue is caused because of the misused @flag. This fixes the code by passing the value of @flag to eeh_unfreeze_pe(). Fixes: 5cfb20b96f6 ("powerpc/eeh: Emulate EEH recovery for VFIO devices") Cc: stable@vger.kernel.org # v3.18+ Signed-off-by: Gavin Shan Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/eeh_driver.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/kernel/eeh_driver.c b/arch/powerpc/kernel/eeh_driver.c index d88573bdd090..b94887165a10 100644 --- a/arch/powerpc/kernel/eeh_driver.c +++ b/arch/powerpc/kernel/eeh_driver.c @@ -545,7 +545,7 @@ static void *eeh_pe_detach_dev(void *data, void *userdata) static void *__eeh_clear_pe_frozen_state(void *data, void *flag) { struct eeh_pe *pe = (struct eeh_pe *)data; - bool *clear_sw_state = flag; + bool clear_sw_state = *(bool *)flag; int i, rc = 1; for (i = 0; rc && i < 3; i++) -- cgit v1.2.3 From f2574030b0e33263b8a1c28fa3c4fa9292283799 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 24 Jan 2017 21:37:20 +1100 Subject: powerpc: Revert the initial stack protector support Unfortunately the stack protector support we merged recently only works on some toolchains. If the toolchain is built without glibc support everything works fine, but if glibc is built then it leads to a panic at boot. The solution is not rc5 material, so revert the support for now. This reverts commits: 6533b7c16ee5 ("powerpc: Initial stack protector (-fstack-protector) support") 902e06eb86cd ("powerpc/32: Change the stack protector canary value per task") Fixes: 6533b7c16ee5 ("powerpc: Initial stack protector (-fstack-protector) support") Signed-off-by: Michael Ellerman --- arch/powerpc/Kconfig | 1 - arch/powerpc/include/asm/stackprotector.h | 40 ------------------------------- arch/powerpc/kernel/Makefile | 4 ---- arch/powerpc/kernel/asm-offsets.c | 3 --- arch/powerpc/kernel/entry_32.S | 6 +---- arch/powerpc/kernel/process.c | 6 ----- 6 files changed, 1 insertion(+), 59 deletions(-) delete mode 100644 arch/powerpc/include/asm/stackprotector.h diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index a8ee573fe610..a46d1c0d14d3 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -164,7 +164,6 @@ config PPC select ARCH_HAS_SCALED_CPUTIME if VIRT_CPU_ACCOUNTING_NATIVE select HAVE_ARCH_HARDENED_USERCOPY select HAVE_KERNEL_GZIP - select HAVE_CC_STACKPROTECTOR config GENERIC_CSUM def_bool CPU_LITTLE_ENDIAN diff --git a/arch/powerpc/include/asm/stackprotector.h b/arch/powerpc/include/asm/stackprotector.h deleted file mode 100644 index 6720190eabec..000000000000 --- a/arch/powerpc/include/asm/stackprotector.h +++ /dev/null @@ -1,40 +0,0 @@ -/* - * GCC stack protector support. - * - * Stack protector works by putting predefined pattern at the start of - * the stack frame and verifying that it hasn't been overwritten when - * returning from the function. The pattern is called stack canary - * and gcc expects it to be defined by a global variable called - * "__stack_chk_guard" on PPC. This unfortunately means that on SMP - * we cannot have a different canary value per task. - */ - -#ifndef _ASM_STACKPROTECTOR_H -#define _ASM_STACKPROTECTOR_H - -#include -#include -#include - -extern unsigned long __stack_chk_guard; - -/* - * Initialize the stackprotector canary value. - * - * NOTE: this must only be called from functions that never return, - * and it must always be inlined. - */ -static __always_inline void boot_init_stack_canary(void) -{ - unsigned long canary; - - /* Try to get a semi random initial value. */ - get_random_bytes(&canary, sizeof(canary)); - canary ^= mftb(); - canary ^= LINUX_VERSION_CODE; - - current->stack_canary = canary; - __stack_chk_guard = current->stack_canary; -} - -#endif /* _ASM_STACKPROTECTOR_H */ diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index 23f8082d7bfa..f4c2b52e58b3 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -19,10 +19,6 @@ CFLAGS_init.o += $(DISABLE_LATENT_ENTROPY_PLUGIN) CFLAGS_btext.o += $(DISABLE_LATENT_ENTROPY_PLUGIN) CFLAGS_prom.o += $(DISABLE_LATENT_ENTROPY_PLUGIN) -# -fstack-protector triggers protection checks in this code, -# but it is being used too early to link to meaningful stack_chk logic. -CFLAGS_prom_init.o += $(call cc-option, -fno-stack-protector) - ifdef CONFIG_FUNCTION_TRACER # Do not trace early boot code CFLAGS_REMOVE_cputable.o = -mno-sched-epilog $(CC_FLAGS_FTRACE) diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c index 0601e6a7297c..195a9fc8f81c 100644 --- a/arch/powerpc/kernel/asm-offsets.c +++ b/arch/powerpc/kernel/asm-offsets.c @@ -91,9 +91,6 @@ int main(void) DEFINE(TI_livepatch_sp, offsetof(struct thread_info, livepatch_sp)); #endif -#ifdef CONFIG_CC_STACKPROTECTOR - DEFINE(TSK_STACK_CANARY, offsetof(struct task_struct, stack_canary)); -#endif DEFINE(KSP, offsetof(struct thread_struct, ksp)); DEFINE(PT_REGS, offsetof(struct thread_struct, regs)); #ifdef CONFIG_BOOKE diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S index 5742dbdbee46..3841d749a430 100644 --- a/arch/powerpc/kernel/entry_32.S +++ b/arch/powerpc/kernel/entry_32.S @@ -674,11 +674,7 @@ BEGIN_FTR_SECTION mtspr SPRN_SPEFSCR,r0 /* restore SPEFSCR reg */ END_FTR_SECTION_IFSET(CPU_FTR_SPE) #endif /* CONFIG_SPE */ -#if defined(CONFIG_CC_STACKPROTECTOR) && !defined(CONFIG_SMP) - lwz r0,TSK_STACK_CANARY(r2) - lis r4,__stack_chk_guard@ha - stw r0,__stack_chk_guard@l(r4) -#endif + lwz r0,_CCR(r1) mtcrf 0xFF,r0 /* r3-r12 are destroyed -- Cort */ diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 04885cec24df..5dd056df0baa 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -64,12 +64,6 @@ #include #include -#ifdef CONFIG_CC_STACKPROTECTOR -#include -unsigned long __stack_chk_guard __read_mostly; -EXPORT_SYMBOL(__stack_chk_guard); -#endif - /* Transactional Memory debug */ #ifdef TM_DEBUG_SW #define TM_DEBUG(x...) printk(KERN_INFO x) -- cgit v1.2.3 From 5d03a2fd2292e71936c4235885c35ccc3c94695b Mon Sep 17 00:00:00 2001 From: Bjørn Mork Date: Tue, 24 Jan 2017 10:31:18 +0100 Subject: USB: serial: option: add device ID for HP lt2523 (Novatel E371) Yet another laptop vendor rebranded Novatel E371. Cc: stable@vger.kernel.org Signed-off-by: Bjørn Mork Signed-off-by: Johan Hovold --- drivers/usb/serial/option.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index 7ce31a4c7e7f..42cc72e54c05 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -2007,6 +2007,7 @@ static const struct usb_device_id option_ids[] = { { USB_DEVICE_AND_INTERFACE_INFO(WETELECOM_VENDOR_ID, WETELECOM_PRODUCT_WMD200, 0xff, 0xff, 0xff) }, { USB_DEVICE_AND_INTERFACE_INFO(WETELECOM_VENDOR_ID, WETELECOM_PRODUCT_6802, 0xff, 0xff, 0xff) }, { USB_DEVICE_AND_INTERFACE_INFO(WETELECOM_VENDOR_ID, WETELECOM_PRODUCT_WMD300, 0xff, 0xff, 0xff) }, + { USB_DEVICE_AND_INTERFACE_INFO(0x03f0, 0x421d, 0xff, 0xff, 0xff) }, /* HP lt2523 (Novatel E371) */ { } /* Terminating entry */ }; MODULE_DEVICE_TABLE(usb, option_ids); -- cgit v1.2.3 From 690e5325b8c7d5db05fc569c0f7b888bb4248272 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 24 Jan 2017 14:50:19 +0100 Subject: block: fix use after free in __blkdev_direct_IO We can't dereference the dio structure after submitting the last bio for this request, as I/O completion might have happened before the code is run. Introduce a local is_sync variable instead. Fixes: 542ff7bf ("block: new direct I/O implementation") Signed-off-by: Christoph Hellwig Reported-by: Matias Bjørling Tested-by: Matias Bjørling Signed-off-by: Jens Axboe --- fs/block_dev.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/block_dev.c b/fs/block_dev.c index 5db5d1340d69..3c47614a4b32 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -331,7 +331,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages) struct blk_plug plug; struct blkdev_dio *dio; struct bio *bio; - bool is_read = (iov_iter_rw(iter) == READ); + bool is_read = (iov_iter_rw(iter) == READ), is_sync; loff_t pos = iocb->ki_pos; blk_qc_t qc = BLK_QC_T_NONE; int ret; @@ -344,7 +344,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages) bio_get(bio); /* extra ref for the completion handler */ dio = container_of(bio, struct blkdev_dio, bio); - dio->is_sync = is_sync_kiocb(iocb); + dio->is_sync = is_sync = is_sync_kiocb(iocb); if (dio->is_sync) dio->waiter = current; else @@ -398,7 +398,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages) } blk_finish_plug(&plug); - if (!dio->is_sync) + if (!is_sync) return -EIOCBQUEUED; for (;;) { -- cgit v1.2.3 From 0d6da872d3e4a60f43c295386d7ff9a4cdcd57e9 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Mon, 23 Jan 2017 22:59:44 +0100 Subject: s390/mm: Fix cmma unused transfer from pgste into pte The last pgtable rework silently disabled the CMMA unused state by setting a local pte variable (a parameter) instead of propagating it back into the caller. Fix it. Fixes: ebde765c0e85 ("s390/mm: uninline ptep_xxx functions from pgtable.h") Cc: stable@vger.kernel.org # v4.6+ Cc: Martin Schwidefsky Cc: Claudio Imbrenda Signed-off-by: Christian Borntraeger Signed-off-by: Martin Schwidefsky --- arch/s390/mm/pgtable.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c index 7a1897c51c54..d56ef26d4681 100644 --- a/arch/s390/mm/pgtable.c +++ b/arch/s390/mm/pgtable.c @@ -202,7 +202,7 @@ static inline pgste_t ptep_xchg_start(struct mm_struct *mm, return pgste; } -static inline void ptep_xchg_commit(struct mm_struct *mm, +static inline pte_t ptep_xchg_commit(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pgste_t pgste, pte_t old, pte_t new) { @@ -220,6 +220,7 @@ static inline void ptep_xchg_commit(struct mm_struct *mm, } else { *ptep = new; } + return old; } pte_t ptep_xchg_direct(struct mm_struct *mm, unsigned long addr, @@ -231,7 +232,7 @@ pte_t ptep_xchg_direct(struct mm_struct *mm, unsigned long addr, preempt_disable(); pgste = ptep_xchg_start(mm, addr, ptep); old = ptep_flush_direct(mm, addr, ptep); - ptep_xchg_commit(mm, addr, ptep, pgste, old, new); + old = ptep_xchg_commit(mm, addr, ptep, pgste, old, new); preempt_enable(); return old; } @@ -246,7 +247,7 @@ pte_t ptep_xchg_lazy(struct mm_struct *mm, unsigned long addr, preempt_disable(); pgste = ptep_xchg_start(mm, addr, ptep); old = ptep_flush_lazy(mm, addr, ptep); - ptep_xchg_commit(mm, addr, ptep, pgste, old, new); + old = ptep_xchg_commit(mm, addr, ptep, pgste, old, new); preempt_enable(); return old; } -- cgit v1.2.3 From 115865fa0826ed18ca04717cf72d0fe874c0fe7f Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 23 Jan 2017 09:29:09 +0100 Subject: mac80211: don't try to sleep in rate_control_rate_init() In my previous patch, I missed that rate_control_rate_init() is called from some places that cannot sleep, so it cannot call ieee80211_recalc_min_chandef(). Remove that call for now to fix the context bug, we'll have to find a different way to fix the minimum channel width issue. Fixes: 96aa2e7cf126 ("mac80211: calculate min channel width correctly") Reported-by: Xiaolong Ye (via lkp-robot) Signed-off-by: Johannes Berg --- net/mac80211/rate.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/net/mac80211/rate.c b/net/mac80211/rate.c index 9e2641d45587..206698bc93f4 100644 --- a/net/mac80211/rate.c +++ b/net/mac80211/rate.c @@ -40,8 +40,6 @@ void rate_control_rate_init(struct sta_info *sta) ieee80211_sta_set_rx_nss(sta); - ieee80211_recalc_min_chandef(sta->sdata); - if (!ref) return; -- cgit v1.2.3 From ad8e66b4a80182174f73487ed25fd2140cf43361 Mon Sep 17 00:00:00 2001 From: Israel Rukshin Date: Wed, 28 Dec 2016 12:48:28 +0200 Subject: IB/srp: fix mr allocation when the device supports sg gaps If the device support arbitrary sg list mapping (device cap IB_DEVICE_SG_GAPS_REG set) we allocate the memory regions with IB_MR_TYPE_SG_GAPS. Fixes: 509c5f33f4f6 ("IB/srp: Prevent mapping failures") Cc: # 4.7+ Signed-off-by: Israel Rukshin Signed-off-by: Max Gurtovoy Reviewed-by: Leon Romanovsky Reviewed-by: Mark Bloch Reviewed-by: Yuval Shaia Reviewed-by: Bart Van Assche Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/srp/ib_srp.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c index 8ddc07123193..0f67cf909462 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.c +++ b/drivers/infiniband/ulp/srp/ib_srp.c @@ -371,6 +371,7 @@ static struct srp_fr_pool *srp_create_fr_pool(struct ib_device *device, struct srp_fr_desc *d; struct ib_mr *mr; int i, ret = -EINVAL; + enum ib_mr_type mr_type; if (pool_size <= 0) goto err; @@ -384,9 +385,13 @@ static struct srp_fr_pool *srp_create_fr_pool(struct ib_device *device, spin_lock_init(&pool->lock); INIT_LIST_HEAD(&pool->free_list); + if (device->attrs.device_cap_flags & IB_DEVICE_SG_GAPS_REG) + mr_type = IB_MR_TYPE_SG_GAPS; + else + mr_type = IB_MR_TYPE_MEM_REG; + for (i = 0, d = &pool->desc[0]; i < pool->size; i++, d++) { - mr = ib_alloc_mr(pd, IB_MR_TYPE_MEM_REG, - max_page_list_len); + mr = ib_alloc_mr(pd, mr_type, max_page_list_len); if (IS_ERR(mr)) { ret = PTR_ERR(mr); if (ret == -ENOMEM) -- cgit v1.2.3 From 0a475ef4226e305bdcffe12b401ca1eab06c4913 Mon Sep 17 00:00:00 2001 From: Israel Rukshin Date: Wed, 4 Jan 2017 15:59:37 +0200 Subject: IB/srp: fix invalid indirect_sg_entries parameter value After setting indirect_sg_entries module_param to huge value (e.g 500,000), srp_alloc_req_data() fails to allocate indirect descriptors for the request ring (kmalloc fails). This commit enforces the maximum value of indirect_sg_entries to be SG_MAX_SEGMENTS as signified in module param description. Fixes: 65e8617fba17 (scsi: rename SCSI_MAX_{SG, SG_CHAIN}_SEGMENTS) Fixes: c07d424d6118 (IB/srp: add support for indirect tables that don't fit in SRP_CMD) Cc: stable@vger.kernel.org # 4.7+ Signed-off-by: Israel Rukshin Signed-off-by: Max Gurtovoy Reviewed-by: Laurence Oberman Reviewed-by: Bart Van Assche -- Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/srp/ib_srp.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c index 0f67cf909462..79bf48477ddb 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.c +++ b/drivers/infiniband/ulp/srp/ib_srp.c @@ -3699,6 +3699,12 @@ static int __init srp_init_module(void) indirect_sg_entries = cmd_sg_entries; } + if (indirect_sg_entries > SG_MAX_SEGMENTS) { + pr_warn("Clamping indirect_sg_entries to %u\n", + SG_MAX_SEGMENTS); + indirect_sg_entries = SG_MAX_SEGMENTS; + } + srp_remove_wq = create_workqueue("srp_remove"); if (!srp_remove_wq) { ret = -ENOMEM; -- cgit v1.2.3 From 1e5db6c31ade4150c2e2b1a21e39f776c38fea39 Mon Sep 17 00:00:00 2001 From: Max Gurtovoy Date: Wed, 18 Jan 2017 00:40:39 +0200 Subject: IB/iser: Fix sg_tablesize calculation For devices that can register page list that is bigger than USHRT_MAX, we actually take the wrong value for sg_tablesize. E.g: for CX4 max_fast_reg_page_list_len is 65536 (bigger than USHRT_MAX) so we set sg_tablesize to 0 by mistake. Therefore, each IO that is bigger than 4k splitted to "< 4k" chunks that cause performance degredation. Remove wrong sg_tablesize assignment, and use the value that was set during address resolution handler with the needed casting. Cc: # v4.5+ Signed-off-by: Max Gurtovoy Reviewed-by: Sagi Grimberg Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/iser/iscsi_iser.c | 7 ------- 1 file changed, 7 deletions(-) diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index 9104e6b8cac9..1c911876556e 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -651,13 +651,6 @@ iscsi_iser_session_create(struct iscsi_endpoint *ep, SHOST_DIX_GUARD_CRC); } - /* - * Limit the sg_tablesize and max_sectors based on the device - * max fastreg page list length. - */ - shost->sg_tablesize = min_t(unsigned short, shost->sg_tablesize, - ib_conn->device->ib_device->attrs.max_fast_reg_page_list_len); - if (iscsi_host_add(shost, ib_conn->device->ib_device->dma_device)) { mutex_unlock(&iser_conn->state_mutex); -- cgit v1.2.3 From 83236f0157feec0f01bf688a1474b889bdcc5ad0 Mon Sep 17 00:00:00 2001 From: Max Gurtovoy Date: Wed, 18 Jan 2017 00:40:40 +0200 Subject: IB/iser: remove unused variable from iser_conn struct max_sectors calculation was fixed in commit: 9c674815d346 ("IB/iser: Fix max_sectors calculation"). Thus, iser_conn variable scsi_max_sectors is not needed anymore. Signed-off-by: Max Gurtovoy Reviewed-by: Sagi Grimberg Tested-by: Raju Rangoju Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/iser/iscsi_iser.c | 4 ++++ drivers/infiniband/ulp/iser/iscsi_iser.h | 2 -- drivers/infiniband/ulp/iser/iser_verbs.c | 13 +------------ 3 files changed, 5 insertions(+), 14 deletions(-) diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index 1c911876556e..e71af717e71b 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -672,6 +672,10 @@ iscsi_iser_session_create(struct iscsi_endpoint *ep, max_fr_sectors = ((shost->sg_tablesize - 1) * PAGE_SIZE) >> 9; shost->max_sectors = min(iser_max_sectors, max_fr_sectors); + iser_dbg("iser_conn %p, sg_tablesize %u, max_sectors %u\n", + iser_conn, shost->sg_tablesize, + shost->max_sectors); + if (cmds_max > max_cmds) { iser_info("cmds_max changed from %u to %u\n", cmds_max, max_cmds); diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h b/drivers/infiniband/ulp/iser/iscsi_iser.h index 0be6a7c5ddb5..9d0b22ad58c1 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.h +++ b/drivers/infiniband/ulp/iser/iscsi_iser.h @@ -496,7 +496,6 @@ struct ib_conn { * @rx_descs: rx buffers array (cyclic buffer) * @num_rx_descs: number of rx descriptors * @scsi_sg_tablesize: scsi host sg_tablesize - * @scsi_max_sectors: scsi host max sectors */ struct iser_conn { struct ib_conn ib_conn; @@ -519,7 +518,6 @@ struct iser_conn { struct iser_rx_desc *rx_descs; u32 num_rx_descs; unsigned short scsi_sg_tablesize; - unsigned int scsi_max_sectors; bool snd_w_inv; }; diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index 8ae7a3beddb7..6a9d1cb548ee 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -707,18 +707,7 @@ iser_calc_scsi_params(struct iser_conn *iser_conn, sup_sg_tablesize = min_t(unsigned, ISCSI_ISER_MAX_SG_TABLESIZE, device->ib_device->attrs.max_fast_reg_page_list_len); - if (sg_tablesize > sup_sg_tablesize) { - sg_tablesize = sup_sg_tablesize; - iser_conn->scsi_max_sectors = sg_tablesize * SIZE_4K / 512; - } else { - iser_conn->scsi_max_sectors = max_sectors; - } - - iser_conn->scsi_sg_tablesize = sg_tablesize; - - iser_dbg("iser_conn %p, sg_tablesize %u, max_sectors %u\n", - iser_conn, iser_conn->scsi_sg_tablesize, - iser_conn->scsi_max_sectors); + iser_conn->scsi_sg_tablesize = min(sg_tablesize, sup_sg_tablesize); } /** -- cgit v1.2.3 From bd00fdf198e2da475a2f4265a83686ab42d998a8 Mon Sep 17 00:00:00 2001 From: Greg Kurz Date: Tue, 24 Jan 2017 17:50:26 +0100 Subject: vfio/spapr: fail tce_iommu_attach_group() when iommu_data is null The recently added mediated VFIO driver doesn't know about powerpc iommu. It thus doesn't register a struct iommu_table_group in the iommu group upon device creation. The iommu_data pointer hence remains null. This causes a kernel oops when userspace tries to set the iommu type of a container associated with a mediated device to VFIO_SPAPR_TCE_v2_IOMMU. [ 82.585440] mtty mtty: MDEV: Registered [ 87.655522] iommu: Adding device 83b8f4f2-509f-382f-3c1e-e6bfe0fa1001 to group 10 [ 87.655527] vfio_mdev 83b8f4f2-509f-382f-3c1e-e6bfe0fa1001: MDEV: group_id = 10 [ 116.297184] Unable to handle kernel paging request for data at address 0x00000030 [ 116.297389] Faulting instruction address: 0xd000000007870524 [ 116.297465] Oops: Kernel access of bad area, sig: 11 [#1] [ 116.297611] SMP NR_CPUS=2048 [ 116.297611] NUMA [ 116.297627] PowerNV ... [ 116.297954] CPU: 33 PID: 7067 Comm: qemu-system-ppc Not tainted 4.10.0-rc5-mdev-test #8 [ 116.297993] task: c000000e7718b680 task.stack: c000000e77214000 [ 116.298025] NIP: d000000007870524 LR: d000000007870518 CTR: 0000000000000000 [ 116.298064] REGS: c000000e77217990 TRAP: 0300 Not tainted (4.10.0-rc5-mdev-test) [ 116.298103] MSR: 9000000000009033 [ 116.298107] CR: 84004444 XER: 00000000 [ 116.298154] CFAR: c00000000000888c DAR: 0000000000000030 DSISR: 40000000 SOFTE: 1 GPR00: d000000007870518 c000000e77217c10 d00000000787b0ed c000000eed2103c0 GPR04: 0000000000000000 0000000000000000 c000000eed2103e0 0000000f24320000 GPR08: 0000000000000104 0000000000000001 0000000000000000 d0000000078729b0 GPR12: c00000000025b7e0 c00000000fe08400 0000000000000001 000001002d31d100 GPR16: 000001002c22c850 00003ffff315c750 0000000043145680 0000000043141bc0 GPR20: ffffffffffffffed fffffffffffff000 0000000020003b65 d000000007706018 GPR24: c000000f16cf0d98 d000000007706000 c000000003f42980 c000000003f42980 GPR28: c000000f1575ac00 c000000003f429c8 0000000000000000 c000000eed2103c0 [ 116.298504] NIP [d000000007870524] tce_iommu_attach_group+0x10c/0x360 [vfio_iommu_spapr_tce] [ 116.298555] LR [d000000007870518] tce_iommu_attach_group+0x100/0x360 [vfio_iommu_spapr_tce] [ 116.298601] Call Trace: [ 116.298610] [c000000e77217c10] [d000000007870518] tce_iommu_attach_group+0x100/0x360 [vfio_iommu_spapr_tce] (unreliable) [ 116.298671] [c000000e77217cb0] [d0000000077033a0] vfio_fops_unl_ioctl+0x278/0x3e0 [vfio] [ 116.298713] [c000000e77217d40] [c0000000002a3ebc] do_vfs_ioctl+0xcc/0x8b0 [ 116.298745] [c000000e77217de0] [c0000000002a4700] SyS_ioctl+0x60/0xc0 [ 116.298782] [c000000e77217e30] [c00000000000b220] system_call+0x38/0xfc [ 116.298812] Instruction dump: [ 116.298828] 7d3f4b78 409effc8 3d220000 e9298020 3c800140 38a00018 608480c0 e8690028 [ 116.298869] 4800249d e8410018 7c7f1b79 41820230 2fa90000 419e0114 e9090020 [ 116.298914] ---[ end trace 1e10b0ced08b9120 ]--- This patch fixes the oops. Reported-by: Vaibhav Jain Signed-off-by: Greg Kurz Signed-off-by: Alex Williamson --- drivers/vfio/vfio_iommu_spapr_tce.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/vfio/vfio_iommu_spapr_tce.c b/drivers/vfio/vfio_iommu_spapr_tce.c index c8823578a1b2..128d10282d16 100644 --- a/drivers/vfio/vfio_iommu_spapr_tce.c +++ b/drivers/vfio/vfio_iommu_spapr_tce.c @@ -1270,6 +1270,10 @@ static int tce_iommu_attach_group(void *iommu_data, /* pr_debug("tce_vfio: Attaching group #%u to iommu %p\n", iommu_group_id(iommu_group), iommu_group); */ table_group = iommu_group_get_iommudata(iommu_group); + if (!table_group) { + ret = -ENODEV; + goto unlock_exit; + } if (tce_groups_attached(container) && (!table_group->ops || !table_group->ops->take_ownership || -- cgit v1.2.3 From b1a27eac7fefff33ccf6acc919fc0725bf9815fb Mon Sep 17 00:00:00 2001 From: Nicolas Iooss Date: Sun, 22 Jan 2017 14:41:22 +0100 Subject: IB/cxgb3: fix misspelling in header guard Use CXGB3_... instead of CXBG3_... Fixes: a85fb3383340 ("IB/cxgb3: Move user vendor structures") Cc: stable@vger.kernel.org # 4.9 Signed-off-by: Nicolas Iooss Reviewed-by: Leon Romanovsky Acked-by: Steve Wise Signed-off-by: Doug Ledford --- include/uapi/rdma/cxgb3-abi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/uapi/rdma/cxgb3-abi.h b/include/uapi/rdma/cxgb3-abi.h index 48a19bda071b..d24eee12128f 100644 --- a/include/uapi/rdma/cxgb3-abi.h +++ b/include/uapi/rdma/cxgb3-abi.h @@ -30,7 +30,7 @@ * SOFTWARE. */ #ifndef CXGB3_ABI_USER_H -#define CXBG3_ABI_USER_H +#define CXGB3_ABI_USER_H #include -- cgit v1.2.3 From f39aac7e839368e3895dff952f3bfa0a22e20060 Mon Sep 17 00:00:00 2001 From: Jingju Hou Date: Sun, 22 Jan 2017 18:20:56 +0800 Subject: net: phy: marvell: Add Wake from LAN support for 88E1510 PHY Signed-off-by: Jingju Hou Signed-off-by: David S. Miller --- drivers/net/phy/marvell.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index 0b78210c0fa7..ed0d235cf850 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -1679,6 +1679,8 @@ static struct phy_driver marvell_drivers[] = { .ack_interrupt = &marvell_ack_interrupt, .config_intr = &marvell_config_intr, .did_interrupt = &m88e1121_did_interrupt, + .get_wol = &m88e1318_get_wol, + .set_wol = &m88e1318_set_wol, .resume = &marvell_resume, .suspend = &marvell_suspend, .get_sset_count = marvell_get_sset_count, -- cgit v1.2.3 From 059aa734824165507c65fd30a55ff000afd14983 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Sun, 22 Jan 2017 14:04:29 -0500 Subject: nfs: Don't increment lock sequence ID after NFS4ERR_MOVED Xuan Qi reports that the Linux NFSv4 client failed to lock a file that was migrated. The steps he observed on the wire: 1. The client sent a LOCK request to the source server 2. The source server replied NFS4ERR_MOVED 3. The client switched to the destination server 4. The client sent the same LOCK request to the destination server with a bumped lock sequence ID 5. The destination server rejected the LOCK request with NFS4ERR_BAD_SEQID RFC 3530 section 8.1.5 provides a list of NFS errors which do not bump a lock sequence ID. However, RFC 3530 is now obsoleted by RFC 7530. In RFC 7530 section 9.1.7, this list has been updated by the addition of NFS4ERR_MOVED. Reported-by: Xuan Qi Signed-off-by: Chuck Lever Cc: stable@vger.kernel.org # v3.7+ Signed-off-by: Trond Myklebust --- include/linux/nfs4.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index bca536341d1a..1b1ca04820a3 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -282,7 +282,7 @@ enum nfsstat4 { static inline bool seqid_mutating_err(u32 err) { - /* rfc 3530 section 8.1.5: */ + /* See RFC 7530, section 9.1.7 */ switch (err) { case NFS4ERR_STALE_CLIENTID: case NFS4ERR_STALE_STATEID: @@ -291,6 +291,7 @@ static inline bool seqid_mutating_err(u32 err) case NFS4ERR_BADXDR: case NFS4ERR_RESOURCE: case NFS4ERR_NOFILEHANDLE: + case NFS4ERR_MOVED: return false; }; return true; -- cgit v1.2.3 From a430607b2ef7c3be090f88c71cfcb1b3988aa7c0 Mon Sep 17 00:00:00 2001 From: Benjamin Coddington Date: Tue, 24 Jan 2017 11:34:20 -0500 Subject: NFSv4.0: always send mode in SETATTR after EXCLUSIVE4 Some nfsv4.0 servers may return a mode for the verifier following an open with EXCLUSIVE4 createmode, but this does not mean the client should skip setting the mode in the following SETATTR. It should only do that for EXCLUSIVE4_1 or UNGAURDED createmode. Fixes: 5334c5bdac92 ("NFS: Send attributes in OPEN request for NFS4_CREATE_EXCLUSIVE4_1") Signed-off-by: Benjamin Coddington Cc: stable@vger.kernel.org # v4.3+ Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 59bb574d7d7c..0a0eaecf9676 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2700,7 +2700,8 @@ static inline void nfs4_exclusive_attrset(struct nfs4_opendata *opendata, sattr->ia_valid |= ATTR_MTIME; /* Except MODE, it seems harmless of setting twice. */ - if ((attrset[1] & FATTR4_WORD1_MODE)) + if (opendata->o_arg.createmode != NFS4_CREATE_EXCLUSIVE && + attrset[1] & FATTR4_WORD1_MODE) sattr->ia_valid &= ~ATTR_MODE; if (attrset[2] & FATTR4_WORD2_SECURITY_LABEL) -- cgit v1.2.3 From 92fdb527eecff7e5eb945a3fbf4743110f5c1171 Mon Sep 17 00:00:00 2001 From: Yuriy Kolerov Date: Wed, 28 Dec 2016 11:46:26 +0300 Subject: ARCv2: MCIP: Deprecate setting of affinity in Device Tree Ignore value of interrupt distribution mode for common interrupts in IDU since setting of affinity using value from Device Tree is deprecated in ARC. Originally it is done in idu_irq_xlate() function and it is semantically wrong and does not guaranty that an affinity value will be set properly. idu_irq_enable() function is better place for initialization of common interrupts. By default send all common interrupts to all available online CPUs. The affinity of common interrupts in IDU must be set manually since in some cases the kernel will not call irq_set_affinity() by itself: 1. When the kernel is not configured with support of SMP. 2. When the kernel is configured with support of SMP but upper interrupt controllers does not support setting of the affinity and cannot propagate it to IDU. Signed-off-by: Yuriy Kolerov Signed-off-by: Vineet Gupta --- .../interrupt-controller/snps,archs-idu-intc.txt | 3 ++ arch/arc/kernel/mcip.c | 52 +++++++++------------- 2 files changed, 25 insertions(+), 30 deletions(-) diff --git a/Documentation/devicetree/bindings/interrupt-controller/snps,archs-idu-intc.txt b/Documentation/devicetree/bindings/interrupt-controller/snps,archs-idu-intc.txt index 0dcb7c7d3e40..944657684d73 100644 --- a/Documentation/devicetree/bindings/interrupt-controller/snps,archs-idu-intc.txt +++ b/Documentation/devicetree/bindings/interrupt-controller/snps,archs-idu-intc.txt @@ -15,6 +15,9 @@ Properties: Second cell specifies the irq distribution mode to cores 0=Round Robin; 1=cpu0, 2=cpu1, 4=cpu2, 8=cpu3 + The second cell in interrupts property is deprecated and may be ignored by + the kernel. + intc accessed via the special ARC AUX register interface, hence "reg" property is not specified. diff --git a/arch/arc/kernel/mcip.c b/arch/arc/kernel/mcip.c index 9274f8ade8c7..9988b427a1e0 100644 --- a/arch/arc/kernel/mcip.c +++ b/arch/arc/kernel/mcip.c @@ -175,7 +175,6 @@ static void idu_irq_unmask(struct irq_data *data) raw_spin_unlock_irqrestore(&mcip_lock, flags); } -#ifdef CONFIG_SMP static int idu_irq_set_affinity(struct irq_data *data, const struct cpumask *cpumask, bool force) @@ -205,12 +204,27 @@ idu_irq_set_affinity(struct irq_data *data, const struct cpumask *cpumask, return IRQ_SET_MASK_OK; } -#endif + +static void idu_irq_enable(struct irq_data *data) +{ + /* + * By default send all common interrupts to all available online CPUs. + * The affinity of common interrupts in IDU must be set manually since + * in some cases the kernel will not call irq_set_affinity() by itself: + * 1. When the kernel is not configured with support of SMP. + * 2. When the kernel is configured with support of SMP but upper + * interrupt controllers does not support setting of the affinity + * and cannot propagate it to IDU. + */ + idu_irq_set_affinity(data, cpu_online_mask, false); + idu_irq_unmask(data); +} static struct irq_chip idu_irq_chip = { .name = "MCIP IDU Intc", .irq_mask = idu_irq_mask, .irq_unmask = idu_irq_unmask, + .irq_enable = idu_irq_enable, #ifdef CONFIG_SMP .irq_set_affinity = idu_irq_set_affinity, #endif @@ -243,36 +257,14 @@ static int idu_irq_xlate(struct irq_domain *d, struct device_node *n, const u32 *intspec, unsigned int intsize, irq_hw_number_t *out_hwirq, unsigned int *out_type) { - irq_hw_number_t hwirq = *out_hwirq = intspec[0]; - int distri = intspec[1]; - unsigned long flags; - + /* + * Ignore value of interrupt distribution mode for common interrupts in + * IDU which resides in intspec[1] since setting an affinity using value + * from Device Tree is deprecated in ARC. + */ + *out_hwirq = intspec[0]; *out_type = IRQ_TYPE_NONE; - /* XXX: validate distribution scheme again online cpu mask */ - if (distri == 0) { - /* 0 - Round Robin to all cpus, otherwise 1 bit per core */ - raw_spin_lock_irqsave(&mcip_lock, flags); - idu_set_dest(hwirq, BIT(num_online_cpus()) - 1); - idu_set_mode(hwirq, IDU_M_TRIG_LEVEL, IDU_M_DISTRI_RR); - raw_spin_unlock_irqrestore(&mcip_lock, flags); - } else { - /* - * DEST based distribution for Level Triggered intr can only - * have 1 CPU, so generalize it to always contain 1 cpu - */ - int cpu = ffs(distri); - - if (cpu != fls(distri)) - pr_warn("IDU irq %lx distri mode set to cpu %x\n", - hwirq, cpu); - - raw_spin_lock_irqsave(&mcip_lock, flags); - idu_set_dest(hwirq, cpu); - idu_set_mode(hwirq, IDU_M_TRIG_LEVEL, IDU_M_DISTRI_DEST); - raw_spin_unlock_irqrestore(&mcip_lock, flags); - } - return 0; } -- cgit v1.2.3 From 3f5c34c6d4688b3b7e1dbc7bbc68a2f03a0d6b0c Mon Sep 17 00:00:00 2001 From: Manuel Lauss Date: Tue, 24 Jan 2017 10:12:00 -0800 Subject: Input: wm97xx - make missing platform data non-fatal Commit 6480af4915d6 ("power_supply: wm97xx_battery: use power_supply_get_drvdata") made wm97xx platform data mandatory, although it's still optional. This patch fixes an oops during driver probe on one of my MIPS boards with a wm9712. Signed-off-by: Manuel Lauss Reviewed-by: Robert Jarzmik Acked-by: Charles Keepax Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/wm97xx-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/touchscreen/wm97xx-core.c b/drivers/input/touchscreen/wm97xx-core.c index 83cf11312fd9..c9d1c91e1887 100644 --- a/drivers/input/touchscreen/wm97xx-core.c +++ b/drivers/input/touchscreen/wm97xx-core.c @@ -682,7 +682,7 @@ static int wm97xx_probe(struct device *dev) } platform_set_drvdata(wm->battery_dev, wm); wm->battery_dev->dev.parent = dev; - wm->battery_dev->dev.platform_data = pdata->batt_pdata; + wm->battery_dev->dev.platform_data = pdata ? pdata->batt_pdata : NULL; ret = platform_device_add(wm->battery_dev); if (ret < 0) goto batt_reg_err; -- cgit v1.2.3 From 7630ea4bda18df2ee1c64dfdca1724a9cc32f920 Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Sun, 22 Jan 2017 17:41:32 +0100 Subject: Documentation: net: phy: improve explanation when to specify the PHY ID The old description basically read like "ethernet-phy-idAAAA.BBBB" can be specified when you know the actual PHY ID. However, specifying this has a side-effect: it forces Linux to bind to a certain PHY driver (the one that matches the ID given in the compatible string), ignoring the ID which is reported by the actual PHY. Whenever a device is shipped with (multiple) different PHYs during it's production lifetime then explicitly specifying "ethernet-phy-idAAAA.BBBB" could break certain revisions of that device. Signed-off-by: Martin Blumenstingl Reviewed-by: Andrew Lunn Acked-by: Rob Herring Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- Documentation/devicetree/bindings/net/phy.txt | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/Documentation/devicetree/bindings/net/phy.txt b/Documentation/devicetree/bindings/net/phy.txt index ff1bc4b1bb3b..fb5056b22685 100644 --- a/Documentation/devicetree/bindings/net/phy.txt +++ b/Documentation/devicetree/bindings/net/phy.txt @@ -19,8 +19,9 @@ Optional Properties: specifications. If neither of these are specified, the default is to assume clause 22. - If the phy's identifier is known then the list may contain an entry - of the form: "ethernet-phy-idAAAA.BBBB" where + If the PHY reports an incorrect ID (or none at all) then the + "compatible" list may contain an entry with the correct PHY ID in the + form: "ethernet-phy-idAAAA.BBBB" where AAAA - The value of the 16 bit Phy Identifier 1 register as 4 hex digits. This is the chip vendor OUI bits 3:18 BBBB - The value of the 16 bit Phy Identifier 2 register as -- cgit v1.2.3 From 6a0b76c04ec157c88ca943debf78a8ee58469f2d Mon Sep 17 00:00:00 2001 From: hayeswang Date: Mon, 23 Jan 2017 14:18:43 +0800 Subject: r8152: don't execute runtime suspend if the tx is not empty Runtime suspend shouldn't be executed if the tx queue is not empty, because the device is not idle. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller --- drivers/net/usb/r8152.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 0e99af090734..e1466b4d2b6c 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -32,7 +32,7 @@ #define NETNEXT_VERSION "08" /* Information for net */ -#define NET_VERSION "6" +#define NET_VERSION "7" #define DRIVER_VERSION "v1." NETNEXT_VERSION "." NET_VERSION #define DRIVER_AUTHOR "Realtek linux nic maintainers " @@ -3574,6 +3574,8 @@ static bool delay_autosuspend(struct r8152 *tp) */ if (!sw_linking && tp->rtl_ops.in_nway(tp)) return true; + else if (!skb_queue_empty(&tp->tx_queue)) + return true; else return false; } -- cgit v1.2.3 From a59b7e0246774e28193126fe7fdbbd0ae9c67dcc Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 23 Jan 2017 11:11:42 +0100 Subject: mlxsw: spectrum_router: Correctly reallocate adjacency entries mlxsw_sp_nexthop_group_mac_update() is called in one of two cases: 1) When the MAC of a nexthop needs to be updated 2) When the size of a nexthop group has changed In the second case the adjacency entries for the nexthop group need to be reallocated from the adjacency table. In this case we must write to the entries the MAC addresses of all the nexthops that should be offloaded and not only those whose MAC changed. Otherwise, these entries would be filled with garbage data, resulting in packet loss. Fixes: a7ff87acd995 ("mlxsw: spectrum_router: Implement next-hop routing") Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 01d0efa9c5c7..9e494a446b7e 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -1172,7 +1172,8 @@ static int mlxsw_sp_nexthop_mac_update(struct mlxsw_sp *mlxsw_sp, u32 adj_index, static int mlxsw_sp_nexthop_group_mac_update(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_nexthop_group *nh_grp) + struct mlxsw_sp_nexthop_group *nh_grp, + bool reallocate) { u32 adj_index = nh_grp->adj_index; /* base */ struct mlxsw_sp_nexthop *nh; @@ -1187,7 +1188,7 @@ mlxsw_sp_nexthop_group_mac_update(struct mlxsw_sp *mlxsw_sp, continue; } - if (nh->update) { + if (nh->update || reallocate) { err = mlxsw_sp_nexthop_mac_update(mlxsw_sp, adj_index, nh); if (err) @@ -1248,7 +1249,8 @@ mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp, /* Nothing was added or removed, so no need to reallocate. Just * update MAC on existing adjacency indexes. */ - err = mlxsw_sp_nexthop_group_mac_update(mlxsw_sp, nh_grp); + err = mlxsw_sp_nexthop_group_mac_update(mlxsw_sp, nh_grp, + false); if (err) { dev_warn(mlxsw_sp->bus_info->dev, "Failed to update neigh MAC in adjacency table.\n"); goto set_trap; @@ -1276,7 +1278,7 @@ mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp, nh_grp->adj_index_valid = 1; nh_grp->adj_index = adj_index; nh_grp->ecmp_size = ecmp_size; - err = mlxsw_sp_nexthop_group_mac_update(mlxsw_sp, nh_grp); + err = mlxsw_sp_nexthop_group_mac_update(mlxsw_sp, nh_grp, true); if (err) { dev_warn(mlxsw_sp->bus_info->dev, "Failed to update neigh MAC in adjacency table.\n"); goto set_trap; -- cgit v1.2.3 From 36425cd67052e3becf325fd4d3ba5691791ef7e4 Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Tue, 24 Jan 2017 10:23:42 -0800 Subject: ARC: udelay: fix inline assembler by adding LP_COUNT to clobber list commit 3c7c7a2fc8811bc ("ARC: Don't use "+l" inline asm constraint") modified the inline assembly to setup LP_COUNT register manually and NOT rely on gcc to do it (with the +l inline assembler contraint hint, now being retired in the compiler) However the fix was flawed as we didn't add LP_COUNT to asm clobber list, meaning gcc doesn't know that LP_COUNT or zero-delay-loops are in action in the inline asm. This resulted in some fun - as nested ZOL loops were being generared | mov lp_count,250000 ;16 # tmp235, | lp .L__GCC__LP14 # <======= OUTER LOOP (gcc generated) | .L14: | ld r2, [r5] # MEM[(volatile u32 *)prephitmp_43], w | dmb 1 | breq r2, -1, @.L21 #, w,, | bbit0 r2,1,@.L13 # w,, | ld r4,[r7] ;25 # loops_per_jiffy, loops_per_jiffy | mpymu r3,r4,r6 #, loops_per_jiffy, tmp234 | | mov lp_count, r3 # <====== INNER LOOP (from inline asm) | lp 1f | nop | 1: | nop_s | .L__GCC__LP14: ; loop end, start is @.L14 #, This caused issues with drivers relying on sane behaviour of udelay friends. With LP_COUNT added to clobber list, gcc doesn't generate the outer loop in say above case. Addresses STAR 9001146134 Reported-by: Joao Pinto Fixes: 3c7c7a2fc8811bc ("ARC: Don't use "+l" inline asm constraint") Cc: stable@vger.kernel.org Signed-off-by: Vineet Gupta --- arch/arc/include/asm/delay.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/arc/include/asm/delay.h b/arch/arc/include/asm/delay.h index a36e8601114d..d5da2115d78a 100644 --- a/arch/arc/include/asm/delay.h +++ b/arch/arc/include/asm/delay.h @@ -26,7 +26,9 @@ static inline void __delay(unsigned long loops) " lp 1f \n" " nop \n" "1: \n" - : : "r"(loops)); + : + : "r"(loops) + : "lp_count"); } extern void __bad_udelay(void); -- cgit v1.2.3 From 517e7610d2ce04d1b8d8b6c6d1a36dcce5cac6ab Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Thu, 19 Jan 2017 17:05:00 -0800 Subject: ARCv2: MCIP: update the BCR per current changes Signed-off-by: Vineet Gupta --- arch/arc/kernel/mcip.c | 3 +-- include/soc/arc/mcip.h | 16 ++++++++-------- 2 files changed, 9 insertions(+), 10 deletions(-) diff --git a/arch/arc/kernel/mcip.c b/arch/arc/kernel/mcip.c index 9988b427a1e0..9f6b68fd4f3b 100644 --- a/arch/arc/kernel/mcip.c +++ b/arch/arc/kernel/mcip.c @@ -93,11 +93,10 @@ static void mcip_probe_n_setup(void) READ_BCR(ARC_REG_MCIP_BCR, mp); sprintf(smp_cpuinfo_buf, - "Extn [SMP]\t: ARConnect (v%d): %d cores with %s%s%s%s%s\n", + "Extn [SMP]\t: ARConnect (v%d): %d cores with %s%s%s%s\n", mp.ver, mp.num_cores, IS_AVAIL1(mp.ipi, "IPI "), IS_AVAIL1(mp.idu, "IDU "), - IS_AVAIL1(mp.llm, "LLM "), IS_AVAIL1(mp.dbg, "DEBUG "), IS_AVAIL1(mp.gfrc, "GFRC")); diff --git a/include/soc/arc/mcip.h b/include/soc/arc/mcip.h index 6902c2a8bd23..4b6b489a8d7c 100644 --- a/include/soc/arc/mcip.h +++ b/include/soc/arc/mcip.h @@ -55,17 +55,17 @@ struct mcip_cmd { struct mcip_bcr { #ifdef CONFIG_CPU_BIG_ENDIAN - unsigned int pad3:8, - idu:1, llm:1, num_cores:6, - iocoh:1, gfrc:1, dbg:1, pad2:1, - msg:1, sem:1, ipi:1, pad:1, + unsigned int pad4:6, pw_dom:1, pad3:1, + idu:1, pad2:1, num_cores:6, + pad:1, gfrc:1, dbg:1, pw:1, + msg:1, sem:1, ipi:1, slv:1, ver:8; #else unsigned int ver:8, - pad:1, ipi:1, sem:1, msg:1, - pad2:1, dbg:1, gfrc:1, iocoh:1, - num_cores:6, llm:1, idu:1, - pad3:8; + slv:1, ipi:1, sem:1, msg:1, + pw:1, dbg:1, gfrc:1, pad:1, + num_cores:6, pad2:1, idu:1, + pad3:1, pw_dom:1, pad4:6; #endif }; -- cgit v1.2.3 From bf02454a741b58682a82c314a9a46bed930ed2f7 Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Thu, 12 Jan 2017 14:30:29 -0800 Subject: ARC: smp-boot: Decouple Non masters waiting API from jump to entry point For run-on-reset SMP configs, non master cores call a routine which waits until Master gives it a "go" signal (currently using a shared mem flag). The same routine then jumps off the well known entry point of all non Master cores i.e. @first_lines_of_secondary This patch moves out the last part into one single place in early boot code. This is better in terms of absraction (the wait API only waits) and returns, leaving out the "jump off to" part. In actual implementation this requires some restructuring of the early boot code as well as Master now jumps to BSS setup explicitly, vs. falling thru into it before. Technically this patch doesn't cause any functional change, it just moves the ugly #ifdef'ry from assembly code to "C" Signed-off-by: Vineet Gupta --- arch/arc/kernel/head.S | 14 +++++++------- arch/arc/kernel/smp.c | 6 ++++-- 2 files changed, 11 insertions(+), 9 deletions(-) diff --git a/arch/arc/kernel/head.S b/arch/arc/kernel/head.S index 689dd867fdff..8b90d25a15cc 100644 --- a/arch/arc/kernel/head.S +++ b/arch/arc/kernel/head.S @@ -71,14 +71,14 @@ ENTRY(stext) GET_CPU_ID r5 cmp r5, 0 mov.nz r0, r5 -#ifdef CONFIG_ARC_SMP_HALT_ON_RESET - ; Non-Master can proceed as system would be booted sufficiently - jnz first_lines_of_secondary -#else + bz .Lmaster_proceed + ; Non-Masters wait for Master to boot enough and bring them up - jnz arc_platform_smp_wait_to_boot -#endif - ; Master falls thru + ; when they resume, tail-call to entry point + mov blink, @first_lines_of_secondary + j arc_platform_smp_wait_to_boot + +.Lmaster_proceed: #endif ; Clear BSS before updating any globals diff --git a/arch/arc/kernel/smp.c b/arch/arc/kernel/smp.c index 88674d972c9d..44a0d21ed342 100644 --- a/arch/arc/kernel/smp.c +++ b/arch/arc/kernel/smp.c @@ -98,14 +98,16 @@ static void arc_default_smp_cpu_kick(int cpu, unsigned long pc) void arc_platform_smp_wait_to_boot(int cpu) { + /* for halt-on-reset, we've waited already */ + if (IS_ENABLED(CONFIG_ARC_SMP_HALT_ON_RESET)) + return; + while (wake_flag != cpu) ; wake_flag = 0; - __asm__ __volatile__("j @first_lines_of_secondary \n"); } - const char *arc_platform_smp_cpuinfo(void) { return plat_smp_ops.info ? : ""; -- cgit v1.2.3 From 7d211c81e97ef8505610ef82e14e302ab415bad1 Mon Sep 17 00:00:00 2001 From: Adit Ranadive Date: Thu, 19 Jan 2017 13:20:39 -0800 Subject: IB/vmw_pvrdma: Don't leak info from alloc_ucontext Clear out the user response struct correctly. Fixes: 29c8d9eba550 ("IB: Add vmw_pvrdma driver") Reported-by: Dan Carpenter Signed-off-by: Adit Ranadive Signed-off-by: Doug Ledford --- drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c index 54891370d18a..c2aa52638dcb 100644 --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c @@ -306,7 +306,7 @@ struct ib_ucontext *pvrdma_alloc_ucontext(struct ib_device *ibdev, union pvrdma_cmd_resp rsp; struct pvrdma_cmd_create_uc *cmd = &req.create_uc; struct pvrdma_cmd_create_uc_resp *resp = &rsp.create_uc_resp; - struct pvrdma_alloc_ucontext_resp uresp; + struct pvrdma_alloc_ucontext_resp uresp = {0}; int ret; void *ptr; -- cgit v1.2.3 From ff89b070b7c98eb6782361310ca7a15186f15b2c Mon Sep 17 00:00:00 2001 From: Adit Ranadive Date: Thu, 19 Jan 2017 13:20:40 -0800 Subject: IB/vmw_pvrdma: Fix incorrect cleanup on pvrdma_pci_probe error path If the interrupt allocation failed we should start freeing the CQ rings rather than unregistering the netdev notifier. Fixes: 29c8d9eba550 ("IB: Add vmw_pvrdma driver") Signed-off-by: Adit Ranadive Reviewed-by: Yuval Shaia Signed-off-by: Doug Ledford --- drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c index 231a1ce1f4be..bd8fbd3d2032 100644 --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c @@ -1029,7 +1029,7 @@ static int pvrdma_pci_probe(struct pci_dev *pdev, if (ret) { dev_err(&pdev->dev, "failed to allocate interrupts\n"); ret = -ENOMEM; - goto err_netdevice; + goto err_free_cq_ring; } /* Allocate UAR table. */ @@ -1092,8 +1092,6 @@ err_free_uar_table: err_free_intrs: pvrdma_free_irq(dev); pvrdma_disable_msi_all(dev); -err_netdevice: - unregister_netdevice_notifier(&dev->nb_netdev); err_free_cq_ring: pvrdma_page_dir_cleanup(dev, &dev->cq_pdir); err_free_async_ring: -- cgit v1.2.3 From d46d29f072accb069cb42b5fbebcc77d9094a785 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Wed, 11 Jan 2017 13:38:52 -0800 Subject: md/raid5-cache: delete meaningless code sector_t is unsigned long, it's never < 0 Reported-by: Julia Lawall Signed-off-by: Shaohua Li --- drivers/md/raid5-cache.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c index 0e8ed2c327b0..e0d6dd1835d0 100644 --- a/drivers/md/raid5-cache.c +++ b/drivers/md/raid5-cache.c @@ -1393,8 +1393,6 @@ static void r5l_do_reclaim(struct r5l_log *log) next_checkpoint = r5c_calculate_new_cp(conf); spin_unlock_irq(&log->io_list_lock); - BUG_ON(reclaimable < 0); - if (reclaimable == 0 || !write_super) return; -- cgit v1.2.3 From 86aa1397ddfde563b3692adadb8b8e32e97b4e5e Mon Sep 17 00:00:00 2001 From: Song Liu Date: Thu, 12 Jan 2017 17:22:41 -0800 Subject: md/r5cache: read data into orig_page for prexor of cached data With write back cache, we use orig_page to do prexor. This patch makes sure we read data into orig_page for it. Flag R5_OrigPageUPTDODATE is added to show whether orig_page has the latest data from raid disk. We introduce a helper function uptodate_for_rmw() to simplify the a couple conditions in handle_stripe_dirtying(). Signed-off-by: Song Liu Signed-off-by: Shaohua Li --- drivers/md/raid5-cache.c | 2 ++ drivers/md/raid5.c | 44 +++++++++++++++++++++++++++++++++++--------- drivers/md/raid5.h | 5 +++++ 3 files changed, 42 insertions(+), 9 deletions(-) diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c index e0d6dd1835d0..95dcaa022e1f 100644 --- a/drivers/md/raid5-cache.c +++ b/drivers/md/raid5-cache.c @@ -2349,6 +2349,8 @@ void r5c_release_extra_page(struct stripe_head *sh) struct page *p = sh->dev[i].orig_page; sh->dev[i].orig_page = sh->dev[i].page; + clear_bit(R5_OrigPageUPTDODATE, &sh->dev[i].flags); + if (!using_disk_info_extra_page) put_page(p); } diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 36c13e4be9c9..7780ae44f355 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -1015,7 +1015,17 @@ again: if (test_bit(R5_SkipCopy, &sh->dev[i].flags)) WARN_ON(test_bit(R5_UPTODATE, &sh->dev[i].flags)); - sh->dev[i].vec.bv_page = sh->dev[i].page; + + if (!op_is_write(op) && + test_bit(R5_InJournal, &sh->dev[i].flags)) + /* + * issuing read for a page in journal, this + * must be preparing for prexor in rmw; read + * the data into orig_page + */ + sh->dev[i].vec.bv_page = sh->dev[i].orig_page; + else + sh->dev[i].vec.bv_page = sh->dev[i].page; bi->bi_vcnt = 1; bi->bi_io_vec[0].bv_len = STRIPE_SIZE; bi->bi_io_vec[0].bv_offset = 0; @@ -2380,6 +2390,13 @@ static void raid5_end_read_request(struct bio * bi) } else if (test_bit(R5_ReadNoMerge, &sh->dev[i].flags)) clear_bit(R5_ReadNoMerge, &sh->dev[i].flags); + if (test_bit(R5_InJournal, &sh->dev[i].flags)) + /* + * end read for a page in journal, this + * must be preparing for prexor in rmw + */ + set_bit(R5_OrigPageUPTDODATE, &sh->dev[i].flags); + if (atomic_read(&rdev->read_errors)) atomic_set(&rdev->read_errors, 0); } else { @@ -3594,6 +3611,21 @@ unhash: break_stripe_batch_list(head_sh, STRIPE_EXPAND_SYNC_FLAGS); } +/* + * For RMW in write back cache, we need extra page in prexor to store the + * old data. This page is stored in dev->orig_page. + * + * This function checks whether we have data for prexor. The exact logic + * is: + * R5_UPTODATE && (!R5_InJournal || R5_OrigPageUPTDODATE) + */ +static inline bool uptodate_for_rmw(struct r5dev *dev) +{ + return (test_bit(R5_UPTODATE, &dev->flags)) && + (!test_bit(R5_InJournal, &dev->flags) || + test_bit(R5_OrigPageUPTDODATE, &dev->flags)); +} + static int handle_stripe_dirtying(struct r5conf *conf, struct stripe_head *sh, struct stripe_head_state *s, @@ -3625,9 +3657,7 @@ static int handle_stripe_dirtying(struct r5conf *conf, if ((dev->towrite || i == sh->pd_idx || i == sh->qd_idx || test_bit(R5_InJournal, &dev->flags)) && !test_bit(R5_LOCKED, &dev->flags) && - !((test_bit(R5_UPTODATE, &dev->flags) && - (!test_bit(R5_InJournal, &dev->flags) || - dev->page != dev->orig_page)) || + !(uptodate_for_rmw(dev) || test_bit(R5_Wantcompute, &dev->flags))) { if (test_bit(R5_Insync, &dev->flags)) rmw++; @@ -3639,7 +3669,6 @@ static int handle_stripe_dirtying(struct r5conf *conf, i != sh->pd_idx && i != sh->qd_idx && !test_bit(R5_LOCKED, &dev->flags) && !(test_bit(R5_UPTODATE, &dev->flags) || - test_bit(R5_InJournal, &dev->flags) || test_bit(R5_Wantcompute, &dev->flags))) { if (test_bit(R5_Insync, &dev->flags)) rcw++; @@ -3693,9 +3722,7 @@ static int handle_stripe_dirtying(struct r5conf *conf, i == sh->pd_idx || i == sh->qd_idx || test_bit(R5_InJournal, &dev->flags)) && !test_bit(R5_LOCKED, &dev->flags) && - !((test_bit(R5_UPTODATE, &dev->flags) && - (!test_bit(R5_InJournal, &dev->flags) || - dev->page != dev->orig_page)) || + !(uptodate_for_rmw(dev) || test_bit(R5_Wantcompute, &dev->flags)) && test_bit(R5_Insync, &dev->flags)) { if (test_bit(STRIPE_PREREAD_ACTIVE, @@ -3722,7 +3749,6 @@ static int handle_stripe_dirtying(struct r5conf *conf, i != sh->pd_idx && i != sh->qd_idx && !test_bit(R5_LOCKED, &dev->flags) && !(test_bit(R5_UPTODATE, &dev->flags) || - test_bit(R5_InJournal, &dev->flags) || test_bit(R5_Wantcompute, &dev->flags))) { rcw++; if (test_bit(R5_Insync, &dev->flags) && diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h index ed8e1362ab36..461df197d157 100644 --- a/drivers/md/raid5.h +++ b/drivers/md/raid5.h @@ -322,6 +322,11 @@ enum r5dev_flags { * data and parity being written are in the journal * device */ + R5_OrigPageUPTDODATE, /* with write back cache, we read old data into + * dev->orig_page for prexor. When this flag is + * set, orig_page contains latest data in the + * raid disk. + */ }; /* -- cgit v1.2.3 From ba02684daf7fb4a827580f909b7c7db61c05ae7d Mon Sep 17 00:00:00 2001 From: Song Liu Date: Thu, 12 Jan 2017 17:22:42 -0800 Subject: md/raid5: move comment of fetch_block to right location Signed-off-by: Song Liu Signed-off-by: Shaohua Li --- drivers/md/raid5.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 7780ae44f355..13d76767c2cf 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3312,13 +3312,6 @@ static int want_replace(struct stripe_head *sh, int disk_idx) return rv; } -/* fetch_block - checks the given member device to see if its data needs - * to be read or computed to satisfy a request. - * - * Returns 1 when no more member devices need to be checked, otherwise returns - * 0 to tell the loop in handle_stripe_fill to continue - */ - static int need_this_block(struct stripe_head *sh, struct stripe_head_state *s, int disk_idx, int disks) { @@ -3409,6 +3402,12 @@ static int need_this_block(struct stripe_head *sh, struct stripe_head_state *s, return 0; } +/* fetch_block - checks the given member device to see if its data needs + * to be read or computed to satisfy a request. + * + * Returns 1 when no more member devices need to be checked, otherwise returns + * 0 to tell the loop in handle_stripe_fill to continue + */ static int fetch_block(struct stripe_head *sh, struct stripe_head_state *s, int disk_idx, int disks) { -- cgit v1.2.3 From a85dd7b8df52e35d8ee3794c65cac5c39128fd80 Mon Sep 17 00:00:00 2001 From: Song Liu Date: Mon, 23 Jan 2017 17:12:57 -0800 Subject: md/r5cache: flush data only stripes in r5l_recovery_log() For safer operation, all arrays start in write-through mode, which has been better tested and is more mature. And actually the write-through/write-mode isn't persistent after array restarted, so we always start array in write-through mode. However, if recovery found data-only stripes before the shutdown (from previous write-back mode), it is not safe to start the array in write-through mode, as write-through mode can not handle stripes with data in write-back cache. To solve this problem, we flush all data-only stripes in r5l_recovery_log(). When r5l_recovery_log() returns, the array starts with empty cache in write-through mode. This logic is implemented in r5c_recovery_flush_data_only_stripes(): 1. enable write back cache 2. flush all stripes 3. wake up conf->mddev->thread 4. wait for all stripes get flushed (reuse wait_for_quiescent) 5. disable write back cache The wait in 4 will be waked up in release_inactive_stripe_list() when conf->active_stripes reaches 0. It is safe to wake up mddev->thread here because all the resource required for the thread has been initialized. Signed-off-by: Song Liu Signed-off-by: Shaohua Li --- drivers/md/md.c | 5 +++++ drivers/md/raid5-cache.c | 56 ++++++++++++++++++++++++++++++++++-------------- 2 files changed, 45 insertions(+), 16 deletions(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 82821ee0d57f..01175dac0db6 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -5291,6 +5291,11 @@ int md_run(struct mddev *mddev) if (start_readonly && mddev->ro == 0) mddev->ro = 2; /* read-only, but switch on first write */ + /* + * NOTE: some pers->run(), for example r5l_recovery_log(), wakes + * up mddev->thread. It is important to initialize critical + * resources for mddev->thread BEFORE calling pers->run(). + */ err = pers->run(mddev); if (err) pr_warn("md: pers->run() failed ...\n"); diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c index 95dcaa022e1f..3d7dda85494c 100644 --- a/drivers/md/raid5-cache.c +++ b/drivers/md/raid5-cache.c @@ -2060,7 +2060,7 @@ static int r5c_recovery_rewrite_data_only_stripes(struct r5l_log *log, struct r5l_recovery_ctx *ctx) { - struct stripe_head *sh, *next; + struct stripe_head *sh; struct mddev *mddev = log->rdev->mddev; struct page *page; sector_t next_checkpoint = MaxSector; @@ -2074,7 +2074,7 @@ r5c_recovery_rewrite_data_only_stripes(struct r5l_log *log, WARN_ON(list_empty(&ctx->cached_list)); - list_for_each_entry_safe(sh, next, &ctx->cached_list, lru) { + list_for_each_entry(sh, &ctx->cached_list, lru) { struct r5l_meta_block *mb; int i; int offset; @@ -2124,14 +2124,39 @@ r5c_recovery_rewrite_data_only_stripes(struct r5l_log *log, ctx->pos = write_pos; ctx->seq += 1; next_checkpoint = sh->log_start; - list_del_init(&sh->lru); - raid5_release_stripe(sh); } log->next_checkpoint = next_checkpoint; __free_page(page); return 0; } +static void r5c_recovery_flush_data_only_stripes(struct r5l_log *log, + struct r5l_recovery_ctx *ctx) +{ + struct mddev *mddev = log->rdev->mddev; + struct r5conf *conf = mddev->private; + struct stripe_head *sh, *next; + + if (ctx->data_only_stripes == 0) + return; + + log->r5c_journal_mode = R5C_JOURNAL_MODE_WRITE_BACK; + + list_for_each_entry_safe(sh, next, &ctx->cached_list, lru) { + r5c_make_stripe_write_out(sh); + set_bit(STRIPE_HANDLE, &sh->state); + list_del_init(&sh->lru); + raid5_release_stripe(sh); + } + + md_wakeup_thread(conf->mddev->thread); + /* reuse conf->wait_for_quiescent in recovery */ + wait_event(conf->wait_for_quiescent, + atomic_read(&conf->active_stripes) == 0); + + log->r5c_journal_mode = R5C_JOURNAL_MODE_WRITE_THROUGH; +} + static int r5l_recovery_log(struct r5l_log *log) { struct mddev *mddev = log->rdev->mddev; @@ -2158,32 +2183,31 @@ static int r5l_recovery_log(struct r5l_log *log) pos = ctx.pos; ctx.seq += 10000; - if (ctx.data_only_stripes == 0) { - log->next_checkpoint = ctx.pos; - r5l_log_write_empty_meta_block(log, ctx.pos, ctx.seq++); - ctx.pos = r5l_ring_add(log, ctx.pos, BLOCK_SECTORS); - } if ((ctx.data_only_stripes == 0) && (ctx.data_parity_stripes == 0)) pr_debug("md/raid:%s: starting from clean shutdown\n", mdname(mddev)); - else { + else pr_debug("md/raid:%s: recovering %d data-only stripes and %d data-parity stripes\n", mdname(mddev), ctx.data_only_stripes, ctx.data_parity_stripes); - if (ctx.data_only_stripes > 0) - if (r5c_recovery_rewrite_data_only_stripes(log, &ctx)) { - pr_err("md/raid:%s: failed to rewrite stripes to journal\n", - mdname(mddev)); - return -EIO; - } + if (ctx.data_only_stripes == 0) { + log->next_checkpoint = ctx.pos; + r5l_log_write_empty_meta_block(log, ctx.pos, ctx.seq++); + ctx.pos = r5l_ring_add(log, ctx.pos, BLOCK_SECTORS); + } else if (r5c_recovery_rewrite_data_only_stripes(log, &ctx)) { + pr_err("md/raid:%s: failed to rewrite stripes to journal\n", + mdname(mddev)); + return -EIO; } log->log_start = ctx.pos; log->seq = ctx.seq; log->last_checkpoint = pos; r5l_write_super(log, pos); + + r5c_recovery_flush_data_only_stripes(log, &ctx); return 0; } -- cgit v1.2.3 From 07e83364845e1e1c7e189a01206a9d7d33831568 Mon Sep 17 00:00:00 2001 From: Song Liu Date: Mon, 23 Jan 2017 17:12:58 -0800 Subject: md/r5cache: shift complex rmw from read path to write path Write back cache requires a complex RMW mechanism, where old data is read into dev->orig_page for prexor, and then xor is done with dev->page. This logic is already implemented in the write path. However, current read path is not awared of this requirement. When the array is optimal, the RMW is not required, as the data are read from raid disks. However, when the target stripe is degraded, complex RMW is required to generate right data. To keep read path as clean as possible, we handle read path by flushing degraded, in-journal stripes before processing reads to missing dev. Specifically, when there is read requests to a degraded stripe with data in journal, handle_stripe_fill() calls r5c_make_stripe_write_out() and exits. Then handle_stripe_dirtying() will do the complex RMW and flush the stripe to RAID disks. After that, read requests are handled. There is one more corner case when there is non-overwrite bio for the missing (or out of sync) dev. handle_stripe_dirtying() will not be able to process the non-overwrite bios without constructing the data in handle_stripe_fill(). This is fixed by delaying non-overwrite bios in handle_stripe_dirtying(). So handle_stripe_fill() works on these bios after the stripe is flushed to raid disks. Signed-off-by: Song Liu Signed-off-by: Shaohua Li --- drivers/md/raid5.c | 49 +++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 45 insertions(+), 4 deletions(-) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 13d76767c2cf..dc83da69ca7c 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -2897,6 +2897,30 @@ sector_t raid5_compute_blocknr(struct stripe_head *sh, int i, int previous) return r_sector; } +/* + * There are cases where we want handle_stripe_dirtying() and + * schedule_reconstruction() to delay towrite to some dev of a stripe. + * + * This function checks whether we want to delay the towrite. Specifically, + * we delay the towrite when: + * + * 1. degraded stripe has a non-overwrite to the missing dev, AND this + * stripe has data in journal (for other devices). + * + * In this case, when reading data for the non-overwrite dev, it is + * necessary to handle complex rmw of write back cache (prexor with + * orig_page, and xor with page). To keep read path simple, we would + * like to flush data in journal to RAID disks first, so complex rmw + * is handled in the write patch (handle_stripe_dirtying). + * + */ +static inline bool delay_towrite(struct r5dev *dev, + struct stripe_head_state *s) +{ + return !test_bit(R5_OVERWRITE, &dev->flags) && + !test_bit(R5_Insync, &dev->flags) && s->injournal; +} + static void schedule_reconstruction(struct stripe_head *sh, struct stripe_head_state *s, int rcw, int expand) @@ -2917,7 +2941,7 @@ schedule_reconstruction(struct stripe_head *sh, struct stripe_head_state *s, for (i = disks; i--; ) { struct r5dev *dev = &sh->dev[i]; - if (dev->towrite) { + if (dev->towrite && !delay_towrite(dev, s)) { set_bit(R5_LOCKED, &dev->flags); set_bit(R5_Wantdrain, &dev->flags); if (!expand) @@ -3494,10 +3518,26 @@ static void handle_stripe_fill(struct stripe_head *sh, * midst of changing due to a write */ if (!test_bit(STRIPE_COMPUTE_RUN, &sh->state) && !sh->check_state && - !sh->reconstruct_state) + !sh->reconstruct_state) { + + /* + * For degraded stripe with data in journal, do not handle + * read requests yet, instead, flush the stripe to raid + * disks first, this avoids handling complex rmw of write + * back cache (prexor with orig_page, and then xor with + * page) in the read path + */ + if (s->injournal && s->failed) { + if (test_bit(STRIPE_R5C_CACHING, &sh->state)) + r5c_make_stripe_write_out(sh); + goto out; + } + for (i = disks; i--; ) if (fetch_block(sh, s, i, disks)) break; + } +out: set_bit(STRIPE_HANDLE, &sh->state); } @@ -3653,7 +3693,8 @@ static int handle_stripe_dirtying(struct r5conf *conf, } else for (i = disks; i--; ) { /* would I have to read this buffer for read_modify_write */ struct r5dev *dev = &sh->dev[i]; - if ((dev->towrite || i == sh->pd_idx || i == sh->qd_idx || + if (((dev->towrite && !delay_towrite(dev, s)) || + i == sh->pd_idx || i == sh->qd_idx || test_bit(R5_InJournal, &dev->flags)) && !test_bit(R5_LOCKED, &dev->flags) && !(uptodate_for_rmw(dev) || @@ -3717,7 +3758,7 @@ static int handle_stripe_dirtying(struct r5conf *conf, for (i = disks; i--; ) { struct r5dev *dev = &sh->dev[i]; - if ((dev->towrite || + if (((dev->towrite && !delay_towrite(dev, s)) || i == sh->pd_idx || i == sh->qd_idx || test_bit(R5_InJournal, &dev->flags)) && !test_bit(R5_LOCKED, &dev->flags) && -- cgit v1.2.3 From 2e38a37f23c98d7fad87ff022670060b8a0e2bf5 Mon Sep 17 00:00:00 2001 From: Song Liu Date: Tue, 24 Jan 2017 10:45:30 -0800 Subject: md/r5cache: disable write back for degraded array write-back cache in degraded mode introduces corner cases to the array. Although we try to cover all these corner cases, it is safer to just disable write-back cache when the array is in degraded mode. In this patch, we disable writeback cache for degraded mode: 1. On device failure, if the array enters degraded mode, raid5_error() will submit async job r5c_disable_writeback_async to disable writeback; 2. In r5c_journal_mode_store(), it is invalid to enable writeback in degraded mode; 3. In r5c_try_caching_write(), stripes with s->failed>0 will be handled in write-through mode. Signed-off-by: Song Liu Signed-off-by: Shaohua Li --- drivers/md/raid5-cache.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++ drivers/md/raid5.c | 15 ++++++++------- drivers/md/raid5.h | 2 ++ 3 files changed, 56 insertions(+), 7 deletions(-) diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c index 3d7dda85494c..302dea3296ba 100644 --- a/drivers/md/raid5-cache.c +++ b/drivers/md/raid5-cache.c @@ -162,6 +162,8 @@ struct r5l_log { /* to submit async io_units, to fulfill ordering of flush */ struct work_struct deferred_io_work; + /* to disable write back during in degraded mode */ + struct work_struct disable_writeback_work; }; /* @@ -611,6 +613,21 @@ static void r5l_submit_io_async(struct work_struct *work) r5l_do_submit_io(log, io); } +static void r5c_disable_writeback_async(struct work_struct *work) +{ + struct r5l_log *log = container_of(work, struct r5l_log, + disable_writeback_work); + struct mddev *mddev = log->rdev->mddev; + + if (log->r5c_journal_mode == R5C_JOURNAL_MODE_WRITE_THROUGH) + return; + pr_info("md/raid:%s: Disabling writeback cache for degraded array.\n", + mdname(mddev)); + mddev_suspend(mddev); + log->r5c_journal_mode = R5C_JOURNAL_MODE_WRITE_THROUGH; + mddev_resume(mddev); +} + static void r5l_submit_current_io(struct r5l_log *log) { struct r5l_io_unit *io = log->current_io; @@ -2269,6 +2286,10 @@ static ssize_t r5c_journal_mode_store(struct mddev *mddev, val > R5C_JOURNAL_MODE_WRITE_BACK) return -EINVAL; + if (raid5_calc_degraded(conf) > 0 && + val == R5C_JOURNAL_MODE_WRITE_BACK) + return -EINVAL; + mddev_suspend(mddev); conf->log->r5c_journal_mode = val; mddev_resume(mddev); @@ -2323,6 +2344,16 @@ int r5c_try_caching_write(struct r5conf *conf, set_bit(STRIPE_R5C_CACHING, &sh->state); } + /* + * When run in degraded mode, array is set to write-through mode. + * This check helps drain pending write safely in the transition to + * write-through mode. + */ + if (s->failed) { + r5c_make_stripe_write_out(sh); + return -EAGAIN; + } + for (i = disks; i--; ) { dev = &sh->dev[i]; /* if non-overwrite, use writing-out phase */ @@ -2579,6 +2610,19 @@ ioerr: return ret; } +void r5c_update_on_rdev_error(struct mddev *mddev) +{ + struct r5conf *conf = mddev->private; + struct r5l_log *log = conf->log; + + if (!log) + return; + + if (raid5_calc_degraded(conf) > 0 && + conf->log->r5c_journal_mode == R5C_JOURNAL_MODE_WRITE_BACK) + schedule_work(&log->disable_writeback_work); +} + int r5l_init_log(struct r5conf *conf, struct md_rdev *rdev) { struct request_queue *q = bdev_get_queue(rdev->bdev); @@ -2651,6 +2695,7 @@ int r5l_init_log(struct r5conf *conf, struct md_rdev *rdev) spin_lock_init(&log->no_space_stripes_lock); INIT_WORK(&log->deferred_io_work, r5l_submit_io_async); + INIT_WORK(&log->disable_writeback_work, r5c_disable_writeback_async); log->r5c_journal_mode = R5C_JOURNAL_MODE_WRITE_THROUGH; INIT_LIST_HEAD(&log->stripe_in_journal_list); @@ -2683,6 +2728,7 @@ io_kc: void r5l_exit_log(struct r5l_log *log) { + flush_work(&log->disable_writeback_work); md_unregister_thread(&log->reclaim_thread); mempool_destroy(log->meta_pool); bioset_free(log->bs); diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index dc83da69ca7c..3c7e106c12a2 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -556,7 +556,7 @@ static struct stripe_head *__find_stripe(struct r5conf *conf, sector_t sector, * of the two sections, and some non-in_sync devices may * be insync in the section most affected by failed devices. */ -static int calc_degraded(struct r5conf *conf) +int raid5_calc_degraded(struct r5conf *conf) { int degraded, degraded2; int i; @@ -619,7 +619,7 @@ static int has_failed(struct r5conf *conf) if (conf->mddev->reshape_position == MaxSector) return conf->mddev->degraded > conf->max_degraded; - degraded = calc_degraded(conf); + degraded = raid5_calc_degraded(conf); if (degraded > conf->max_degraded) return 1; return 0; @@ -2555,7 +2555,7 @@ static void raid5_error(struct mddev *mddev, struct md_rdev *rdev) spin_lock_irqsave(&conf->device_lock, flags); clear_bit(In_sync, &rdev->flags); - mddev->degraded = calc_degraded(conf); + mddev->degraded = raid5_calc_degraded(conf); spin_unlock_irqrestore(&conf->device_lock, flags); set_bit(MD_RECOVERY_INTR, &mddev->recovery); @@ -2569,6 +2569,7 @@ static void raid5_error(struct mddev *mddev, struct md_rdev *rdev) bdevname(rdev->bdev, b), mdname(mddev), conf->raid_disks - mddev->degraded); + r5c_update_on_rdev_error(mddev); } /* @@ -7091,7 +7092,7 @@ static int raid5_run(struct mddev *mddev) /* * 0 for a fully functional array, 1 or 2 for a degraded array. */ - mddev->degraded = calc_degraded(conf); + mddev->degraded = raid5_calc_degraded(conf); if (has_failed(conf)) { pr_crit("md/raid:%s: not enough operational devices (%d/%d failed)\n", @@ -7338,7 +7339,7 @@ static int raid5_spare_active(struct mddev *mddev) } } spin_lock_irqsave(&conf->device_lock, flags); - mddev->degraded = calc_degraded(conf); + mddev->degraded = raid5_calc_degraded(conf); spin_unlock_irqrestore(&conf->device_lock, flags); print_raid5_conf(conf); return count; @@ -7698,7 +7699,7 @@ static int raid5_start_reshape(struct mddev *mddev) * pre and post number of devices. */ spin_lock_irqsave(&conf->device_lock, flags); - mddev->degraded = calc_degraded(conf); + mddev->degraded = raid5_calc_degraded(conf); spin_unlock_irqrestore(&conf->device_lock, flags); } mddev->raid_disks = conf->raid_disks; @@ -7786,7 +7787,7 @@ static void raid5_finish_reshape(struct mddev *mddev) } else { int d; spin_lock_irq(&conf->device_lock); - mddev->degraded = calc_degraded(conf); + mddev->degraded = raid5_calc_degraded(conf); spin_unlock_irq(&conf->device_lock); for (d = conf->raid_disks ; d < conf->raid_disks - mddev->delta_disks; diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h index 461df197d157..1440fa26e296 100644 --- a/drivers/md/raid5.h +++ b/drivers/md/raid5.h @@ -758,6 +758,7 @@ extern sector_t raid5_compute_sector(struct r5conf *conf, sector_t r_sector, extern struct stripe_head * raid5_get_active_stripe(struct r5conf *conf, sector_t sector, int previous, int noblock, int noquiesce); +extern int raid5_calc_degraded(struct r5conf *conf); extern int r5l_init_log(struct r5conf *conf, struct md_rdev *rdev); extern void r5l_exit_log(struct r5l_log *log); extern int r5l_write_stripe(struct r5l_log *log, struct stripe_head *head_sh); @@ -786,4 +787,5 @@ extern void r5c_flush_cache(struct r5conf *conf, int num); extern void r5c_check_stripe_cache_usage(struct r5conf *conf); extern void r5c_check_cached_full_stripe(struct r5conf *conf); extern struct md_sysfs_entry r5c_journal_mode; +extern void r5c_update_on_rdev_error(struct mddev *mddev); #endif -- cgit v1.2.3 From 0fb44559ffd67de8517098b81f675fa0210f13f0 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Mon, 23 Jan 2017 11:17:35 -0800 Subject: af_unix: move unix_mknod() out of bindlock Dmitry reported a deadlock scenario: unix_bind() path: u->bindlock ==> sb_writer do_splice() path: sb_writer ==> pipe->mutex ==> u->bindlock In the unix_bind() code path, unix_mknod() does not have to be done with u->bindlock held, since it is a pure fs operation, so we can just move unix_mknod() out. Reported-by: Dmitry Vyukov Tested-by: Dmitry Vyukov Cc: Rainer Weikusat Cc: Al Viro Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- net/unix/af_unix.c | 27 ++++++++++++++++----------- 1 file changed, 16 insertions(+), 11 deletions(-) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 127656ebe7be..cef79873b09d 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -995,6 +995,7 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) unsigned int hash; struct unix_address *addr; struct hlist_head *list; + struct path path = { NULL, NULL }; err = -EINVAL; if (sunaddr->sun_family != AF_UNIX) @@ -1010,9 +1011,20 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) goto out; addr_len = err; + if (sun_path[0]) { + umode_t mode = S_IFSOCK | + (SOCK_INODE(sock)->i_mode & ~current_umask()); + err = unix_mknod(sun_path, mode, &path); + if (err) { + if (err == -EEXIST) + err = -EADDRINUSE; + goto out; + } + } + err = mutex_lock_interruptible(&u->bindlock); if (err) - goto out; + goto out_put; err = -EINVAL; if (u->addr) @@ -1029,16 +1041,6 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) atomic_set(&addr->refcnt, 1); if (sun_path[0]) { - struct path path; - umode_t mode = S_IFSOCK | - (SOCK_INODE(sock)->i_mode & ~current_umask()); - err = unix_mknod(sun_path, mode, &path); - if (err) { - if (err == -EEXIST) - err = -EADDRINUSE; - unix_release_addr(addr); - goto out_up; - } addr->hash = UNIX_HASH_SIZE; hash = d_backing_inode(path.dentry)->i_ino & (UNIX_HASH_SIZE - 1); spin_lock(&unix_table_lock); @@ -1065,6 +1067,9 @@ out_unlock: spin_unlock(&unix_table_lock); out_up: mutex_unlock(&u->bindlock); +out_put: + if (err) + path_put(&path); out: return err; } -- cgit v1.2.3 From d0fa28f00052391b5df328f502fbbdd4444938b7 Mon Sep 17 00:00:00 2001 From: Michael S. Tsirkin Date: Mon, 23 Jan 2017 21:37:52 +0200 Subject: virtio_net: fix PAGE_SIZE > 64k I don't have any guests with PAGE_SIZE > 64k but the code seems to be clearly broken in that case as PAGE_SIZE / MERGEABLE_BUFFER_ALIGN will need more than 8 bit and so the code in mergeable_ctx_to_buf_address does not give us the actual true size. Cc: John Fastabend Signed-off-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 347424351ade..3d1519ea0669 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -48,8 +48,16 @@ module_param(gso, bool, 0444); */ DECLARE_EWMA(pkt_len, 1, 64) +/* With mergeable buffers we align buffer address and use the low bits to + * encode its true size. Buffer size is up to 1 page so we need to align to + * square root of page size to ensure we reserve enough bits to encode the true + * size. + */ +#define MERGEABLE_BUFFER_MIN_ALIGN_SHIFT ((PAGE_SHIFT + 1) / 2) + /* Minimum alignment for mergeable packet buffers. */ -#define MERGEABLE_BUFFER_ALIGN max(L1_CACHE_BYTES, 256) +#define MERGEABLE_BUFFER_ALIGN max(L1_CACHE_BYTES, \ + 1 << MERGEABLE_BUFFER_MIN_ALIGN_SHIFT) #define VIRTNET_DRIVER_VERSION "1.0.0" -- cgit v1.2.3 From 21b995a9cb093fff33ec91d7cb3822b882a90a1e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 23 Jan 2017 16:43:05 -0800 Subject: ip6_tunnel: must reload ipv6h in ip6ip6_tnl_xmit() Since ip6_tnl_parse_tlv_enc_lim() can call pskb_may_pull(), we must reload any pointer that was related to skb->head (or skb->data), or risk use after free. Fixes: c12b395a4664 ("gre: Support GRE over IPv6") Signed-off-by: Eric Dumazet Cc: Dmitry Kozlov Signed-off-by: David S. Miller --- net/ipv6/ip6_gre.c | 3 +++ net/ipv6/ip6_tunnel.c | 2 ++ 2 files changed, 5 insertions(+) diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 75b6108234dd..558631860d91 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -582,6 +582,9 @@ static inline int ip6gre_xmit_ipv6(struct sk_buff *skb, struct net_device *dev) return -1; offset = ip6_tnl_parse_tlv_enc_lim(skb, skb_network_header(skb)); + /* ip6_tnl_parse_tlv_enc_lim() might have reallocated skb->head */ + ipv6h = ipv6_hdr(skb); + if (offset > 0) { struct ipv6_tlv_tnl_enc_lim *tel; tel = (struct ipv6_tlv_tnl_enc_lim *)&skb_network_header(skb)[offset]; diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 753d6d0860fb..02923f956ac8 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -1303,6 +1303,8 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) fl6.flowlabel = key->label; } else { offset = ip6_tnl_parse_tlv_enc_lim(skb, skb_network_header(skb)); + /* ip6_tnl_parse_tlv_enc_lim() might have reallocated skb->head */ + ipv6h = ipv6_hdr(skb); if (offset > 0) { struct ipv6_tlv_tnl_enc_lim *tel; -- cgit v1.2.3 From fbfa743a9d2a0ffa24251764f10afc13eb21e739 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 23 Jan 2017 16:43:06 -0800 Subject: ipv6: fix ip6_tnl_parse_tlv_enc_lim() This function suffers from multiple issues. First one is that pskb_may_pull() may reallocate skb->head, so the 'raw' pointer needs either to be reloaded or not used at all. Second issue is that NEXTHDR_DEST handling does not validate that the options are present in skb->data, so we might read garbage or access non existent memory. With help from Willem de Bruijn. Signed-off-by: Eric Dumazet Reported-by: Dmitry Vyukov Cc: Willem de Bruijn Signed-off-by: David S. Miller --- net/ipv6/ip6_tunnel.c | 34 ++++++++++++++++++++++------------ 1 file changed, 22 insertions(+), 12 deletions(-) diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 02923f956ac8..ff8ee06491c3 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -400,18 +400,19 @@ ip6_tnl_dev_uninit(struct net_device *dev) __u16 ip6_tnl_parse_tlv_enc_lim(struct sk_buff *skb, __u8 *raw) { - const struct ipv6hdr *ipv6h = (const struct ipv6hdr *) raw; - __u8 nexthdr = ipv6h->nexthdr; - __u16 off = sizeof(*ipv6h); + const struct ipv6hdr *ipv6h = (const struct ipv6hdr *)raw; + unsigned int nhoff = raw - skb->data; + unsigned int off = nhoff + sizeof(*ipv6h); + u8 next, nexthdr = ipv6h->nexthdr; while (ipv6_ext_hdr(nexthdr) && nexthdr != NEXTHDR_NONE) { - __u16 optlen = 0; struct ipv6_opt_hdr *hdr; - if (raw + off + sizeof(*hdr) > skb->data && - !pskb_may_pull(skb, raw - skb->data + off + sizeof (*hdr))) + u16 optlen; + + if (!pskb_may_pull(skb, off + sizeof(*hdr))) break; - hdr = (struct ipv6_opt_hdr *) (raw + off); + hdr = (struct ipv6_opt_hdr *)(skb->data + off); if (nexthdr == NEXTHDR_FRAGMENT) { struct frag_hdr *frag_hdr = (struct frag_hdr *) hdr; if (frag_hdr->frag_off) @@ -422,20 +423,29 @@ __u16 ip6_tnl_parse_tlv_enc_lim(struct sk_buff *skb, __u8 *raw) } else { optlen = ipv6_optlen(hdr); } + /* cache hdr->nexthdr, since pskb_may_pull() might + * invalidate hdr + */ + next = hdr->nexthdr; if (nexthdr == NEXTHDR_DEST) { - __u16 i = off + 2; + u16 i = 2; + + /* Remember : hdr is no longer valid at this point. */ + if (!pskb_may_pull(skb, off + optlen)) + break; + while (1) { struct ipv6_tlv_tnl_enc_lim *tel; /* No more room for encapsulation limit */ - if (i + sizeof (*tel) > off + optlen) + if (i + sizeof(*tel) > optlen) break; - tel = (struct ipv6_tlv_tnl_enc_lim *) &raw[i]; + tel = (struct ipv6_tlv_tnl_enc_lim *) skb->data + off + i; /* return index of option if found and valid */ if (tel->type == IPV6_TLV_TNL_ENCAP_LIMIT && tel->length == 1) - return i; + return i + off - nhoff; /* else jump to next option */ if (tel->type) i += tel->length + 2; @@ -443,7 +453,7 @@ __u16 ip6_tnl_parse_tlv_enc_lim(struct sk_buff *skb, __u8 *raw) i++; } } - nexthdr = hdr->nexthdr; + nexthdr = next; off += optlen; } return 0; -- cgit v1.2.3 From 8b3f9337e17aaf710c79e65fd0a3c572a075f498 Mon Sep 17 00:00:00 2001 From: Roopa Prabhu Date: Mon, 23 Jan 2017 20:44:32 -0800 Subject: vxlan: don't flush static fdb entries on admin down This patch skips flushing static fdb entries in ndo_stop, but flushes all fdb entries during vxlan device delete. This is consistent with the bridge driver fdb Signed-off-by: Roopa Prabhu Signed-off-by: David S. Miller --- drivers/net/vxlan.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 8a79cfcac889..746a819f0a41 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2354,7 +2354,7 @@ static int vxlan_open(struct net_device *dev) } /* Purge the forwarding table */ -static void vxlan_flush(struct vxlan_dev *vxlan) +static void vxlan_flush(struct vxlan_dev *vxlan, bool do_all) { unsigned int h; @@ -2364,6 +2364,8 @@ static void vxlan_flush(struct vxlan_dev *vxlan) hlist_for_each_safe(p, n, &vxlan->fdb_head[h]) { struct vxlan_fdb *f = container_of(p, struct vxlan_fdb, hlist); + if (!do_all && (f->state & (NUD_PERMANENT | NUD_NOARP))) + continue; /* the all_zeros_mac entry is deleted at vxlan_uninit */ if (!is_zero_ether_addr(f->eth_addr)) vxlan_fdb_destroy(vxlan, f); @@ -2385,7 +2387,7 @@ static int vxlan_stop(struct net_device *dev) del_timer_sync(&vxlan->age_timer); - vxlan_flush(vxlan); + vxlan_flush(vxlan, false); vxlan_sock_release(vxlan); return ret; @@ -3058,6 +3060,8 @@ static void vxlan_dellink(struct net_device *dev, struct list_head *head) struct vxlan_dev *vxlan = netdev_priv(dev); struct vxlan_net *vn = net_generic(vxlan->net, vxlan_net_id); + vxlan_flush(vxlan, true); + spin_lock(&vn->sock_lock); if (!hlist_unhashed(&vxlan->hlist)) hlist_del_rcu(&vxlan->hlist); -- cgit v1.2.3 From efb5f68f32995c146944a9d4257c3cf8eae2c4a1 Mon Sep 17 00:00:00 2001 From: Balakrishnan Raman Date: Mon, 23 Jan 2017 20:44:33 -0800 Subject: vxlan: do not age static remote mac entries Mac aging is applicable only for dynamically learnt remote mac entries. Check for user configured static remote mac entries and skip aging. Signed-off-by: Balakrishnan Raman Signed-off-by: Roopa Prabhu Signed-off-by: David S. Miller --- drivers/net/vxlan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 746a819f0a41..50b62db213b0 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2268,7 +2268,7 @@ static void vxlan_cleanup(unsigned long arg) = container_of(p, struct vxlan_fdb, hlist); unsigned long timeout; - if (f->state & NUD_PERMANENT) + if (f->state & (NUD_PERMANENT | NUD_NOARP)) continue; timeout = f->used + vxlan->cfg.age_interval * HZ; -- cgit v1.2.3 From 23d28a859fb847fd7fcfbd31acb3b160abb5d6ae Mon Sep 17 00:00:00 2001 From: Thomas Huth Date: Tue, 24 Jan 2017 07:28:41 +0100 Subject: ibmveth: Add a proper check for the availability of the checksum features When using the ibmveth driver in a KVM/QEMU based VM, it currently always prints out a scary error message like this when it is started: ibmveth 71000003 (unregistered net_device): unable to change checksum offload settings. 1 rc=-2 ret_attr=71000003 This happens because the driver always tries to enable the checksum offloading without checking for the availability of this feature first. QEMU does not support checksum offloading for the spapr-vlan device, thus we always get the error message here. According to the LoPAPR specification, the "ibm,illan-options" property of the corresponding device tree node should be checked first to see whether the H_ILLAN_ATTRIUBTES hypercall and thus the checksum offloading feature is available. Thus let's do this in the ibmveth driver, too, so that the error message is really only limited to cases where something goes wrong, and does not occur if the feature is just missing. Signed-off-by: Thomas Huth Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmveth.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c index a831f947ca8c..309f5c66083c 100644 --- a/drivers/net/ethernet/ibm/ibmveth.c +++ b/drivers/net/ethernet/ibm/ibmveth.c @@ -1601,8 +1601,11 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id) netdev->netdev_ops = &ibmveth_netdev_ops; netdev->ethtool_ops = &netdev_ethtool_ops; SET_NETDEV_DEV(netdev, &dev->dev); - netdev->hw_features = NETIF_F_SG | NETIF_F_RXCSUM | - NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM; + netdev->hw_features = NETIF_F_SG; + if (vio_get_attribute(dev, "ibm,illan-options", NULL) != NULL) { + netdev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | + NETIF_F_RXCSUM; + } netdev->features |= netdev->hw_features; -- cgit v1.2.3 From 83d230eb5c638949350f4761acdfc0af5cb1bc00 Mon Sep 17 00:00:00 2001 From: Darrick J. Wong Date: Mon, 23 Jan 2017 19:43:00 -0800 Subject: xfs: verify dirblocklog correctly sb_dirblklog is added to sb_blocklog to compute the directory block size in bytes. Therefore, we must compare the sum of both those values against XFS_MAX_BLOCKSIZE_LOG, not just dirblklog. Signed-off-by: Darrick J. Wong Reviewed-by: Eric Sandeen Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_sb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index 2580262e4ea0..584ec896a533 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -242,7 +242,7 @@ xfs_mount_validate_sb( sbp->sb_blocklog < XFS_MIN_BLOCKSIZE_LOG || sbp->sb_blocklog > XFS_MAX_BLOCKSIZE_LOG || sbp->sb_blocksize != (1 << sbp->sb_blocklog) || - sbp->sb_dirblklog > XFS_MAX_BLOCKSIZE_LOG || + sbp->sb_dirblklog + sbp->sb_blocklog > XFS_MAX_BLOCKSIZE_LOG || sbp->sb_inodesize < XFS_DINODE_MIN_SIZE || sbp->sb_inodesize > XFS_DINODE_MAX_SIZE || sbp->sb_inodelog < XFS_DINODE_MIN_LOG || -- cgit v1.2.3 From 5b9f57516337b523f7466a53939aaaea7b78141b Mon Sep 17 00:00:00 2001 From: Bjørn Mork Date: Tue, 24 Jan 2017 10:45:38 +0100 Subject: qmi_wwan/cdc_ether: add device ID for HP lt2523 (Novatel E371) WWAN card Another rebranded Novatel E371. qmi_wwan should drive this device, while cdc_ether should ignore it. Even though the USB descriptors are plain CDC-ETHER that USB interface is a QMI interface. Ref commit 7fdb7846c9ca ("qmi_wwan/cdc_ether: add device IDs for Dell 5804 (Novatel E371) WWAN card") Cc: Dan Williams Signed-off-by: Bjørn Mork Signed-off-by: David S. Miller --- drivers/net/usb/cdc_ether.c | 8 ++++++++ drivers/net/usb/qmi_wwan.c | 7 +++++++ 2 files changed, 15 insertions(+) diff --git a/drivers/net/usb/cdc_ether.c b/drivers/net/usb/cdc_ether.c index fe7b2886cb6b..86144f9a80ee 100644 --- a/drivers/net/usb/cdc_ether.c +++ b/drivers/net/usb/cdc_ether.c @@ -531,6 +531,7 @@ static const struct driver_info wwan_info = { #define SAMSUNG_VENDOR_ID 0x04e8 #define LENOVO_VENDOR_ID 0x17ef #define NVIDIA_VENDOR_ID 0x0955 +#define HP_VENDOR_ID 0x03f0 static const struct usb_device_id products[] = { /* BLACKLIST !! @@ -677,6 +678,13 @@ static const struct usb_device_id products[] = { .driver_info = 0, }, +/* HP lt2523 (Novatel E371) - handled by qmi_wwan */ +{ + USB_DEVICE_AND_INTERFACE_INFO(HP_VENDOR_ID, 0x421d, USB_CLASS_COMM, + USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE), + .driver_info = 0, +}, + /* AnyDATA ADU960S - handled by qmi_wwan */ { USB_DEVICE_AND_INTERFACE_INFO(0x16d5, 0x650a, USB_CLASS_COMM, diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index 6fe1cdb0174f..24d5272cdce5 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -654,6 +654,13 @@ static const struct usb_device_id products[] = { USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&qmi_wwan_info, }, + { /* HP lt2523 (Novatel E371) */ + USB_DEVICE_AND_INTERFACE_INFO(0x03f0, 0x421d, + USB_CLASS_COMM, + USB_CDC_SUBCLASS_ETHERNET, + USB_CDC_PROTO_NONE), + .driver_info = (unsigned long)&qmi_wwan_info, + }, { /* HP lt4112 LTE/HSPA+ Gobi 4G Module (Huawei me906e) */ USB_DEVICE_AND_INTERFACE_INFO(0x03f0, 0x581d, USB_CLASS_VENDOR_SPEC, 1, 7), .driver_info = (unsigned long)&qmi_wwan_info, -- cgit v1.2.3 From f1db5c101cd48b5555ed9e061dcc49ed329812ea Mon Sep 17 00:00:00 2001 From: Tobias Regnery Date: Tue, 24 Jan 2017 14:34:22 +0100 Subject: alx: fix wrong condition to free descriptor memory The condition to free the descriptor memory is wrong, we want to free the memory if it is set and not if it is unset. Invert the test to fix this issue. Fixes: b0999223f224b ("alx: add ability to allocate and free alx_napi structures") Signed-off-by: Tobias Regnery Signed-off-by: David S. Miller --- drivers/net/ethernet/atheros/alx/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c index c8f525574d68..765306bd78c2 100644 --- a/drivers/net/ethernet/atheros/alx/main.c +++ b/drivers/net/ethernet/atheros/alx/main.c @@ -703,7 +703,7 @@ static void alx_free_rings(struct alx_priv *alx) if (alx->qnapi[0] && alx->qnapi[0]->rxq) kfree(alx->qnapi[0]->rxq->bufs); - if (!alx->descmem.virt) + if (alx->descmem.virt) dma_free_coherent(&alx->hw.pdev->dev, alx->descmem.size, alx->descmem.virt, -- cgit v1.2.3 From 37187a016c37d7e550544544dba25399ce4589c9 Mon Sep 17 00:00:00 2001 From: Tobias Regnery Date: Tue, 24 Jan 2017 14:34:23 +0100 Subject: alx: fix fallback to msi or legacy interrupts If requesting msi-x interrupts fails we should fall back to msi or legacy interrupts. However alx_realloc_ressources don't call alx_init_intr, so we fail to set the right number of tx queues. This results in watchdog timeouts and a nonfunctional adapter. Fixes: d768319cd427 ("alx: enable multiple tx queues") Signed-off-by: Tobias Regnery Signed-off-by: David S. Miller --- drivers/net/ethernet/atheros/alx/main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c index 765306bd78c2..75cbd46e429d 100644 --- a/drivers/net/ethernet/atheros/alx/main.c +++ b/drivers/net/ethernet/atheros/alx/main.c @@ -984,6 +984,7 @@ static int alx_realloc_resources(struct alx_priv *alx) alx_free_rings(alx); alx_free_napis(alx); alx_disable_advanced_intr(alx); + alx_init_intr(alx, false); err = alx_alloc_napis(alx); if (err) -- cgit v1.2.3 From 185aceefd80f98dc5b9d73eb6cbb70739a5ce4ea Mon Sep 17 00:00:00 2001 From: Tobias Regnery Date: Tue, 24 Jan 2017 14:34:24 +0100 Subject: alx: work around hardware bug in interrupt fallback path If requesting msi-x interrupts fails in alx_request_irq we fall back to a single tx queue and msi or legacy interrupts. Currently the adapter stops working in this case and we get tx watchdog timeouts. For reasons unknown the adapter gets confused when we load the dma adresses to the chip in alx_init_ring_ptrs twice: the first time with multiple queues and the second time in the fallback case with a single queue. To fix this move the the call to alx_reinit_rings (which calls alx_init_ring_ptrs) after alx_request_irq. At this time it is clear how much tx queues we have and which dma addresses we use. Fixes: d768319cd427 ("alx: enable multiple tx queues") Signed-off-by: Tobias Regnery Signed-off-by: David S. Miller --- drivers/net/ethernet/atheros/alx/main.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c index 75cbd46e429d..7dcc907a449d 100644 --- a/drivers/net/ethernet/atheros/alx/main.c +++ b/drivers/net/ethernet/atheros/alx/main.c @@ -685,8 +685,6 @@ static int alx_alloc_rings(struct alx_priv *alx) return -ENOMEM; } - alx_reinit_rings(alx); - return 0; } @@ -1242,6 +1240,12 @@ static int __alx_open(struct alx_priv *alx, bool resume) if (err) goto out_free_rings; + /* must be called after alx_request_irq because the chip stops working + * if we copy the dma addresses in alx_init_ring_ptrs twice when + * requesting msi-x interrupts failed + */ + alx_reinit_rings(alx); + netif_set_real_num_tx_queues(alx->dev, alx->num_txq); netif_set_real_num_rx_queues(alx->dev, alx->num_rxq); -- cgit v1.2.3 From c929ea0b910355e1876c64431f3d5802f95b3d75 Mon Sep 17 00:00:00 2001 From: Kinglong Mee Date: Fri, 20 Jan 2017 16:48:39 +0800 Subject: SUNRPC: cleanup ida information when removing sunrpc module After removing sunrpc module, I get many kmemleak information as, unreferenced object 0xffff88003316b1e0 (size 544): comm "gssproxy", pid 2148, jiffies 4294794465 (age 4200.081s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] kmemleak_alloc+0x4a/0xa0 [] kmem_cache_alloc+0x15e/0x1f0 [] ida_pre_get+0xaa/0x150 [] ida_simple_get+0xad/0x180 [] nlmsvc_lookup_host+0x4ab/0x7f0 [lockd] [] lockd+0x4d/0x270 [lockd] [] param_set_timeout+0x55/0x100 [lockd] [] svc_defer+0x114/0x3f0 [sunrpc] [] svc_defer+0x2d7/0x3f0 [sunrpc] [] rpc_show_info+0x8a/0x110 [sunrpc] [] proc_reg_write+0x7f/0xc0 [] __vfs_write+0xdf/0x3c0 [] vfs_write+0xef/0x240 [] SyS_write+0xad/0x130 [] entry_SYSCALL_64_fastpath+0x1a/0xa9 [] 0xffffffffffffffff I found, the ida information (dynamic memory) isn't cleanup. Signed-off-by: Kinglong Mee Fixes: 2f048db4680a ("SUNRPC: Add an identifier for struct rpc_clnt") Cc: stable@vger.kernel.org # v3.12+ Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 1 + net/sunrpc/clnt.c | 5 +++++ net/sunrpc/sunrpc_syms.c | 1 + 3 files changed, 7 insertions(+) diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index 85cc819676e8..333ad11b3dd9 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -216,5 +216,6 @@ void rpc_clnt_xprt_switch_put(struct rpc_clnt *); void rpc_clnt_xprt_switch_add_xprt(struct rpc_clnt *, struct rpc_xprt *); bool rpc_clnt_xprt_switch_has_addr(struct rpc_clnt *clnt, const struct sockaddr *sap); +void rpc_cleanup_clids(void); #endif /* __KERNEL__ */ #endif /* _LINUX_SUNRPC_CLNT_H */ diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 1efbe48e794f..1dc9f3bac099 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -336,6 +336,11 @@ out: static DEFINE_IDA(rpc_clids); +void rpc_cleanup_clids(void) +{ + ida_destroy(&rpc_clids); +} + static int rpc_alloc_clid(struct rpc_clnt *clnt) { int clid; diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c index d1c330a7953a..c73de181467a 100644 --- a/net/sunrpc/sunrpc_syms.c +++ b/net/sunrpc/sunrpc_syms.c @@ -119,6 +119,7 @@ out: static void __exit cleanup_sunrpc(void) { + rpc_cleanup_clids(); rpcauth_remove_module(); cleanup_socket_xprt(); svc_cleanup_xprt_sock(); -- cgit v1.2.3 From d3f4aadd614c4627244452ad64eaf351179f2c31 Mon Sep 17 00:00:00 2001 From: Amrani, Ram Date: Mon, 26 Dec 2016 08:40:57 +0200 Subject: RDMA/core: Add the function ib_mtu_int_to_enum As the functionality to convert the MTU from a number to enum_ib_mtu is ubiquitous, define a dedicated function and remove the duplicated code. Signed-off-by: Ram Amrani Reviewed-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/cxgb3/iwch_provider.c | 11 +---------- drivers/infiniband/hw/cxgb4/provider.c | 11 +---------- drivers/infiniband/hw/i40iw/i40iw_verbs.c | 11 +---------- drivers/infiniband/hw/nes/nes_verbs.c | 12 +----------- include/rdma/ib_verbs.h | 14 ++++++++++++++ 5 files changed, 18 insertions(+), 41 deletions(-) diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c index 9d5fe1853da4..6262dc035f3c 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.c +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c @@ -1135,16 +1135,7 @@ static int iwch_query_port(struct ib_device *ibdev, memset(props, 0, sizeof(struct ib_port_attr)); props->max_mtu = IB_MTU_4096; - if (netdev->mtu >= 4096) - props->active_mtu = IB_MTU_4096; - else if (netdev->mtu >= 2048) - props->active_mtu = IB_MTU_2048; - else if (netdev->mtu >= 1024) - props->active_mtu = IB_MTU_1024; - else if (netdev->mtu >= 512) - props->active_mtu = IB_MTU_512; - else - props->active_mtu = IB_MTU_256; + props->active_mtu = ib_mtu_int_to_enum(netdev->mtu); if (!netif_carrier_ok(netdev)) props->state = IB_PORT_DOWN; diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c index fa64f5d93b11..3345e1c312f7 100644 --- a/drivers/infiniband/hw/cxgb4/provider.c +++ b/drivers/infiniband/hw/cxgb4/provider.c @@ -373,16 +373,7 @@ static int c4iw_query_port(struct ib_device *ibdev, u8 port, memset(props, 0, sizeof(struct ib_port_attr)); props->max_mtu = IB_MTU_4096; - if (netdev->mtu >= 4096) - props->active_mtu = IB_MTU_4096; - else if (netdev->mtu >= 2048) - props->active_mtu = IB_MTU_2048; - else if (netdev->mtu >= 1024) - props->active_mtu = IB_MTU_1024; - else if (netdev->mtu >= 512) - props->active_mtu = IB_MTU_512; - else - props->active_mtu = IB_MTU_256; + props->active_mtu = ib_mtu_int_to_enum(netdev->mtu); if (!netif_carrier_ok(netdev)) props->state = IB_PORT_DOWN; diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c index 29e97df9e1a7..4c000d60d5c6 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c @@ -100,16 +100,7 @@ static int i40iw_query_port(struct ib_device *ibdev, memset(props, 0, sizeof(*props)); props->max_mtu = IB_MTU_4096; - if (netdev->mtu >= 4096) - props->active_mtu = IB_MTU_4096; - else if (netdev->mtu >= 2048) - props->active_mtu = IB_MTU_2048; - else if (netdev->mtu >= 1024) - props->active_mtu = IB_MTU_1024; - else if (netdev->mtu >= 512) - props->active_mtu = IB_MTU_512; - else - props->active_mtu = IB_MTU_256; + props->active_mtu = ib_mtu_int_to_enum(netdev->mtu); props->lid = 1; if (netif_carrier_ok(iwdev->netdev)) diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c index aff9fb14768b..5a31f3c6a421 100644 --- a/drivers/infiniband/hw/nes/nes_verbs.c +++ b/drivers/infiniband/hw/nes/nes_verbs.c @@ -478,17 +478,7 @@ static int nes_query_port(struct ib_device *ibdev, u8 port, struct ib_port_attr memset(props, 0, sizeof(*props)); props->max_mtu = IB_MTU_4096; - - if (netdev->mtu >= 4096) - props->active_mtu = IB_MTU_4096; - else if (netdev->mtu >= 2048) - props->active_mtu = IB_MTU_2048; - else if (netdev->mtu >= 1024) - props->active_mtu = IB_MTU_1024; - else if (netdev->mtu >= 512) - props->active_mtu = IB_MTU_512; - else - props->active_mtu = IB_MTU_256; + props->active_mtu = ib_mtu_int_to_enum(netdev->mtu); props->lid = 1; props->lmc = 0; diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 958a24d8fae7..b567e4452a47 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -352,6 +352,20 @@ static inline int ib_mtu_enum_to_int(enum ib_mtu mtu) } } +static inline enum ib_mtu ib_mtu_int_to_enum(int mtu) +{ + if (mtu >= 4096) + return IB_MTU_4096; + else if (mtu >= 2048) + return IB_MTU_2048; + else if (mtu >= 1024) + return IB_MTU_1024; + else if (mtu >= 512) + return IB_MTU_512; + else + return IB_MTU_256; +} + enum ib_port_state { IB_PORT_NOP = 0, IB_PORT_DOWN = 1, -- cgit v1.2.3 From 097b615965fb1af714fbc2311f68839b1086ebcb Mon Sep 17 00:00:00 2001 From: Amrani, Ram Date: Mon, 26 Dec 2016 08:40:58 +0200 Subject: RDMA/qedr: Fix MTU returned from QP query MTU value returned from QP query should include overhead. Signed-off-by: Ram Amrani Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/verbs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 57c8de208077..84dcd73bd57b 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -2016,7 +2016,7 @@ int qedr_query_qp(struct ib_qp *ibqp, qp_attr->qp_state = qedr_get_ibqp_state(params.state); qp_attr->cur_qp_state = qedr_get_ibqp_state(params.state); - qp_attr->path_mtu = iboe_get_mtu(params.mtu); + qp_attr->path_mtu = ib_mtu_int_to_enum(params.mtu); qp_attr->path_mig_state = IB_MIG_MIGRATED; qp_attr->rq_psn = params.rq_psn; qp_attr->sq_psn = params.sq_psn; -- cgit v1.2.3 From 20f5e10ef8bcf29a915642245b66e5a132e38fc4 Mon Sep 17 00:00:00 2001 From: Amrani, Ram Date: Tue, 24 Jan 2017 12:01:31 +0200 Subject: RDMA/qedr: Add uapi header qedr-abi.h Signed-off-by: Ram Amrani Signed-off-by: Doug Ledford --- include/uapi/rdma/Kbuild | 1 + 1 file changed, 1 insertion(+) diff --git a/include/uapi/rdma/Kbuild b/include/uapi/rdma/Kbuild index 82bdf5626859..bb68cb1b04ed 100644 --- a/include/uapi/rdma/Kbuild +++ b/include/uapi/rdma/Kbuild @@ -16,3 +16,4 @@ header-y += nes-abi.h header-y += ocrdma-abi.h header-y += hns-abi.h header-y += vmw_pvrdma-abi.h +header-y += qedr-abi.h -- cgit v1.2.3 From 865cea40b69741c3da2574176876463233b2b67c Mon Sep 17 00:00:00 2001 From: Ram Amrani Date: Tue, 24 Jan 2017 13:50:34 +0200 Subject: RDMA/qedr: Return success when not changing QP state If the user is requesting us to change the QP state to the same state that it is already in, return success instead of failure. Signed-off-by: Ram Amrani Signed-off-by: Michal Kalderon Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/verbs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 84dcd73bd57b..27d90a82d731 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -1657,7 +1657,7 @@ static int qedr_update_qp_state(struct qedr_dev *dev, int status = 0; if (new_state == qp->state) - return 1; + return 0; switch (qp->state) { case QED_ROCE_QP_STATE_RESET: -- cgit v1.2.3 From 59e8970b3798e4cbe575ed9cf4d53098760a2a86 Mon Sep 17 00:00:00 2001 From: Ram Amrani Date: Tue, 24 Jan 2017 13:50:35 +0200 Subject: RDMA/qedr: Return max inline data in QP query result Return the maximum supported amount of inline data, not the qp's current configured inline data size, when filling out the results of a query qp call. Signed-off-by: Ram Amrani Signed-off-by: Michal Kalderon Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/verbs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 27d90a82d731..bb907b14c988 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -2028,7 +2028,7 @@ int qedr_query_qp(struct ib_qp *ibqp, qp_attr->cap.max_recv_wr = qp->rq.max_wr; qp_attr->cap.max_send_sge = qp->sq.max_sges; qp_attr->cap.max_recv_sge = qp->rq.max_sges; - qp_attr->cap.max_inline_data = qp->max_inline_data; + qp_attr->cap.max_inline_data = ROCE_REQ_MAX_INLINE_DATA_SIZE; qp_init_attr->cap = qp_attr->cap; memcpy(&qp_attr->ah_attr.grh.dgid.raw[0], ¶ms.dgid.bytes[0], -- cgit v1.2.3 From 91bff997db2ec04f9ba761a55c21642f9803b06c Mon Sep 17 00:00:00 2001 From: Ram Amrani Date: Tue, 24 Jan 2017 13:50:36 +0200 Subject: RDMA/qedr: Remove CQ spinlock from CM completion handlers There is only a single event queue that triggers the completion events for the RDMA CM and it is being processed serially. This means that inherently there can no parallelism of CQ completion handler callbacks, hence the lock is redundant. Signed-off-by: Ram Amrani Signed-off-by: Michal Kalderon Reviewed-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/qedr.h | 3 --- drivers/infiniband/hw/qedr/qedr_cm.c | 10 ++-------- 2 files changed, 2 insertions(+), 11 deletions(-) diff --git a/drivers/infiniband/hw/qedr/qedr.h b/drivers/infiniband/hw/qedr/qedr.h index 620badd7d4fb..94319abb0df2 100644 --- a/drivers/infiniband/hw/qedr/qedr.h +++ b/drivers/infiniband/hw/qedr/qedr.h @@ -251,9 +251,6 @@ struct qedr_cq { u16 icid; - /* Lock to protect completion handler */ - spinlock_t comp_handler_lock; - /* Lock to protect multiplem CQ's */ spinlock_t cq_lock; u8 arm_flags; diff --git a/drivers/infiniband/hw/qedr/qedr_cm.c b/drivers/infiniband/hw/qedr/qedr_cm.c index 63890ebb72bd..00361f310d15 100644 --- a/drivers/infiniband/hw/qedr/qedr_cm.c +++ b/drivers/infiniband/hw/qedr/qedr_cm.c @@ -87,11 +87,8 @@ void qedr_ll2_tx_cb(void *_qdev, struct qed_roce_ll2_packet *pkt) qedr_inc_sw_gsi_cons(&qp->sq); spin_unlock_irqrestore(&qp->q_lock, flags); - if (cq->ibcq.comp_handler) { - spin_lock_irqsave(&cq->comp_handler_lock, flags); + if (cq->ibcq.comp_handler) (*cq->ibcq.comp_handler) (&cq->ibcq, cq->ibcq.cq_context); - spin_unlock_irqrestore(&cq->comp_handler_lock, flags); - } } void qedr_ll2_rx_cb(void *_dev, struct qed_roce_ll2_packet *pkt, @@ -113,11 +110,8 @@ void qedr_ll2_rx_cb(void *_dev, struct qed_roce_ll2_packet *pkt, spin_unlock_irqrestore(&qp->q_lock, flags); - if (cq->ibcq.comp_handler) { - spin_lock_irqsave(&cq->comp_handler_lock, flags); + if (cq->ibcq.comp_handler) (*cq->ibcq.comp_handler) (&cq->ibcq, cq->ibcq.cq_context); - spin_unlock_irqrestore(&cq->comp_handler_lock, flags); - } } static void qedr_destroy_gsi_cq(struct qedr_dev *dev, -- cgit v1.2.3 From c78c31496111f497b4a03f955c100091185da8b6 Mon Sep 17 00:00:00 2001 From: Ram Amrani Date: Tue, 24 Jan 2017 13:50:37 +0200 Subject: RDMA/qedr: Don't spam dmesg if QP is in error state It is normal to flush CQEs if the QP is in error state. Hence there's no use in printing a message per CQE to dmesg. Signed-off-by: Ram Amrani Signed-off-by: Michal Kalderon Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/verbs.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index bb907b14c988..0688dce54f92 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -3234,9 +3234,10 @@ static int qedr_poll_cq_req(struct qedr_dev *dev, IB_WC_SUCCESS, 0); break; case RDMA_CQE_REQ_STS_WORK_REQUEST_FLUSHED_ERR: - DP_ERR(dev, - "Error: POLL CQ with RDMA_CQE_REQ_STS_WORK_REQUEST_FLUSHED_ERR. CQ icid=0x%x, QP icid=0x%x\n", - cq->icid, qp->icid); + if (qp->state != QED_ROCE_QP_STATE_ERR) + DP_ERR(dev, + "Error: POLL CQ with RDMA_CQE_REQ_STS_WORK_REQUEST_FLUSHED_ERR. CQ icid=0x%x, QP icid=0x%x\n", + cq->icid, qp->icid); cnt = process_req(dev, qp, cq, num_entries, wc, req->sq_cons, IB_WC_WR_FLUSH_ERR, 1); break; -- cgit v1.2.3 From 933e6dcaa0f65eb2f624ad760274020874a1f35e Mon Sep 17 00:00:00 2001 From: Ram Amrani Date: Tue, 24 Jan 2017 13:50:38 +0200 Subject: RDMA/qedr: Don't reset QP when queues aren't flushed Fail QP state transition from error to reset if SQ/RQ are not empty and still in the process of flushing out the queued work entries. Signed-off-by: Ram Amrani Signed-off-by: Michal Kalderon Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/verbs.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 0688dce54f92..3dae9641f821 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -1733,6 +1733,14 @@ static int qedr_update_qp_state(struct qedr_dev *dev, /* ERR->XXX */ switch (new_state) { case QED_ROCE_QP_STATE_RESET: + if ((qp->rq.prod != qp->rq.cons) || + (qp->sq.prod != qp->sq.cons)) { + DP_NOTICE(dev, + "Error->Reset with rq/sq not empty rq.prod=%x rq.cons=%x sq.prod=%x sq.cons=%x\n", + qp->rq.prod, qp->rq.cons, qp->sq.prod, + qp->sq.cons); + status = -EINVAL; + } break; default: status = -EINVAL; -- cgit v1.2.3 From 27a4b1a6d6fcf09314359bacefa1e106927ae21b Mon Sep 17 00:00:00 2001 From: Ram Amrani Date: Tue, 24 Jan 2017 13:51:39 +0200 Subject: RDMA/qedr: Mark three functions as static mark qedr_get_state_from_ibqp(), __qedr_alloc_mr() and __qedr_post_send() as static since they are only used in the same file. Signed-off-by: Ram Amrani Signed-off-by: Ariel Elior Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/verbs.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 3dae9641f821..1b9d55965479 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -1600,7 +1600,7 @@ err0: return ERR_PTR(-EFAULT); } -enum ib_qp_state qedr_get_ibqp_state(enum qed_roce_qp_state qp_state) +static enum ib_qp_state qedr_get_ibqp_state(enum qed_roce_qp_state qp_state) { switch (qp_state) { case QED_ROCE_QP_STATE_RESET: @@ -1621,7 +1621,8 @@ enum ib_qp_state qedr_get_ibqp_state(enum qed_roce_qp_state qp_state) return IB_QPS_ERR; } -enum qed_roce_qp_state qedr_get_state_from_ibqp(enum ib_qp_state qp_state) +static enum qed_roce_qp_state qedr_get_state_from_ibqp( + enum ib_qp_state qp_state) { switch (qp_state) { case IB_QPS_RESET: @@ -2310,7 +2311,8 @@ int qedr_dereg_mr(struct ib_mr *ib_mr) return rc; } -struct qedr_mr *__qedr_alloc_mr(struct ib_pd *ibpd, int max_page_list_len) +static struct qedr_mr *__qedr_alloc_mr(struct ib_pd *ibpd, + int max_page_list_len) { struct qedr_pd *pd = get_qedr_pd(ibpd); struct qedr_dev *dev = get_qedr_dev(ibpd->device); @@ -2712,7 +2714,7 @@ static int qedr_prepare_reg(struct qedr_qp *qp, return 0; } -enum ib_wc_opcode qedr_ib_to_wc_opcode(enum ib_wr_opcode opcode) +static enum ib_wc_opcode qedr_ib_to_wc_opcode(enum ib_wr_opcode opcode) { switch (opcode) { case IB_WR_RDMA_WRITE: @@ -2737,7 +2739,7 @@ enum ib_wc_opcode qedr_ib_to_wc_opcode(enum ib_wr_opcode opcode) } } -inline bool qedr_can_post_send(struct qedr_qp *qp, struct ib_send_wr *wr) +static inline bool qedr_can_post_send(struct qedr_qp *qp, struct ib_send_wr *wr) { int wq_is_full, err_wr, pbl_is_full; struct qedr_dev *dev = qp->dev; @@ -2774,7 +2776,7 @@ inline bool qedr_can_post_send(struct qedr_qp *qp, struct ib_send_wr *wr) return true; } -int __qedr_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, +static int __qedr_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, struct ib_send_wr **bad_wr) { struct qedr_dev *dev = get_qedr_dev(ibqp->device); -- cgit v1.2.3 From 1a59075197976611bacaa383a6673f9e57e9e98b Mon Sep 17 00:00:00 2001 From: Ram Amrani Date: Tue, 24 Jan 2017 13:51:40 +0200 Subject: RDMA/qedr: Fix formatting Remove standalone ';'. List function's parameters in a single line. Signed-off-by: Ram Amrani Signed-off-by: Ariel Elior Reviewed-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/main.c | 3 +-- drivers/infiniband/hw/qedr/verbs.c | 1 - 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/qedr/main.c b/drivers/infiniband/hw/qedr/main.c index 7b74d09a8217..eac0bfc1fc99 100644 --- a/drivers/infiniband/hw/qedr/main.c +++ b/drivers/infiniband/hw/qedr/main.c @@ -576,8 +576,7 @@ static int qedr_set_device_attr(struct qedr_dev *dev) return 0; } -void qedr_unaffiliated_event(void *context, - u8 event_code) +void qedr_unaffiliated_event(void *context, u8 event_code) { pr_err("unaffiliated event not implemented yet\n"); } diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 1b9d55965479..c4f4c2131a45 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -1874,7 +1874,6 @@ int qedr_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, qp_params.sgid.dwords[2], qp_params.sgid.dwords[3]); DP_DEBUG(dev, QEDR_MSG_QP, "remote_mac=[%pM]\n", qp_params.remote_mac_addr); -; qp_params.mtu = qp->mtu; qp_params.lb_indication = false; -- cgit v1.2.3 From af2b14b8b8ae21b0047a52c767ac8b44f435a280 Mon Sep 17 00:00:00 2001 From: Ram Amrani Date: Tue, 24 Jan 2017 13:51:41 +0200 Subject: RDMA/qedr: Fix RDMA CM loopback The loopback logic in RDMA CM packets compares Ethernet addresses and was accidently inverse. Signed-off-by: Ram Amrani Signed-off-by: Ariel Elior Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/qedr_cm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/qedr/qedr_cm.c b/drivers/infiniband/hw/qedr/qedr_cm.c index 00361f310d15..a9a8d8745d2e 100644 --- a/drivers/infiniband/hw/qedr/qedr_cm.c +++ b/drivers/infiniband/hw/qedr/qedr_cm.c @@ -398,9 +398,9 @@ static inline int qedr_gsi_build_packet(struct qedr_dev *dev, } if (ether_addr_equal(udh.eth.smac_h, udh.eth.dmac_h)) - packet->tx_dest = QED_ROCE_LL2_TX_DEST_NW; - else packet->tx_dest = QED_ROCE_LL2_TX_DEST_LB; + else + packet->tx_dest = QED_ROCE_LL2_TX_DEST_NW; packet->roce_mode = roce_mode; memcpy(packet->header.vaddr, ud_header_buffer, header_size); -- cgit v1.2.3 From 9c1e0228ab35e52d30abf4b5629c28350833fbcb Mon Sep 17 00:00:00 2001 From: Ram Amrani Date: Tue, 24 Jan 2017 13:51:42 +0200 Subject: RDMA/qedr: Fix and simplify memory leak in PD alloc Free the PD if no internal resources were available. Move userspace code under the relevant 'if'. Signed-off-by: Ram Amrani Signed-off-by: Ariel Elior Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/verbs.c | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-) diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index c4f4c2131a45..c7d6c9a783bd 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -471,8 +471,6 @@ struct ib_pd *qedr_alloc_pd(struct ib_device *ibdev, struct ib_ucontext *context, struct ib_udata *udata) { struct qedr_dev *dev = get_qedr_dev(ibdev); - struct qedr_ucontext *uctx = NULL; - struct qedr_alloc_pd_uresp uresp; struct qedr_pd *pd; u16 pd_id; int rc; @@ -489,21 +487,33 @@ struct ib_pd *qedr_alloc_pd(struct ib_device *ibdev, if (!pd) return ERR_PTR(-ENOMEM); - dev->ops->rdma_alloc_pd(dev->rdma_ctx, &pd_id); + rc = dev->ops->rdma_alloc_pd(dev->rdma_ctx, &pd_id); + if (rc) + goto err; - uresp.pd_id = pd_id; pd->pd_id = pd_id; if (udata && context) { + struct qedr_alloc_pd_uresp uresp; + + uresp.pd_id = pd_id; + rc = ib_copy_to_udata(udata, &uresp, sizeof(uresp)); - if (rc) + if (rc) { DP_ERR(dev, "copy error pd_id=0x%x.\n", pd_id); - uctx = get_qedr_ucontext(context); - uctx->pd = pd; - pd->uctx = uctx; + dev->ops->rdma_dealloc_pd(dev->rdma_ctx, pd_id); + goto err; + } + + pd->uctx = get_qedr_ucontext(context); + pd->uctx->pd = pd; } return &pd->ibpd; + +err: + kfree(pd); + return ERR_PTR(rc); } int qedr_dealloc_pd(struct ib_pd *ibpd) -- cgit v1.2.3 From f449c7a2d822c2d81b5bcb2c50eec80796766726 Mon Sep 17 00:00:00 2001 From: Ram Amrani Date: Tue, 24 Jan 2017 13:51:43 +0200 Subject: RDMA/qedr: Dispatch port active event from qedr_add Relying on qede to trigger qedr on startup is problematic. When probing both if qedr loads slowly then qede can assume qedr is missing and not trigger it. This patch adds a triggering from qedr and protects against a race via an atomic bit. Signed-off-by: Ram Amrani Signed-off-by: Ariel Elior Signed-off-by: Doug Ledford --- drivers/infiniband/hw/qedr/main.c | 20 ++++++++++++++------ drivers/infiniband/hw/qedr/qedr.h | 5 +++++ 2 files changed, 19 insertions(+), 6 deletions(-) diff --git a/drivers/infiniband/hw/qedr/main.c b/drivers/infiniband/hw/qedr/main.c index eac0bfc1fc99..3ac8aa5ef37d 100644 --- a/drivers/infiniband/hw/qedr/main.c +++ b/drivers/infiniband/hw/qedr/main.c @@ -791,6 +791,9 @@ static struct qedr_dev *qedr_add(struct qed_dev *cdev, struct pci_dev *pdev, if (device_create_file(&dev->ibdev.dev, qedr_attributes[i])) goto sysfs_err; + if (!test_and_set_bit(QEDR_ENET_STATE_BIT, &dev->enet_state)) + qedr_ib_dispatch_event(dev, QEDR_PORT, IB_EVENT_PORT_ACTIVE); + DP_DEBUG(dev, QEDR_MSG_INIT, "qedr driver loaded successfully\n"); return dev; @@ -823,11 +826,10 @@ static void qedr_remove(struct qedr_dev *dev) ib_dealloc_device(&dev->ibdev); } -static int qedr_close(struct qedr_dev *dev) +static void qedr_close(struct qedr_dev *dev) { - qedr_ib_dispatch_event(dev, 1, IB_EVENT_PORT_ERR); - - return 0; + if (test_and_clear_bit(QEDR_ENET_STATE_BIT, &dev->enet_state)) + qedr_ib_dispatch_event(dev, QEDR_PORT, IB_EVENT_PORT_ERR); } static void qedr_shutdown(struct qedr_dev *dev) @@ -836,6 +838,12 @@ static void qedr_shutdown(struct qedr_dev *dev) qedr_remove(dev); } +static void qedr_open(struct qedr_dev *dev) +{ + if (!test_and_set_bit(QEDR_ENET_STATE_BIT, &dev->enet_state)) + qedr_ib_dispatch_event(dev, QEDR_PORT, IB_EVENT_PORT_ACTIVE); +} + static void qedr_mac_address_change(struct qedr_dev *dev) { union ib_gid *sgid = &dev->sgid_tbl[0]; @@ -862,7 +870,7 @@ static void qedr_mac_address_change(struct qedr_dev *dev) ether_addr_copy(dev->gsi_ll2_mac_address, dev->ndev->dev_addr); - qedr_ib_dispatch_event(dev, 1, IB_EVENT_GID_CHANGE); + qedr_ib_dispatch_event(dev, QEDR_PORT, IB_EVENT_GID_CHANGE); if (rc) DP_ERR(dev, "Error updating mac filter\n"); @@ -876,7 +884,7 @@ static void qedr_notify(struct qedr_dev *dev, enum qede_roce_event event) { switch (event) { case QEDE_UP: - qedr_ib_dispatch_event(dev, 1, IB_EVENT_PORT_ACTIVE); + qedr_open(dev); break; case QEDE_DOWN: qedr_close(dev); diff --git a/drivers/infiniband/hw/qedr/qedr.h b/drivers/infiniband/hw/qedr/qedr.h index 94319abb0df2..bb32e4792ec9 100644 --- a/drivers/infiniband/hw/qedr/qedr.h +++ b/drivers/infiniband/hw/qedr/qedr.h @@ -113,6 +113,8 @@ struct qedr_device_attr { struct qed_rdma_events events; }; +#define QEDR_ENET_STATE_BIT (0) + struct qedr_dev { struct ib_device ibdev; struct qed_dev *cdev; @@ -153,6 +155,8 @@ struct qedr_dev { struct qedr_cq *gsi_sqcq; struct qedr_cq *gsi_rqcq; struct qedr_qp *gsi_qp; + + unsigned long enet_state; }; #define QEDR_MAX_SQ_PBL (0x8000) @@ -188,6 +192,7 @@ struct qedr_dev { #define QEDR_ROCE_MAX_CNQ_SIZE (0x4000) #define QEDR_MAX_PORT (1) +#define QEDR_PORT (1) #define QEDR_UVERBS(CMD_NAME) (1ull << IB_USER_VERBS_CMD_##CMD_NAME) -- cgit v1.2.3 From 828f6fa65ce7e80f77f5ab12942e44eb3d9d174e Mon Sep 17 00:00:00 2001 From: Kenneth Lee Date: Thu, 5 Jan 2017 15:00:05 +0800 Subject: IB/umem: Release pid in error and ODP flow 1. Release pid before enter odp flow 2. Release pid when fail to allocate memory Fixes: 87773dd56d54 ("IB: ib_umem_release() should decrement mm->pinned_vm from ib_umem_get") Fixes: 8ada2c1c0c1d ("IB/core: Add support for on demand paging regions") Signed-off-by: Kenneth Lee Reviewed-by: Haggai Eran Reviewed-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/core/umem.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index 1e62a5f0cb28..4609b921f899 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -134,6 +134,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, IB_ACCESS_REMOTE_ATOMIC | IB_ACCESS_MW_BIND)); if (access & IB_ACCESS_ON_DEMAND) { + put_pid(umem->pid); ret = ib_umem_odp_get(context, umem); if (ret) { kfree(umem); @@ -149,6 +150,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, page_list = (struct page **) __get_free_page(GFP_KERNEL); if (!page_list) { + put_pid(umem->pid); kfree(umem); return ERR_PTR(-ENOMEM); } -- cgit v1.2.3 From 5ce6b04ce96896e8a79e6f60740ced911eaac7a4 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sun, 22 Jan 2017 22:10:32 +0800 Subject: netfilter: nft_log: restrict the log prefix length to 127 First, log prefix will be truncated to NF_LOG_PREFIXLEN-1, i.e. 127, at nf_log_packet(), so the extra part is useless. Second, after adding a log rule with a very very long prefix, we will fail to dump the nft rules after this _special_ one, but acctually, they do exist. For example: # name_65000=$(printf "%0.sQ" {1..65000}) # nft add rule filter output log prefix "$name_65000" # nft add rule filter output counter # nft add rule filter output counter # nft list chain filter output table ip filter { chain output { type filter hook output priority 0; policy accept; } } So now, restrict the log prefix length to NF_LOG_PREFIXLEN-1. Fixes: 96518518cc41 ("netfilter: add nftables") Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/nf_log.h | 2 ++ net/netfilter/nf_log.c | 1 - net/netfilter/nft_log.c | 3 ++- 3 files changed, 4 insertions(+), 2 deletions(-) diff --git a/include/uapi/linux/netfilter/nf_log.h b/include/uapi/linux/netfilter/nf_log.h index 8be21e02387d..d0b5fa91ff54 100644 --- a/include/uapi/linux/netfilter/nf_log.h +++ b/include/uapi/linux/netfilter/nf_log.h @@ -9,4 +9,6 @@ #define NF_LOG_MACDECODE 0x20 /* Decode MAC header */ #define NF_LOG_MASK 0x2f +#define NF_LOG_PREFIXLEN 128 + #endif /* _NETFILTER_NF_LOG_H */ diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index 3dca90dc24ad..ffb9e8ada899 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -13,7 +13,6 @@ /* Internal logging interface, which relies on the real LOG target modules */ -#define NF_LOG_PREFIXLEN 128 #define NFLOGGER_NAME_LEN 64 static struct nf_logger __rcu *loggers[NFPROTO_NUMPROTO][NF_LOG_TYPE_MAX] __read_mostly; diff --git a/net/netfilter/nft_log.c b/net/netfilter/nft_log.c index 6271e40a3dd6..6f6e64423643 100644 --- a/net/netfilter/nft_log.c +++ b/net/netfilter/nft_log.c @@ -39,7 +39,8 @@ static void nft_log_eval(const struct nft_expr *expr, static const struct nla_policy nft_log_policy[NFTA_LOG_MAX + 1] = { [NFTA_LOG_GROUP] = { .type = NLA_U16 }, - [NFTA_LOG_PREFIX] = { .type = NLA_STRING }, + [NFTA_LOG_PREFIX] = { .type = NLA_STRING, + .len = NF_LOG_PREFIXLEN - 1 }, [NFTA_LOG_SNAPLEN] = { .type = NLA_U32 }, [NFTA_LOG_QTHRESHOLD] = { .type = NLA_U16 }, [NFTA_LOG_LEVEL] = { .type = NLA_U32 }, -- cgit v1.2.3 From 35d0ac9070ef619e3bf44324375878a1c540387b Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 24 Jan 2017 00:51:32 +0100 Subject: netfilter: nf_tables: fix set->nelems counting with no NLM_F_EXCL If the element exists and no NLM_F_EXCL is specified, do not bump set->nelems, otherwise we leak one set element slot. This problem amplifies if the set is full since the abort path always decrements the counter for the -ENFILE case too, giving one spare extra slot. Fix this by moving set->nelems update to nft_add_set_elem() after successful element insertion. Moreover, remove the element if the set is full so there is no need to rely on the abort path to undo things anymore. Fixes: c016c7e45ddf ("netfilter: nf_tables: honor NLM_F_EXCL flag in set element insertion") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index b84c7b25219b..831a9a16f563 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -3745,10 +3745,18 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set, goto err5; } + if (set->size && + !atomic_add_unless(&set->nelems, 1, set->size + set->ndeact)) { + err = -ENFILE; + goto err6; + } + nft_trans_elem(trans) = elem; list_add_tail(&trans->list, &ctx->net->nft.commit_list); return 0; +err6: + set->ops->remove(set, &elem); err5: kfree(trans); err4: @@ -3795,15 +3803,9 @@ static int nf_tables_newsetelem(struct net *net, struct sock *nlsk, return -EBUSY; nla_for_each_nested(attr, nla[NFTA_SET_ELEM_LIST_ELEMENTS], rem) { - if (set->size && - !atomic_add_unless(&set->nelems, 1, set->size + set->ndeact)) - return -ENFILE; - err = nft_add_set_elem(&ctx, set, attr, nlh->nlmsg_flags); - if (err < 0) { - atomic_dec(&set->nelems); + if (err < 0) break; - } } return err; } -- cgit v1.2.3 From de70185de0333783154863278ac87bfbbc54e384 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 24 Jan 2017 00:51:41 +0100 Subject: netfilter: nf_tables: deconstify walk callback function The flush operation needs to modify set and element objects, so let's deconstify this. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 6 +++--- net/netfilter/nf_tables_api.c | 24 ++++++++++++------------ net/netfilter/nft_set_hash.c | 2 +- net/netfilter/nft_set_rbtree.c | 2 +- 4 files changed, 17 insertions(+), 17 deletions(-) diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 924325c46aab..7dfdb517f0be 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -207,9 +207,9 @@ struct nft_set_iter { unsigned int skip; int err; int (*fn)(const struct nft_ctx *ctx, - const struct nft_set *set, + struct nft_set *set, const struct nft_set_iter *iter, - const struct nft_set_elem *elem); + struct nft_set_elem *elem); }; /** @@ -301,7 +301,7 @@ struct nft_set_ops { void (*remove)(const struct nft_set *set, const struct nft_set_elem *elem); void (*walk)(const struct nft_ctx *ctx, - const struct nft_set *set, + struct nft_set *set, struct nft_set_iter *iter); unsigned int (*privsize)(const struct nlattr * const nla[]); diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 831a9a16f563..5bd0068320fb 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -3087,9 +3087,9 @@ static int nf_tables_delset(struct net *net, struct sock *nlsk, } static int nf_tables_bind_check_setelem(const struct nft_ctx *ctx, - const struct nft_set *set, + struct nft_set *set, const struct nft_set_iter *iter, - const struct nft_set_elem *elem) + struct nft_set_elem *elem) { const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv); enum nft_registers dreg; @@ -3308,9 +3308,9 @@ struct nft_set_dump_args { }; static int nf_tables_dump_setelem(const struct nft_ctx *ctx, - const struct nft_set *set, + struct nft_set *set, const struct nft_set_iter *iter, - const struct nft_set_elem *elem) + struct nft_set_elem *elem) { struct nft_set_dump_args *args; @@ -3322,7 +3322,7 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb) { struct net *net = sock_net(skb->sk); u8 genmask = nft_genmask_cur(net); - const struct nft_set *set; + struct nft_set *set; struct nft_set_dump_args args; struct nft_ctx ctx; struct nlattr *nla[NFTA_SET_ELEM_LIST_MAX + 1]; @@ -3890,9 +3890,9 @@ err1: } static int nft_flush_set(const struct nft_ctx *ctx, - const struct nft_set *set, + struct nft_set *set, const struct nft_set_iter *iter, - const struct nft_set_elem *elem) + struct nft_set_elem *elem) { struct nft_trans *trans; int err; @@ -3907,8 +3907,8 @@ static int nft_flush_set(const struct nft_ctx *ctx, goto err1; } - nft_trans_elem_set(trans) = (struct nft_set *)set; - nft_trans_elem(trans) = *((struct nft_set_elem *)elem); + nft_trans_elem_set(trans) = set; + nft_trans_elem(trans) = *elem; list_add_tail(&trans->list, &ctx->net->nft.commit_list); return 0; @@ -5019,9 +5019,9 @@ static int nf_tables_check_loops(const struct nft_ctx *ctx, const struct nft_chain *chain); static int nf_tables_loop_check_setelem(const struct nft_ctx *ctx, - const struct nft_set *set, + struct nft_set *set, const struct nft_set_iter *iter, - const struct nft_set_elem *elem) + struct nft_set_elem *elem) { const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv); const struct nft_data *data; @@ -5045,7 +5045,7 @@ static int nf_tables_check_loops(const struct nft_ctx *ctx, { const struct nft_rule *rule; const struct nft_expr *expr, *last; - const struct nft_set *set; + struct nft_set *set; struct nft_set_binding *binding; struct nft_set_iter iter; diff --git a/net/netfilter/nft_set_hash.c b/net/netfilter/nft_set_hash.c index 1e20e2bbb6d9..e36069fb76ae 100644 --- a/net/netfilter/nft_set_hash.c +++ b/net/netfilter/nft_set_hash.c @@ -212,7 +212,7 @@ static void nft_hash_remove(const struct nft_set *set, rhashtable_remove_fast(&priv->ht, &he->node, nft_hash_params); } -static void nft_hash_walk(const struct nft_ctx *ctx, const struct nft_set *set, +static void nft_hash_walk(const struct nft_ctx *ctx, struct nft_set *set, struct nft_set_iter *iter) { struct nft_hash *priv = nft_set_priv(set); diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c index 08376e50f6cd..f06f55ee516d 100644 --- a/net/netfilter/nft_set_rbtree.c +++ b/net/netfilter/nft_set_rbtree.c @@ -221,7 +221,7 @@ static void *nft_rbtree_deactivate(const struct net *net, } static void nft_rbtree_walk(const struct nft_ctx *ctx, - const struct nft_set *set, + struct nft_set *set, struct nft_set_iter *iter) { const struct nft_rbtree *priv = nft_set_priv(set); -- cgit v1.2.3 From b2c11e4b9536ebab6b39929e1fe15f57039ab445 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 24 Jan 2017 00:51:48 +0100 Subject: netfilter: nf_tables: bump set->ndeact on set flush Add missing set->ndeact update on each deactivated element from the set flush path. Otherwise, sets with fixed size break after flush since accounting breaks. # nft add set x y { type ipv4_addr\; size 2\; } # nft add element x y { 1.1.1.1 } # nft add element x y { 1.1.1.2 } # nft flush set x y # nft add element x y { 1.1.1.1 } :1:1-28: Error: Could not process rule: Too many open files in system Fixes: 8411b6442e59 ("netfilter: nf_tables: support for set flushing") Reported-by: Elise Lennion Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 5bd0068320fb..1b913760f205 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -3906,6 +3906,7 @@ static int nft_flush_set(const struct nft_ctx *ctx, err = -ENOENT; goto err1; } + set->ndeact++; nft_trans_elem_set(trans) = set; nft_trans_elem(trans) = *elem; -- cgit v1.2.3 From 93f955aad4bacee5acebad141d1a03cd51f27b4e Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 24 Jan 2017 13:00:43 +0100 Subject: tipc: fix nametbl_lock soft lockup at node/link events We trigger a soft lockup as we grab nametbl_lock twice if the node has a pending node up/down or link up/down event while: - we process an incoming named message in tipc_named_rcv() and perform an tipc_update_nametbl(). - we have pending backlog items in the name distributor queue during a nametable update using tipc_nametbl_publish() or tipc_nametbl_withdraw(). The following are the call chain associated: tipc_named_rcv() Grabs nametbl_lock tipc_update_nametbl() (publish/withdraw) tipc_node_subscribe()/unsubscribe() tipc_node_write_unlock() << lockup occurs if an outstanding node/link event exits, as we grabs nametbl_lock again >> tipc_nametbl_withdraw() Grab nametbl_lock tipc_named_process_backlog() tipc_update_nametbl() << rest as above >> The function tipc_node_write_unlock(), in addition to releasing the lock processes the outstanding node/link up/down events. To do this, we need to grab the nametbl_lock again leading to the lockup. In this commit we fix the soft lockup by introducing a fast variant of node_unlock(), where we just release the lock. We adapt the node_subscribe()/node_unsubscribe() to use the fast variants. Reported-and-Tested-by: John Thompson Acked-by: Ying Xue Acked-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/node.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/net/tipc/node.c b/net/tipc/node.c index 9d2f4c2b08ab..27753325e06e 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -263,6 +263,11 @@ static void tipc_node_write_lock(struct tipc_node *n) write_lock_bh(&n->lock); } +static void tipc_node_write_unlock_fast(struct tipc_node *n) +{ + write_unlock_bh(&n->lock); +} + static void tipc_node_write_unlock(struct tipc_node *n) { struct net *net = n->net; @@ -417,7 +422,7 @@ void tipc_node_subscribe(struct net *net, struct list_head *subscr, u32 addr) } tipc_node_write_lock(n); list_add_tail(subscr, &n->publ_list); - tipc_node_write_unlock(n); + tipc_node_write_unlock_fast(n); tipc_node_put(n); } @@ -435,7 +440,7 @@ void tipc_node_unsubscribe(struct net *net, struct list_head *subscr, u32 addr) } tipc_node_write_lock(n); list_del_init(subscr); - tipc_node_write_unlock(n); + tipc_node_write_unlock_fast(n); tipc_node_put(n); } -- cgit v1.2.3 From d094c4d5f5c7e1b225e94227ca3f007be3adc4e8 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 24 Jan 2017 13:00:44 +0100 Subject: tipc: add subscription refcount to avoid invalid delete Until now, the subscribers keep track of the subscriptions using reference count at subscriber level. At subscription cancel or subscriber delete, we delete the subscription only if the timer was pending for the subscription. This approach is incorrect as: 1. del_timer() is not SMP safe, if on CPU0 the check for pending timer returns true but CPU1 might schedule the timer callback thereby deleting the subscription. Thus when CPU0 is scheduled, it deletes an invalid subscription. 2. We export tipc_subscrp_report_overlap(), which accesses the subscription pointer multiple times. Meanwhile the subscription timer can expire thereby freeing the subscription and we might continue to access the subscription pointer leading to memory violations. In this commit, we introduce subscription refcount to avoid deleting an invalid subscription. Reported-and-Tested-by: John Thompson Acked-by: Ying Xue Acked-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/subscr.c | 124 ++++++++++++++++++++++++++++++------------------------ net/tipc/subscr.h | 1 + 2 files changed, 71 insertions(+), 54 deletions(-) diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index 0dd02244e21d..9d94e65d0894 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -54,6 +54,8 @@ struct tipc_subscriber { static void tipc_subscrp_delete(struct tipc_subscription *sub); static void tipc_subscrb_put(struct tipc_subscriber *subscriber); +static void tipc_subscrp_put(struct tipc_subscription *subscription); +static void tipc_subscrp_get(struct tipc_subscription *subscription); /** * htohl - convert value to endianness used by destination @@ -123,6 +125,7 @@ void tipc_subscrp_report_overlap(struct tipc_subscription *sub, u32 found_lower, { struct tipc_name_seq seq; + tipc_subscrp_get(sub); tipc_subscrp_convert_seq(&sub->evt.s.seq, sub->swap, &seq); if (!tipc_subscrp_check_overlap(&seq, found_lower, found_upper)) return; @@ -132,30 +135,23 @@ void tipc_subscrp_report_overlap(struct tipc_subscription *sub, u32 found_lower, tipc_subscrp_send_event(sub, found_lower, found_upper, event, port_ref, node); + tipc_subscrp_put(sub); } static void tipc_subscrp_timeout(unsigned long data) { struct tipc_subscription *sub = (struct tipc_subscription *)data; - struct tipc_subscriber *subscriber = sub->subscriber; /* Notify subscriber of timeout */ tipc_subscrp_send_event(sub, sub->evt.s.seq.lower, sub->evt.s.seq.upper, TIPC_SUBSCR_TIMEOUT, 0, 0); - spin_lock_bh(&subscriber->lock); - tipc_subscrp_delete(sub); - spin_unlock_bh(&subscriber->lock); - - tipc_subscrb_put(subscriber); + tipc_subscrp_put(sub); } static void tipc_subscrb_kref_release(struct kref *kref) { - struct tipc_subscriber *subcriber = container_of(kref, - struct tipc_subscriber, kref); - - kfree(subcriber); + kfree(container_of(kref,struct tipc_subscriber, kref)); } static void tipc_subscrb_put(struct tipc_subscriber *subscriber) @@ -168,6 +164,59 @@ static void tipc_subscrb_get(struct tipc_subscriber *subscriber) kref_get(&subscriber->kref); } +static void tipc_subscrp_kref_release(struct kref *kref) +{ + struct tipc_subscription *sub = container_of(kref, + struct tipc_subscription, + kref); + struct tipc_net *tn = net_generic(sub->net, tipc_net_id); + struct tipc_subscriber *subscriber = sub->subscriber; + + spin_lock_bh(&subscriber->lock); + tipc_nametbl_unsubscribe(sub); + list_del(&sub->subscrp_list); + atomic_dec(&tn->subscription_count); + spin_unlock_bh(&subscriber->lock); + kfree(sub); + tipc_subscrb_put(subscriber); +} + +static void tipc_subscrp_put(struct tipc_subscription *subscription) +{ + kref_put(&subscription->kref, tipc_subscrp_kref_release); +} + +static void tipc_subscrp_get(struct tipc_subscription *subscription) +{ + kref_get(&subscription->kref); +} + +/* tipc_subscrb_subscrp_delete - delete a specific subscription or all + * subscriptions for a given subscriber. + */ +static void tipc_subscrb_subscrp_delete(struct tipc_subscriber *subscriber, + struct tipc_subscr *s) +{ + struct list_head *subscription_list = &subscriber->subscrp_list; + struct tipc_subscription *sub, *temp; + + spin_lock_bh(&subscriber->lock); + list_for_each_entry_safe(sub, temp, subscription_list, subscrp_list) { + if (s && memcmp(s, &sub->evt.s, sizeof(struct tipc_subscr))) + continue; + + tipc_subscrp_get(sub); + spin_unlock_bh(&subscriber->lock); + tipc_subscrp_delete(sub); + tipc_subscrp_put(sub); + spin_lock_bh(&subscriber->lock); + + if (s) + break; + } + spin_unlock_bh(&subscriber->lock); +} + static struct tipc_subscriber *tipc_subscrb_create(int conid) { struct tipc_subscriber *subscriber; @@ -177,8 +226,8 @@ static struct tipc_subscriber *tipc_subscrb_create(int conid) pr_warn("Subscriber rejected, no memory\n"); return NULL; } - kref_init(&subscriber->kref); INIT_LIST_HEAD(&subscriber->subscrp_list); + kref_init(&subscriber->kref); subscriber->conid = conid; spin_lock_init(&subscriber->lock); @@ -187,55 +236,22 @@ static struct tipc_subscriber *tipc_subscrb_create(int conid) static void tipc_subscrb_delete(struct tipc_subscriber *subscriber) { - struct tipc_subscription *sub, *temp; - u32 timeout; - - spin_lock_bh(&subscriber->lock); - /* Destroy any existing subscriptions for subscriber */ - list_for_each_entry_safe(sub, temp, &subscriber->subscrp_list, - subscrp_list) { - timeout = htohl(sub->evt.s.timeout, sub->swap); - if ((timeout == TIPC_WAIT_FOREVER) || del_timer(&sub->timer)) { - tipc_subscrp_delete(sub); - tipc_subscrb_put(subscriber); - } - } - spin_unlock_bh(&subscriber->lock); - + tipc_subscrb_subscrp_delete(subscriber, NULL); tipc_subscrb_put(subscriber); } static void tipc_subscrp_delete(struct tipc_subscription *sub) { - struct tipc_net *tn = net_generic(sub->net, tipc_net_id); + u32 timeout = htohl(sub->evt.s.timeout, sub->swap); - tipc_nametbl_unsubscribe(sub); - list_del(&sub->subscrp_list); - kfree(sub); - atomic_dec(&tn->subscription_count); + if (timeout == TIPC_WAIT_FOREVER || del_timer(&sub->timer)) + tipc_subscrp_put(sub); } static void tipc_subscrp_cancel(struct tipc_subscr *s, struct tipc_subscriber *subscriber) { - struct tipc_subscription *sub, *temp; - u32 timeout; - - spin_lock_bh(&subscriber->lock); - /* Find first matching subscription, exit if not found */ - list_for_each_entry_safe(sub, temp, &subscriber->subscrp_list, - subscrp_list) { - if (!memcmp(s, &sub->evt.s, sizeof(struct tipc_subscr))) { - timeout = htohl(sub->evt.s.timeout, sub->swap); - if ((timeout == TIPC_WAIT_FOREVER) || - del_timer(&sub->timer)) { - tipc_subscrp_delete(sub); - tipc_subscrb_put(subscriber); - } - break; - } - } - spin_unlock_bh(&subscriber->lock); + tipc_subscrb_subscrp_delete(subscriber, s); } static struct tipc_subscription *tipc_subscrp_create(struct net *net, @@ -272,6 +288,7 @@ static struct tipc_subscription *tipc_subscrp_create(struct net *net, sub->swap = swap; memcpy(&sub->evt.s, s, sizeof(*s)); atomic_inc(&tn->subscription_count); + kref_init(&sub->kref); return sub; } @@ -288,17 +305,16 @@ static void tipc_subscrp_subscribe(struct net *net, struct tipc_subscr *s, spin_lock_bh(&subscriber->lock); list_add(&sub->subscrp_list, &subscriber->subscrp_list); - tipc_subscrb_get(subscriber); sub->subscriber = subscriber; tipc_nametbl_subscribe(sub); + tipc_subscrb_get(subscriber); spin_unlock_bh(&subscriber->lock); + setup_timer(&sub->timer, tipc_subscrp_timeout, (unsigned long)sub); timeout = htohl(sub->evt.s.timeout, swap); - if (timeout == TIPC_WAIT_FOREVER) - return; - setup_timer(&sub->timer, tipc_subscrp_timeout, (unsigned long)sub); - mod_timer(&sub->timer, jiffies + msecs_to_jiffies(timeout)); + if (timeout != TIPC_WAIT_FOREVER) + mod_timer(&sub->timer, jiffies + msecs_to_jiffies(timeout)); } /* Handle one termination request for the subscriber */ diff --git a/net/tipc/subscr.h b/net/tipc/subscr.h index be60103082c9..ffdc214c117a 100644 --- a/net/tipc/subscr.h +++ b/net/tipc/subscr.h @@ -57,6 +57,7 @@ struct tipc_subscriber; * @evt: template for events generated by subscription */ struct tipc_subscription { + struct kref kref; struct tipc_subscriber *subscriber; struct net *net; struct timer_list timer; -- cgit v1.2.3 From fc0adfc8fd18b61b6f7a3f28b429e134d6f3a008 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 24 Jan 2017 13:00:45 +0100 Subject: tipc: fix connection refcount error Until now, the generic server framework maintains the connection id's per subscriber in server's conn_idr. At tipc_close_conn, we remove the connection id from the server list, but the connection is valid until we call the refcount cleanup. Hence we have a window where the server allocates the same connection to an new subscriber leading to inconsistent reference count. We have another refcount warning we grab the refcount in tipc_conn_lookup() for connections with flag with CF_CONNECTED not set. This usually occurs at shutdown when the we stop the topology server and withdraw TIPC_CFG_SRV publication thereby triggering a withdraw message to subscribers. In this commit, we: 1. remove the connection from the server list at recount cleanup. 2. grab the refcount for a connection only if CF_CONNECTED is set. Tested-by: John Thompson Acked-by: Ying Xue Acked-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/server.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/net/tipc/server.c b/net/tipc/server.c index 215849ce453d..2e803601aa99 100644 --- a/net/tipc/server.c +++ b/net/tipc/server.c @@ -91,7 +91,8 @@ static void tipc_sock_release(struct tipc_conn *con); static void tipc_conn_kref_release(struct kref *kref) { struct tipc_conn *con = container_of(kref, struct tipc_conn, kref); - struct sockaddr_tipc *saddr = con->server->saddr; + struct tipc_server *s = con->server; + struct sockaddr_tipc *saddr = s->saddr; struct socket *sock = con->sock; struct sock *sk; @@ -106,6 +107,11 @@ static void tipc_conn_kref_release(struct kref *kref) tipc_sock_release(con); sock_release(sock); con->sock = NULL; + + spin_lock_bh(&s->idr_lock); + idr_remove(&s->conn_idr, con->conid); + s->idr_in_use--; + spin_unlock_bh(&s->idr_lock); } tipc_clean_outqueues(con); @@ -128,8 +134,10 @@ static struct tipc_conn *tipc_conn_lookup(struct tipc_server *s, int conid) spin_lock_bh(&s->idr_lock); con = idr_find(&s->conn_idr, conid); - if (con) + if (con && test_bit(CF_CONNECTED, &con->flags)) conn_get(con); + else + con = NULL; spin_unlock_bh(&s->idr_lock); return con; } @@ -198,15 +206,8 @@ static void tipc_sock_release(struct tipc_conn *con) static void tipc_close_conn(struct tipc_conn *con) { - struct tipc_server *s = con->server; - if (test_and_clear_bit(CF_CONNECTED, &con->flags)) { - spin_lock_bh(&s->idr_lock); - idr_remove(&s->conn_idr, con->conid); - s->idr_in_use--; - spin_unlock_bh(&s->idr_lock); - /* We shouldn't flush pending works as we may be in the * thread. In fact the races with pending rx/tx work structs * are harmless for us here as we have already deleted this -- cgit v1.2.3 From 9dc3abdd1f7ea524e8552e0a3ef01219892ed1f4 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 24 Jan 2017 13:00:46 +0100 Subject: tipc: fix nametbl_lock soft lockup at module exit Commit 333f796235a527 ("tipc: fix a race condition leading to subscriber refcnt bug") reveals a soft lockup while acquiring nametbl_lock. Before commit 333f796235a527, we call tipc_conn_shutdown() from tipc_close_conn() in the context of tipc_topsrv_stop(). In that context, we are allowed to grab the nametbl_lock. Commit 333f796235a527, moved tipc_conn_release (renamed from tipc_conn_shutdown) to the connection refcount cleanup. This allows either tipc_nametbl_withdraw() or tipc_topsrv_stop() to the cleanup. Since tipc_exit_net() first calls tipc_topsrv_stop() and then tipc_nametble_withdraw() increases the chances for the later to perform the connection cleanup. The soft lockup occurs in the call chain of tipc_nametbl_withdraw(), when it performs the tipc_conn_kref_release() as it tries to grab nametbl_lock again while holding it already. tipc_nametbl_withdraw() grabs nametbl_lock tipc_nametbl_remove_publ() tipc_subscrp_report_overlap() tipc_subscrp_send_event() tipc_conn_sendmsg() << if (con->flags != CF_CONNECTED) we do conn_put(), triggering the cleanup as refcount=0. >> tipc_conn_kref_release tipc_sock_release tipc_conn_release tipc_subscrb_delete tipc_subscrp_delete tipc_nametbl_unsubscribe << Soft Lockup >> The previous changes in this series fixes the race conditions fixed by commit 333f796235a527. Hence we can now revert the commit. Fixes: 333f796235a52727 ("tipc: fix a race condition leading to subscriber refcnt bug") Reported-and-Tested-by: John Thompson Acked-by: Ying Xue Acked-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/server.c | 16 +++++----------- 1 file changed, 5 insertions(+), 11 deletions(-) diff --git a/net/tipc/server.c b/net/tipc/server.c index 2e803601aa99..826cde2c401e 100644 --- a/net/tipc/server.c +++ b/net/tipc/server.c @@ -86,7 +86,6 @@ struct outqueue_entry { static void tipc_recv_work(struct work_struct *work); static void tipc_send_work(struct work_struct *work); static void tipc_clean_outqueues(struct tipc_conn *con); -static void tipc_sock_release(struct tipc_conn *con); static void tipc_conn_kref_release(struct kref *kref) { @@ -104,7 +103,6 @@ static void tipc_conn_kref_release(struct kref *kref) } saddr->scope = -TIPC_NODE_SCOPE; kernel_bind(sock, (struct sockaddr *)saddr, sizeof(*saddr)); - tipc_sock_release(con); sock_release(sock); con->sock = NULL; @@ -194,19 +192,15 @@ static void tipc_unregister_callbacks(struct tipc_conn *con) write_unlock_bh(&sk->sk_callback_lock); } -static void tipc_sock_release(struct tipc_conn *con) +static void tipc_close_conn(struct tipc_conn *con) { struct tipc_server *s = con->server; - if (con->conid) - s->tipc_conn_release(con->conid, con->usr_data); - - tipc_unregister_callbacks(con); -} - -static void tipc_close_conn(struct tipc_conn *con) -{ if (test_and_clear_bit(CF_CONNECTED, &con->flags)) { + tipc_unregister_callbacks(con); + + if (con->conid) + s->tipc_conn_release(con->conid, con->usr_data); /* We shouldn't flush pending works as we may be in the * thread. In fact the races with pending rx/tx work structs -- cgit v1.2.3 From 4c887aa65d38633885010277f3482400681be719 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 24 Jan 2017 13:00:47 +0100 Subject: tipc: ignore requests when the connection state is not CONNECTED In tipc_conn_sendmsg(), we first queue the request to the outqueue followed by the connection state check. If the connection is not connected, we should not queue this message. In this commit, we reject the messages if the connection state is not CF_CONNECTED. Acked-by: Ying Xue Acked-by: Jon Maloy Tested-by: John Thompson Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/server.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/net/tipc/server.c b/net/tipc/server.c index 826cde2c401e..04ff441b8065 100644 --- a/net/tipc/server.c +++ b/net/tipc/server.c @@ -453,6 +453,11 @@ int tipc_conn_sendmsg(struct tipc_server *s, int conid, if (!con) return -EINVAL; + if (!test_bit(CF_CONNECTED, &con->flags)) { + conn_put(con); + return 0; + } + e = tipc_alloc_entry(data, len); if (!e) { conn_put(con); @@ -466,12 +471,8 @@ int tipc_conn_sendmsg(struct tipc_server *s, int conid, list_add_tail(&e->list, &con->outqueue); spin_unlock_bh(&con->outqueue_lock); - if (test_bit(CF_CONNECTED, &con->flags)) { - if (!queue_work(s->send_wq, &con->swork)) - conn_put(con); - } else { + if (!queue_work(s->send_wq, &con->swork)) conn_put(con); - } return 0; } @@ -495,7 +496,7 @@ static void tipc_send_to_sock(struct tipc_conn *con) int ret; spin_lock_bh(&con->outqueue_lock); - while (1) { + while (test_bit(CF_CONNECTED, &con->flags)) { e = list_entry(con->outqueue.next, struct outqueue_entry, list); if ((struct list_head *) e == &con->outqueue) -- cgit v1.2.3 From 35e22e49a5d6a741ebe7f2dd280b2052c3003ef7 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 24 Jan 2017 13:00:48 +0100 Subject: tipc: fix cleanup at module unload In tipc_server_stop(), we iterate over the connections with limiting factor as server's idr_in_use. We ignore the fact that this variable is decremented in tipc_close_conn(), leading to premature exit. In this commit, we iterate until the we have no connections left. Acked-by: Ying Xue Acked-by: Jon Maloy Tested-by: John Thompson Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/server.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/net/tipc/server.c b/net/tipc/server.c index 04ff441b8065..3cd6402e812c 100644 --- a/net/tipc/server.c +++ b/net/tipc/server.c @@ -619,14 +619,12 @@ int tipc_server_start(struct tipc_server *s) void tipc_server_stop(struct tipc_server *s) { struct tipc_conn *con; - int total = 0; int id; spin_lock_bh(&s->idr_lock); - for (id = 0; total < s->idr_in_use; id++) { + for (id = 0; s->idr_in_use; id++) { con = idr_find(&s->conn_idr, id); if (con) { - total++; spin_unlock_bh(&s->idr_lock); tipc_close_conn(con); spin_lock_bh(&s->idr_lock); -- cgit v1.2.3 From f39f775218a7520e3700de2003c84a042c3b5972 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Thu, 19 Jan 2017 15:25:58 +0200 Subject: IB/rxe: Fix rxe dev insertion to rxe_dev_list The first argument of list_add_tail is the new item and the second is the head of the list. Fix the code to pass arguments in the right order, otherwise not all the rxe devices will be removed during teardown. Fixes: 8700e3e7c4857 ('Soft RoCE driver') Signed-off-by: Maor Gottlieb Reviewed-by: Moni Shoua Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/sw/rxe/rxe_net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c index 342e78163613..4abdeb359fb4 100644 --- a/drivers/infiniband/sw/rxe/rxe_net.c +++ b/drivers/infiniband/sw/rxe/rxe_net.c @@ -555,7 +555,7 @@ struct rxe_dev *rxe_net_add(struct net_device *ndev) } spin_lock_bh(&dev_list_lock); - list_add_tail(&rxe_dev_list, &rxe->list); + list_add_tail(&rxe->list, &rxe_dev_list); spin_unlock_bh(&dev_list_lock); return rxe; } -- cgit v1.2.3 From 2d4b21e0a2913612274a69a3ba1bfee4cffc6e77 Mon Sep 17 00:00:00 2001 From: Yonatan Cohen Date: Thu, 19 Jan 2017 15:25:59 +0200 Subject: IB/rxe: Prevent from completer to operate on non valid QP On UD QP completer tasklet is scheduled for each packet sent. If it is followed by a destroy_qp(), the kernel panic will happen as the completer tries to operate on a destroyed QP. Fixes: 8700e3e7c485 ("Soft RoCE driver") Signed-off-by: Yonatan Cohen Reviewed-by: Moni Shoua Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/sw/rxe/rxe_qp.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/infiniband/sw/rxe/rxe_qp.c b/drivers/infiniband/sw/rxe/rxe_qp.c index 486d576e55bc..44b2108253bd 100644 --- a/drivers/infiniband/sw/rxe/rxe_qp.c +++ b/drivers/infiniband/sw/rxe/rxe_qp.c @@ -813,8 +813,7 @@ void rxe_qp_destroy(struct rxe_qp *qp) del_timer_sync(&qp->rnr_nak_timer); rxe_cleanup_task(&qp->req.task); - if (qp_type(qp) == IB_QPT_RC) - rxe_cleanup_task(&qp->comp.task); + rxe_cleanup_task(&qp->comp.task); /* flush out any receive wr's or pending requests */ __rxe_do_task(&qp->req.task); -- cgit v1.2.3 From 88ff7334f25909802140e690c0e16433e485b0a0 Mon Sep 17 00:00:00 2001 From: Robert Shearman Date: Tue, 24 Jan 2017 16:26:47 +0000 Subject: net: Specify the owning module for lwtunnel ops Modules implementing lwtunnel ops should not be allowed to unload while there is state alive using those ops, so specify the owning module for all lwtunnel ops. Signed-off-by: Robert Shearman Signed-off-by: David S. Miller --- include/net/lwtunnel.h | 2 ++ net/core/lwt_bpf.c | 1 + net/ipv4/ip_tunnel_core.c | 2 ++ net/ipv6/ila/ila_lwt.c | 1 + net/ipv6/seg6_iptunnel.c | 1 + net/mpls/mpls_iptunnel.c | 1 + 6 files changed, 8 insertions(+) diff --git a/include/net/lwtunnel.h b/include/net/lwtunnel.h index 0b585f1fd340..73dd87647460 100644 --- a/include/net/lwtunnel.h +++ b/include/net/lwtunnel.h @@ -44,6 +44,8 @@ struct lwtunnel_encap_ops { int (*get_encap_size)(struct lwtunnel_state *lwtstate); int (*cmp_encap)(struct lwtunnel_state *a, struct lwtunnel_state *b); int (*xmit)(struct sk_buff *skb); + + struct module *owner; }; #ifdef CONFIG_LWTUNNEL diff --git a/net/core/lwt_bpf.c b/net/core/lwt_bpf.c index 71bb3e2eca08..b3eef90b2df9 100644 --- a/net/core/lwt_bpf.c +++ b/net/core/lwt_bpf.c @@ -386,6 +386,7 @@ static const struct lwtunnel_encap_ops bpf_encap_ops = { .fill_encap = bpf_fill_encap_info, .get_encap_size = bpf_encap_nlsize, .cmp_encap = bpf_encap_cmp, + .owner = THIS_MODULE, }; static int __init bpf_lwt_init(void) diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index fed3d29f9eb3..0fd1976ab63b 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -313,6 +313,7 @@ static const struct lwtunnel_encap_ops ip_tun_lwt_ops = { .fill_encap = ip_tun_fill_encap_info, .get_encap_size = ip_tun_encap_nlsize, .cmp_encap = ip_tun_cmp_encap, + .owner = THIS_MODULE, }; static const struct nla_policy ip6_tun_policy[LWTUNNEL_IP6_MAX + 1] = { @@ -403,6 +404,7 @@ static const struct lwtunnel_encap_ops ip6_tun_lwt_ops = { .fill_encap = ip6_tun_fill_encap_info, .get_encap_size = ip6_tun_encap_nlsize, .cmp_encap = ip_tun_cmp_encap, + .owner = THIS_MODULE, }; void __init ip_tunnel_core_init(void) diff --git a/net/ipv6/ila/ila_lwt.c b/net/ipv6/ila/ila_lwt.c index a7bc54ab46e2..13b5e85fe0d5 100644 --- a/net/ipv6/ila/ila_lwt.c +++ b/net/ipv6/ila/ila_lwt.c @@ -238,6 +238,7 @@ static const struct lwtunnel_encap_ops ila_encap_ops = { .fill_encap = ila_fill_encap_info, .get_encap_size = ila_encap_nlsize, .cmp_encap = ila_encap_cmp, + .owner = THIS_MODULE, }; int ila_lwt_init(void) diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c index 1d60cb132835..c46f8cbf5ab5 100644 --- a/net/ipv6/seg6_iptunnel.c +++ b/net/ipv6/seg6_iptunnel.c @@ -422,6 +422,7 @@ static const struct lwtunnel_encap_ops seg6_iptun_ops = { .fill_encap = seg6_fill_encap_info, .get_encap_size = seg6_encap_nlsize, .cmp_encap = seg6_encap_cmp, + .owner = THIS_MODULE, }; int __init seg6_iptunnel_init(void) diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c index 2f7ccd934416..1d281c1ff7c1 100644 --- a/net/mpls/mpls_iptunnel.c +++ b/net/mpls/mpls_iptunnel.c @@ -215,6 +215,7 @@ static const struct lwtunnel_encap_ops mpls_iptun_ops = { .fill_encap = mpls_fill_encap_info, .get_encap_size = mpls_encap_nlsize, .cmp_encap = mpls_encap_cmp, + .owner = THIS_MODULE, }; static int __init mpls_iptunnel_init(void) -- cgit v1.2.3 From 85c814016ce3b371016c2c054a905fa2492f5a65 Mon Sep 17 00:00:00 2001 From: Robert Shearman Date: Tue, 24 Jan 2017 16:26:48 +0000 Subject: lwtunnel: Fix oops on state free after encap module unload When attempting to free lwtunnel state after the module for the encap has been unloaded an oops occurs: BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 IP: lwtstate_free+0x18/0x40 [..] task: ffff88003e372380 task.stack: ffffc900001fc000 RIP: 0010:lwtstate_free+0x18/0x40 RSP: 0018:ffff88003fd83e88 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff88002bbb3380 RCX: ffff88000c91a300 [..] Call Trace: free_fib_info_rcu+0x195/0x1a0 ? rt_fibinfo_free+0x50/0x50 rcu_process_callbacks+0x2d3/0x850 ? rcu_process_callbacks+0x296/0x850 __do_softirq+0xe4/0x4cb irq_exit+0xb0/0xc0 smp_apic_timer_interrupt+0x3d/0x50 apic_timer_interrupt+0x93/0xa0 [..] Code: e8 6e c6 fc ff 89 d8 5b 5d c3 bb de ff ff ff eb f4 66 90 66 66 66 66 90 55 48 89 e5 53 0f b7 07 48 89 fb 48 8b 04 c5 00 81 d5 81 <48> 8b 40 08 48 85 c0 74 13 ff d0 48 8d 7b 20 be 20 00 00 00 e8 The problem is after the module for the encap can be unloaded the corresponding ops is removed and is thus NULL here. Modules implementing lwtunnel ops should not be allowed to unload while there is state alive using those ops, so grab the module reference for the ops on creating lwtunnel state and of course release the reference when freeing the state. Fixes: 1104d9ba443a ("lwtunnel: Add destroy state operation") Signed-off-by: Robert Shearman Signed-off-by: David S. Miller --- net/core/lwtunnel.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/core/lwtunnel.c b/net/core/lwtunnel.c index 47b1dd65947b..c23465005f2f 100644 --- a/net/core/lwtunnel.c +++ b/net/core/lwtunnel.c @@ -115,8 +115,11 @@ int lwtunnel_build_state(struct net_device *dev, u16 encap_type, ret = -EOPNOTSUPP; rcu_read_lock(); ops = rcu_dereference(lwtun_encaps[encap_type]); - if (likely(ops && ops->build_state)) + if (likely(ops && ops->build_state && try_module_get(ops->owner))) { ret = ops->build_state(dev, encap, family, cfg, lws); + if (ret) + module_put(ops->owner); + } rcu_read_unlock(); return ret; @@ -194,6 +197,7 @@ void lwtstate_free(struct lwtunnel_state *lws) } else { kfree(lws); } + module_put(ops->owner); } EXPORT_SYMBOL(lwtstate_free); -- cgit v1.2.3 From 78f824d4312a8944f5340c6b161bba3bf2c81096 Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Tue, 21 Jun 2016 14:24:33 +0530 Subject: ARCv2: smp-boot: wake_flag polling by non-Masters needs to be uncached This is needed on HS38 cores, for setting up IO-Coherency aperture properly The polling could perturb the caches and coherecy fabric which could be wrong in the small window when Master is setting up IOC aperture etc in arc_cache_init() We do it only for ARCv2 based builds to not affect EZChip ARCompact based platform. Signed-off-by: Vineet Gupta --- arch/arc/kernel/smp.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) diff --git a/arch/arc/kernel/smp.c b/arch/arc/kernel/smp.c index 44a0d21ed342..2afbafadb6ab 100644 --- a/arch/arc/kernel/smp.c +++ b/arch/arc/kernel/smp.c @@ -90,10 +90,23 @@ void __init smp_cpus_done(unsigned int max_cpus) */ static volatile int wake_flag; +#ifdef CONFIG_ISA_ARCOMPACT + +#define __boot_read(f) f +#define __boot_write(f, v) f = v + +#else + +#define __boot_read(f) arc_read_uncached_32(&f) +#define __boot_write(f, v) arc_write_uncached_32(&f, v) + +#endif + static void arc_default_smp_cpu_kick(int cpu, unsigned long pc) { BUG_ON(cpu == 0); - wake_flag = cpu; + + __boot_write(wake_flag, cpu); } void arc_platform_smp_wait_to_boot(int cpu) @@ -102,10 +115,10 @@ void arc_platform_smp_wait_to_boot(int cpu) if (IS_ENABLED(CONFIG_ARC_SMP_HALT_ON_RESET)) return; - while (wake_flag != cpu) + while (__boot_read(wake_flag) != cpu) ; - wake_flag = 0; + __boot_write(wake_flag, 0); } const char *arc_platform_smp_cpuinfo(void) -- cgit v1.2.3 From 0516ffd88fa0d006ee80389ce14a9ca5ae45e845 Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Thu, 19 Jan 2017 10:43:53 +0000 Subject: vhost/vsock: handle vhost_vq_init_access() error Propagate the error when vhost_vq_init_access() fails and set vq->private_data to NULL. Signed-off-by: Stefan Hajnoczi Signed-off-by: Michael S. Tsirkin --- drivers/vhost/vsock.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c index bbbf588540ed..ce5e63d2c66a 100644 --- a/drivers/vhost/vsock.c +++ b/drivers/vhost/vsock.c @@ -373,6 +373,7 @@ static void vhost_vsock_handle_rx_kick(struct vhost_work *work) static int vhost_vsock_start(struct vhost_vsock *vsock) { + struct vhost_virtqueue *vq; size_t i; int ret; @@ -383,19 +384,20 @@ static int vhost_vsock_start(struct vhost_vsock *vsock) goto err; for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) { - struct vhost_virtqueue *vq = &vsock->vqs[i]; + vq = &vsock->vqs[i]; mutex_lock(&vq->mutex); if (!vhost_vq_access_ok(vq)) { ret = -EFAULT; - mutex_unlock(&vq->mutex); goto err_vq; } if (!vq->private_data) { vq->private_data = vsock; - vhost_vq_init_access(vq); + ret = vhost_vq_init_access(vq); + if (ret) + goto err_vq; } mutex_unlock(&vq->mutex); @@ -405,8 +407,11 @@ static int vhost_vsock_start(struct vhost_vsock *vsock) return 0; err_vq: + vq->private_data = NULL; + mutex_unlock(&vq->mutex); + for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) { - struct vhost_virtqueue *vq = &vsock->vqs[i]; + vq = &vsock->vqs[i]; mutex_lock(&vq->mutex); vq->private_data = NULL; -- cgit v1.2.3 From f7f6634d23830ff74335734fbdb28ea109c1f349 Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Tue, 10 Jan 2017 17:51:17 +0000 Subject: virtio_mmio: Set DMA masks appropriately Once DMA API usage is enabled, it becomes apparent that virtio-mmio is inadvertently relying on the default 32-bit DMA mask, which leads to problems like rapidly exhausting SWIOTLB bounce buffers. Ensure that we set the appropriate 64-bit DMA mask whenever possible, with the coherent mask suitably limited for the legacy vring as per a0be1db4304f ("virtio_pci: Limit DMA mask to 44 bits for legacy virtio devices"). Cc: Andy Lutomirski Cc: Michael S. Tsirkin Reported-by: Jean-Philippe Brucker Fixes: b42111382f0e ("virtio_mmio: Use the DMA API if enabled") Signed-off-by: Robin Murphy Signed-off-by: Michael S. Tsirkin --- drivers/virtio/virtio_mmio.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/drivers/virtio/virtio_mmio.c b/drivers/virtio/virtio_mmio.c index d47a2fcef818..c71fde5fe835 100644 --- a/drivers/virtio/virtio_mmio.c +++ b/drivers/virtio/virtio_mmio.c @@ -59,6 +59,7 @@ #define pr_fmt(fmt) "virtio-mmio: " fmt #include +#include #include #include #include @@ -498,6 +499,7 @@ static int virtio_mmio_probe(struct platform_device *pdev) struct virtio_mmio_device *vm_dev; struct resource *mem; unsigned long magic; + int rc; mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (!mem) @@ -547,9 +549,25 @@ static int virtio_mmio_probe(struct platform_device *pdev) } vm_dev->vdev.id.vendor = readl(vm_dev->base + VIRTIO_MMIO_VENDOR_ID); - if (vm_dev->version == 1) + if (vm_dev->version == 1) { writel(PAGE_SIZE, vm_dev->base + VIRTIO_MMIO_GUEST_PAGE_SIZE); + rc = dma_set_mask(&pdev->dev, DMA_BIT_MASK(64)); + /* + * In the legacy case, ensure our coherently-allocated virtio + * ring will be at an address expressable as a 32-bit PFN. + */ + if (!rc) + dma_set_coherent_mask(&pdev->dev, + DMA_BIT_MASK(32 + PAGE_SHIFT)); + } else { + rc = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)); + } + if (rc) + rc = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32)); + if (rc) + dev_warn(&pdev->dev, "Failed to enable 64-bit or 32-bit DMA. Trying to continue, but this might not work.\n"); + platform_set_drvdata(pdev, vm_dev); return register_virtio_device(&vm_dev->vdev); -- cgit v1.2.3 From c7070619f3408d9a0dffbed9149e6f00479cf43b Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Fri, 20 Jan 2017 10:33:32 +0000 Subject: vring: Force use of DMA API for ARM-based systems with legacy devices Booting Linux on an ARM fastmodel containing an SMMU emulation results in an unexpected I/O page fault from the legacy virtio-blk PCI device: [ 1.211721] arm-smmu-v3 2b400000.smmu: event 0x10 received: [ 1.211800] arm-smmu-v3 2b400000.smmu: 0x00000000fffff010 [ 1.211880] arm-smmu-v3 2b400000.smmu: 0x0000020800000000 [ 1.211959] arm-smmu-v3 2b400000.smmu: 0x00000008fa081002 [ 1.212075] arm-smmu-v3 2b400000.smmu: 0x0000000000000000 [ 1.212155] arm-smmu-v3 2b400000.smmu: event 0x10 received: [ 1.212234] arm-smmu-v3 2b400000.smmu: 0x00000000fffff010 [ 1.212314] arm-smmu-v3 2b400000.smmu: 0x0000020800000000 [ 1.212394] arm-smmu-v3 2b400000.smmu: 0x00000008fa081000 [ 1.212471] arm-smmu-v3 2b400000.smmu: 0x0000000000000000 This is because the legacy virtio-blk device is behind an SMMU, so we have consequently swizzled its DMA ops and configured the SMMU to translate accesses. This then requires the vring code to use the DMA API to establish translations, otherwise all transactions will result in fatal faults and termination. Given that ARM-based systems only see an SMMU if one is really present (the topology is all described by firmware tables such as device-tree or IORT), then we can safely use the DMA API for all legacy virtio devices. Modern devices can advertise the prescense of an IOMMU using the VIRTIO_F_IOMMU_PLATFORM feature flag. Cc: Andy Lutomirski Cc: Michael S. Tsirkin Cc: Fixes: 876945dbf649 ("arm64: Hook up IOMMU dma_ops") Signed-off-by: Will Deacon Signed-off-by: Michael S. Tsirkin Acked-by: Marc Zyngier --- drivers/virtio/virtio_ring.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 409aeaa49246..7e38ed79c3fc 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -159,6 +159,13 @@ static bool vring_use_dma_api(struct virtio_device *vdev) if (xen_domain()) return true; + /* + * On ARM-based machines, the DMA ops will do the right thing, + * so always use them with legacy devices. + */ + if (IS_ENABLED(CONFIG_ARM) || IS_ENABLED(CONFIG_ARM64)) + return !virtio_has_feature(vdev, VIRTIO_F_VERSION_1); + return false; } -- cgit v1.2.3 From 3feb479cea37fc623cf4e705631b2e679cbfbd7a Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Mon, 23 Jan 2017 13:13:58 -0200 Subject: Revert "thermal: thermal_hwmon: Convert to hwmon_device_register_with_info()" This reverts commit 7611fb68062f ("thermal: thermal_hwmon: Convert to hwmon_device_register_with_info()"). Pavel Machek reported breakage in the Nokia N900 due to this commit. We can revisit a proper fix for the warning later. Reported-by: Pavel Machek Signed-off-by: Fabio Estevam Acked-by: Guenter Roeck Acked-by: Pavel Machek Signed-off-by: Zhang Rui --- drivers/thermal/thermal_hwmon.c | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) diff --git a/drivers/thermal/thermal_hwmon.c b/drivers/thermal/thermal_hwmon.c index c4a508a124dc..541af5946203 100644 --- a/drivers/thermal/thermal_hwmon.c +++ b/drivers/thermal/thermal_hwmon.c @@ -58,6 +58,14 @@ static LIST_HEAD(thermal_hwmon_list); static DEFINE_MUTEX(thermal_hwmon_list_lock); +static ssize_t +name_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct thermal_hwmon_device *hwmon = dev_get_drvdata(dev); + return sprintf(buf, "%s\n", hwmon->type); +} +static DEVICE_ATTR_RO(name); + static ssize_t temp_input_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -157,12 +165,15 @@ int thermal_add_hwmon_sysfs(struct thermal_zone_device *tz) INIT_LIST_HEAD(&hwmon->tz_list); strlcpy(hwmon->type, tz->type, THERMAL_NAME_LENGTH); - hwmon->device = hwmon_device_register_with_info(NULL, hwmon->type, - hwmon, NULL, NULL); + hwmon->device = hwmon_device_register(NULL); if (IS_ERR(hwmon->device)) { result = PTR_ERR(hwmon->device); goto free_mem; } + dev_set_drvdata(hwmon->device, hwmon); + result = device_create_file(hwmon->device, &dev_attr_name); + if (result) + goto free_mem; register_sys_interface: temp = kzalloc(sizeof(*temp), GFP_KERNEL); @@ -211,8 +222,10 @@ int thermal_add_hwmon_sysfs(struct thermal_zone_device *tz) free_temp_mem: kfree(temp); unregister_name: - if (new_hwmon_device) + if (new_hwmon_device) { + device_remove_file(hwmon->device, &dev_attr_name); hwmon_device_unregister(hwmon->device); + } free_mem: if (new_hwmon_device) kfree(hwmon); @@ -254,6 +267,7 @@ void thermal_remove_hwmon_sysfs(struct thermal_zone_device *tz) list_del(&hwmon->node); mutex_unlock(&thermal_hwmon_list_lock); + device_remove_file(hwmon->device, &dev_attr_name); hwmon_device_unregister(hwmon->device); kfree(hwmon); } -- cgit v1.2.3 From bdbfd5196d24a6d0845b549eba6ce8e6fa8bb3d0 Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Tue, 24 Jan 2017 12:53:45 -0700 Subject: drm/i915/gvt/kvmgt: mdev ABI is available_instances, not available_instance Per the ABI specification[1], each mdev_supported_types entry should have an available_instances, with an "s", not available_instance. [1] Documentation/ABI/testing/sysfs-bus-vfio-mdev Signed-off-by: Alex Williamson Signed-off-by: Zhenyu Wang --- drivers/gpu/drm/i915/gvt/kvmgt.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index 0c9234a87a20..3f656e3a6e5a 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -230,8 +230,8 @@ static struct intel_vgpu_type *intel_gvt_find_vgpu_type(struct intel_gvt *gvt, return NULL; } -static ssize_t available_instance_show(struct kobject *kobj, struct device *dev, - char *buf) +static ssize_t available_instances_show(struct kobject *kobj, + struct device *dev, char *buf) { struct intel_vgpu_type *type; unsigned int num = 0; @@ -269,12 +269,12 @@ static ssize_t description_show(struct kobject *kobj, struct device *dev, type->fence); } -static MDEV_TYPE_ATTR_RO(available_instance); +static MDEV_TYPE_ATTR_RO(available_instances); static MDEV_TYPE_ATTR_RO(device_api); static MDEV_TYPE_ATTR_RO(description); static struct attribute *type_attrs[] = { - &mdev_type_attr_available_instance.attr, + &mdev_type_attr_available_instances.attr, &mdev_type_attr_device_api.attr, &mdev_type_attr_description.attr, NULL, -- cgit v1.2.3 From 7283accfaef66e6a64f7d3ec0672596dd8e5b144 Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Tue, 24 Jan 2017 13:15:43 -0700 Subject: drm/i915/gvt: Fix kmem_cache_create() name According to kmem_cache_sanity_check(), spaces are not allowed in the name of a cache and results in a kernel oops with CONFIG_DEBUG_VM. Convert to underscores. Signed-off-by: Alex Williamson Signed-off-by: Zhenyu Wang --- drivers/gpu/drm/i915/gvt/execlist.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c index fb852c51d00e..34083731669d 100644 --- a/drivers/gpu/drm/i915/gvt/execlist.c +++ b/drivers/gpu/drm/i915/gvt/execlist.c @@ -798,7 +798,7 @@ int intel_vgpu_init_execlist(struct intel_vgpu *vgpu) INIT_LIST_HEAD(&vgpu->workload_q_head[i]); } - vgpu->workloads = kmem_cache_create("gvt-g vgpu workload", + vgpu->workloads = kmem_cache_create("gvt-g_vgpu_workload", sizeof(struct intel_vgpu_workload), 0, SLAB_HWCACHE_ALIGN, NULL); -- cgit v1.2.3 From ba7addcd805e5c83e201b118a2693b921a980b44 Mon Sep 17 00:00:00 2001 From: Zhenyu Wang Date: Wed, 25 Jan 2017 10:30:02 +0800 Subject: MAINTAINERS: update new mail list for intel gvt driver We've moved to lists.freedesktop.org from lists.01.org. Update info in MAINTAINERS. Signed-off-by: Zhenyu Wang --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 5f0420a0da5b..5bd03d5cb28b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4153,7 +4153,7 @@ F: Documentation/gpu/i915.rst INTEL GVT-g DRIVERS (Intel GPU Virtualization) M: Zhenyu Wang M: Zhi Wang -L: igvt-g-dev@lists.01.org +L: intel-gvt-dev@lists.freedesktop.org L: intel-gfx@lists.freedesktop.org W: https://01.org/igvt-g T: git https://github.com/01org/gvt-linux.git -- cgit v1.2.3 From b5fa0f7f88edcde37df1807fdf9ff10ec787a60e Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 24 Jan 2017 16:36:57 +1100 Subject: powerpc: Fix build failure with clang due to BUILD_BUG_ON() Anton says: In commit 4db7327194db ("powerpc: Add option to use jump label for cpu_has_feature()") and commit c12e6f24d413 ("powerpc: Add option to use jump label for mmu_has_feature()") we added: BUILD_BUG_ON(!__builtin_constant_p(feature)) to cpu_has_feature() and mmu_has_feature() in order to catch usage issues (such as cpu_has_feature(cpu_has_feature(X), which has happened once in the past). Unfortunately LLVM isn't smart enough to resolve this, and it errors out. I work around it in my clang/LLVM builds of the kernel, but I have just discovered that it causes a lot of issues for the bcc (eBPF) trace tool (which uses LLVM). For now just #ifdef it away for clang builds. Fixes: 4db7327194db ("powerpc: Add option to use jump label for cpu_has_feature()") Fixes: c12e6f24d413 ("powerpc: Add option to use jump label for mmu_has_feature()") Cc: stable@vger.kernel.org # v4.8+ Reported-by: Anton Blanchard Tested-by: Naveen N. Rao Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/cpu_has_feature.h | 2 ++ arch/powerpc/include/asm/mmu.h | 2 ++ 2 files changed, 4 insertions(+) diff --git a/arch/powerpc/include/asm/cpu_has_feature.h b/arch/powerpc/include/asm/cpu_has_feature.h index b312b152461b..6e834caa3720 100644 --- a/arch/powerpc/include/asm/cpu_has_feature.h +++ b/arch/powerpc/include/asm/cpu_has_feature.h @@ -23,7 +23,9 @@ static __always_inline bool cpu_has_feature(unsigned long feature) { int i; +#ifndef __clang__ /* clang can't cope with this */ BUILD_BUG_ON(!__builtin_constant_p(feature)); +#endif #ifdef CONFIG_JUMP_LABEL_FEATURE_CHECK_DEBUG if (!static_key_initialized) { diff --git a/arch/powerpc/include/asm/mmu.h b/arch/powerpc/include/asm/mmu.h index a34c764ca8dd..233a7e8cc8e3 100644 --- a/arch/powerpc/include/asm/mmu.h +++ b/arch/powerpc/include/asm/mmu.h @@ -160,7 +160,9 @@ static __always_inline bool mmu_has_feature(unsigned long feature) { int i; +#ifndef __clang__ /* clang can't cope with this */ BUILD_BUG_ON(!__builtin_constant_p(feature)); +#endif #ifdef CONFIG_JUMP_LABEL_FEATURE_CHECK_DEBUG if (!static_key_initialized) { -- cgit v1.2.3 From ae4a3e028bb8b59e7cfeb0cc9ef03d885182ce8b Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Thu, 19 Jan 2017 08:49:07 -0800 Subject: dmaengine: cppi41: Fix runtime PM timeouts with USB mass storage Commit fdea2d09b997 ("dmaengine: cppi41: Add basic PM runtime support") added runtime PM support for cppi41, but had corner case issues. Some of the issues were fixed with commit 098de42ad670 ("dmaengine: cppi41: Fix unpaired pm runtime when only a USB hub is connected"). That fix however caused a new regression where we can get error -115 messages with USB on BeagleBone when connecting a USB mass storage device to a hub. This is because when connecting a USB mass storage device to a hub, the initial DMA transfers can take over 200ms to complete and cppi41 autosuspend delay times out. To fix the issue, we want to implement refcounting for chan_busy array that contains the active dma transfers. Increasing the autosuspend delay won't help as that the delay could be potentially seconds, and it's best to let the USB subsystem to deal with the timeouts on errors. The earlier attempt for runtime PM was buggy as the pm_runtime_get/put() calls could get unpaired easily as they did not follow the state of the chan_busy array as described in commit 098de42ad670 ("dmaengine: cppi41: Fix unpaired pm runtime when only a USB hub is connected". Let's fix the issue by adding pm_runtime_get() to where a new transfer is added to the chan_busy array, and calls to pm_runtime_put() where chan_busy array entry is cleared. This prevents any autosuspend timeouts from happening while dma transfers are active. Fixes: 098de42ad670 ("dmaengine: cppi41: Fix unpaired pm runtime when only a USB hub is connected") Fixes: fdea2d09b997 ("dmaengine: cppi41: Add basic PM runtime support") Cc: Andy Shevchenko Cc: Bin Liu Cc: Grygorii Strashko Cc: Kevin Hilman Cc: Patrick Titiano Cc: Sergei Shtylyov Signed-off-by: Tony Lindgren Tested-by: Bin Liu Signed-off-by: Vinod Koul --- drivers/dma/cppi41.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/dma/cppi41.c b/drivers/dma/cppi41.c index d5ba43a87a68..7de4fdf86a6a 100644 --- a/drivers/dma/cppi41.c +++ b/drivers/dma/cppi41.c @@ -257,6 +257,10 @@ static struct cppi41_channel *desc_to_chan(struct cppi41_dd *cdd, u32 desc) BUG_ON(desc_num >= ALLOC_DECS_NUM); c = cdd->chan_busy[desc_num]; cdd->chan_busy[desc_num] = NULL; + + /* Usecount for chan_busy[], paired with push_desc_queue() */ + pm_runtime_put(cdd->ddev.dev); + return c; } @@ -447,6 +451,15 @@ static void push_desc_queue(struct cppi41_channel *c) */ __iowmb(); + /* + * DMA transfers can take at least 200ms to complete with USB mass + * storage connected. To prevent autosuspend timeouts, we must use + * pm_runtime_get/put() when chan_busy[] is modified. This will get + * cleared in desc_to_chan() or cppi41_stop_chan() depending on the + * outcome of the transfer. + */ + pm_runtime_get(cdd->ddev.dev); + desc_phys = lower_32_bits(c->desc_phys); desc_num = (desc_phys - cdd->descs_phys) / sizeof(struct cppi41_desc); WARN_ON(cdd->chan_busy[desc_num]); @@ -705,6 +718,9 @@ static int cppi41_stop_chan(struct dma_chan *chan) WARN_ON(!cdd->chan_busy[desc_num]); cdd->chan_busy[desc_num] = NULL; + /* Usecount for chan_busy[], paired with push_desc_queue() */ + pm_runtime_put(cdd->ddev.dev); + return 0; } -- cgit v1.2.3 From 362f4562466c3b9490e733e06999025638310d4a Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Thu, 19 Jan 2017 08:49:08 -0800 Subject: dmaengine: cppi41: Fix oops in cppi41_runtime_resume Commit fdea2d09b997 ("dmaengine: cppi41: Add basic PM runtime support") together with recent MUSB changes allowed USB and DMA on BeagleBone to idle when no cable is connected. But looks like few corner case issues still remain. Looks like just by re-plugging USB cable about ten or so times on BeagleBone when configured in USB peripheral mode we can get warnings and eventually trigger an oops in cppi41 DMA: WARNING: CPU: 0 PID: 14 at drivers/dma/cppi41.c:1154 cppi41_runtime_suspend+ x28/0x38 [cppi41] ... WARNING: CPU: 0 PID: 14 at drivers/dma/cppi41.c:452 push_desc_queue+0x94/0x9c [cppi41] ... Unable to handle kernel NULL pointer dereference at virtual address 00000104 pgd = c0004000 [00000104] *pgd=00000000 Internal error: Oops: 805 [#1] SMP ARM ... [] (cppi41_runtime_resume [cppi41]) from [] (__rpm_callback+0xc0/0x214) [] (__rpm_callback) from [] (rpm_callback+0x20/0x80) [] (rpm_callback) from [] (rpm_resume+0x504/0x78c) [] (rpm_resume) from [] (pm_runtime_work+0x60/0xa8) [] (pm_runtime_work) from [] (process_one_work+0x2b4/0x808) This is because of a race with runtime PM and cppi41_dma_issue_pending() as reported by Alexandre Bailon in earlier set of patches. Based on mailing list discussions we however came to the conclusion that a different fix from Alexandre's fix is needed in order to guarantee that DMA is really active when we try to use it. To fix the issue, we need to add a driver specific flag as we otherwise can have -EINPROGRESS state set by runtime PM and can't rely on pm_runtime_active() to tell us when we can use the DMA. And we need to make sure the DMA transfers get triggered in the queued order. So let's always queue the transfers, then flush the queue from both cppi41_dma_issue_pending() and cppi41_runtime_resume() as suggested by Grygorii Strashko in an earlier example patch. For reference, this is also documented in Documentation/power/runtime_pm.txt in the example at the end of the file as pointed out by Grygorii Strashko . Based on earlier patches from Alexandre Bailon and Grygorii Strashko modified based on testing and what was discussed on the mailing lists. Fixes: fdea2d09b997 ("dmaengine: cppi41: Add basic PM runtime support") Cc: Andy Shevchenko Cc: Bin Liu Cc: Grygorii Strashko Cc: Kevin Hilman Cc: Patrick Titiano Cc: Sergei Shtylyov Reported-by: Alexandre Bailon Signed-off-by: Tony Lindgren Tested-by: Bin Liu Signed-off-by: Vinod Koul --- drivers/dma/cppi41.c | 40 +++++++++++++++++++++++++--------------- 1 file changed, 25 insertions(+), 15 deletions(-) diff --git a/drivers/dma/cppi41.c b/drivers/dma/cppi41.c index 7de4fdf86a6a..55c1782e3623 100644 --- a/drivers/dma/cppi41.c +++ b/drivers/dma/cppi41.c @@ -153,6 +153,8 @@ struct cppi41_dd { /* context for suspend/resume */ unsigned int dma_tdfdq; + + bool is_suspended; }; #define FIST_COMPLETION_QUEUE 93 @@ -470,20 +472,26 @@ static void push_desc_queue(struct cppi41_channel *c) cppi_writel(reg, cdd->qmgr_mem + QMGR_QUEUE_D(c->q_num)); } -static void pending_desc(struct cppi41_channel *c) +/* + * Caller must hold cdd->lock to prevent push_desc_queue() + * getting called out of order. We have both cppi41_dma_issue_pending() + * and cppi41_runtime_resume() call this function. + */ +static void cppi41_run_queue(struct cppi41_dd *cdd) { - struct cppi41_dd *cdd = c->cdd; - unsigned long flags; + struct cppi41_channel *c, *_c; - spin_lock_irqsave(&cdd->lock, flags); - list_add_tail(&c->node, &cdd->pending); - spin_unlock_irqrestore(&cdd->lock, flags); + list_for_each_entry_safe(c, _c, &cdd->pending, node) { + push_desc_queue(c); + list_del(&c->node); + } } static void cppi41_dma_issue_pending(struct dma_chan *chan) { struct cppi41_channel *c = to_cpp41_chan(chan); struct cppi41_dd *cdd = c->cdd; + unsigned long flags; int error; error = pm_runtime_get(cdd->ddev.dev); @@ -495,10 +503,11 @@ static void cppi41_dma_issue_pending(struct dma_chan *chan) return; } - if (likely(pm_runtime_active(cdd->ddev.dev))) - push_desc_queue(c); - else - pending_desc(c); + spin_lock_irqsave(&cdd->lock, flags); + list_add_tail(&c->node, &cdd->pending); + if (!cdd->is_suspended) + cppi41_run_queue(cdd); + spin_unlock_irqrestore(&cdd->lock, flags); pm_runtime_mark_last_busy(cdd->ddev.dev); pm_runtime_put_autosuspend(cdd->ddev.dev); @@ -1166,8 +1175,12 @@ static int __maybe_unused cppi41_resume(struct device *dev) static int __maybe_unused cppi41_runtime_suspend(struct device *dev) { struct cppi41_dd *cdd = dev_get_drvdata(dev); + unsigned long flags; + spin_lock_irqsave(&cdd->lock, flags); + cdd->is_suspended = true; WARN_ON(!list_empty(&cdd->pending)); + spin_unlock_irqrestore(&cdd->lock, flags); return 0; } @@ -1175,14 +1188,11 @@ static int __maybe_unused cppi41_runtime_suspend(struct device *dev) static int __maybe_unused cppi41_runtime_resume(struct device *dev) { struct cppi41_dd *cdd = dev_get_drvdata(dev); - struct cppi41_channel *c, *_c; unsigned long flags; spin_lock_irqsave(&cdd->lock, flags); - list_for_each_entry_safe(c, _c, &cdd->pending, node) { - push_desc_queue(c); - list_del(&c->node); - } + cdd->is_suspended = false; + cppi41_run_queue(cdd); spin_unlock_irqrestore(&cdd->lock, flags); return 0; -- cgit v1.2.3 From 6610d0edf6dc7ee97e46ab3a538a565c79d26199 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Fri, 20 Jan 2017 12:07:53 -0800 Subject: dmaengine: cppi41: Clean up pointless warnings With patches "dmaengine: cppi41: Fix runtime PM timeouts with USB mass storage", and "dmaengine: cppi41: Fix oops in cppi41_runtime_resume", the pm_runtime_get/put() in cppi41_irq() is no longer needed. We now guarantee that cppi41 is enabled when dma is in use. We can still get pointless error -115 when musb is configured as a usb peripheral. That's because we should now check for the state of is_suspended instead. Let's just remove the now useless code and replace it with a WARN(). Signed-off-by: Tony Lindgren Signed-off-by: Vinod Koul --- drivers/dma/cppi41.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/drivers/dma/cppi41.c b/drivers/dma/cppi41.c index 55c1782e3623..200828c60db9 100644 --- a/drivers/dma/cppi41.c +++ b/drivers/dma/cppi41.c @@ -323,12 +323,12 @@ static irqreturn_t cppi41_irq(int irq, void *data) while (val) { u32 desc, len; - int error; - error = pm_runtime_get(cdd->ddev.dev); - if (error < 0) - dev_err(cdd->ddev.dev, "%s pm runtime get: %i\n", - __func__, error); + /* + * This should never trigger, see the comments in + * push_desc_queue() + */ + WARN_ON(cdd->is_suspended); q_num = __fls(val); val &= ~(1 << q_num); @@ -349,9 +349,6 @@ static irqreturn_t cppi41_irq(int irq, void *data) c->residue = pd_trans_len(c->desc->pd6) - len; dma_cookie_complete(&c->txd); dmaengine_desc_get_callback_invoke(&c->txd, NULL); - - pm_runtime_mark_last_busy(cdd->ddev.dev); - pm_runtime_put_autosuspend(cdd->ddev.dev); } } return IRQ_HANDLED; -- cgit v1.2.3 From 407788b51db6f6aab499d02420082f436abf3238 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Tue, 24 Jan 2017 09:18:57 -0600 Subject: usb: musb: Fix host mode error -71 regression Commit 467d5c980709 ("usb: musb: Implement session bit based runtime PM for musb-core") started implementing musb generic runtime PM support by introducing devctl register session bit based state control. This caused a regression where if a USB mass storage device is connected to a USB hub, we can get: usb 1-1: reset high-speed USB device number 2 using musb-hdrc usb 1-1: device descriptor read/64, error -71 usb 1-1.1: new high-speed USB device number 4 using musb-hdrc This is because before the USB storage device is connected, musb is in OTG_STATE_A_SUSPEND. And we currently only set need_finish_resume in musb_stage0_irq() and the related code calling finish_resume_work in musb_resume() and musb_runtime_resume() never gets called. To fix the issue, we can call schedule_delayed_work() directly in musb_stage0_irq() to have finish_resume_work run. And we should no longer never get interrupts when when suspended. We have changed musb to no longer need pm_runtime_irqsafe(). The need_finish_resume flag was added in commit 9298b4aad37e ("usb: musb: fix device hotplug behind hub") and no longer applies as far as I can tell. So let's just remove the earlier code that no longer is needed. Fixes: 467d5c980709 ("usb: musb: Implement session bit based runtime PM for musb-core") Reported-by: Bin Liu Signed-off-by: Tony Lindgren Signed-off-by: Bin Liu Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/musb/musb_core.c | 15 ++------------- drivers/usb/musb/musb_core.h | 1 - 2 files changed, 2 insertions(+), 14 deletions(-) diff --git a/drivers/usb/musb/musb_core.c b/drivers/usb/musb/musb_core.c index fca288bbc800..cd40467be3c5 100644 --- a/drivers/usb/musb/musb_core.c +++ b/drivers/usb/musb/musb_core.c @@ -594,11 +594,11 @@ static irqreturn_t musb_stage0_irq(struct musb *musb, u8 int_usb, | MUSB_PORT_STAT_RESUME; musb->rh_timer = jiffies + msecs_to_jiffies(USB_RESUME_TIMEOUT); - musb->need_finish_resume = 1; - musb->xceiv->otg->state = OTG_STATE_A_HOST; musb->is_active = 1; musb_host_resume_root_hub(musb); + schedule_delayed_work(&musb->finish_resume_work, + msecs_to_jiffies(USB_RESUME_TIMEOUT)); break; case OTG_STATE_B_WAIT_ACON: musb->xceiv->otg->state = OTG_STATE_B_PERIPHERAL; @@ -2710,11 +2710,6 @@ static int musb_resume(struct device *dev) mask = MUSB_DEVCTL_BDEVICE | MUSB_DEVCTL_FSDEV | MUSB_DEVCTL_LSDEV; if ((devctl & mask) != (musb->context.devctl & mask)) musb->port1_status = 0; - if (musb->need_finish_resume) { - musb->need_finish_resume = 0; - schedule_delayed_work(&musb->finish_resume_work, - msecs_to_jiffies(USB_RESUME_TIMEOUT)); - } /* * The USB HUB code expects the device to be in RPM_ACTIVE once it came @@ -2766,12 +2761,6 @@ static int musb_runtime_resume(struct device *dev) musb_restore_context(musb); - if (musb->need_finish_resume) { - musb->need_finish_resume = 0; - schedule_delayed_work(&musb->finish_resume_work, - msecs_to_jiffies(USB_RESUME_TIMEOUT)); - } - spin_lock_irqsave(&musb->lock, flags); error = musb_run_resume_work(musb); if (error) diff --git a/drivers/usb/musb/musb_core.h b/drivers/usb/musb/musb_core.h index ade902ea1221..ce5a18c98c6d 100644 --- a/drivers/usb/musb/musb_core.h +++ b/drivers/usb/musb/musb_core.h @@ -410,7 +410,6 @@ struct musb { /* is_suspended means USB B_PERIPHERAL suspend */ unsigned is_suspended:1; - unsigned need_finish_resume :1; /* may_wakeup means remote wakeup is enabled */ unsigned may_wakeup:1; -- cgit v1.2.3 From 3ba7b7795b7e8889af1377904c55c7fae9e0c775 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Tue, 24 Jan 2017 09:18:58 -0600 Subject: usb: musb: Fix external abort on non-linefetch for musb_irq_work() While testing musb host mode cable plugging on a BeagleBone, I came across this error: Unhandled fault: external abort on non-linefetch (0x1008) at 0xd1dcfc60 ... [] (musb_default_readb [musb_hdrc]) from [] (musb_irq_work+0x1c/0x180 [musb_hdrc]) [] (musb_irq_work [musb_hdrc]) from [] (process_one_work+0x2b4/0x808) [] (process_one_work) from [] (worker_thread+0x3c/0x550) [] (worker_thread) from [] (kthread+0x104/0x148) [] (kthread) from [] (ret_from_fork+0x14/0x24) Signed-off-by: Tony Lindgren Signed-off-by: Bin Liu Signed-off-by: Greg Kroah-Hartman --- drivers/usb/musb/musb_core.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/usb/musb/musb_core.c b/drivers/usb/musb/musb_core.c index cd40467be3c5..772f15821242 100644 --- a/drivers/usb/musb/musb_core.c +++ b/drivers/usb/musb/musb_core.c @@ -1925,6 +1925,14 @@ static void musb_pm_runtime_check_session(struct musb *musb) static void musb_irq_work(struct work_struct *data) { struct musb *musb = container_of(data, struct musb, irq_work.work); + int error; + + error = pm_runtime_get_sync(musb->controller); + if (error < 0) { + dev_err(musb->controller, "Could not enable: %i\n", error); + + return; + } musb_pm_runtime_check_session(musb); @@ -1932,6 +1940,9 @@ static void musb_irq_work(struct work_struct *data) musb->xceiv_old_state = musb->xceiv->otg->state; sysfs_notify(&musb->controller->kobj, NULL, "mode"); } + + pm_runtime_mark_last_busy(musb->controller); + pm_runtime_put_autosuspend(musb->controller); } static void musb_recover_from_babble(struct musb *musb) -- cgit v1.2.3 From d9b2997e4a0a874e452df7cdd7de5a54502bd0aa Mon Sep 17 00:00:00 2001 From: Lukáš Lalinský Date: Fri, 20 Jan 2017 19:46:34 +0100 Subject: USB: Add quirk for WORLDE easykey.25 MIDI keyboard Add a quirk for WORLDE easykey.25 MIDI keyboard (idVendor=0218, idProduct=0401). The device reports that it has config string descriptor at index 3, but when the system selects the configuration and tries to get the description, it returns a -EPROTO error, the communication restarts and this keeps repeating over and over again. Not requesting the string descriptor makes the device work correctly. Relevant info from Wireshark: [...] CONFIGURATION DESCRIPTOR bLength: 9 bDescriptorType: 0x02 (CONFIGURATION) wTotalLength: 101 bNumInterfaces: 2 bConfigurationValue: 1 iConfiguration: 3 Configuration bmAttributes: 0xc0 SELF-POWERED NO REMOTE-WAKEUP 1... .... = Must be 1: Must be 1 for USB 1.1 and higher .1.. .... = Self-Powered: This device is SELF-POWERED ..0. .... = Remote Wakeup: This device does NOT support remote wakeup bMaxPower: 50 (100mA) [...] 45 0.369104 host 2.38.0 USB 64 GET DESCRIPTOR Request STRING [...] URB setup bmRequestType: 0x80 1... .... = Direction: Device-to-host .00. .... = Type: Standard (0x00) ...0 0000 = Recipient: Device (0x00) bRequest: GET DESCRIPTOR (6) Descriptor Index: 0x03 bDescriptorType: 0x03 Language Id: English (United States) (0x0409) wLength: 255 46 0.369255 2.38.0 host USB 64 GET DESCRIPTOR Response STRING[Malformed Packet] [...] Frame 46: 64 bytes on wire (512 bits), 64 bytes captured (512 bits) on interface 0 USB URB [Source: 2.38.0] [Destination: host] URB id: 0xffff88021f62d480 URB type: URB_COMPLETE ('C') URB transfer type: URB_CONTROL (0x02) Endpoint: 0x80, Direction: IN Device: 38 URB bus id: 2 Device setup request: not relevant ('-') Data: present (0) URB sec: 1484896277 URB usec: 455031 URB status: Protocol error (-EPROTO) (-71) URB length [bytes]: 0 Data length [bytes]: 0 [Request in: 45] [Time from request: 0.000151000 seconds] Unused Setup Header Interval: 0 Start frame: 0 Copy of Transfer Flags: 0x00000200 Number of ISO descriptors: 0 [Malformed Packet: USB] [Expert Info (Error/Malformed): Malformed Packet (Exception occurred)] [Malformed Packet (Exception occurred)] [Severity level: Error] [Group: Malformed] Signed-off-by: Lukáš Lalinský Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/quirks.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/usb/core/quirks.c b/drivers/usb/core/quirks.c index d2e50a27140c..24f9f98968a5 100644 --- a/drivers/usb/core/quirks.c +++ b/drivers/usb/core/quirks.c @@ -37,6 +37,10 @@ static const struct usb_device_id usb_quirk_list[] = { /* CBM - Flash disk */ { USB_DEVICE(0x0204, 0x6025), .driver_info = USB_QUIRK_RESET_RESUME }, + /* WORLDE easy key (easykey.25) MIDI controller */ + { USB_DEVICE(0x0218, 0x0401), .driver_info = + USB_QUIRK_CONFIG_INTF_STRINGS }, + /* HP 5300/5370C scanner */ { USB_DEVICE(0x03f0, 0x0701), .driver_info = USB_QUIRK_STRING_FETCH_255 }, -- cgit v1.2.3 From 91539eb1fda2d530d3b268eef542c5414e54bf1a Mon Sep 17 00:00:00 2001 From: Iago Abal Date: Wed, 11 Jan 2017 14:00:21 +0100 Subject: dmaengine: pl330: fix double lock The static bug finder EBA (http://www.iagoabal.eu/eba/) reported the following double-lock bug: Double lock: 1. spin_lock_irqsave(pch->lock, flags) at pl330_free_chan_resources:2236; 2. call to function `pl330_release_channel' immediately after; 3. call to function `dma_pl330_rqcb' in line 1753; 4. spin_lock_irqsave(pch->lock, flags) at dma_pl330_rqcb:1505. I have fixed it as suggested by Marek Szyprowski. First, I have replaced `pch->lock' with `pl330->lock' in functions `pl330_alloc_chan_resources' and `pl330_free_chan_resources'. This avoids the double-lock by acquiring a different lock than `dma_pl330_rqcb'. NOTE that, as a result, `pl330_free_chan_resources' executes `list_splice_tail_init' on `pch->work_list' under lock `pl330->lock', whereas in the rest of the code `pch->work_list' is protected by `pch->lock'. I don't know if this may cause race conditions. Similarly `pch->cyclic' is written by `pl330_alloc_chan_resources' under `pl330->lock' but read by `pl330_tx_submit' under `pch->lock'. Second, I have removed locking from `pl330_request_channel' and `pl330_release_channel' functions. Function `pl330_request_channel' is only called from `pl330_alloc_chan_resources', so the lock is already held. Function `pl330_release_channel' is called from `pl330_free_chan_resources', which already holds the lock, and from `pl330_del'. Function `pl330_del' is called in an error path of `pl330_probe' and at the end of `pl330_remove', but I assume that there cannot be concurrent accesses to the protected data at those points. Signed-off-by: Iago Abal Reviewed-by: Marek Szyprowski Signed-off-by: Vinod Koul --- drivers/dma/pl330.c | 19 ++++++------------- 1 file changed, 6 insertions(+), 13 deletions(-) diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c index 740bbb942594..7539f73df9e0 100644 --- a/drivers/dma/pl330.c +++ b/drivers/dma/pl330.c @@ -1699,7 +1699,6 @@ static bool _chan_ns(const struct pl330_dmac *pl330, int i) static struct pl330_thread *pl330_request_channel(struct pl330_dmac *pl330) { struct pl330_thread *thrd = NULL; - unsigned long flags; int chans, i; if (pl330->state == DYING) @@ -1707,8 +1706,6 @@ static struct pl330_thread *pl330_request_channel(struct pl330_dmac *pl330) chans = pl330->pcfg.num_chan; - spin_lock_irqsave(&pl330->lock, flags); - for (i = 0; i < chans; i++) { thrd = &pl330->channels[i]; if ((thrd->free) && (!_manager_ns(thrd) || @@ -1726,8 +1723,6 @@ static struct pl330_thread *pl330_request_channel(struct pl330_dmac *pl330) thrd = NULL; } - spin_unlock_irqrestore(&pl330->lock, flags); - return thrd; } @@ -1745,7 +1740,6 @@ static inline void _free_event(struct pl330_thread *thrd, int ev) static void pl330_release_channel(struct pl330_thread *thrd) { struct pl330_dmac *pl330; - unsigned long flags; if (!thrd || thrd->free) return; @@ -1757,10 +1751,8 @@ static void pl330_release_channel(struct pl330_thread *thrd) pl330 = thrd->dmac; - spin_lock_irqsave(&pl330->lock, flags); _free_event(thrd, thrd->ev); thrd->free = true; - spin_unlock_irqrestore(&pl330->lock, flags); } /* Initialize the structure for PL330 configuration, that can be used @@ -2122,20 +2114,20 @@ static int pl330_alloc_chan_resources(struct dma_chan *chan) struct pl330_dmac *pl330 = pch->dmac; unsigned long flags; - spin_lock_irqsave(&pch->lock, flags); + spin_lock_irqsave(&pl330->lock, flags); dma_cookie_init(chan); pch->cyclic = false; pch->thread = pl330_request_channel(pl330); if (!pch->thread) { - spin_unlock_irqrestore(&pch->lock, flags); + spin_unlock_irqrestore(&pl330->lock, flags); return -ENOMEM; } tasklet_init(&pch->task, pl330_tasklet, (unsigned long) pch); - spin_unlock_irqrestore(&pch->lock, flags); + spin_unlock_irqrestore(&pl330->lock, flags); return 1; } @@ -2238,12 +2230,13 @@ static int pl330_pause(struct dma_chan *chan) static void pl330_free_chan_resources(struct dma_chan *chan) { struct dma_pl330_chan *pch = to_pchan(chan); + struct pl330_dmac *pl330 = pch->dmac; unsigned long flags; tasklet_kill(&pch->task); pm_runtime_get_sync(pch->dmac->ddma.dev); - spin_lock_irqsave(&pch->lock, flags); + spin_lock_irqsave(&pl330->lock, flags); pl330_release_channel(pch->thread); pch->thread = NULL; @@ -2251,7 +2244,7 @@ static void pl330_free_chan_resources(struct dma_chan *chan) if (pch->cyclic) list_splice_tail_init(&pch->work_list, &pch->dmac->desc_pool); - spin_unlock_irqrestore(&pch->lock, flags); + spin_unlock_irqrestore(&pl330->lock, flags); pm_runtime_mark_last_busy(pch->dmac->ddma.dev); pm_runtime_put_autosuspend(pch->dmac->ddma.dev); } -- cgit v1.2.3 From b17c1bba9cec1727451b906d9a0c209774624873 Mon Sep 17 00:00:00 2001 From: Rui Miguel Silva Date: Mon, 23 Jan 2017 16:32:57 +0000 Subject: staging: greybus: timesync: validate platform state callback When tearingdown timesync, and not in arche platform, the state platform callback is not initialized. That will trigger the following NULL dereferencing. CallTrace: ? gb_timesync_platform_unlock_bus+0x11/0x20 [greybus] gb_timesync_teardown+0x85/0xc0 [greybus] gb_timesync_svc_remove+0xab/0x190 [greybus] gb_svc_del+0x29/0x110 [greybus] gb_hd_del+0x14/0x20 [greybus] ap_disconnect+0x24/0x60 [gb_es2] usb_unbind_interface+0x7a/0x2c0 __device_release_driver+0x96/0x150 device_release_driver+0x1e/0x30 bus_remove_device+0xe7/0x130 device_del+0x116/0x230 usb_disable_device+0x97/0x1f0 usb_disconnect+0x80/0x260 hub_event+0x5ca/0x10e0 process_one_work+0x126/0x3b0 worker_thread+0x55/0x4c0 ? process_one_work+0x3b0/0x3b0 kthread+0xc4/0xe0 ? kthread_park+0xb0/0xb0 ret_from_fork+0x22/0x30 So, fix that by adding checks before use the callback. Fixes: 970dc85bd95d ("greybus: timesync: Add timesync core driver") Cc: # 4.9.x Signed-off-by: Rui Miguel Silva Reviewed-by: Viresh Kumar Reviewed-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman --- drivers/staging/greybus/timesync_platform.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/staging/greybus/timesync_platform.c b/drivers/staging/greybus/timesync_platform.c index 113f3d6c4b3a..27f75b17679b 100644 --- a/drivers/staging/greybus/timesync_platform.c +++ b/drivers/staging/greybus/timesync_platform.c @@ -45,12 +45,18 @@ u32 gb_timesync_platform_get_clock_rate(void) int gb_timesync_platform_lock_bus(struct gb_timesync_svc *pdata) { + if (!arche_platform_change_state_cb) + return 0; + return arche_platform_change_state_cb(ARCHE_PLATFORM_STATE_TIME_SYNC, pdata); } void gb_timesync_platform_unlock_bus(void) { + if (!arche_platform_change_state_cb) + return; + arche_platform_change_state_cb(ARCHE_PLATFORM_STATE_ACTIVE, NULL); } -- cgit v1.2.3 From b78671591a10218ab18bbea120fd05df7a002e88 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sun, 15 Jan 2017 12:58:25 +0000 Subject: drm/i915: Avoid drm_atomic_state_put(NULL) in intel_display_resume intel_display_resume() may be called without an atomic state to restore, i.e. dev_priv->modeset_reset_restore state is NULL. One such case is following a lid open/close event and the forced modeset in intel_lid_notify(). Reported-by: Stefan Seyfried Tested-by: Stefan Seyfried Fixes: 0853695c3ba4 ("drm: Add reference counting to drm_atomic_state") Signed-off-by: Chris Wilson Cc: Daniel Vetter Cc: Jani Nikula Cc: # v4.10-rc1+ Link: http://patchwork.freedesktop.org/patch/msgid/20170115125825.18597-1-chris@chris-wilson.co.uk Reviewed-by: Ander Conselvan de Oliveira (cherry picked from commit 3c5e37f169cb67cbd03c6116fbc93e0805815d29) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_display.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 8d702cf1a616..1a01d4189c3f 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -17027,7 +17027,8 @@ void intel_display_resume(struct drm_device *dev) if (ret) DRM_ERROR("Restoring old state failed with %i\n", ret); - drm_atomic_state_put(state); + if (state) + drm_atomic_state_put(state); } void intel_modeset_gem_init(struct drm_device *dev) -- cgit v1.2.3 From 27892bbdc9233f33bf0f44e08aab8f12e0dec142 Mon Sep 17 00:00:00 2001 From: Clint Taylor Date: Wed, 18 Jan 2017 13:38:43 -0800 Subject: drm/i915: prevent crash with .disable_display parameter The .disable_display parameter was causing a fatal crash when fbdev was dereferenced during driver init. V1: protection in i915_drv.c V2: Moved protection to intel_fbdev.c Fixes: 43cee314345a ("drm/i915/fbdev: Limit the global async-domain synchronization") Testcase: igt/drv_module_reload/basic-no-display Cc: Chris Wilson Signed-off-by: Clint Taylor Link: http://patchwork.freedesktop.org/patch/msgid/1484775523-29428-1-git-send-email-clinton.a.taylor@intel.com Reviewed-by: Chris Wilson Cc: Lukas Wunner Cc: Daniel Vetter Cc: Jani Nikula Cc: # v4.8+ Signed-off-by: Chris Wilson (cherry picked from commit 5b8cd0755f8a06a851c436a013e7be0823fb155a) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_fbdev.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c index beb08982dc0b..8cf2d80f2254 100644 --- a/drivers/gpu/drm/i915/intel_fbdev.c +++ b/drivers/gpu/drm/i915/intel_fbdev.c @@ -742,6 +742,9 @@ void intel_fbdev_initial_config_async(struct drm_device *dev) { struct intel_fbdev *ifbdev = to_i915(dev)->fbdev; + if (!ifbdev) + return; + ifbdev->cookie = async_schedule(intel_fbdev_initial_config, ifbdev); } -- cgit v1.2.3 From a38a7bd1766b42ea0ed14b99be23a653922ed5c8 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Thu, 19 Jan 2017 11:37:49 +0000 Subject: drm/i915: Release temporary load-detect state upon switching After we call drm_atomic_commit() on the load-detect state, we can free our local reference. Upon restore, we only apply and free the previous state. Fixes: 0853695c3ba4 ("drm: Add reference counting to drm_atomic_state") Signed-off-by: Chris Wilson Cc: Chris Wilson Cc: Daniel Vetter Cc: # v4.10-rc1+ Link: http://patchwork.freedesktop.org/patch/msgid/20170119113749.2517-1-chris@chris-wilson.co.uk Reviewed-by: Ville Syrjälä (cherry picked from commit 7abbd11f344aa7abe29befb218774a1ea26018ac) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_display.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 1a01d4189c3f..f313d62a7e74 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -11246,6 +11246,7 @@ found: } old->restore_state = restore_state; + drm_atomic_state_put(state); /* let the connector get through one full cycle before testing */ intel_wait_for_vblank(dev_priv, intel_crtc->pipe); -- cgit v1.2.3 From c34f078675f505c4437919bb1897b1351f16a050 Mon Sep 17 00:00:00 2001 From: Ander Conselvan de Oliveira Date: Fri, 20 Jan 2017 16:28:42 +0200 Subject: drm/i915: Don't leak edid in intel_crt_detect_ddc() In the path where intel_crt_detect_ddc() detects a CRT, if would return true without freeing the edid. Fixes: a2bd1f541f19 ("drm/i915: check whether we actually received an edid in detect_ddc") Cc: Chris Wilson Cc: Daniel Vetter Cc: Daniel Vetter Cc: Jani Nikula Cc: intel-gfx@lists.freedesktop.org Cc: # v3.6+ Signed-off-by: Ander Conselvan de Oliveira Reviewed-by: Ville Syrjälä Reviewed-by: Jani Nikula Link: http://patchwork.freedesktop.org/patch/msgid/1484922525-6131-1-git-send-email-ander.conselvan.de.oliveira@intel.com (cherry picked from commit c96b63a6a7ac4bd670ec2e663793a9a31418b790) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_crt.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c index 86ecec5601d4..588470eb8d39 100644 --- a/drivers/gpu/drm/i915/intel_crt.c +++ b/drivers/gpu/drm/i915/intel_crt.c @@ -499,6 +499,7 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector) struct drm_i915_private *dev_priv = to_i915(crt->base.base.dev); struct edid *edid; struct i2c_adapter *i2c; + bool ret = false; BUG_ON(crt->base.type != INTEL_OUTPUT_ANALOG); @@ -515,17 +516,17 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector) */ if (!is_digital) { DRM_DEBUG_KMS("CRT detected via DDC:0x50 [EDID]\n"); - return true; + ret = true; + } else { + DRM_DEBUG_KMS("CRT not detected via DDC:0x50 [EDID reports a digital panel]\n"); } - - DRM_DEBUG_KMS("CRT not detected via DDC:0x50 [EDID reports a digital panel]\n"); } else { DRM_DEBUG_KMS("CRT not detected via DDC:0x50 [no valid EDID found]\n"); } kfree(edid); - return false; + return ret; } static enum drm_connector_status -- cgit v1.2.3 From 21d6e0bde50713922a6520ef84e5fd245b05d468 Mon Sep 17 00:00:00 2001 From: Ander Conselvan de Oliveira Date: Fri, 20 Jan 2017 16:28:43 +0200 Subject: drm/i915: Don't init hpd polling for vlv and chv from runtime_suspend() An error in the condition for avoiding the call to intel_hpd_poll_init() for valleyview and cherryview from intel_runtime_suspend() caused it to be called unconditionally. Fix it. Fixes: 19625e85c6ec ("drm/i915: Enable polling when we don't have hpd") Cc: stable@vger.kernel.org Cc: Ville Syrjälä Cc: Daniel Vetter Cc: Lyude Cc: Daniel Vetter Cc: Jani Nikula Cc: intel-gfx@lists.freedesktop.org Cc: # v4.9+ Signed-off-by: Ander Conselvan de Oliveira Reviewed-by: Ville Syrjälä Link: http://patchwork.freedesktop.org/patch/msgid/1484922525-6131-2-git-send-email-ander.conselvan.de.oliveira@intel.com (cherry picked from commit 04313b00b79405f86d815100f85c47a2ee5b8ca0) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 445fec9c2841..b2c4a0b8a627 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -2378,7 +2378,7 @@ static int intel_runtime_suspend(struct device *kdev) assert_forcewakes_inactive(dev_priv); - if (!IS_VALLEYVIEW(dev_priv) || !IS_CHERRYVIEW(dev_priv)) + if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv)) intel_hpd_poll_init(dev_priv); DRM_DEBUG_KMS("Device suspended\n"); -- cgit v1.2.3 From 3781bd6e7d64d5f5bea9fdee11ab9460a700c0e4 Mon Sep 17 00:00:00 2001 From: Ander Conselvan de Oliveira Date: Fri, 20 Jan 2017 16:28:44 +0200 Subject: drm/i915: Fix calculation of rotated x and y offsets for planar formats Parameters tile_size, tile_width and tile_height were passed in the wrong order to _intel_adjust_tile_offset() when calculating the rotated offsets. This doesn't fix any user visible bug, since for packed formats new and old offset are the same and the rotated offsets are within a tile before they are fed to _intel_adjust_tile_offset(). In that case, the offsets are unchanged. That is not true for planar formats, but those are currently not supported. Fixes: 66a2d927cb0e ("drm/i915: Make intel_adjust_tile_offset() work for linear buffers") Cc: Ville Syrjälä Cc: Sivakumar Thulasimani Cc: Daniel Vetter Cc: Jani Nikula Cc: intel-gfx@lists.freedesktop.org Cc: # v4.9+ Signed-off-by: Ander Conselvan de Oliveira Reviewed-by: Ville Syrjälä Link: http://patchwork.freedesktop.org/patch/msgid/1484922525-6131-3-git-send-email-ander.conselvan.de.oliveira@intel.com (cherry picked from commit 46a1bd289507dfcc428fb9daf65421ed6be6af8b) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_display.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index f313d62a7e74..7a9040fc6e5a 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2585,8 +2585,9 @@ intel_fill_fb_info(struct drm_i915_private *dev_priv, * We only keep the x/y offsets, so push all of the * gtt offset into the x/y offsets. */ - _intel_adjust_tile_offset(&x, &y, tile_size, - tile_width, tile_height, pitch_tiles, + _intel_adjust_tile_offset(&x, &y, + tile_width, tile_height, + tile_size, pitch_tiles, gtt_offset_rotated * tile_size, 0); gtt_offset_rotated += rot_info->plane[i].width * rot_info->plane[i].height; -- cgit v1.2.3 From 6d1d427a4e24c403b4adf928d61994bdaa0ca03a Mon Sep 17 00:00:00 2001 From: Ander Conselvan de Oliveira Date: Fri, 20 Jan 2017 16:28:45 +0200 Subject: drm/i915: Check for NULL atomic state in intel_crtc_disable_noatomic() In intel_crtc_disable_noatomic(), bail on a failure to allocate an atomic state to avoid a NULL pointer dereference. Fixes: 4a80655827af ("drm/i915: Pass atomic state to crtc enable/disable functions") Cc: Maarten Lankhorst Cc: Daniel Vetter Cc: Daniel Vetter Cc: Jani Nikula Cc: intel-gfx@lists.freedesktop.org Cc: # v4.9+ Signed-off-by: Ander Conselvan de Oliveira Reviewed-by: Ville Syrjälä Link: http://patchwork.freedesktop.org/patch/msgid/1484922525-6131-4-git-send-email-ander.conselvan.de.oliveira@intel.com (cherry picked from commit 31bb2ef97ea9db343348f9b5ccaa9bb6f48fc655) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_display.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 7a9040fc6e5a..f6dd073afdc9 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -6850,6 +6850,12 @@ static void intel_crtc_disable_noatomic(struct drm_crtc *crtc) } state = drm_atomic_state_alloc(crtc->dev); + if (!state) { + DRM_DEBUG_KMS("failed to disable [CRTC:%d:%s], out of memory", + crtc->base.id, crtc->name); + return; + } + state->acquire_ctx = crtc->dev->mode_config.acquire_ctx; /* Everything's already locked, -EDEADLK can't happen. */ -- cgit v1.2.3 From 6f0f02dc56f18760b46dc1bf5b3f7386869d4162 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Mon, 23 Jan 2017 21:29:39 +0000 Subject: drm/i915: Move atomic state free from out of fence release Fences are required to support being released from under an atomic context. The drm_atomic_state struct may take a mutex when being released and so we cannot drop a reference to the drm_atomic_state from the fence release path directly, and so we need to defer that unreference to a worker. [ 326.576697] WARNING: CPU: 2 PID: 366 at kernel/sched/core.c:7737 __might_sleep+0x5d/0x80 [ 326.576816] do not call blocking ops when !TASK_RUNNING; state=1 set at [] intel_breadcrumbs_signaler+0x59/0x270 [i915] [ 326.576818] Modules linked in: rfcomm fuse snd_hda_codec_hdmi bnep snd_hda_codec_realtek snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hwdep snd_hda_core snd_pcm snd_seq_midi snd_seq_midi_event snd_rawmidi snd_seq snd_seq_device snd_timer input_leds led_class snd punit_atom_debug btusb btrtl btbcm btintel intel_rapl bluetooth i915 drm_kms_helper syscopyarea sysfillrect iwlwifi sysimgblt soundcore fb_sys_fops mei_txe cfg80211 drm pwm_lpss_platform pwm_lpss pinctrl_cherryview fjes acpi_pad parport_pc ppdev parport autofs4 [ 326.576899] CPU: 2 PID: 366 Comm: i915/signal:0 Tainted: G U 4.10.0-rc3-patser+ #5030 [ 326.576902] Hardware name: /NUC5PPYB, BIOS PYBSWCEL.86A.0031.2015.0601.1712 06/01/2015 [ 326.576905] Call Trace: [ 326.576920] dump_stack+0x4d/0x6d [ 326.576926] __warn+0xc0/0xe0 [ 326.576931] warn_slowpath_fmt+0x5a/0x80 [ 326.577004] ? intel_breadcrumbs_signaler+0x59/0x270 [i915] [ 326.577075] ? intel_breadcrumbs_signaler+0x59/0x270 [i915] [ 326.577079] __might_sleep+0x5d/0x80 [ 326.577087] mutex_lock+0x1b/0x40 [ 326.577133] drm_property_free_blob+0x1e/0x80 [drm] [ 326.577167] ? drm_property_destroy+0xe0/0xe0 [drm] [ 326.577200] drm_mode_object_unreference+0x5c/0x70 [drm] [ 326.577233] drm_property_unreference_blob+0xe/0x10 [drm] [ 326.577260] __drm_atomic_helper_crtc_destroy_state+0x14/0x40 [drm_kms_helper] [ 326.577278] drm_atomic_helper_crtc_destroy_state+0x10/0x20 [drm_kms_helper] [ 326.577352] intel_crtc_destroy_state+0x9/0x10 [i915] [ 326.577388] drm_atomic_state_default_clear+0xea/0x1d0 [drm] [ 326.577462] intel_atomic_state_clear+0xd/0x20 [i915] [ 326.577497] drm_atomic_state_clear+0x1a/0x30 [drm] [ 326.577532] __drm_atomic_state_free+0x13/0x60 [drm] [ 326.577607] intel_atomic_commit_ready+0x6f/0x78 [i915] [ 326.577670] i915_sw_fence_release+0x3a/0x50 [i915] [ 326.577733] dma_i915_sw_fence_wake+0x39/0x80 [i915] [ 326.577741] dma_fence_signal+0xda/0x120 [ 326.577812] ? intel_breadcrumbs_signaler+0x59/0x270 [i915] [ 326.577884] intel_breadcrumbs_signaler+0xb1/0x270 [i915] [ 326.577889] kthread+0x127/0x130 [ 326.577961] ? intel_engine_remove_wait+0x1a0/0x1a0 [i915] [ 326.577964] ? kthread_stop+0x120/0x120 [ 326.577970] ret_from_fork+0x22/0x30 Fixes: c004a90b7263 ("drm/i915: Restore nonblocking awaits for modesetting") Reported-by: Maarten Lankhorst Signed-off-by: Chris Wilson Cc: Chris Wilson Cc: Joonas Lahtinen Cc: Maarten Lankhorst Cc: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/20170123212939.30345-1-chris@chris-wilson.co.uk Cc: # v4.10-rc1+ Reviewed-by: Joonas Lahtinen (cherry picked from commit eb955eee27d9dc176871540c43c9070ee4701642) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_drv.h | 5 +++++ drivers/gpu/drm/i915/intel_display.c | 28 ++++++++++++++++++++++++++-- drivers/gpu/drm/i915/intel_drv.h | 2 ++ 3 files changed, 33 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 243224aeabf8..69bc3b0c4390 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1977,6 +1977,11 @@ struct drm_i915_private { struct i915_frontbuffer_tracking fb_tracking; + struct intel_atomic_helper { + struct llist_head free_list; + struct work_struct free_work; + } atomic_helper; + u16 orig_clock; bool mchbar_need_disable; diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index f6dd073afdc9..77f7b1d849a4 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -14523,8 +14523,14 @@ intel_atomic_commit_ready(struct i915_sw_fence *fence, break; case FENCE_FREE: - drm_atomic_state_put(&state->base); - break; + { + struct intel_atomic_helper *helper = + &to_i915(state->base.dev)->atomic_helper; + + if (llist_add(&state->freed, &helper->free_list)) + schedule_work(&helper->free_work); + break; + } } return NOTIFY_DONE; @@ -16403,6 +16409,18 @@ fail: drm_modeset_acquire_fini(&ctx); } +static void intel_atomic_helper_free_state(struct work_struct *work) +{ + struct drm_i915_private *dev_priv = + container_of(work, typeof(*dev_priv), atomic_helper.free_work); + struct intel_atomic_state *state, *next; + struct llist_node *freed; + + freed = llist_del_all(&dev_priv->atomic_helper.free_list); + llist_for_each_entry_safe(state, next, freed, freed) + drm_atomic_state_put(&state->base); +} + int intel_modeset_init(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); @@ -16422,6 +16440,9 @@ int intel_modeset_init(struct drm_device *dev) dev->mode_config.funcs = &intel_mode_funcs; + INIT_WORK(&dev_priv->atomic_helper.free_work, + intel_atomic_helper_free_state); + intel_init_quirks(dev); intel_init_pm(dev_priv); @@ -17106,6 +17127,9 @@ void intel_modeset_cleanup(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + flush_work(&dev_priv->atomic_helper.free_work); + WARN_ON(!llist_empty(&dev_priv->atomic_helper.free_list)); + intel_disable_gt_powersave(dev_priv); /* diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index cd132c216a67..cd72ae171eeb 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -370,6 +370,8 @@ struct intel_atomic_state { struct skl_wm_values wm_results; struct i915_sw_fence commit_ready; + + struct llist_node freed; }; struct intel_plane_state { -- cgit v1.2.3 From 2f5db26c2ecb248bdc319feb2990453cb02fc950 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Fri, 20 Jan 2017 13:51:23 -0800 Subject: drm/i915: reinstate call to trace_i915_vma_bind The call went away in: commit 3b16525cc4c1a43e9053cfdc414356eea24bdfad Author: Chris Wilson Date: Thu Aug 4 16:32:25 2016 +0100 drm/i915: Split insertion/binding of an object into the VM It is useful to have this trace as it pairs nicely with the vma_unbind one to track vma activity. Added inside the i915_vma_bind function (was outside before) to keep a similar placement as trace_i915_vma_unbind. v2: print bind_flags instead of flags (Chris) Fixes: 3b16525cc4c1 ("drm/i915: Split insertion/binding of an object into the VM") Cc: Chris Wilson Signed-off-by: Daniele Ceraolo Spurio Link: http://patchwork.freedesktop.org/patch/msgid/1484949083-11430-1-git-send-email-daniele.ceraolospurio@intel.com Reviewed-by: Chris Wilson Signed-off-by: Chris Wilson (cherry picked from commit 6146e6da5c961735dacf9b6c0c8b5f1382193ee2) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_vma.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index a792dcb902b5..e924a9516079 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -185,6 +185,7 @@ int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level cache_level, return ret; } + trace_i915_vma_bind(vma, bind_flags); ret = vma->vm->bind_vma(vma, cache_level, bind_flags); if (ret) return ret; -- cgit v1.2.3 From 83e526f2a2fa4b2e82b6bd3ddbb26b70acfa8947 Mon Sep 17 00:00:00 2001 From: Vincent Pelletier Date: Wed, 18 Jan 2017 00:57:44 +0000 Subject: usb: gadget: f_fs: Assorted buffer overflow checks. OS descriptor head, when flagged as provided, is accessed without checking if it fits in provided buffer. Verify length before access. Also, there are other places where buffer length it checked after accessing offsets which are potentially past the end. Check buffer length before as well to fail cleanly. Signed-off-by: Vincent Pelletier Acked-by: Felipe Balbi Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/function/f_fs.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c index 5490fc51638e..fd80c1b9c823 100644 --- a/drivers/usb/gadget/function/f_fs.c +++ b/drivers/usb/gadget/function/f_fs.c @@ -2269,6 +2269,8 @@ static int __ffs_data_do_os_desc(enum ffs_os_desc_type type, if (len < sizeof(*d) || h->interface >= ffs->interfaces_count) return -EINVAL; length = le32_to_cpu(d->dwSize); + if (len < length) + return -EINVAL; type = le32_to_cpu(d->dwPropertyDataType); if (type < USB_EXT_PROP_UNICODE || type > USB_EXT_PROP_UNICODE_MULTI) { @@ -2277,6 +2279,11 @@ static int __ffs_data_do_os_desc(enum ffs_os_desc_type type, return -EINVAL; } pnl = le16_to_cpu(d->wPropertyNameLength); + if (length < 14 + pnl) { + pr_vdebug("invalid os descriptor length: %d pnl:%d (descriptor %d)\n", + length, pnl, type); + return -EINVAL; + } pdl = le32_to_cpu(*(u32 *)((u8 *)data + 10 + pnl)); if (length != 14 + pnl + pdl) { pr_vdebug("invalid os descriptor length: %d pnl:%d pdl:%d (descriptor %d)\n", @@ -2363,6 +2370,9 @@ static int __ffs_data_got_descs(struct ffs_data *ffs, } } if (flags & (1 << i)) { + if (len < 4) { + goto error; + } os_descs_count = get_unaligned_le32(data); data += 4; len -= 4; @@ -2435,7 +2445,8 @@ static int __ffs_data_got_strings(struct ffs_data *ffs, ENTER(); - if (unlikely(get_unaligned_le32(data) != FUNCTIONFS_STRINGS_MAGIC || + if (unlikely(len < 16 || + get_unaligned_le32(data) != FUNCTIONFS_STRINGS_MAGIC || get_unaligned_le32(data + 4) != len)) goto error; str_count = get_unaligned_le32(data + 8); -- cgit v1.2.3 From fd25ea29093e275195d0ae8b2573021a1c98959f Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Sun, 22 Jan 2017 13:24:05 +0100 Subject: Revert "ACPI / video: Add force_native quirk for HP Pavilion dv6" Revert commit 6276e53fa8c0 (ACPI / video: Add force_native quirk for HP Pavilion dv6). In the commit message for the quirk this revert removes I wrote: "Note that there are quite a few HP Pavilion dv6 variants, some woth ATI and some with NVIDIA hybrid gfx, both seem to need this quirk to have working backlight control. There are also some versions with only Intel integrated gfx, these may not need this quirk, but it should not hurt there." Unfortunately that seems wrong, I've already received 2 reports of this commit causing regressions on some dv6 variants (at least one of which actually has a nvidia GPU). So it seems that HP has made a mess here by using the same model-name both in marketing and in the DMI data for many different variants. Some of which need acpi_backlight=native for functional backlight control (as the quirk this commit reverts was doing), where as others are broken by it. So lets get back to the old sitation so as to avoid regressing on models which used to work without any kernel cmdline arguments before. Fixes: 6276e53fa8c0 (ACPI / video: Add force_native quirk for HP Pavilion dv6) Signed-off-by: Hans de Goede Signed-off-by: Rafael J. Wysocki --- drivers/acpi/video_detect.c | 11 ----------- 1 file changed, 11 deletions(-) diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c index 02ded25c82e4..7f48156cbc0c 100644 --- a/drivers/acpi/video_detect.c +++ b/drivers/acpi/video_detect.c @@ -305,17 +305,6 @@ static const struct dmi_system_id video_detect_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "Dell System XPS L702X"), }, }, - { - /* https://bugzilla.redhat.com/show_bug.cgi?id=1204476 */ - /* https://bugs.launchpad.net/ubuntu/+source/linux-lts-trusty/+bug/1416940 */ - .callback = video_detect_force_native, - .ident = "HP Pavilion dv6", - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"), - DMI_MATCH(DMI_PRODUCT_NAME, "HP Pavilion dv6 Notebook PC"), - }, - }, - { }, }; -- cgit v1.2.3 From 4dfa2b84118fd6c95202ae87e62adf5000ccd4d0 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 25 Jan 2017 07:49:34 -0800 Subject: xfs: only update mount/resv fields on success in __xfs_ag_resv_init Try to reserve the blocks first and only then update the fields in or hanging off the mount structure. This way we can call __xfs_ag_resv_init again after a previous failure. Signed-off-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag_resv.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag_resv.c b/fs/xfs/libxfs/xfs_ag_resv.c index d346d42c54d1..94234bff40dc 100644 --- a/fs/xfs/libxfs/xfs_ag_resv.c +++ b/fs/xfs/libxfs/xfs_ag_resv.c @@ -200,22 +200,27 @@ __xfs_ag_resv_init( struct xfs_mount *mp = pag->pag_mount; struct xfs_ag_resv *resv; int error; + xfs_extlen_t reserved; - resv = xfs_perag_resv(pag, type); if (used > ask) ask = used; - resv->ar_asked = ask; - resv->ar_reserved = resv->ar_orig_reserved = ask - used; - mp->m_ag_max_usable -= ask; + reserved = ask - used; - trace_xfs_ag_resv_init(pag, type, ask); - - error = xfs_mod_fdblocks(mp, -(int64_t)resv->ar_reserved, true); - if (error) + error = xfs_mod_fdblocks(mp, -(int64_t)reserved, true); + if (error) { trace_xfs_ag_resv_init_error(pag->pag_mount, pag->pag_agno, error, _RET_IP_); + return error; + } - return error; + mp->m_ag_max_usable -= ask; + + resv = xfs_perag_resv(pag, type); + resv->ar_asked = ask; + resv->ar_reserved = resv->ar_orig_reserved = reserved; + + trace_xfs_ag_resv_init(pag, type, ask); + return 0; } /* Create a per-AG block reservation. */ -- cgit v1.2.3 From 76d771b4cbe33c581bd6ca2710c120be51172440 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 25 Jan 2017 07:49:35 -0800 Subject: xfs: use per-AG reservations for the finobt Currently we try to rely on the global reserved block pool for block allocations for the free inode btree, but I have customer reports (fairly complex workload, need to find an easier reproducer) where that is not enough as the AG where we free an inode that requires a new finobt block is entirely full. This causes us to cancel a dirty transaction and thus a file system shutdown. I think the right way to guard against this is to treat the finot the same way as the refcount btree and have a per-AG reservations for the possible worst case size of it, and the patch below implements that. Note that this could increase mount times with large finobt trees. In an ideal world we would have added a field for the number of finobt fields to the AGI, similar to what we did for the refcount blocks. We should do add it next time we rev the AGI or AGF format by adding new fields. Signed-off-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_ag_resv.c | 47 ++++++++++++++++++--- fs/xfs/libxfs/xfs_ialloc_btree.c | 90 ++++++++++++++++++++++++++++++++++++++-- fs/xfs/libxfs/xfs_ialloc_btree.h | 3 ++ fs/xfs/xfs_inode.c | 23 +++++----- fs/xfs/xfs_mount.h | 1 + 5 files changed, 144 insertions(+), 20 deletions(-) diff --git a/fs/xfs/libxfs/xfs_ag_resv.c b/fs/xfs/libxfs/xfs_ag_resv.c index 94234bff40dc..33db69be4832 100644 --- a/fs/xfs/libxfs/xfs_ag_resv.c +++ b/fs/xfs/libxfs/xfs_ag_resv.c @@ -39,6 +39,7 @@ #include "xfs_rmap_btree.h" #include "xfs_btree.h" #include "xfs_refcount_btree.h" +#include "xfs_ialloc_btree.h" /* * Per-AG Block Reservations @@ -210,6 +211,9 @@ __xfs_ag_resv_init( if (error) { trace_xfs_ag_resv_init_error(pag->pag_mount, pag->pag_agno, error, _RET_IP_); + xfs_warn(mp, +"Per-AG reservation for AG %u failed. Filesystem may run out of space.", + pag->pag_agno); return error; } @@ -228,6 +232,8 @@ int xfs_ag_resv_init( struct xfs_perag *pag) { + struct xfs_mount *mp = pag->pag_mount; + xfs_agnumber_t agno = pag->pag_agno; xfs_extlen_t ask; xfs_extlen_t used; int error = 0; @@ -236,23 +242,45 @@ xfs_ag_resv_init( if (pag->pag_meta_resv.ar_asked == 0) { ask = used = 0; - error = xfs_refcountbt_calc_reserves(pag->pag_mount, - pag->pag_agno, &ask, &used); + error = xfs_refcountbt_calc_reserves(mp, agno, &ask, &used); if (error) goto out; - error = __xfs_ag_resv_init(pag, XFS_AG_RESV_METADATA, - ask, used); + error = xfs_finobt_calc_reserves(mp, agno, &ask, &used); if (error) goto out; + + error = __xfs_ag_resv_init(pag, XFS_AG_RESV_METADATA, + ask, used); + if (error) { + /* + * Because we didn't have per-AG reservations when the + * finobt feature was added we might not be able to + * reserve all needed blocks. Warn and fall back to the + * old and potentially buggy code in that case, but + * ensure we do have the reservation for the refcountbt. + */ + ask = used = 0; + + mp->m_inotbt_nores = true; + + error = xfs_refcountbt_calc_reserves(mp, agno, &ask, + &used); + if (error) + goto out; + + error = __xfs_ag_resv_init(pag, XFS_AG_RESV_METADATA, + ask, used); + if (error) + goto out; + } } /* Create the AGFL metadata reservation */ if (pag->pag_agfl_resv.ar_asked == 0) { ask = used = 0; - error = xfs_rmapbt_calc_reserves(pag->pag_mount, pag->pag_agno, - &ask, &used); + error = xfs_rmapbt_calc_reserves(mp, agno, &ask, &used); if (error) goto out; @@ -261,9 +289,16 @@ xfs_ag_resv_init( goto out; } +#ifdef DEBUG + /* need to read in the AGF for the ASSERT below to work */ + error = xfs_alloc_pagf_init(pag->pag_mount, NULL, pag->pag_agno, 0); + if (error) + return error; + ASSERT(xfs_perag_resv(pag, XFS_AG_RESV_METADATA)->ar_reserved + xfs_perag_resv(pag, XFS_AG_RESV_AGFL)->ar_reserved <= pag->pagf_freeblks + pag->pagf_flcount); +#endif out: return error; } diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c index 0fd086d03d41..7c471881c9a6 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.c +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c @@ -82,11 +82,12 @@ xfs_finobt_set_root( } STATIC int -xfs_inobt_alloc_block( +__xfs_inobt_alloc_block( struct xfs_btree_cur *cur, union xfs_btree_ptr *start, union xfs_btree_ptr *new, - int *stat) + int *stat, + enum xfs_ag_resv_type resv) { xfs_alloc_arg_t args; /* block allocation args */ int error; /* error return value */ @@ -103,6 +104,7 @@ xfs_inobt_alloc_block( args.maxlen = 1; args.prod = 1; args.type = XFS_ALLOCTYPE_NEAR_BNO; + args.resv = resv; error = xfs_alloc_vextent(&args); if (error) { @@ -122,6 +124,27 @@ xfs_inobt_alloc_block( return 0; } +STATIC int +xfs_inobt_alloc_block( + struct xfs_btree_cur *cur, + union xfs_btree_ptr *start, + union xfs_btree_ptr *new, + int *stat) +{ + return __xfs_inobt_alloc_block(cur, start, new, stat, XFS_AG_RESV_NONE); +} + +STATIC int +xfs_finobt_alloc_block( + struct xfs_btree_cur *cur, + union xfs_btree_ptr *start, + union xfs_btree_ptr *new, + int *stat) +{ + return __xfs_inobt_alloc_block(cur, start, new, stat, + XFS_AG_RESV_METADATA); +} + STATIC int xfs_inobt_free_block( struct xfs_btree_cur *cur, @@ -328,7 +351,7 @@ static const struct xfs_btree_ops xfs_finobt_ops = { .dup_cursor = xfs_inobt_dup_cursor, .set_root = xfs_finobt_set_root, - .alloc_block = xfs_inobt_alloc_block, + .alloc_block = xfs_finobt_alloc_block, .free_block = xfs_inobt_free_block, .get_minrecs = xfs_inobt_get_minrecs, .get_maxrecs = xfs_inobt_get_maxrecs, @@ -480,3 +503,64 @@ xfs_inobt_rec_check_count( return 0; } #endif /* DEBUG */ + +static xfs_extlen_t +xfs_inobt_max_size( + struct xfs_mount *mp) +{ + /* Bail out if we're uninitialized, which can happen in mkfs. */ + if (mp->m_inobt_mxr[0] == 0) + return 0; + + return xfs_btree_calc_size(mp, mp->m_inobt_mnr, + (uint64_t)mp->m_sb.sb_agblocks * mp->m_sb.sb_inopblock / + XFS_INODES_PER_CHUNK); +} + +static int +xfs_inobt_count_blocks( + struct xfs_mount *mp, + xfs_agnumber_t agno, + xfs_btnum_t btnum, + xfs_extlen_t *tree_blocks) +{ + struct xfs_buf *agbp; + struct xfs_btree_cur *cur; + int error; + + error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp); + if (error) + return error; + + cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno, btnum); + error = xfs_btree_count_blocks(cur, tree_blocks); + xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR); + xfs_buf_relse(agbp); + + return error; +} + +/* + * Figure out how many blocks to reserve and how many are used by this btree. + */ +int +xfs_finobt_calc_reserves( + struct xfs_mount *mp, + xfs_agnumber_t agno, + xfs_extlen_t *ask, + xfs_extlen_t *used) +{ + xfs_extlen_t tree_len = 0; + int error; + + if (!xfs_sb_version_hasfinobt(&mp->m_sb)) + return 0; + + error = xfs_inobt_count_blocks(mp, agno, XFS_BTNUM_FINO, &tree_len); + if (error) + return error; + + *ask += xfs_inobt_max_size(mp); + *used += tree_len; + return 0; +} diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.h b/fs/xfs/libxfs/xfs_ialloc_btree.h index bd88453217ce..aa81e2e63f3f 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.h +++ b/fs/xfs/libxfs/xfs_ialloc_btree.h @@ -72,4 +72,7 @@ int xfs_inobt_rec_check_count(struct xfs_mount *, #define xfs_inobt_rec_check_count(mp, rec) 0 #endif /* DEBUG */ +int xfs_finobt_calc_reserves(struct xfs_mount *mp, xfs_agnumber_t agno, + xfs_extlen_t *ask, xfs_extlen_t *used); + #endif /* __XFS_IALLOC_BTREE_H__ */ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index b9557795eb74..de32f0fe47c8 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1792,22 +1792,23 @@ xfs_inactive_ifree( int error; /* - * The ifree transaction might need to allocate blocks for record - * insertion to the finobt. We don't want to fail here at ENOSPC, so - * allow ifree to dip into the reserved block pool if necessary. - * - * Freeing large sets of inodes generally means freeing inode chunks, - * directory and file data blocks, so this should be relatively safe. - * Only under severe circumstances should it be possible to free enough - * inodes to exhaust the reserve block pool via finobt expansion while - * at the same time not creating free space in the filesystem. + * We try to use a per-AG reservation for any block needed by the finobt + * tree, but as the finobt feature predates the per-AG reservation + * support a degraded file system might not have enough space for the + * reservation at mount time. In that case try to dip into the reserved + * pool and pray. * * Send a warning if the reservation does happen to fail, as the inode * now remains allocated and sits on the unlinked list until the fs is * repaired. */ - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_ifree, - XFS_IFREE_SPACE_RES(mp), 0, XFS_TRANS_RESERVE, &tp); + if (unlikely(mp->m_inotbt_nores)) { + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_ifree, + XFS_IFREE_SPACE_RES(mp), 0, XFS_TRANS_RESERVE, + &tp); + } else { + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_ifree, 0, 0, 0, &tp); + } if (error) { if (error == -ENOSPC) { xfs_warn_ratelimited(mp, diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 84f785218907..7f351f706b7a 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -140,6 +140,7 @@ typedef struct xfs_mount { int m_fixedfsid[2]; /* unchanged for life of FS */ uint m_dmevmask; /* DMI events for this FS */ __uint64_t m_flags; /* global mount flags */ + bool m_inotbt_nores; /* no per-AG finobt resv. */ int m_ialloc_inos; /* inodes in inode allocation */ int m_ialloc_blks; /* blocks in inode allocation */ int m_ialloc_min_blks;/* min blocks in sparse inode -- cgit v1.2.3 From 5a93790d4e2df73e30c965ec6e49be82fc3ccfce Mon Sep 17 00:00:00 2001 From: Brian Foster Date: Wed, 25 Jan 2017 07:53:43 -0800 Subject: xfs: remove racy hasattr check from attr ops xfs_attr_[get|remove]() have unlocked attribute fork checks to optimize away a lock cycle in cases where the fork does not exist or is otherwise empty. This check is not safe, however, because an attribute fork short form to extent format conversion includes a transient state that causes the xfs_inode_hasattr() check to fail. Specifically, xfs_attr_shortform_to_leaf() creates an empty extent format attribute fork and then adds the existing shortform attributes to it. This means that lookup of an existing xattr can spuriously return -ENOATTR when racing against a setxattr that causes the associated format conversion. This was originally reproduced by an untar on a particularly configured glusterfs volume, but can also be reproduced on demand with properly crafted xattr requests. The format conversion occurs under the exclusive ilock. xfs_attr_get() and xfs_attr_remove() already have the proper locking and checks further down in the functions to handle this situation correctly. Drop the unlocked checks to avoid the spurious failure and rely on the existing logic. Signed-off-by: Brian Foster Reviewed-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_attr.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index af1ecb19121e..6622d46ddec3 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -131,9 +131,6 @@ xfs_attr_get( if (XFS_FORCED_SHUTDOWN(ip->i_mount)) return -EIO; - if (!xfs_inode_hasattr(ip)) - return -ENOATTR; - error = xfs_attr_args_init(&args, ip, name, flags); if (error) return error; @@ -392,9 +389,6 @@ xfs_attr_remove( if (XFS_FORCED_SHUTDOWN(dp->i_mount)) return -EIO; - if (!xfs_inode_hasattr(dp)) - return -ENOATTR; - error = xfs_attr_args_init(&args, dp, name, flags); if (error) return error; -- cgit v1.2.3 From 493611ebd62673f39e2f52c2561182c558a21cb6 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 25 Jan 2017 08:59:43 -0800 Subject: xfs: extsize hints are not unlikely in xfs_bmap_btalloc With COW files they are the hotpath, just like for files with the extent size hint attribute. We really shouldn't micro-manage anything but failure cases with unlikely. Additionally Arnd Bergmann recently reported that one of these two unlikely annotations causes link failures together with an upcoming kernel instrumentation patch, so let's get rid of it ASAP. Signed-off-by: Christoph Hellwig Reported-by: Arnd Bergmann Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_bmap.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index ab82dd4a4980..bfc00de5c6f1 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3629,7 +3629,7 @@ xfs_bmap_btalloc( align = xfs_get_cowextsz_hint(ap->ip); else if (xfs_alloc_is_userdata(ap->datatype)) align = xfs_get_extsz_hint(ap->ip); - if (unlikely(align)) { + if (align) { error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev, align, 0, ap->eof, 0, ap->conv, &ap->offset, &ap->length); @@ -3701,7 +3701,7 @@ xfs_bmap_btalloc( args.minlen = ap->minlen; } /* apply extent size hints if obtained earlier */ - if (unlikely(align)) { + if (align) { args.prod = align; if ((args.mod = (xfs_extlen_t)do_mod(ap->offset, args.prod))) args.mod = (xfs_extlen_t)(args.prod - args.mod); -- cgit v1.2.3 From 6f29a130613191d3c6335169febe002cba00edf5 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 24 Jan 2017 14:01:53 +0800 Subject: sctp: sctp_addr_id2transport should verify the addr before looking up assoc sctp_addr_id2transport is a function for sockopt to look up assoc by address. As the address is from userspace, it can be a v4-mapped v6 address. But in sctp protocol stack, it always handles a v4-mapped v6 address as a v4 address. So it's necessary to convert it to a v4 address before looking up assoc by address. This patch is to fix it by calling sctp_verify_addr in which it can do this conversion before calling sctp_endpoint_lookup_assoc, just like what sctp_sendmsg and __sctp_connect do for the address from users. Signed-off-by: Xin Long Acked-by: Neil Horman Signed-off-by: David S. Miller --- net/sctp/socket.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 318c6786d653..37eeab7899fc 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -235,8 +235,12 @@ static struct sctp_transport *sctp_addr_id2transport(struct sock *sk, sctp_assoc_t id) { struct sctp_association *addr_asoc = NULL, *id_asoc = NULL; - struct sctp_transport *transport; + struct sctp_af *af = sctp_get_af_specific(addr->ss_family); union sctp_addr *laddr = (union sctp_addr *)addr; + struct sctp_transport *transport; + + if (sctp_verify_addr(sk, laddr, af->sockaddr_len)) + return NULL; addr_asoc = sctp_endpoint_lookup_assoc(sctp_sk(sk)->ep, laddr, -- cgit v1.2.3 From 5207f3996338e1db71363fe381c81aaf1e54e4e3 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 24 Jan 2017 14:05:16 +0800 Subject: sctp: sctp gso should set feature with NETIF_F_SG when calling skb_segment Now sctp gso puts segments into skb's frag_list, then processes these segments in skb_segment. But skb_segment handles them only when gs is enabled, as it's in the same branch with skb's frags. Although almost all the NICs support sg other than some old ones, but since commit 1e16aa3ddf86 ("net: gso: use feature flag argument in all protocol gso handlers"), features &= skb->dev->hw_enc_features, and xfrm_output_gso call skb_segment with features = 0, which means sctp gso would call skb_segment with sg = 0, and skb_segment would not work as expected. This patch is to fix it by setting features param with NETIF_F_SG when calling skb_segment so that it can go the right branch to process the skb's frag_list. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- net/sctp/offload.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sctp/offload.c b/net/sctp/offload.c index 7e869d0cca69..4f5a2b580aa5 100644 --- a/net/sctp/offload.c +++ b/net/sctp/offload.c @@ -68,7 +68,7 @@ static struct sk_buff *sctp_gso_segment(struct sk_buff *skb, goto out; } - segs = skb_segment(skb, features | NETIF_F_HW_CSUM); + segs = skb_segment(skb, features | NETIF_F_HW_CSUM | NETIF_F_SG); if (IS_ERR(segs)) goto out; -- cgit v1.2.3 From 56d806222ace4c3aeae516cd7a855340fb2839d8 Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Tue, 24 Jan 2017 21:49:41 -0500 Subject: tcp: correct memory barrier usage in tcp_check_space() sock_reset_flag() maps to __clear_bit() not the atomic version clear_bit(). Thus, we need smp_mb(), smp_mb__after_atomic() is not sufficient. Fixes: 3c7151275c0c ("tcp: add memory barriers to write space paths") Cc: Eric Dumazet Cc: Oleg Nesterov Signed-off-by: Jason Baron Acked-by: Eric Dumazet Reported-by: Oleg Nesterov Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 6c790754ae3e..41dcbd568cbe 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5078,7 +5078,7 @@ static void tcp_check_space(struct sock *sk) if (sock_flag(sk, SOCK_QUEUE_SHRUNK)) { sock_reset_flag(sk, SOCK_QUEUE_SHRUNK); /* pairs with tcp_poll() */ - smp_mb__after_atomic(); + smp_mb(); if (sk->sk_socket && test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { tcp_new_space(sk); -- cgit v1.2.3 From a551ee94ea723b4af9b827c7460f108bc13425ee Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 25 Jan 2017 02:55:07 -0500 Subject: bnxt_en: Fix bnxt_reset() in the slow path task. In bnxt_sp_task(), we set a bit BNXT_STATE_IN_SP_TASK so that bnxt_close() will synchronize and wait for bnxt_sp_task() to finish. Some functions in bnxt_sp_task() require us to clear BNXT_STATE_IN_SP_TASK and then acquire rtnl_lock() to prevent race conditions. There are some bugs related to this logic. This patch refactors the code to have common bnxt_rtnl_lock_sp() and bnxt_rtnl_unlock_sp() to handle the RTNL and the clearing/setting of the bit. Multiple functions will need the same logic. We also need to move bnxt_reset() to the end of bnxt_sp_task(). Functions that clear BNXT_STATE_IN_SP_TASK must be the last functions to be called in bnxt_sp_task(). The common scheme will handle the condition properly. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 38 ++++++++++++++++++++----------- 1 file changed, 25 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 53e686fdf2ce..30d7d643bdbe 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -6200,23 +6200,32 @@ bnxt_restart_timer: mod_timer(&bp->timer, jiffies + bp->current_interval); } -/* Only called from bnxt_sp_task() */ -static void bnxt_reset(struct bnxt *bp, bool silent) +static void bnxt_rtnl_lock_sp(struct bnxt *bp) { - /* bnxt_reset_task() calls bnxt_close_nic() which waits - * for BNXT_STATE_IN_SP_TASK to clear. - * If there is a parallel dev_close(), bnxt_close() may be holding + /* We are called from bnxt_sp_task which has BNXT_STATE_IN_SP_TASK + * set. If the device is being closed, bnxt_close() may be holding * rtnl() and waiting for BNXT_STATE_IN_SP_TASK to clear. So we * must clear BNXT_STATE_IN_SP_TASK before holding rtnl(). */ clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state); rtnl_lock(); - if (test_bit(BNXT_STATE_OPEN, &bp->state)) - bnxt_reset_task(bp, silent); +} + +static void bnxt_rtnl_unlock_sp(struct bnxt *bp) +{ set_bit(BNXT_STATE_IN_SP_TASK, &bp->state); rtnl_unlock(); } +/* Only called from bnxt_sp_task() */ +static void bnxt_reset(struct bnxt *bp, bool silent) +{ + bnxt_rtnl_lock_sp(bp); + if (test_bit(BNXT_STATE_OPEN, &bp->state)) + bnxt_reset_task(bp, silent); + bnxt_rtnl_unlock_sp(bp); +} + static void bnxt_cfg_ntp_filters(struct bnxt *); static void bnxt_sp_task(struct work_struct *work) @@ -6266,18 +6275,21 @@ static void bnxt_sp_task(struct work_struct *work) bnxt_hwrm_tunnel_dst_port_free( bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE); } - if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event)) - bnxt_reset(bp, false); - - if (test_and_clear_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event)) - bnxt_reset(bp, true); - if (test_and_clear_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event)) bnxt_get_port_module_status(bp); if (test_and_clear_bit(BNXT_PERIODIC_STATS_SP_EVENT, &bp->sp_event)) bnxt_hwrm_port_qstats(bp); + /* These functions below will clear BNXT_STATE_IN_SP_TASK. They + * must be the last functions to be called before exiting. + */ + if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event)) + bnxt_reset(bp, false); + + if (test_and_clear_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event)) + bnxt_reset(bp, true); + smp_mb__before_atomic(); clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state); } -- cgit v1.2.3 From 0eaa24b971ae251ae9d3be23f77662a655532063 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 25 Jan 2017 02:55:08 -0500 Subject: bnxt_en: Fix RTNL lock usage on bnxt_update_link(). bnxt_update_link() is called from multiple code paths. Most callers, such as open, ethtool, already hold RTNL. Only the caller bnxt_sp_task() does not. So it is a bug to take RTNL inside bnxt_update_link(). Fix it by removing the RTNL inside bnxt_update_link(). The function now expects the caller to always hold RTNL. In bnxt_sp_task(), call bnxt_rtnl_lock_sp() before calling bnxt_update_link(). We also need to move the call to the end of bnxt_sp_task() since it will be clearing the BNXT_STATE_IN_SP_TASK bit. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 37 +++++++++++++++---------------- 1 file changed, 18 insertions(+), 19 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 30d7d643bdbe..69b0fce4f5b3 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -5314,17 +5314,12 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state) if ((link_info->support_auto_speeds | diff) != link_info->support_auto_speeds) { /* An advertised speed is no longer supported, so we need to - * update the advertisement settings. See bnxt_reset() for - * comments about the rtnl_lock() sequence below. + * update the advertisement settings. Caller holds RTNL + * so we can modify link settings. */ - clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state); - rtnl_lock(); link_info->advertising = link_info->support_auto_speeds; - if (test_bit(BNXT_STATE_OPEN, &bp->state) && - (link_info->autoneg & BNXT_AUTONEG_SPEED)) + if (link_info->autoneg & BNXT_AUTONEG_SPEED) bnxt_hwrm_set_link_setting(bp, true, false); - set_bit(BNXT_STATE_IN_SP_TASK, &bp->state); - rtnl_unlock(); } return 0; } @@ -6231,7 +6226,6 @@ static void bnxt_cfg_ntp_filters(struct bnxt *); static void bnxt_sp_task(struct work_struct *work) { struct bnxt *bp = container_of(work, struct bnxt, sp_task); - int rc; set_bit(BNXT_STATE_IN_SP_TASK, &bp->state); smp_mb__after_atomic(); @@ -6245,16 +6239,6 @@ static void bnxt_sp_task(struct work_struct *work) if (test_and_clear_bit(BNXT_RX_NTP_FLTR_SP_EVENT, &bp->sp_event)) bnxt_cfg_ntp_filters(bp); - if (test_and_clear_bit(BNXT_LINK_CHNG_SP_EVENT, &bp->sp_event)) { - if (test_and_clear_bit(BNXT_LINK_SPEED_CHNG_SP_EVENT, - &bp->sp_event)) - bnxt_hwrm_phy_qcaps(bp); - - rc = bnxt_update_link(bp, true); - if (rc) - netdev_err(bp->dev, "SP task can't update link (rc: %x)\n", - rc); - } if (test_and_clear_bit(BNXT_HWRM_EXEC_FWD_REQ_SP_EVENT, &bp->sp_event)) bnxt_hwrm_exec_fwd_req(bp); if (test_and_clear_bit(BNXT_VXLAN_ADD_PORT_SP_EVENT, &bp->sp_event)) { @@ -6284,6 +6268,21 @@ static void bnxt_sp_task(struct work_struct *work) /* These functions below will clear BNXT_STATE_IN_SP_TASK. They * must be the last functions to be called before exiting. */ + if (test_and_clear_bit(BNXT_LINK_CHNG_SP_EVENT, &bp->sp_event)) { + int rc = 0; + + if (test_and_clear_bit(BNXT_LINK_SPEED_CHNG_SP_EVENT, + &bp->sp_event)) + bnxt_hwrm_phy_qcaps(bp); + + bnxt_rtnl_lock_sp(bp); + if (test_bit(BNXT_STATE_OPEN, &bp->state)) + rc = bnxt_update_link(bp, true); + bnxt_rtnl_unlock_sp(bp); + if (rc) + netdev_err(bp->dev, "SP task can't update link (rc: %x)\n", + rc); + } if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event)) bnxt_reset(bp, false); -- cgit v1.2.3 From 90c694bb71819fb5bd3501ac397307d7e41ddeca Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 25 Jan 2017 02:55:09 -0500 Subject: bnxt_en: Fix RTNL lock usage on bnxt_get_port_module_status(). bnxt_get_port_module_status() calls bnxt_update_link() which expects RTNL to be held. In bnxt_sp_task() that does not hold RTNL, we need to call it with a prior call to bnxt_rtnl_lock_sp() and the call needs to be moved to the end of bnxt_sp_task(). Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 69b0fce4f5b3..4fcc6a84a087 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -6259,9 +6259,6 @@ static void bnxt_sp_task(struct work_struct *work) bnxt_hwrm_tunnel_dst_port_free( bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE); } - if (test_and_clear_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event)) - bnxt_get_port_module_status(bp); - if (test_and_clear_bit(BNXT_PERIODIC_STATS_SP_EVENT, &bp->sp_event)) bnxt_hwrm_port_qstats(bp); @@ -6283,6 +6280,12 @@ static void bnxt_sp_task(struct work_struct *work) netdev_err(bp->dev, "SP task can't update link (rc: %x)\n", rc); } + if (test_and_clear_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event)) { + bnxt_rtnl_lock_sp(bp); + if (test_bit(BNXT_STATE_OPEN, &bp->state)) + bnxt_get_port_module_status(bp); + bnxt_rtnl_unlock_sp(bp); + } if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event)) bnxt_reset(bp, false); -- cgit v1.2.3 From 61976fff20f92aceecc3670f6168bfc57a79e047 Mon Sep 17 00:00:00 2001 From: John Crispin Date: Wed, 25 Jan 2017 09:20:54 +0100 Subject: Documentation: devicetree: change the mediatek ethernet compatible string When the binding was defined, I was not aware that mt2701 was an earlier version of the SoC. For sake of consistency, the ethernet driver should use mt2701 inside the compat string as this is the earliest SoC with the ethernet core. The ethernet driver is currently of no real use until we finish and upstream the DSA driver. There are no users of this binding yet. It should be safe to fix this now before it is too late and we need to provide backward compatibility for the mt7623-eth compat string. Reported-by: Sean Wang Signed-off-by: John Crispin Reviewed-by: Matthias Brugger Signed-off-by: David S. Miller --- Documentation/devicetree/bindings/net/mediatek-net.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/net/mediatek-net.txt b/Documentation/devicetree/bindings/net/mediatek-net.txt index c010fafc66a8..c7194e87d5f4 100644 --- a/Documentation/devicetree/bindings/net/mediatek-net.txt +++ b/Documentation/devicetree/bindings/net/mediatek-net.txt @@ -7,7 +7,7 @@ have dual GMAC each represented by a child node.. * Ethernet controller node Required properties: -- compatible: Should be "mediatek,mt7623-eth" +- compatible: Should be "mediatek,mt2701-eth" - reg: Address and length of the register set for the device - interrupts: Should contain the three frame engines interrupts in numeric order. These are fe_int0, fe_int1 and fe_int2. -- cgit v1.2.3 From 8b901f6bbcf12a20e43105d161bedde093431e61 Mon Sep 17 00:00:00 2001 From: John Crispin Date: Wed, 25 Jan 2017 09:20:55 +0100 Subject: net-next: ethernet: mediatek: change the compatible string When the binding was defined, I was not aware that mt2701 was an earlier version of the SoC. For sake of consistency, the ethernet driver should use mt2701 inside the compat string as this is the earliest SoC with the ethernet core. The ethernet driver is currently of no real use until we finish and upstream the DSA driver. There are no users of this binding yet. It should be safe to fix this now before it is too late and we need to provide backward compatibility for the mt7623-eth compat string. Reported-by: Sean Wang Signed-off-by: John Crispin Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 3dd87889e67e..1c29c86f8709 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -2517,7 +2517,7 @@ static int mtk_remove(struct platform_device *pdev) } const struct of_device_id of_mtk_match[] = { - { .compatible = "mediatek,mt7623-eth" }, + { .compatible = "mediatek,mt2701-eth" }, {}, }; MODULE_DEVICE_TABLE(of, of_mtk_match); -- cgit v1.2.3 From 8a87fca8dd5879eb05a0903cb7ea4fd2a3876ae0 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Wed, 25 Jan 2017 11:39:48 +0100 Subject: net: phy: leds: Clear phy_num_led_triggers on failure to avoid crash phy_attach_direct() ignores errors returned by phy_led_triggers_register(). I think that's OK, as LED triggers can be considered a non-critical feature. However, this causes problems later: - phy_led_trigger_change_speed() will access the array phy_device.phy_led_triggers, which has been freed in the error path of phy_led_triggers_register(), which may lead to a crash. - phy_led_triggers_unregister() will access the same array, leading to crashes during s2ram or poweroff, like: Unable to handle kernel NULL pointer dereference at virtual address 00000000 ... [] (__list_del_entry_valid) from [] (led_trigger_unregister+0x34/0xcc) [] (led_trigger_unregister) from [] (phy_led_triggers_unregister+0x28/0x34) [] (phy_led_triggers_unregister) from [] (phy_detach+0x30/0x74) [] (phy_detach) from [] (sh_eth_close+0x64/0x9c) [] (sh_eth_close) from [] (dpm_run_callback+0x48/0xc8) or: list_del corruption. prev->next should be dede6540, but was 2e323931 ------------[ cut here ]------------ kernel BUG at lib/list_debug.c:52! ... [] (__list_del_entry_valid) from [] (led_trigger_unregister+0x34/0xcc) [] (led_trigger_unregister) from [] (phy_led_triggers_unregister+0x28/0x34) [] (phy_led_triggers_unregister) from [] (phy_detach+0x30/0x74) [] (phy_detach) from [] (sh_eth_close+0x6c/0xa4) [] (sh_eth_close) from [] (__dev_close_many+0xac/0xd0) To fix this, clear phy_device.phy_num_led_triggers in the error path of phy_led_triggers_register() fails. Note that the "No phy led trigger registered for speed" message will still be printed on link speed changes, which is a good cue that something went wrong with the LED triggers. Fixes: 2e0bc452f4721520 ("net: phy: leds: add support for led triggers on phy link state change") Signed-off-by: Geert Uytterhoeven Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/phy_led_triggers.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/phy_led_triggers.c b/drivers/net/phy/phy_led_triggers.c index fa62bdf2f526..3f619e7371e9 100644 --- a/drivers/net/phy/phy_led_triggers.c +++ b/drivers/net/phy/phy_led_triggers.c @@ -102,8 +102,10 @@ int phy_led_triggers_register(struct phy_device *phy) sizeof(struct phy_led_trigger) * phy->phy_num_led_triggers, GFP_KERNEL); - if (!phy->phy_led_triggers) - return -ENOMEM; + if (!phy->phy_led_triggers) { + err = -ENOMEM; + goto out_clear; + } for (i = 0; i < phy->phy_num_led_triggers; i++) { err = phy_led_trigger_register(phy, &phy->phy_led_triggers[i], @@ -120,6 +122,8 @@ out_unreg: while (i--) phy_led_trigger_unregister(&phy->phy_led_triggers[i]); devm_kfree(&phy->mdio.dev, phy->phy_led_triggers); +out_clear: + phy->phy_num_led_triggers = 0; return err; } EXPORT_SYMBOL_GPL(phy_led_triggers_register); -- cgit v1.2.3 From d6f8cfa3dea294eabf8f302e90176dd6381fb66e Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Wed, 25 Jan 2017 11:39:49 +0100 Subject: net: phy: leds: Break dependency of phy.h on phy_led_triggers.h includes , which is not really needed. Drop the include from , and add it to all users that didn't include it explicitly. Suggested-by: Andrew Lunn Signed-off-by: Geert Uytterhoeven Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/phy/phy.c | 1 + drivers/net/phy/phy_led_triggers.c | 1 + include/linux/phy.h | 1 - 3 files changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index e687a9cb4a37..7cc1b7dcfe05 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/net/phy/phy_led_triggers.c b/drivers/net/phy/phy_led_triggers.c index 3f619e7371e9..94ca42e630bb 100644 --- a/drivers/net/phy/phy_led_triggers.c +++ b/drivers/net/phy/phy_led_triggers.c @@ -12,6 +12,7 @@ */ #include #include +#include #include static struct phy_led_trigger *phy_speed_to_led_trigger(struct phy_device *phy, diff --git a/include/linux/phy.h b/include/linux/phy.h index f7d95f644eed..7fc1105605bf 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -25,7 +25,6 @@ #include #include #include -#include #include -- cgit v1.2.3 From 3c880eb0205222bb062970085ebedc73ec8dfd14 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Wed, 25 Jan 2017 11:39:50 +0100 Subject: net: phy: leds: Fix truncated LED trigger names Commit 4567d686f5c6d955 ("phy: increase size of MII_BUS_ID_SIZE and bus_id") increased the size of MII bus IDs, but forgot to update the private definition in . This may cause: 1. Truncation of LED trigger names, 2. Duplicate LED trigger names, 3. Failures registering LED triggers, 4. Crashes due to bad error handling in the LED trigger failure path. To fix this, and prevent the definitions going out of sync again in the future, let the PHY LED trigger code use the existing MII_BUS_ID_SIZE definition. Example: - Before I had triggers "ee700000.etherne:01:100Mbps" and "ee700000.etherne:01:10Mbps", - After the increase of MII_BUS_ID_SIZE, both became "ee700000.ethernet-ffffffff:01:" => FAIL, - Now, the triggers are "ee700000.ethernet-ffffffff:01:100Mbps" and "ee700000.ethernet-ffffffff:01:10Mbps", which are unique again. Fixes: 4567d686f5c6d955 ("phy: increase size of MII_BUS_ID_SIZE and bus_id") Fixes: 2e0bc452f4721520 ("net: phy: leds: add support for led triggers on phy link state change") Signed-off-by: Geert Uytterhoeven Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- include/linux/phy_led_triggers.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/phy_led_triggers.h b/include/linux/phy_led_triggers.h index a2daea0a37d2..b37b05bfd1a6 100644 --- a/include/linux/phy_led_triggers.h +++ b/include/linux/phy_led_triggers.h @@ -18,11 +18,11 @@ struct phy_device; #ifdef CONFIG_LED_TRIGGER_PHY #include +#include #define PHY_LED_TRIGGER_SPEED_SUFFIX_SIZE 10 -#define PHY_MII_BUS_ID_SIZE (20 - 3) -#define PHY_LINK_LED_TRIGGER_NAME_SIZE (PHY_MII_BUS_ID_SIZE + \ +#define PHY_LINK_LED_TRIGGER_NAME_SIZE (MII_BUS_ID_SIZE + \ FIELD_SIZEOF(struct mdio_device, addr)+\ PHY_LED_TRIGGER_SPEED_SUFFIX_SIZE) -- cgit v1.2.3 From f154be241d22298d2b63c9b613f619fa1086ea75 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Wed, 25 Jan 2017 09:10:41 -0800 Subject: net: dsa: Bring back device detaching in dsa_slave_suspend() Commit 448b4482c671 ("net: dsa: Add lockdep class to tx queues to avoid lockdep splat") removed the netif_device_detach() call done in dsa_slave_suspend() which is necessary, and paired with a corresponding netif_device_attach(), bring it back. Fixes: 448b4482c671 ("net: dsa: Add lockdep class to tx queues to avoid lockdep splat") Signed-off-by: Florian Fainelli Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- net/dsa/slave.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/dsa/slave.c b/net/dsa/slave.c index ba1b6b9630d2..7d4596110851 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -1201,6 +1201,8 @@ int dsa_slave_suspend(struct net_device *slave_dev) { struct dsa_slave_priv *p = netdev_priv(slave_dev); + netif_device_detach(slave_dev); + if (p->phy) { phy_stop(p->phy); p->old_pause = -1; -- cgit v1.2.3 From 54a07c7bb0da0343734c78212bbe9f3735394962 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Thu, 26 Jan 2017 06:44:03 +1000 Subject: Revert "drm/probe-helpers: Drop locking from poll_enable" This reverts commit 3846fd9b86001bea171943cc3bb9222cb6da6b42. There were some precursor commits missing for this around connector locking, we should probably merge Lyude's nouveau avoid the problem patch. --- drivers/gpu/drm/drm_probe_helper.c | 51 ++++++++++++++++++++++-------------- drivers/gpu/drm/i915/intel_hotplug.c | 4 +-- include/drm/drm_crtc_helper.h | 1 + 3 files changed, 34 insertions(+), 22 deletions(-) diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c index b452a7ccd84b..cf8f0128c161 100644 --- a/drivers/gpu/drm/drm_probe_helper.c +++ b/drivers/gpu/drm/drm_probe_helper.c @@ -115,27 +115,24 @@ static int drm_helper_probe_add_cmdline_mode(struct drm_connector *connector) #define DRM_OUTPUT_POLL_PERIOD (10*HZ) /** - * drm_kms_helper_poll_enable - re-enable output polling. + * drm_kms_helper_poll_enable_locked - re-enable output polling. * @dev: drm_device * - * This function re-enables the output polling work, after it has been - * temporarily disabled using drm_kms_helper_poll_disable(), for example over - * suspend/resume. + * This function re-enables the output polling work without + * locking the mode_config mutex. * - * Drivers can call this helper from their device resume implementation. It is - * an error to call this when the output polling support has not yet been set - * up. - * - * Note that calls to enable and disable polling must be strictly ordered, which - * is automatically the case when they're only call from suspend/resume - * callbacks. + * This is like drm_kms_helper_poll_enable() however it is to be + * called from a context where the mode_config mutex is locked + * already. */ -void drm_kms_helper_poll_enable(struct drm_device *dev) +void drm_kms_helper_poll_enable_locked(struct drm_device *dev) { bool poll = false; struct drm_connector *connector; unsigned long delay = DRM_OUTPUT_POLL_PERIOD; + WARN_ON(!mutex_is_locked(&dev->mode_config.mutex)); + if (!dev->mode_config.poll_enabled || !drm_kms_helper_poll) return; @@ -163,7 +160,7 @@ void drm_kms_helper_poll_enable(struct drm_device *dev) if (poll) schedule_delayed_work(&dev->mode_config.output_poll_work, delay); } -EXPORT_SYMBOL(drm_kms_helper_poll_enable); +EXPORT_SYMBOL(drm_kms_helper_poll_enable_locked); static enum drm_connector_status drm_connector_detect(struct drm_connector *connector, bool force) @@ -290,7 +287,7 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector, /* Re-enable polling in case the global poll config changed. */ if (drm_kms_helper_poll != dev->mode_config.poll_running) - drm_kms_helper_poll_enable(dev); + drm_kms_helper_poll_enable_locked(dev); dev->mode_config.poll_running = drm_kms_helper_poll; @@ -482,12 +479,8 @@ out: * This function disables the output polling work. * * Drivers can call this helper from their device suspend implementation. It is - * not an error to call this even when output polling isn't enabled or already - * disabled. Polling is re-enabled by calling drm_kms_helper_poll_enable(). - * - * Note that calls to enable and disable polling must be strictly ordered, which - * is automatically the case when they're only call from suspend/resume - * callbacks. + * not an error to call this even when output polling isn't enabled or arlready + * disabled. */ void drm_kms_helper_poll_disable(struct drm_device *dev) { @@ -497,6 +490,24 @@ void drm_kms_helper_poll_disable(struct drm_device *dev) } EXPORT_SYMBOL(drm_kms_helper_poll_disable); +/** + * drm_kms_helper_poll_enable - re-enable output polling. + * @dev: drm_device + * + * This function re-enables the output polling work. + * + * Drivers can call this helper from their device resume implementation. It is + * an error to call this when the output polling support has not yet been set + * up. + */ +void drm_kms_helper_poll_enable(struct drm_device *dev) +{ + mutex_lock(&dev->mode_config.mutex); + drm_kms_helper_poll_enable_locked(dev); + mutex_unlock(&dev->mode_config.mutex); +} +EXPORT_SYMBOL(drm_kms_helper_poll_enable); + /** * drm_kms_helper_poll_init - initialize and enable output polling * @dev: drm_device diff --git a/drivers/gpu/drm/i915/intel_hotplug.c b/drivers/gpu/drm/i915/intel_hotplug.c index b62e3f8ad415..3d546c019de0 100644 --- a/drivers/gpu/drm/i915/intel_hotplug.c +++ b/drivers/gpu/drm/i915/intel_hotplug.c @@ -180,7 +180,7 @@ static void intel_hpd_irq_storm_disable(struct drm_i915_private *dev_priv) /* Enable polling and queue hotplug re-enabling. */ if (hpd_disabled) { - drm_kms_helper_poll_enable(dev); + drm_kms_helper_poll_enable_locked(dev); mod_delayed_work(system_wq, &dev_priv->hotplug.reenable_work, msecs_to_jiffies(HPD_STORM_REENABLE_DELAY)); } @@ -511,7 +511,7 @@ static void i915_hpd_poll_init_work(struct work_struct *work) } if (enabled) - drm_kms_helper_poll_enable(dev); + drm_kms_helper_poll_enable_locked(dev); mutex_unlock(&dev->mode_config.mutex); diff --git a/include/drm/drm_crtc_helper.h b/include/drm/drm_crtc_helper.h index d026f5017c33..982c299e435a 100644 --- a/include/drm/drm_crtc_helper.h +++ b/include/drm/drm_crtc_helper.h @@ -73,5 +73,6 @@ extern void drm_kms_helper_hotplug_event(struct drm_device *dev); extern void drm_kms_helper_poll_disable(struct drm_device *dev); extern void drm_kms_helper_poll_enable(struct drm_device *dev); +extern void drm_kms_helper_poll_enable_locked(struct drm_device *dev); #endif -- cgit v1.2.3 From 1372cef1c697d8aac0cc923f8aa2c37d790ec9ed Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Wed, 25 Jan 2017 19:30:09 +0000 Subject: regulator: fixed: Revert support for ACPI interface This reverts commit 13bed58ce874 (regulator: fixed: add support for ACPI interface). While there does appear to be a practical need to manage regulators on ACPI systems, using ad-hoc properties to describe regulators to the kernel presents a number of problems (especially should ACPI gain first class support for such things), and there are ongoing discussions as to how to manage this. Until there is a rough consensus, revert commit 13bed58ce8748d43, which hasn't been in a released kernel yet as discussed in [1] and the surrounding thread. [1] http://lkml.kernel.org/r/20170125184949.x2wkoo7kbaaajkjk@sirena.org.uk Signed-off-by: Mark Rutland Cc: Liam Girdwood Cc: Lorenzo Pieralisi Cc: Lu Baolu Cc: Mark Brown Cc: Rafael J. Wysocki Cc: linux-kernel@vger.kernel.org Signed-off-by: Mark Brown --- drivers/regulator/fixed.c | 46 ---------------------------------------------- 1 file changed, 46 deletions(-) diff --git a/drivers/regulator/fixed.c b/drivers/regulator/fixed.c index a43b0e8a438d..988a7472c2ab 100644 --- a/drivers/regulator/fixed.c +++ b/drivers/regulator/fixed.c @@ -30,9 +30,6 @@ #include #include #include -#include -#include -#include struct fixed_voltage_data { struct regulator_desc desc; @@ -97,44 +94,6 @@ of_get_fixed_voltage_config(struct device *dev, return config; } -/** - * acpi_get_fixed_voltage_config - extract fixed_voltage_config structure info - * @dev: device requesting for fixed_voltage_config - * @desc: regulator description - * - * Populates fixed_voltage_config structure by extracting data through ACPI - * interface, returns a pointer to the populated structure of NULL if memory - * alloc fails. - */ -static struct fixed_voltage_config * -acpi_get_fixed_voltage_config(struct device *dev, - const struct regulator_desc *desc) -{ - struct fixed_voltage_config *config; - const char *supply_name; - struct gpio_desc *gpiod; - int ret; - - config = devm_kzalloc(dev, sizeof(*config), GFP_KERNEL); - if (!config) - return ERR_PTR(-ENOMEM); - - ret = device_property_read_string(dev, "supply-name", &supply_name); - if (!ret) - config->supply_name = supply_name; - - gpiod = gpiod_get(dev, "gpio", GPIOD_ASIS); - if (IS_ERR(gpiod)) - return ERR_PTR(-ENODEV); - - config->gpio = desc_to_gpio(gpiod); - config->enable_high = device_property_read_bool(dev, - "enable-active-high"); - gpiod_put(gpiod); - - return config; -} - static struct regulator_ops fixed_voltage_ops = { }; @@ -155,11 +114,6 @@ static int reg_fixed_voltage_probe(struct platform_device *pdev) &drvdata->desc); if (IS_ERR(config)) return PTR_ERR(config); - } else if (ACPI_HANDLE(&pdev->dev)) { - config = acpi_get_fixed_voltage_config(&pdev->dev, - &drvdata->desc); - if (IS_ERR(config)) - return PTR_ERR(config); } else { config = dev_get_platdata(&pdev->dev); } -- cgit v1.2.3 From 0e1929dedea36781e25902118c93edd8d8f09af1 Mon Sep 17 00:00:00 2001 From: Mike Looijmans Date: Mon, 16 Jan 2017 15:49:38 +0100 Subject: i2c: i2c-cadence: Initialize configuration before probing devices The cadence I2C driver calls cdns_i2c_writereg(..) to setup a workaround in the controller, but did so after calling i2c_add_adapter() which starts probing devices on the bus. Change the order so that the configuration is completely finished before using the adapter. Signed-off-by: Mike Looijmans Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-cadence.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/i2c/busses/i2c-cadence.c b/drivers/i2c/busses/i2c-cadence.c index 686971263bef..45d6771fac8c 100644 --- a/drivers/i2c/busses/i2c-cadence.c +++ b/drivers/i2c/busses/i2c-cadence.c @@ -962,10 +962,6 @@ static int cdns_i2c_probe(struct platform_device *pdev) goto err_clk_dis; } - ret = i2c_add_adapter(&id->adap); - if (ret < 0) - goto err_clk_dis; - /* * Cadence I2C controller has a bug wherein it generates * invalid read transaction after HW timeout in master receiver mode. @@ -975,6 +971,10 @@ static int cdns_i2c_probe(struct platform_device *pdev) */ cdns_i2c_writereg(CDNS_I2C_TIMEOUT_MAX, CDNS_I2C_TIME_OUT_OFFSET); + ret = i2c_add_adapter(&id->adap); + if (ret < 0) + goto err_clk_dis; + dev_info(&pdev->dev, "%u kHz mmio %08lx irq %d\n", id->i2c_clk / 1000, (unsigned long)r_mem->start, id->irq); -- cgit v1.2.3 From b9b487e494712c8e5905b724e12f5ef17e9ae6f9 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 25 Jan 2017 12:00:29 -0500 Subject: Revert "drm/radeon: always apply pci shutdown callbacks" This seems to break reboot on some evergreen systems. bugs: https://bugs.freedesktop.org/show_bug.cgi?id=99524 https://bugzilla.kernel.org/show_bug.cgi?id=192271 This reverts commit a481daa88fd4d6b54f25348972bba10b5f6a84d0. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/radeon/radeon_drv.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index 00ea0002b539..e0c143b865f3 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -366,11 +366,10 @@ static void radeon_pci_shutdown(struct pci_dev *pdev) { /* if we are running in a VM, make sure the device - * torn down properly on reboot/shutdown. - * unfortunately we can't detect certain - * hypervisors so just do this all the time. + * torn down properly on reboot/shutdown */ - radeon_pci_remove(pdev); + if (radeon_device_is_virtual()) + radeon_pci_remove(pdev); } static int radeon_pmops_suspend(struct device *dev) -- cgit v1.2.3 From e13fe92bb58cf9b8f709ec18267ffc9e6ffeb016 Mon Sep 17 00:00:00 2001 From: Gao Pan Date: Tue, 17 Jan 2017 18:20:55 +0800 Subject: i2c: imx-lpi2c: add VLLS mode support When system enters VLLS mode, module power is turned off. As a result, all registers are reset to HW default value. After exiting VLLS mode, registers are still in default mode. As a result, the pinctrl settings are incorrect, which will affect the module function. The patch recovers the pinctrl setting when exit VLLS mode. Signed-off-by: Gao Pan Reviewed-by: Vladimir Zapolskiy [wsa: added missing include] Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-imx-lpi2c.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/drivers/i2c/busses/i2c-imx-lpi2c.c b/drivers/i2c/busses/i2c-imx-lpi2c.c index c62b7cd475f8..3310f2e0dbd3 100644 --- a/drivers/i2c/busses/i2c-imx-lpi2c.c +++ b/drivers/i2c/busses/i2c-imx-lpi2c.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include #include @@ -636,12 +637,31 @@ static int lpi2c_imx_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM_SLEEP +static int lpi2c_imx_suspend(struct device *dev) +{ + pinctrl_pm_select_sleep_state(dev); + + return 0; +} + +static int lpi2c_imx_resume(struct device *dev) +{ + pinctrl_pm_select_default_state(dev); + + return 0; +} +#endif + +static SIMPLE_DEV_PM_OPS(imx_lpi2c_pm, lpi2c_imx_suspend, lpi2c_imx_resume); + static struct platform_driver lpi2c_imx_driver = { .probe = lpi2c_imx_probe, .remove = lpi2c_imx_remove, .driver = { .name = DRIVER_NAME, .of_match_table = lpi2c_imx_of_match, + .pm = &imx_lpi2c_pm, }, }; -- cgit v1.2.3 From 26afec39306926654e9cd320f19bbf3685bb0997 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Thu, 26 Jan 2017 09:38:31 +0800 Subject: r8152: avoid start_xmit to call napi_schedule during autosuspend Adjust the setting of the flag of SELECTIVE_SUSPEND to prevent start_xmit() from calling napi_schedule() directly during runtime suspend. After calling napi_disable() or clearing the flag of WORK_ENABLE, scheduling the napi is useless. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller --- drivers/net/usb/r8152.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index e1466b4d2b6c..23bef8e6e0bd 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -3585,10 +3585,15 @@ static int rtl8152_rumtime_suspend(struct r8152 *tp) struct net_device *netdev = tp->netdev; int ret = 0; + set_bit(SELECTIVE_SUSPEND, &tp->flags); + smp_mb__after_atomic(); + if (netif_running(netdev) && test_bit(WORK_ENABLE, &tp->flags)) { u32 rcr = 0; if (delay_autosuspend(tp)) { + clear_bit(SELECTIVE_SUSPEND, &tp->flags); + smp_mb__after_atomic(); ret = -EBUSY; goto out1; } @@ -3605,6 +3610,8 @@ static int rtl8152_rumtime_suspend(struct r8152 *tp) if (!(ocp_data & RXFIFO_EMPTY)) { rxdy_gated_en(tp, false); ocp_write_dword(tp, MCU_TYPE_PLA, PLA_RCR, rcr); + clear_bit(SELECTIVE_SUSPEND, &tp->flags); + smp_mb__after_atomic(); ret = -EBUSY; goto out1; } @@ -3624,8 +3631,6 @@ static int rtl8152_rumtime_suspend(struct r8152 *tp) } } - set_bit(SELECTIVE_SUSPEND, &tp->flags); - out1: return ret; } @@ -3681,12 +3686,13 @@ static int rtl8152_resume(struct usb_interface *intf) if (netif_running(tp->netdev) && tp->netdev->flags & IFF_UP) { if (test_bit(SELECTIVE_SUSPEND, &tp->flags)) { tp->rtl_ops.autosuspend_en(tp, false); - clear_bit(SELECTIVE_SUSPEND, &tp->flags); napi_disable(&tp->napi); set_bit(WORK_ENABLE, &tp->flags); if (netif_carrier_ok(tp->netdev)) rtl_start_rx(tp); napi_enable(&tp->napi); + clear_bit(SELECTIVE_SUSPEND, &tp->flags); + smp_mb__after_atomic(); } else { tp->rtl_ops.up(tp); netif_carrier_off(tp->netdev); -- cgit v1.2.3 From de9bf29dd6e4a8a874cb92f8901aed50a9d0b1d3 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Thu, 26 Jan 2017 09:38:32 +0800 Subject: r8152: avoid start_xmit to schedule napi when napi is disabled Stop the tx when the napi is disabled to prevent napi_schedule() is called. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller --- drivers/net/usb/r8152.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 23bef8e6e0bd..ec882be4cfaf 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -3155,10 +3155,13 @@ static void set_carrier(struct r8152 *tp) if (!netif_carrier_ok(netdev)) { tp->rtl_ops.enable(tp); set_bit(RTL8152_SET_RX_MODE, &tp->flags); + netif_stop_queue(netdev); napi_disable(&tp->napi); netif_carrier_on(netdev); rtl_start_rx(tp); napi_enable(&tp->napi); + netif_wake_queue(netdev); + netif_info(tp, link, netdev, "carrier on\n"); } } else { if (netif_carrier_ok(netdev)) { @@ -3166,6 +3169,7 @@ static void set_carrier(struct r8152 *tp) napi_disable(&tp->napi); tp->rtl_ops.disable(tp); napi_enable(&tp->napi); + netif_info(tp, link, netdev, "carrier off\n"); } } } @@ -3515,12 +3519,12 @@ static int rtl8152_pre_reset(struct usb_interface *intf) if (!netif_running(netdev)) return 0; + netif_stop_queue(netdev); napi_disable(&tp->napi); clear_bit(WORK_ENABLE, &tp->flags); usb_kill_urb(tp->intr_urb); cancel_delayed_work_sync(&tp->schedule); if (netif_carrier_ok(netdev)) { - netif_stop_queue(netdev); mutex_lock(&tp->control); tp->rtl_ops.disable(tp); mutex_unlock(&tp->control); @@ -3548,10 +3552,10 @@ static int rtl8152_post_reset(struct usb_interface *intf) rtl_start_rx(tp); rtl8152_set_rx_mode(netdev); mutex_unlock(&tp->control); - netif_wake_queue(netdev); } napi_enable(&tp->napi); + netif_wake_queue(netdev); usb_submit_urb(tp->intr_urb, GFP_KERNEL); return 0; -- cgit v1.2.3 From 248b213ad908b88db15941202ef7cb7eb137c1a0 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Thu, 26 Jan 2017 09:38:33 +0800 Subject: r8152: re-schedule napi for tx Re-schedule napi after napi_complete() for tx, if it is necessay. In r8152_poll(), if the tx is completed after tx_bottom() and before napi_complete(), the scheduling of napi would be lost. Then, no one handles the next tx until the next napi_schedule() is called. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller --- drivers/net/usb/r8152.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index ec882be4cfaf..4785d2b9d80f 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -1936,6 +1936,9 @@ static int r8152_poll(struct napi_struct *napi, int budget) napi_complete(napi); if (!list_empty(&tp->rx_done)) napi_schedule(napi); + else if (!skb_queue_empty(&tp->tx_queue) && + !list_empty(&tp->tx_free)) + napi_schedule(napi); } return work_done; -- cgit v1.2.3 From 7489bdadb7d17d3c81e39b85688500f700beb790 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Thu, 26 Jan 2017 09:38:34 +0800 Subject: r8152: check rx after napi is enabled Schedule the napi after napi_enable() for rx, if it is necessary. If the rx is completed when napi is disabled, the sheduling of napi would be lost. Then, no one handles the rx packet until next napi is scheduled. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller --- drivers/net/usb/r8152.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 4785d2b9d80f..ad42295356dd 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -32,7 +32,7 @@ #define NETNEXT_VERSION "08" /* Information for net */ -#define NET_VERSION "7" +#define NET_VERSION "8" #define DRIVER_VERSION "v1." NETNEXT_VERSION "." NET_VERSION #define DRIVER_AUTHOR "Realtek linux nic maintainers " @@ -3561,6 +3561,9 @@ static int rtl8152_post_reset(struct usb_interface *intf) netif_wake_queue(netdev); usb_submit_urb(tp->intr_urb, GFP_KERNEL); + if (!list_empty(&tp->rx_done)) + napi_schedule(&tp->napi); + return 0; } @@ -3700,6 +3703,8 @@ static int rtl8152_resume(struct usb_interface *intf) napi_enable(&tp->napi); clear_bit(SELECTIVE_SUSPEND, &tp->flags); smp_mb__after_atomic(); + if (!list_empty(&tp->rx_done)) + napi_schedule(&tp->napi); } else { tp->rtl_ops.up(tp); netif_carrier_off(tp->netdev); -- cgit v1.2.3 From b68df015609eac67f045c155cb3195e5a1061d66 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Wed, 25 Jan 2017 18:22:48 -0800 Subject: virtio_net: use dev_kfree_skb for small buffer XDP receive In the small buffer case during driver unload we currently use put_page instead of dev_kfree_skb. Resolve this by adding a check for virtnet mode when checking XDP queue type. Also name the function so that the code reads correctly to match the additional check. Fixes: bb91accf2733 ("virtio-net: XDP support for small buffers") Signed-off-by: John Fastabend Acked-by: Jason Wang Acked-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 3d1519ea0669..4c6e9b43f1d3 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -1898,8 +1898,12 @@ static void free_receive_page_frags(struct virtnet_info *vi) put_page(vi->rq[i].alloc_frag.page); } -static bool is_xdp_queue(struct virtnet_info *vi, int q) +static bool is_xdp_raw_buffer_queue(struct virtnet_info *vi, int q) { + /* For small receive mode always use kfree_skb variants */ + if (!vi->mergeable_rx_bufs) + return false; + if (q < (vi->curr_queue_pairs - vi->xdp_queue_pairs)) return false; else if (q < vi->curr_queue_pairs) @@ -1916,7 +1920,7 @@ static void free_unused_bufs(struct virtnet_info *vi) for (i = 0; i < vi->max_queue_pairs; i++) { struct virtqueue *vq = vi->sq[i].vq; while ((buf = virtqueue_detach_unused_buf(vq)) != NULL) { - if (!is_xdp_queue(vi, i)) + if (!is_xdp_raw_buffer_queue(vi, i)) dev_kfree_skb(buf); else put_page(virt_to_head_page(buf)); -- cgit v1.2.3 From 529ec6ac26656378435eb0396a780f017d51e105 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 25 Jan 2017 14:56:36 -0800 Subject: virtio_net: reject XDP programs using header adjustment commit 17bedab27231 ("bpf: xdp: Allow head adjustment in XDP prog") added a new XDP helper to prepend and remove data from a frame. Make virtio_net reject programs making use of this helper until proper support is added. Signed-off-by: Jakub Kicinski Acked-by: John Fastabend Acked-by: Jason Wang Acked-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 4c6e9b43f1d3..765c2d6358da 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -1715,6 +1715,11 @@ static int virtnet_xdp_set(struct net_device *dev, struct bpf_prog *prog) u16 xdp_qp = 0, curr_qp; int i, err; + if (prog && prog->xdp_adjust_head) { + netdev_warn(dev, "Does not support bpf_xdp_adjust_head()\n"); + return -EOPNOTSUPP; + } + if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_GUEST_TSO4) || virtio_has_feature(vi->vdev, VIRTIO_NET_F_GUEST_TSO6) || virtio_has_feature(vi->vdev, VIRTIO_NET_F_GUEST_ECN) || -- cgit v1.2.3 From 2aa6ba7b5ad3189cc27f14540aa2f57f0ed8df4b Mon Sep 17 00:00:00 2001 From: Darrick J. Wong Date: Wed, 25 Jan 2017 20:24:57 -0800 Subject: xfs: clear _XBF_PAGES from buffers when readahead page If we try to allocate memory pages to back an xfs_buf that we're trying to read, it's possible that we'll be so short on memory that the page allocation fails. For a blocking read we'll just wait, but for readahead we simply dump all the pages we've collected so far. Unfortunately, after dumping the pages we neglect to clear the _XBF_PAGES state, which means that the subsequent call to xfs_buf_free thinks that b_pages still points to pages we own. It then double-frees the b_pages pages. This results in screaming about negative page refcounts from the memory manager, which xfs oughtn't be triggering. To reproduce this case, mount a filesystem where the size of the inodes far outweighs the availalble memory (a ~500M inode filesystem on a VM with 300MB memory did the trick here) and run bulkstat in parallel with other memory eating processes to put a huge load on the system. The "check summary" phase of xfs_scrub also works for this purpose. Signed-off-by: Darrick J. Wong Reviewed-by: Eric Sandeen --- fs/xfs/xfs_buf.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index 7f0a01f7b592..ac3b4db519df 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -422,6 +422,7 @@ retry: out_free_pages: for (i = 0; i < bp->b_page_count; i++) __free_page(bp->b_pages[i]); + bp->b_flags &= ~_XBF_PAGES; return error; } -- cgit v1.2.3 From c81e55e057b6458aac6d96a6429ef021b7f6f62c Mon Sep 17 00:00:00 2001 From: James Smart Date: Sat, 24 Dec 2016 09:46:43 -0800 Subject: nvmet_fc: correct logic in disconnect queue LS handling Correct logic in disconnect queue LS handling. Rework so that queue searching and error reporting is above the section to send back a ls rjt Signed-off-by: James Smart Signed-off-by: Sagi Grimberg --- drivers/nvme/target/fc.c | 36 ++++++++++++++++++++++-------------- 1 file changed, 22 insertions(+), 14 deletions(-) diff --git a/drivers/nvme/target/fc.c b/drivers/nvme/target/fc.c index 173e842f19c9..ba57f9852bde 100644 --- a/drivers/nvme/target/fc.c +++ b/drivers/nvme/target/fc.c @@ -1314,7 +1314,7 @@ nvmet_fc_ls_disconnect(struct nvmet_fc_tgtport *tgtport, (struct fcnvme_ls_disconnect_rqst *)iod->rqstbuf; struct fcnvme_ls_disconnect_acc *acc = (struct fcnvme_ls_disconnect_acc *)iod->rspbuf; - struct nvmet_fc_tgt_queue *queue; + struct nvmet_fc_tgt_queue *queue = NULL; struct nvmet_fc_tgt_assoc *assoc; int ret = 0; bool del_assoc = false; @@ -1348,7 +1348,18 @@ nvmet_fc_ls_disconnect(struct nvmet_fc_tgtport *tgtport, assoc = nvmet_fc_find_target_assoc(tgtport, be64_to_cpu(rqst->associd.association_id)); iod->assoc = assoc; - if (!assoc) + if (assoc) { + if (rqst->discon_cmd.scope == + FCNVME_DISCONN_CONNECTION) { + queue = nvmet_fc_find_target_queue(tgtport, + be64_to_cpu( + rqst->discon_cmd.id)); + if (!queue) { + nvmet_fc_tgt_a_put(assoc); + ret = VERR_NO_CONN; + } + } + } else ret = VERR_NO_ASSOC; } @@ -1373,21 +1384,18 @@ nvmet_fc_ls_disconnect(struct nvmet_fc_tgtport *tgtport, FCNVME_LS_DISCONNECT); - if (rqst->discon_cmd.scope == FCNVME_DISCONN_CONNECTION) { - queue = nvmet_fc_find_target_queue(tgtport, - be64_to_cpu(rqst->discon_cmd.id)); - if (queue) { - int qid = queue->qid; + /* are we to delete a Connection ID (queue) */ + if (queue) { + int qid = queue->qid; - nvmet_fc_delete_target_queue(queue); + nvmet_fc_delete_target_queue(queue); - /* release the get taken by find_target_queue */ - nvmet_fc_tgt_q_put(queue); + /* release the get taken by find_target_queue */ + nvmet_fc_tgt_q_put(queue); - /* tear association down if io queue terminated */ - if (!qid) - del_assoc = true; - } + /* tear association down if io queue terminated */ + if (!qid) + del_assoc = true; } /* release get taken in nvmet_fc_find_target_assoc */ -- cgit v1.2.3 From 344770b07b7ae70639ebf110010eb6156a6e55e9 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Sun, 27 Nov 2016 22:29:17 +0200 Subject: nvmet: delete controllers deletion upon subsystem release No reason for them to be kept around if we are deleting the subsystem, so instead of passively wait for the host to disconnect, actively delete the controllers. Signed-off-by: Sagi Grimberg Reviewed-by: Johannes Thumshirn Reviewed-by: Max Gurtovoy Reviewed-by: Christoph Hellwig --- drivers/nvme/target/configfs.c | 1 + drivers/nvme/target/core.c | 10 ++++++++++ drivers/nvme/target/nvmet.h | 1 + 3 files changed, 12 insertions(+) diff --git a/drivers/nvme/target/configfs.c b/drivers/nvme/target/configfs.c index 6f5074153dcd..be8c800078e2 100644 --- a/drivers/nvme/target/configfs.c +++ b/drivers/nvme/target/configfs.c @@ -631,6 +631,7 @@ static void nvmet_subsys_release(struct config_item *item) { struct nvmet_subsys *subsys = to_subsys(item); + nvmet_subsys_del_ctrls(subsys); nvmet_subsys_put(subsys); } diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index b1d66ed655c9..4a367549eb93 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -935,6 +935,16 @@ static void nvmet_subsys_free(struct kref *ref) kfree(subsys); } +void nvmet_subsys_del_ctrls(struct nvmet_subsys *subsys) +{ + struct nvmet_ctrl *ctrl; + + mutex_lock(&subsys->lock); + list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) + ctrl->ops->delete_ctrl(ctrl); + mutex_unlock(&subsys->lock); +} + void nvmet_subsys_put(struct nvmet_subsys *subsys) { kref_put(&subsys->ref, nvmet_subsys_free); diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h index 23d5eb1c944f..cc7ad06b43a7 100644 --- a/drivers/nvme/target/nvmet.h +++ b/drivers/nvme/target/nvmet.h @@ -282,6 +282,7 @@ void nvmet_ctrl_put(struct nvmet_ctrl *ctrl); struct nvmet_subsys *nvmet_subsys_alloc(const char *subsysnqn, enum nvme_subsys_type type); void nvmet_subsys_put(struct nvmet_subsys *subsys); +void nvmet_subsys_del_ctrls(struct nvmet_subsys *subsys); struct nvmet_ns *nvmet_find_namespace(struct nvmet_ctrl *ctrl, __le32 nsid); void nvmet_put_namespace(struct nvmet_ns *ns); -- cgit v1.2.3 From 06406d81a2d7cfb8abcc4fa6cdfeb8e5897007c5 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Sun, 1 Jan 2017 13:41:56 +0200 Subject: nvmet: cancel fatal error and flush async work before free controller Make sure they are not running and we can free the controller safely. Signed-off-by: Roy Shterman Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/target/core.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index 4a367549eb93..a327a43f8166 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -816,6 +816,9 @@ static void nvmet_ctrl_free(struct kref *ref) list_del(&ctrl->subsys_entry); mutex_unlock(&subsys->lock); + flush_work(&ctrl->async_event_work); + cancel_work_sync(&ctrl->fatal_err_work); + ida_simple_remove(&subsys->cntlid_ida, ctrl->cntlid); nvmet_subsys_put(subsys); -- cgit v1.2.3 From 23a8ed4a624324dc696c328f09bd502c4a3816f0 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Sun, 1 Jan 2017 13:18:26 +0200 Subject: nvmet: Call fatal_error from keep-alive timout expiration We only need to call delete_ctrl once, so given that both keep-alive timeout and any other fatal error can trigger it, just make sure we only call delete_ctrl once. Signed-off-by: Sagi Grimberg Reviewed-by: Christoph Hellwig --- drivers/nvme/target/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index a327a43f8166..fc5ba2f9e15f 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -200,7 +200,7 @@ static void nvmet_keep_alive_timer(struct work_struct *work) pr_err("ctrl %d keep-alive timer (%d seconds) expired!\n", ctrl->cntlid, ctrl->kato); - ctrl->ops->delete_ctrl(ctrl); + nvmet_ctrl_fatal_error(ctrl); } static void nvmet_start_keep_alive_timer(struct nvmet_ctrl *ctrl) -- cgit v1.2.3 From 748ff8408f8e208f279ba221e5c12612fbb4dddb Mon Sep 17 00:00:00 2001 From: Parav Pandit Date: Thu, 19 Jan 2017 09:55:08 -0600 Subject: nvmet-rdma: Fix missing dma sync to nvme data structures This patch performs dma sync operations on nvme_command and nvme_completion. nvme_command is synced (a) on receiving of the recv queue completion for cpu access. (b) before posting recv wqe back to rdma adapter for device access. nvme_completion is synced (a) on receiving of the recv queue completion of associated nvme_command for cpu access. (b) before posting send wqe to rdma adapter for device access. This patch is generated for git://git.infradead.org/nvme-fabrics.git Branch: nvmf-4.10 Signed-off-by: Parav Pandit Reviewed-by: Max Gurtovoy --- drivers/nvme/target/rdma.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c index 8c3760a78ac0..60990220bd83 100644 --- a/drivers/nvme/target/rdma.c +++ b/drivers/nvme/target/rdma.c @@ -438,6 +438,10 @@ static int nvmet_rdma_post_recv(struct nvmet_rdma_device *ndev, { struct ib_recv_wr *bad_wr; + ib_dma_sync_single_for_device(ndev->device, + cmd->sge[0].addr, cmd->sge[0].length, + DMA_FROM_DEVICE); + if (ndev->srq) return ib_post_srq_recv(ndev->srq, &cmd->wr, &bad_wr); return ib_post_recv(cmd->queue->cm_id->qp, &cmd->wr, &bad_wr); @@ -538,6 +542,11 @@ static void nvmet_rdma_queue_response(struct nvmet_req *req) first_wr = &rsp->send_wr; nvmet_rdma_post_recv(rsp->queue->dev, rsp->cmd); + + ib_dma_sync_single_for_device(rsp->queue->dev->device, + rsp->send_sge.addr, rsp->send_sge.length, + DMA_TO_DEVICE); + if (ib_post_send(cm_id->qp, first_wr, &bad_wr)) { pr_err("sending cmd response failed\n"); nvmet_rdma_release_rsp(rsp); @@ -698,6 +707,14 @@ static void nvmet_rdma_handle_command(struct nvmet_rdma_queue *queue, cmd->n_rdma = 0; cmd->req.port = queue->port; + + ib_dma_sync_single_for_cpu(queue->dev->device, + cmd->cmd->sge[0].addr, cmd->cmd->sge[0].length, + DMA_FROM_DEVICE); + ib_dma_sync_single_for_cpu(queue->dev->device, + cmd->send_sge.addr, cmd->send_sge.length, + DMA_TO_DEVICE); + if (!nvmet_req_init(&cmd->req, &queue->nvme_cq, &queue->nvme_sq, &nvmet_rdma_ops)) return; -- cgit v1.2.3 From 19e420bb4076ace670addc55300e3b8c4a02dfc6 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 19 Jan 2017 16:55:57 +0100 Subject: nvme-fc: use blk_rq_nr_phys_segments Without this deallocate won't work properly due to the mismatch of the bio/request size and the actual payload size. Signed-off-by: Christoph Hellwig Reviewed-by: James Smart Reviewed-by: Johannes Thumshirn Signed-off-by: Sagi Grimberg --- drivers/nvme/host/fc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index fcc9dcfdf675..e65041c640cb 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -1663,13 +1663,13 @@ nvme_fc_map_data(struct nvme_fc_ctrl *ctrl, struct request *rq, return 0; freq->sg_table.sgl = freq->first_sgl; - ret = sg_alloc_table_chained(&freq->sg_table, rq->nr_phys_segments, - freq->sg_table.sgl); + ret = sg_alloc_table_chained(&freq->sg_table, + blk_rq_nr_phys_segments(rq), freq->sg_table.sgl); if (ret) return -ENOMEM; op->nents = blk_rq_map_sg(rq->q, rq, freq->sg_table.sgl); - WARN_ON(op->nents > rq->nr_phys_segments); + WARN_ON(op->nents > blk_rq_nr_phys_segments(rq)); dir = (rq_data_dir(rq) == WRITE) ? DMA_TO_DEVICE : DMA_FROM_DEVICE; freq->sg_cnt = fc_dma_map_sg(ctrl->lport->dev, freq->sg_table.sgl, op->nents, dir); -- cgit v1.2.3 From 08965c2eba135bdfb6e86cf25308e01421c7e0ce Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Wed, 25 Jan 2017 13:43:56 -0800 Subject: Revert "sd: remove __data_len hack for WRITE SAME" This patch reverts commit f80de881d8df and avoids that sending a WRITE SAME command to the iSCSI initiator triggers the following: BUG: unable to handle kernel NULL pointer dereference at 0000000000000014 TARGET_CORE[iSCSI]: Expected Transfer Length: 260096 does not match SCSI CDB Length: 512 for SAM Opcode: 0x41 IP: iscsi_tcp_segment_done+0x20b/0x310 [libiscsi_tcp] Oops: 0000 [#1] SMP Modules linked in: target_core_user uio target_core_iblock target_core_file iscsi_target_mod target_core_mod netconsole configfs crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel aes_x86_64 crypto_simd cryptd glue_helper virtio_console virtio_rng virtio_balloon serio_raw i2c_piix4 acpi_cpufreq button iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ext4 jbd2 mbcache virtio_blk virtio_net psmouse floppy drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm drm virtio_pci CPU: 2 PID: 5 Comm: kworker/u8:0 Not tainted 4.10.0-rc5-debug+ #3 Workqueue: iscsi_q_0 iscsi_xmitworker [libiscsi] RIP: 0010:iscsi_tcp_segment_done+0x20b/0x310 [libiscsi_tcp] Call Trace: iscsi_sw_tcp_xmit_segment+0x84/0x120 [iscsi_tcp] iscsi_sw_tcp_pdu_xmit+0x51/0x180 [iscsi_tcp] iscsi_tcp_task_xmit+0xb3/0x290 [libiscsi_tcp] iscsi_xmit_task+0x4e/0xc0 [libiscsi] iscsi_xmitworker+0x243/0x330 [libiscsi] process_one_work+0x1d8/0x4b0 worker_thread+0x49/0x4a0 kthread+0x102/0x140 Fixes: f80de881d8df ("sd: remove __data_len hack for WRITE SAME") Signed-off-by: Bart Van Assche Cc: Hannes Reinecke Cc: Sagi Grimberg Cc: Jens Axboe Cc: Lee Duncan Cc: Chris Leech Acked-by: Christoph Hellwig Acked-by: Martin K. Petersen Signed-off-by: Jens Axboe --- drivers/scsi/sd.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 0b09638fa39b..1f5d92a25a49 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -836,6 +836,7 @@ static int sd_setup_write_same_cmnd(struct scsi_cmnd *cmd) struct bio *bio = rq->bio; sector_t sector = blk_rq_pos(rq); unsigned int nr_sectors = blk_rq_sectors(rq); + unsigned int nr_bytes = blk_rq_bytes(rq); int ret; if (sdkp->device->no_write_same) @@ -868,7 +869,21 @@ static int sd_setup_write_same_cmnd(struct scsi_cmnd *cmd) cmd->transfersize = sdp->sector_size; cmd->allowed = SD_MAX_RETRIES; - return scsi_init_io(cmd); + + /* + * For WRITE SAME the data transferred via the DATA OUT buffer is + * different from the amount of data actually written to the target. + * + * We set up __data_len to the amount of data transferred via the + * DATA OUT buffer so that blk_rq_map_sg sets up the proper S/G list + * to transfer a single sector of data first, but then reset it to + * the amount of data to be written right after so that the I/O path + * knows how much to actually write. + */ + rq->__data_len = sdp->sector_size; + ret = scsi_init_io(cmd); + rq->__data_len = nr_bytes; + return ret; } static int sd_setup_flush_cmnd(struct scsi_cmnd *cmd) -- cgit v1.2.3 From ff9f8a7cf935468a94d9927c68b00daae701667e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 25 Jan 2017 18:20:55 -0800 Subject: sysctl: fix proc_doulongvec_ms_jiffies_minmax() We perform the conversion between kernel jiffies and ms only when exporting kernel value to user space. We need to do the opposite operation when value is written by user. Only matters when HZ != 1000 Signed-off-by: Eric Dumazet Cc: stable@vger.kernel.org Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 8dbaec0e4f7f..1aea594a54db 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2475,6 +2475,7 @@ static int __do_proc_doulongvec_minmax(void *data, struct ctl_table *table, int break; if (neg) continue; + val = convmul * val / convdiv; if ((min && val < *min) || (max && val > *max)) continue; *i = val; -- cgit v1.2.3 From c364b6d0b6cda1cd5d9ab689489adda3e82529aa Mon Sep 17 00:00:00 2001 From: Darrick J. Wong Date: Thu, 26 Jan 2017 09:50:30 -0800 Subject: xfs: fix bmv_count confusion w/ shared extents In a bmapx call, bmv_count is the total size of the array, including the zeroth element that userspace uses to supply the search key. The output array starts at offset 1 so that we can set up the user for the next invocation. Since we now can split an extent into multiple bmap records due to shared/unshared status, we have to be careful that we don't overflow the output array. In the original patch f86f403794b ("xfs: teach get_bmapx about shared extents and the CoW fork") I used cur_ext (the output index) to check for overflows, albeit with an off-by-one error. Since nexleft no longer describes the number of unfilled slots in the output, we can rip all that out and use cur_ext for the overflow check directly. Failure to do this causes heap corruption in bmapx callers such as xfs_io and xfs_scrub. xfs/328 can reproduce this problem. Reviewed-by: Eric Sandeen Signed-off-by: Darrick J. Wong --- fs/xfs/xfs_bmap_util.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index b9abce524c33..c1417919ab0a 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -528,7 +528,6 @@ xfs_getbmap( xfs_bmbt_irec_t *map; /* buffer for user's data */ xfs_mount_t *mp; /* file system mount point */ int nex; /* # of user extents can do */ - int nexleft; /* # of user extents left */ int subnex; /* # of bmapi's can do */ int nmap; /* number of map entries */ struct getbmapx *out; /* output structure */ @@ -686,10 +685,8 @@ xfs_getbmap( goto out_free_map; } - nexleft = nex; - do { - nmap = (nexleft > subnex) ? subnex : nexleft; + nmap = (nex> subnex) ? subnex : nex; error = xfs_bmapi_read(ip, XFS_BB_TO_FSBT(mp, bmv->bmv_offset), XFS_BB_TO_FSB(mp, bmv->bmv_length), map, &nmap, bmapi_flags); @@ -697,8 +694,8 @@ xfs_getbmap( goto out_free_map; ASSERT(nmap <= subnex); - for (i = 0; i < nmap && nexleft && bmv->bmv_length && - cur_ext < bmv->bmv_count; i++) { + for (i = 0; i < nmap && bmv->bmv_length && + cur_ext < bmv->bmv_count - 1; i++) { out[cur_ext].bmv_oflags = 0; if (map[i].br_state == XFS_EXT_UNWRITTEN) out[cur_ext].bmv_oflags |= BMV_OF_PREALLOC; @@ -760,16 +757,27 @@ xfs_getbmap( continue; } + /* + * In order to report shared extents accurately, + * we report each distinct shared/unshared part + * of a single bmbt record using multiple bmap + * extents. To make that happen, we iterate the + * same map array item multiple times, each + * time trimming out the subextent that we just + * reported. + * + * Because of this, we must check the out array + * index (cur_ext) directly against bmv_count-1 + * to avoid overflows. + */ if (inject_map.br_startblock != NULLFSBLOCK) { map[i] = inject_map; i--; - } else - nexleft--; + } bmv->bmv_entries++; cur_ext++; } - } while (nmap && nexleft && bmv->bmv_length && - cur_ext < bmv->bmv_count); + } while (nmap && bmv->bmv_length && cur_ext < bmv->bmv_count - 1); out_free_map: kmem_free(map); -- cgit v1.2.3 From 406dab8450ec76eca88a1af2fc15d18a2b36ca49 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 26 Jan 2017 15:14:52 -0500 Subject: nfs: Fix "Don't increment lock sequence ID after NFS4ERR_MOVED" Lock sequence IDs are bumped in decode_lock by calling nfs_increment_seqid(). nfs_increment_sequid() does not use the seqid_mutating_err() function fixed in commit 059aa7348241 ("Don't increment lock sequence ID after NFS4ERR_MOVED"). Fixes: 059aa7348241 ("Don't increment lock sequence ID after ...") Signed-off-by: Chuck Lever Tested-by: Xuan Qi Cc: stable@vger.kernel.org # v3.7+ Signed-off-by: Trond Myklebust --- fs/nfs/nfs4state.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 90e6193ce6be..daeb94e3acd4 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1091,6 +1091,7 @@ static void nfs_increment_seqid(int status, struct nfs_seqid *seqid) case -NFS4ERR_BADXDR: case -NFS4ERR_RESOURCE: case -NFS4ERR_NOFILEHANDLE: + case -NFS4ERR_MOVED: /* Non-seqid mutating errors */ return; }; -- cgit v1.2.3 From ee6625a948d2e47267ec8fd97307fdd67d0f8a5b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 26 Jan 2017 15:50:41 -0500 Subject: pNFS: Fix a reference leak in _pnfs_return_layout IF NFS_LAYOUT_RETURN_REQUESTED is not set, then we currently exit without freeing the list of invalidated layout segments, leading to a reference leak. Reported-by: Olga Kornievskaia Fixes: 24408f5282 ("pNFS: Fix bugs in _pnfs_return_layout") Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 59554f3adf29..dd042498ce7c 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1200,10 +1200,10 @@ _pnfs_return_layout(struct inode *ino) send = pnfs_prepare_layoutreturn(lo, &stateid, NULL); spin_unlock(&ino->i_lock); - pnfs_free_lseg_list(&tmp_list); if (send) status = pnfs_send_layoutreturn(lo, &stateid, IOMODE_ANY, true); out_put_layout_hdr: + pnfs_free_lseg_list(&tmp_list); pnfs_put_layout_hdr(lo); out: dprintk("<-- %s status: %d\n", __func__, status); -- cgit v1.2.3 From 282e4637bc1c0b338708bcebd09d31c69abec070 Mon Sep 17 00:00:00 2001 From: Jason Gerecke Date: Thu, 26 Jan 2017 09:06:22 -0800 Subject: HID: wacom: Fix poor prox handling in 'wacom_pl_irq' Commit 025bcc1 performed cleanup work on the 'wacom_pl_irq' function, making it follow the standards used in the rest of the codebase. The change unintiontionally allowed the function to send input events from reports that are not marked as being in prox. This can cause problems as the report values for X, Y, etc. are not guaranteed to be correct. In particular, occasionally the tablet will send a report with these values set to zero. If such a report is received it can caus an unexpected jump in the XY position. This patch surrounds more of the processing code with a proximity check, preventing these zeroed reports from overwriting the current state. To be safe, only the tool type and ABS_MISC events should be reported when the pen is marked as being out of prox. Fixes: 025bcc1540 ("HID: wacom: Simplify 'wacom_pl_irq'") Signed-off-by: Jason Gerecke Reviewed-by: Ping Cheng Signed-off-by: Jiri Kosina --- drivers/hid/wacom_wac.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) diff --git a/drivers/hid/wacom_wac.c b/drivers/hid/wacom_wac.c index 0884dc9554fd..672145b0d8f5 100644 --- a/drivers/hid/wacom_wac.c +++ b/drivers/hid/wacom_wac.c @@ -166,19 +166,21 @@ static int wacom_pl_irq(struct wacom_wac *wacom) wacom->id[0] = STYLUS_DEVICE_ID; } - pressure = (signed char)((data[7] << 1) | ((data[4] >> 2) & 1)); - if (features->pressure_max > 255) - pressure = (pressure << 1) | ((data[4] >> 6) & 1); - pressure += (features->pressure_max + 1) / 2; - - input_report_abs(input, ABS_X, data[3] | (data[2] << 7) | ((data[1] & 0x03) << 14)); - input_report_abs(input, ABS_Y, data[6] | (data[5] << 7) | ((data[4] & 0x03) << 14)); - input_report_abs(input, ABS_PRESSURE, pressure); - - input_report_key(input, BTN_TOUCH, data[4] & 0x08); - input_report_key(input, BTN_STYLUS, data[4] & 0x10); - /* Only allow the stylus2 button to be reported for the pen tool. */ - input_report_key(input, BTN_STYLUS2, (wacom->tool[0] == BTN_TOOL_PEN) && (data[4] & 0x20)); + if (prox) { + pressure = (signed char)((data[7] << 1) | ((data[4] >> 2) & 1)); + if (features->pressure_max > 255) + pressure = (pressure << 1) | ((data[4] >> 6) & 1); + pressure += (features->pressure_max + 1) / 2; + + input_report_abs(input, ABS_X, data[3] | (data[2] << 7) | ((data[1] & 0x03) << 14)); + input_report_abs(input, ABS_Y, data[6] | (data[5] << 7) | ((data[4] & 0x03) << 14)); + input_report_abs(input, ABS_PRESSURE, pressure); + + input_report_key(input, BTN_TOUCH, data[4] & 0x08); + input_report_key(input, BTN_STYLUS, data[4] & 0x10); + /* Only allow the stylus2 button to be reported for the pen tool. */ + input_report_key(input, BTN_STYLUS2, (wacom->tool[0] == BTN_TOOL_PEN) && (data[4] & 0x20)); + } if (!prox) wacom->id[0] = 0; -- cgit v1.2.3 From ed9ab4287f96e66340e0390e2c583f2f9110cba0 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Thu, 26 Jan 2017 17:34:40 +0000 Subject: HID: usbhid: Quirk a AMI virtual mouse and keyboard with ALWAYS_POLL Quirking the following AMI USB device with ALWAYS_POLL fixes an AMI virtual keyboard and mouse from not responding and timing out when it is attached to a ppc64el Power 8 system and when we have some rapid open/closes on the mouse device. usb 1-3: new high-speed USB device number 2 using xhci_hcd usb 1-3: New USB device found, idVendor=046b, idProduct=ff01 usb 1-3: New USB device strings: Mfr=1, Product=2, SerialNumber=3 usb 1-3: Product: Virtual Hub usb 1-3: Manufacturer: American Megatrends Inc. usb 1-3: SerialNumber: serial usb 1-3.3: new high-speed USB device number 3 using xhci_hcd usb 1-3.3: New USB device found, idVendor=046b, idProduct=ff31 usb 1-3.3: New USB device strings: Mfr=1, Product=2, SerialNumber=3 usb 1-3.3: Product: Virtual HardDisk Device usb 1-3.3: Manufacturer: American Megatrends Inc. usb 1-3.4: new low-speed USB device number 4 using xhci_hcd usb 1-3.4: New USB device found, idVendor=046b, idProduct=ff10 usb 1-3.4: New USB device strings: Mfr=1, Product=2, SerialNumber=0 usb 1-3.4: Product: Virtual Keyboard and Mouse usb 1-3.4: Manufacturer: American Megatrends Inc. With the quirk I have not been able to trigger the issue with half an hour of saturation soak testing. Signed-off-by: Colin Ian King Signed-off-by: Jiri Kosina --- drivers/hid/hid-ids.h | 3 +++ drivers/hid/usbhid/hid-quirks.c | 1 + 2 files changed, 4 insertions(+) diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index f46f2c5117fa..350accfee8e8 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -76,6 +76,9 @@ #define USB_VENDOR_ID_ALPS_JP 0x044E #define HID_DEVICE_ID_ALPS_U1_DUAL 0x120B +#define USB_VENDOR_ID_AMI 0x046b +#define USB_DEVICE_ID_AMI_VIRT_KEYBOARD_AND_MOUSE 0xff10 + #define USB_VENDOR_ID_ANTON 0x1130 #define USB_DEVICE_ID_ANTON_TOUCH_PAD 0x3101 diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c index e9d6cc7cdfc5..30a2977e2645 100644 --- a/drivers/hid/usbhid/hid-quirks.c +++ b/drivers/hid/usbhid/hid-quirks.c @@ -57,6 +57,7 @@ static const struct hid_blacklist { { USB_VENDOR_ID_AIREN, USB_DEVICE_ID_AIREN_SLIMPLUS, HID_QUIRK_NOGET }, { USB_VENDOR_ID_AKAI, USB_DEVICE_ID_AKAI_MPKMINI2, HID_QUIRK_NO_INIT_REPORTS }, { USB_VENDOR_ID_AKAI_09E8, USB_DEVICE_ID_AKAI_09E8_MIDIMIX, HID_QUIRK_NO_INIT_REPORTS }, + { USB_VENDOR_ID_AMI, USB_DEVICE_ID_AMI_VIRT_KEYBOARD_AND_MOUSE, HID_QUIRK_ALWAYS_POLL }, { USB_VENDOR_ID_ATEN, USB_DEVICE_ID_ATEN_UC100KM, HID_QUIRK_NOGET }, { USB_VENDOR_ID_ATEN, USB_DEVICE_ID_ATEN_CS124U, HID_QUIRK_NOGET }, { USB_VENDOR_ID_ATEN, USB_DEVICE_ID_ATEN_2PORTKVM, HID_QUIRK_NOGET }, -- cgit v1.2.3 From 877a021e08ccb6434718c0cc781fdf943c884cc0 Mon Sep 17 00:00:00 2001 From: Ardinartsev Nikita Date: Thu, 26 Jan 2017 16:54:42 +0300 Subject: HID: hid-lg: Fix immediate disconnection of Logitech Rumblepad 2 With NOGET quirk Logitech F510 is now fully workable in dinput mode including rumble effects (according to fftest). Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=117091 [jkosina@suse.cz: fix patch format] Signed-off-by: Ardinartsev Nikita Acked-by: Benjamin Tissoires Signed-off-by: Jiri Kosina --- drivers/hid/hid-lg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hid/hid-lg.c b/drivers/hid/hid-lg.c index c5c5fbe9d605..52026dc94d5c 100644 --- a/drivers/hid/hid-lg.c +++ b/drivers/hid/hid-lg.c @@ -872,7 +872,7 @@ static const struct hid_device_id lg_devices[] = { { HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_WINGMAN_FFG), .driver_data = LG_NOGET | LG_FF4 }, { HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_RUMBLEPAD2), - .driver_data = LG_FF2 }, + .driver_data = LG_NOGET | LG_FF2 }, { HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_FLIGHT_SYSTEM_G940), .driver_data = LG_FF3 }, { HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_SPACENAVIGATOR), -- cgit v1.2.3 From 07cd12945551b63ecb1a349d50a6d69d1d6feb4a Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 26 Jan 2017 16:47:28 -0500 Subject: cgroup: don't online subsystems before cgroup_name/path() are operational While refactoring cgroup creation, a5bca2152036 ("cgroup: factor out cgroup_create() out of cgroup_mkdir()") incorrectly onlined subsystems before the new cgroup is associated with it kernfs_node. This is fine for cgroup proper but cgroup_name/path() depend on the associated kernfs_node and if a subsystem makes the new cgroup_subsys_state visible, which they're allowed to after onlining, it can lead to NULL dereference. The current code performs cgroup creation and subsystem onlining in cgroup_create() and cgroup_mkdir() makes the cgroup and subsystems visible afterwards. There's no reason to online the subsystems early and we can simply drop cgroup_apply_control_enable() call from cgroup_create() so that the subsystems are onlined and made visible at the same time. Signed-off-by: Tejun Heo Reported-by: Konstantin Khlebnikov Fixes: a5bca2152036 ("cgroup: factor out cgroup_create() out of cgroup_mkdir()") Cc: stable@vger.kernel.org # v4.6+ --- kernel/cgroup.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 2ee9ec3051b2..688dd02af985 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -5221,6 +5221,11 @@ err_free_css: return ERR_PTR(err); } +/* + * The returned cgroup is fully initialized including its control mask, but + * it isn't associated with its kernfs_node and doesn't have the control + * mask applied. + */ static struct cgroup *cgroup_create(struct cgroup *parent) { struct cgroup_root *root = parent->root; @@ -5288,11 +5293,6 @@ static struct cgroup *cgroup_create(struct cgroup *parent) cgroup_propagate_control(cgrp); - /* @cgrp doesn't have dir yet so the following will only create csses */ - ret = cgroup_apply_control_enable(cgrp); - if (ret) - goto out_destroy; - return cgrp; out_cancel_ref: @@ -5300,9 +5300,6 @@ out_cancel_ref: out_free_cgrp: kfree(cgrp); return ERR_PTR(ret); -out_destroy: - cgroup_destroy_locked(cgrp); - return ERR_PTR(ret); } static int cgroup_mkdir(struct kernfs_node *parent_kn, const char *name, -- cgit v1.2.3 From 586655d278ba08af7b198b93217746f9a506ee8a Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Wed, 25 Jan 2017 00:44:16 +0100 Subject: rtc: jz4740: make the driver buildable as a module again By using kernel_halt() instead of machine_halt(), we can make the driver build as a module. However, jz4740 platforms not loading this module will not be able to power off. Suggested-by: Geert Uytterhoeven Signed-off-by: Alexandre Belloni Revert "rtc: jz4740: make the driver builtin only" This reverts commit b9168c539c0b2de756aaffd380384dbde8adbe07. --- drivers/rtc/Kconfig | 5 ++++- drivers/rtc/rtc-jz4740.c | 12 ++++++++++-- 2 files changed, 14 insertions(+), 3 deletions(-) diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig index c93c5a8fba32..5dc673dc9487 100644 --- a/drivers/rtc/Kconfig +++ b/drivers/rtc/Kconfig @@ -1551,12 +1551,15 @@ config RTC_DRV_MPC5121 will be called rtc-mpc5121. config RTC_DRV_JZ4740 - bool "Ingenic JZ4740 SoC" + tristate "Ingenic JZ4740 SoC" depends on MACH_INGENIC || COMPILE_TEST help If you say yes here you get support for the Ingenic JZ47xx SoCs RTC controllers. + This driver can also be buillt as a module. If so, the module + will be called rtc-jz4740. + config RTC_DRV_LPC24XX tristate "NXP RTC for LPC178x/18xx/408x/43xx" depends on ARCH_LPC18XX || COMPILE_TEST diff --git a/drivers/rtc/rtc-jz4740.c b/drivers/rtc/rtc-jz4740.c index 72918c1ba092..64989afffa3d 100644 --- a/drivers/rtc/rtc-jz4740.c +++ b/drivers/rtc/rtc-jz4740.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -294,7 +295,7 @@ static void jz4740_rtc_power_off(void) JZ_REG_RTC_RESET_COUNTER, reset_counter_ticks); jz4740_rtc_poweroff(dev_for_power_off); - machine_halt(); + kernel_halt(); } static const struct of_device_id jz4740_rtc_of_match[] = { @@ -302,6 +303,7 @@ static const struct of_device_id jz4740_rtc_of_match[] = { { .compatible = "ingenic,jz4780-rtc", .data = (void *)ID_JZ4780 }, {}, }; +MODULE_DEVICE_TABLE(of, jz4740_rtc_of_match); static int jz4740_rtc_probe(struct platform_device *pdev) { @@ -429,6 +431,7 @@ static const struct platform_device_id jz4740_rtc_ids[] = { { "jz4780-rtc", ID_JZ4780 }, {} }; +MODULE_DEVICE_TABLE(platform, jz4740_rtc_ids); static struct platform_driver jz4740_rtc_driver = { .probe = jz4740_rtc_probe, @@ -440,4 +443,9 @@ static struct platform_driver jz4740_rtc_driver = { .id_table = jz4740_rtc_ids, }; -builtin_platform_driver(jz4740_rtc_driver); +module_platform_driver(jz4740_rtc_driver); + +MODULE_AUTHOR("Lars-Peter Clausen "); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("RTC driver for the JZ4740 SoC\n"); +MODULE_ALIAS("platform:jz4740-rtc"); -- cgit v1.2.3 From a47b70ea86bdeb3091341f5ae3ef580f1a1ad822 Mon Sep 17 00:00:00 2001 From: Kazuya Mizuguchi Date: Thu, 26 Jan 2017 14:29:27 +0100 Subject: ravb: unmap descriptors when freeing rings "swiotlb buffer is full" errors occur after repeated initialisation of a device - f.e. suspend/resume or ip link set up/down. This is because memory mapped using dma_map_single() in ravb_ring_format() and ravb_start_xmit() is not released. Resolve this problem by unmapping descriptors when freeing rings. Fixes: c156633f1353 ("Renesas Ethernet AVB driver proper") Signed-off-by: Kazuya Mizuguchi [simon: reworked] Signed-off-by: Simon Horman Acked-by: Sergei Shtylyov Signed-off-by: David S. Miller --- drivers/net/ethernet/renesas/ravb_main.c | 112 ++++++++++++++++++------------- 1 file changed, 64 insertions(+), 48 deletions(-) diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 89ac1e3f6175..301f48755093 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -179,6 +179,49 @@ static struct mdiobb_ops bb_ops = { .get_mdio_data = ravb_get_mdio_data, }; +/* Free TX skb function for AVB-IP */ +static int ravb_tx_free(struct net_device *ndev, int q, bool free_txed_only) +{ + struct ravb_private *priv = netdev_priv(ndev); + struct net_device_stats *stats = &priv->stats[q]; + struct ravb_tx_desc *desc; + int free_num = 0; + int entry; + u32 size; + + for (; priv->cur_tx[q] - priv->dirty_tx[q] > 0; priv->dirty_tx[q]++) { + bool txed; + + entry = priv->dirty_tx[q] % (priv->num_tx_ring[q] * + NUM_TX_DESC); + desc = &priv->tx_ring[q][entry]; + txed = desc->die_dt == DT_FEMPTY; + if (free_txed_only && !txed) + break; + /* Descriptor type must be checked before all other reads */ + dma_rmb(); + size = le16_to_cpu(desc->ds_tagl) & TX_DS; + /* Free the original skb. */ + if (priv->tx_skb[q][entry / NUM_TX_DESC]) { + dma_unmap_single(ndev->dev.parent, le32_to_cpu(desc->dptr), + size, DMA_TO_DEVICE); + /* Last packet descriptor? */ + if (entry % NUM_TX_DESC == NUM_TX_DESC - 1) { + entry /= NUM_TX_DESC; + dev_kfree_skb_any(priv->tx_skb[q][entry]); + priv->tx_skb[q][entry] = NULL; + if (txed) + stats->tx_packets++; + } + free_num++; + } + if (txed) + stats->tx_bytes += size; + desc->die_dt = DT_EEMPTY; + } + return free_num; +} + /* Free skb's and DMA buffers for Ethernet AVB */ static void ravb_ring_free(struct net_device *ndev, int q) { @@ -194,19 +237,21 @@ static void ravb_ring_free(struct net_device *ndev, int q) kfree(priv->rx_skb[q]); priv->rx_skb[q] = NULL; - /* Free TX skb ringbuffer */ - if (priv->tx_skb[q]) { - for (i = 0; i < priv->num_tx_ring[q]; i++) - dev_kfree_skb(priv->tx_skb[q][i]); - } - kfree(priv->tx_skb[q]); - priv->tx_skb[q] = NULL; - /* Free aligned TX buffers */ kfree(priv->tx_align[q]); priv->tx_align[q] = NULL; if (priv->rx_ring[q]) { + for (i = 0; i < priv->num_rx_ring[q]; i++) { + struct ravb_ex_rx_desc *desc = &priv->rx_ring[q][i]; + + if (!dma_mapping_error(ndev->dev.parent, + le32_to_cpu(desc->dptr))) + dma_unmap_single(ndev->dev.parent, + le32_to_cpu(desc->dptr), + PKT_BUF_SZ, + DMA_FROM_DEVICE); + } ring_size = sizeof(struct ravb_ex_rx_desc) * (priv->num_rx_ring[q] + 1); dma_free_coherent(ndev->dev.parent, ring_size, priv->rx_ring[q], @@ -215,12 +260,20 @@ static void ravb_ring_free(struct net_device *ndev, int q) } if (priv->tx_ring[q]) { + ravb_tx_free(ndev, q, false); + ring_size = sizeof(struct ravb_tx_desc) * (priv->num_tx_ring[q] * NUM_TX_DESC + 1); dma_free_coherent(ndev->dev.parent, ring_size, priv->tx_ring[q], priv->tx_desc_dma[q]); priv->tx_ring[q] = NULL; } + + /* Free TX skb ringbuffer. + * SKBs are freed by ravb_tx_free() call above. + */ + kfree(priv->tx_skb[q]); + priv->tx_skb[q] = NULL; } /* Format skb and descriptor buffer for Ethernet AVB */ @@ -431,44 +484,6 @@ static int ravb_dmac_init(struct net_device *ndev) return 0; } -/* Free TX skb function for AVB-IP */ -static int ravb_tx_free(struct net_device *ndev, int q) -{ - struct ravb_private *priv = netdev_priv(ndev); - struct net_device_stats *stats = &priv->stats[q]; - struct ravb_tx_desc *desc; - int free_num = 0; - int entry; - u32 size; - - for (; priv->cur_tx[q] - priv->dirty_tx[q] > 0; priv->dirty_tx[q]++) { - entry = priv->dirty_tx[q] % (priv->num_tx_ring[q] * - NUM_TX_DESC); - desc = &priv->tx_ring[q][entry]; - if (desc->die_dt != DT_FEMPTY) - break; - /* Descriptor type must be checked before all other reads */ - dma_rmb(); - size = le16_to_cpu(desc->ds_tagl) & TX_DS; - /* Free the original skb. */ - if (priv->tx_skb[q][entry / NUM_TX_DESC]) { - dma_unmap_single(ndev->dev.parent, le32_to_cpu(desc->dptr), - size, DMA_TO_DEVICE); - /* Last packet descriptor? */ - if (entry % NUM_TX_DESC == NUM_TX_DESC - 1) { - entry /= NUM_TX_DESC; - dev_kfree_skb_any(priv->tx_skb[q][entry]); - priv->tx_skb[q][entry] = NULL; - stats->tx_packets++; - } - free_num++; - } - stats->tx_bytes += size; - desc->die_dt = DT_EEMPTY; - } - return free_num; -} - static void ravb_get_tx_tstamp(struct net_device *ndev) { struct ravb_private *priv = netdev_priv(ndev); @@ -902,7 +917,7 @@ static int ravb_poll(struct napi_struct *napi, int budget) spin_lock_irqsave(&priv->lock, flags); /* Clear TX interrupt */ ravb_write(ndev, ~mask, TIS); - ravb_tx_free(ndev, q); + ravb_tx_free(ndev, q, true); netif_wake_subqueue(ndev, q); mmiowb(); spin_unlock_irqrestore(&priv->lock, flags); @@ -1567,7 +1582,8 @@ static netdev_tx_t ravb_start_xmit(struct sk_buff *skb, struct net_device *ndev) priv->cur_tx[q] += NUM_TX_DESC; if (priv->cur_tx[q] - priv->dirty_tx[q] > - (priv->num_tx_ring[q] - 1) * NUM_TX_DESC && !ravb_tx_free(ndev, q)) + (priv->num_tx_ring[q] - 1) * NUM_TX_DESC && + !ravb_tx_free(ndev, q, true)) netif_stop_subqueue(ndev, q); exit: -- cgit v1.2.3 From 67ade058ef2c65a3e56878af9c293ec76722a2e5 Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Wed, 25 Jan 2017 17:06:38 -0800 Subject: Btrfs: remove old tree_root case in btrfs_read_locked_inode() As Jeff explained in c2951f32d36c ("btrfs: remove old tree_root dirent processing in btrfs_real_readdir()"), supporting this old format is no longer necessary since the Btrfs magic number has been updated since we changed to the current format. There are other places where we still handle this old format, but since this is part of a fix that is going to stable, I'm only removing this one for now. Cc: # 4.9.x Signed-off-by: Omar Sandoval Reviewed-by: David Sterba Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 177e7284909c..e2b961e20ec1 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3835,10 +3835,7 @@ cache_acl: break; case S_IFDIR: inode->i_fop = &btrfs_dir_file_operations; - if (root == fs_info->tree_root) - inode->i_op = &btrfs_dir_ro_inode_operations; - else - inode->i_op = &btrfs_dir_inode_operations; + inode->i_op = &btrfs_dir_inode_operations; break; case S_IFLNK: inode->i_op = &btrfs_symlink_inode_operations; -- cgit v1.2.3 From 1fdf41941b8010691679638f8d0c8d08cfee7726 Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Wed, 25 Jan 2017 17:06:39 -0800 Subject: Btrfs: disable xattr operations on subvolume directories When you snapshot a subvolume containing a subvolume, you get a placeholder directory where the subvolume would be. These directory inodes have ->i_ops set to btrfs_dir_ro_inode_operations. Previously, these i_ops didn't include the xattr operation callbacks. The conversion to xattr_handlers missed this case, leading to bogus attempts to set xattrs on these inodes. This manifested itself as failures when running delayed inodes. To fix this, clear IOP_XATTR in ->i_opflags on these inodes. Fixes: 6c6ef9f26e59 ("xattr: Stop calling {get,set,remove}xattr inode operations") Cc: Andreas Gruenbacher Reported-by: Chris Murphy Tested-by: Chris Murphy Cc: # 4.9.x Signed-off-by: Omar Sandoval Reviewed-by: David Sterba Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index e2b961e20ec1..16694325a449 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -5718,6 +5718,7 @@ static struct inode *new_simple_dir(struct super_block *s, inode->i_ino = BTRFS_EMPTY_SUBVOL_DIR_OBJECTID; inode->i_op = &btrfs_dir_ro_inode_operations; + inode->i_opflags &= ~IOP_XATTR; inode->i_fop = &simple_dir_operations; inode->i_mode = S_IFDIR | S_IRUGO | S_IWUSR | S_IXUGO; inode->i_mtime = current_time(inode); -- cgit v1.2.3 From 57b59ed2e5b91e958843609c7884794e29e6c4cb Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Wed, 25 Jan 2017 17:06:40 -0800 Subject: Btrfs: remove ->{get, set}_acl() from btrfs_dir_ro_inode_operations Subvolume directory inodes can't have ACLs. Cc: # 4.9.x Signed-off-by: Omar Sandoval Reviewed-by: David Sterba Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 16694325a449..8a8e719778bd 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -10592,8 +10592,6 @@ static const struct inode_operations btrfs_dir_inode_operations = { static const struct inode_operations btrfs_dir_ro_inode_operations = { .lookup = btrfs_lookup, .permission = btrfs_permission, - .get_acl = btrfs_get_acl, - .set_acl = btrfs_set_acl, .update_time = btrfs_update_time, }; -- cgit v1.2.3 From 6c971c09f38704513c426ba6515f22fb3d6c87d5 Mon Sep 17 00:00:00 2001 From: Y.C. Chen Date: Thu, 26 Jan 2017 09:45:40 +0800 Subject: drm/ast: Fixed system hanged if disable P2A The original ast driver will access some BMC configuration through P2A bridge that can be disabled since AST2300 and after. It will cause system hanged if P2A bridge is disabled. Here is the update to fix it. Signed-off-by: Y.C. Chen Signed-off-by: Dave Airlie --- drivers/gpu/drm/ast/ast_drv.h | 1 + drivers/gpu/drm/ast/ast_main.c | 157 ++++++++++++++++++++++------------------- drivers/gpu/drm/ast/ast_post.c | 18 +++-- 3 files changed, 97 insertions(+), 79 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 908011d2c8f5..7abda94fc2cf 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -113,6 +113,7 @@ struct ast_private { struct ttm_bo_kmap_obj cache_kmap; int next_cursor; bool support_wide_screen; + bool DisableP2A; enum ast_tx_chip tx_chip_type; u8 dp501_maxclk; diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index f75c6421db62..533e762d036d 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -124,6 +124,12 @@ static int ast_detect_chip(struct drm_device *dev, bool *need_post) } else *need_post = false; + /* Check P2A Access */ + ast->DisableP2A = true; + data = ast_read32(ast, 0xf004); + if (data != 0xFFFFFFFF) + ast->DisableP2A = false; + /* Check if we support wide screen */ switch (ast->chip) { case AST1180: @@ -140,15 +146,17 @@ static int ast_detect_chip(struct drm_device *dev, bool *need_post) ast->support_wide_screen = true; else { ast->support_wide_screen = false; - /* Read SCU7c (silicon revision register) */ - ast_write32(ast, 0xf004, 0x1e6e0000); - ast_write32(ast, 0xf000, 0x1); - data = ast_read32(ast, 0x1207c); - data &= 0x300; - if (ast->chip == AST2300 && data == 0x0) /* ast1300 */ - ast->support_wide_screen = true; - if (ast->chip == AST2400 && data == 0x100) /* ast1400 */ - ast->support_wide_screen = true; + if (ast->DisableP2A == false) { + /* Read SCU7c (silicon revision register) */ + ast_write32(ast, 0xf004, 0x1e6e0000); + ast_write32(ast, 0xf000, 0x1); + data = ast_read32(ast, 0x1207c); + data &= 0x300; + if (ast->chip == AST2300 && data == 0x0) /* ast1300 */ + ast->support_wide_screen = true; + if (ast->chip == AST2400 && data == 0x100) /* ast1400 */ + ast->support_wide_screen = true; + } } break; } @@ -216,80 +224,81 @@ static int ast_get_dram_info(struct drm_device *dev) uint32_t data, data2; uint32_t denum, num, div, ref_pll; - ast_write32(ast, 0xf004, 0x1e6e0000); - ast_write32(ast, 0xf000, 0x1); - - - ast_write32(ast, 0x10000, 0xfc600309); - - do { - if (pci_channel_offline(dev->pdev)) - return -EIO; - } while (ast_read32(ast, 0x10000) != 0x01); - data = ast_read32(ast, 0x10004); - - if (data & 0x40) + if (ast->DisableP2A) + { ast->dram_bus_width = 16; + ast->dram_type = AST_DRAM_1Gx16; + ast->mclk = 396; + } else - ast->dram_bus_width = 32; + { + ast_write32(ast, 0xf004, 0x1e6e0000); + ast_write32(ast, 0xf000, 0x1); + data = ast_read32(ast, 0x10004); + + if (data & 0x40) + ast->dram_bus_width = 16; + else + ast->dram_bus_width = 32; + + if (ast->chip == AST2300 || ast->chip == AST2400) { + switch (data & 0x03) { + case 0: + ast->dram_type = AST_DRAM_512Mx16; + break; + default: + case 1: + ast->dram_type = AST_DRAM_1Gx16; + break; + case 2: + ast->dram_type = AST_DRAM_2Gx16; + break; + case 3: + ast->dram_type = AST_DRAM_4Gx16; + break; + } + } else { + switch (data & 0x0c) { + case 0: + case 4: + ast->dram_type = AST_DRAM_512Mx16; + break; + case 8: + if (data & 0x40) + ast->dram_type = AST_DRAM_1Gx16; + else + ast->dram_type = AST_DRAM_512Mx32; + break; + case 0xc: + ast->dram_type = AST_DRAM_1Gx32; + break; + } + } - if (ast->chip == AST2300 || ast->chip == AST2400) { - switch (data & 0x03) { - case 0: - ast->dram_type = AST_DRAM_512Mx16; - break; - default: - case 1: - ast->dram_type = AST_DRAM_1Gx16; - break; - case 2: - ast->dram_type = AST_DRAM_2Gx16; - break; + data = ast_read32(ast, 0x10120); + data2 = ast_read32(ast, 0x10170); + if (data2 & 0x2000) + ref_pll = 14318; + else + ref_pll = 12000; + + denum = data & 0x1f; + num = (data & 0x3fe0) >> 5; + data = (data & 0xc000) >> 14; + switch (data) { case 3: - ast->dram_type = AST_DRAM_4Gx16; - break; - } - } else { - switch (data & 0x0c) { - case 0: - case 4: - ast->dram_type = AST_DRAM_512Mx16; + div = 0x4; break; - case 8: - if (data & 0x40) - ast->dram_type = AST_DRAM_1Gx16; - else - ast->dram_type = AST_DRAM_512Mx32; + case 2: + case 1: + div = 0x2; break; - case 0xc: - ast->dram_type = AST_DRAM_1Gx32; + default: + div = 0x1; break; } + ast->mclk = ref_pll * (num + 2) / (denum + 2) * (div * 1000); } - - data = ast_read32(ast, 0x10120); - data2 = ast_read32(ast, 0x10170); - if (data2 & 0x2000) - ref_pll = 14318; - else - ref_pll = 12000; - - denum = data & 0x1f; - num = (data & 0x3fe0) >> 5; - data = (data & 0xc000) >> 14; - switch (data) { - case 3: - div = 0x4; - break; - case 2: - case 1: - div = 0x2; - break; - default: - div = 0x1; - break; - } - ast->mclk = ref_pll * (num + 2) / (denum + 2) * (div * 1000); return 0; } diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c index 810c51d92b99..5331ee1df086 100644 --- a/drivers/gpu/drm/ast/ast_post.c +++ b/drivers/gpu/drm/ast/ast_post.c @@ -379,12 +379,20 @@ void ast_post_gpu(struct drm_device *dev) ast_open_key(ast); ast_set_def_ext_reg(dev); - if (ast->chip == AST2300 || ast->chip == AST2400) - ast_init_dram_2300(dev); - else - ast_init_dram_reg(dev); + if (ast->DisableP2A == false) + { + if (ast->chip == AST2300 || ast->chip == AST2400) + ast_init_dram_2300(dev); + else + ast_init_dram_reg(dev); - ast_init_3rdtx(dev); + ast_init_3rdtx(dev); + } + else + { + if (ast->tx_chip_type != AST_TX_NONE) + ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xa3, 0xcf, 0x80); /* Enable DVO */ + } } /* AST 2300 DRAM settings */ -- cgit v1.2.3 From cae9ff036eea577856d5b12860b4c79c5e71db4a Mon Sep 17 00:00:00 2001 From: Lyude Paul Date: Wed, 11 Jan 2017 21:25:23 -0500 Subject: drm/nouveau: Don't enabling polling twice on runtime resume As it turns out, on cards that actually have CRTCs on them we're already calling drm_kms_helper_poll_enable(drm_dev) from nouveau_display_resume() before we call it in nouveau_pmops_runtime_resume(). This leads us to accidentally trying to enable polling twice, which results in a potential deadlock between the RPM locks and drm_dev->mode_config.mutex if we end up trying to enable polling the second time while output_poll_execute is running and holding the mode_config lock. As such, make sure we only enable polling in nouveau_pmops_runtime_resume() if we need to. This fixes hangs observed on the ThinkPad W541 Signed-off-by: Lyude Cc: Hans de Goede Cc: Kilian Singer Cc: Lukas Wunner Cc: David Airlie Signed-off-by: Dave Airlie --- drivers/gpu/drm/nouveau/nouveau_display.c | 3 ++- drivers/gpu/drm/nouveau/nouveau_drm.c | 5 ++++- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c index cef08da1da4e..6a157763dfc3 100644 --- a/drivers/gpu/drm/nouveau/nouveau_display.c +++ b/drivers/gpu/drm/nouveau/nouveau_display.c @@ -411,7 +411,8 @@ nouveau_display_init(struct drm_device *dev) return ret; /* enable polling for external displays */ - drm_kms_helper_poll_enable(dev); + if (!dev->mode_config.poll_enabled) + drm_kms_helper_poll_enable(dev); /* enable hotplug interrupts */ list_for_each_entry(connector, &dev->mode_config.connector_list, head) { diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 59348fc41c77..bc85a45f91cd 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -773,7 +773,10 @@ nouveau_pmops_runtime_resume(struct device *dev) pci_set_master(pdev); ret = nouveau_do_resume(drm_dev, true); - drm_kms_helper_poll_enable(drm_dev); + + if (!drm_dev->mode_config.poll_enabled) + drm_kms_helper_poll_enable(drm_dev); + /* do magic */ nvif_mask(&device->object, 0x088488, (1 << 25), (1 << 25)); vga_switcheroo_set_dynamic_switch(pdev, VGA_SWITCHEROO_ON); -- cgit v1.2.3 From 15266ae38fe09dae07bd8812cb7a7717b1e1d992 Mon Sep 17 00:00:00 2001 From: Lyude Paul Date: Wed, 11 Jan 2017 21:25:24 -0500 Subject: drm/nouveau: Handle fbcon suspend/resume in seperate worker Resuming from RPM can happen while already holding dev->mode_config.mutex. This means we can't actually handle fbcon in any RPM resume workers, since restoring fbcon requires grabbing dev->mode_config.mutex again. So move the fbcon suspend/resume code into it's own worker, and rely on that instead to avoid deadlocking. This fixes more deadlocks for runtime suspending the GPU on the ThinkPad W541. Reproduction recipe: - Get a machine with both optimus and a nvidia card with connectors attached to it - Wait for the nvidia GPU to suspend - Attempt to manually reprobe any of the connectors on the nvidia GPU using sysfs - *deadlock* [airlied: use READ_ONCE to address Hans's comment] Signed-off-by: Lyude Cc: Hans de Goede Cc: Kilian Singer Cc: Lukas Wunner Cc: David Airlie Signed-off-by: Dave Airlie --- drivers/gpu/drm/nouveau/nouveau_drv.h | 2 ++ drivers/gpu/drm/nouveau/nouveau_fbcon.c | 43 ++++++++++++++++++++++++++------- 2 files changed, 36 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h index 8d5ed5bfdacb..42c1fa53d431 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drv.h +++ b/drivers/gpu/drm/nouveau/nouveau_drv.h @@ -165,6 +165,8 @@ struct nouveau_drm { struct backlight_device *backlight; struct list_head bl_connectors; struct work_struct hpd_work; + struct work_struct fbcon_work; + int fbcon_new_state; #ifdef CONFIG_ACPI struct notifier_block acpi_nb; #endif diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c index 2f2a3dcd4ad7..fa2d0a978ccc 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c +++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c @@ -470,19 +470,43 @@ static const struct drm_fb_helper_funcs nouveau_fbcon_helper_funcs = { .fb_probe = nouveau_fbcon_create, }; +static void +nouveau_fbcon_set_suspend_work(struct work_struct *work) +{ + struct nouveau_drm *drm = container_of(work, typeof(*drm), fbcon_work); + int state = READ_ONCE(drm->fbcon_new_state); + + if (state == FBINFO_STATE_RUNNING) + pm_runtime_get_sync(drm->dev->dev); + + console_lock(); + if (state == FBINFO_STATE_RUNNING) + nouveau_fbcon_accel_restore(drm->dev); + drm_fb_helper_set_suspend(&drm->fbcon->helper, state); + if (state != FBINFO_STATE_RUNNING) + nouveau_fbcon_accel_save_disable(drm->dev); + console_unlock(); + + if (state == FBINFO_STATE_RUNNING) { + pm_runtime_mark_last_busy(drm->dev->dev); + pm_runtime_put_sync(drm->dev->dev); + } +} + void nouveau_fbcon_set_suspend(struct drm_device *dev, int state) { struct nouveau_drm *drm = nouveau_drm(dev); - if (drm->fbcon) { - console_lock(); - if (state == FBINFO_STATE_RUNNING) - nouveau_fbcon_accel_restore(dev); - drm_fb_helper_set_suspend(&drm->fbcon->helper, state); - if (state != FBINFO_STATE_RUNNING) - nouveau_fbcon_accel_save_disable(dev); - console_unlock(); - } + + if (!drm->fbcon) + return; + + drm->fbcon_new_state = state; + /* Since runtime resume can happen as a result of a sysfs operation, + * it's possible we already have the console locked. So handle fbcon + * init/deinit from a seperate work thread + */ + schedule_work(&drm->fbcon_work); } int @@ -502,6 +526,7 @@ nouveau_fbcon_init(struct drm_device *dev) return -ENOMEM; drm->fbcon = fbcon; + INIT_WORK(&drm->fbcon_work, nouveau_fbcon_set_suspend_work); drm_fb_helper_prepare(dev, &fbcon->helper, &nouveau_fbcon_helper_funcs); -- cgit v1.2.3 From 191e885a2e130e639bb0c8ee350d7047294f2ce6 Mon Sep 17 00:00:00 2001 From: Luis R. Rodriguez Date: Wed, 25 Jan 2017 10:31:52 -0800 Subject: firmware: fix NULL pointer dereference in __fw_load_abort() Since commit 5d47ec02c37ea6 ("firmware: Correct handling of fw_state_wait() return value") fw_load_abort() could be called twice and lead us to a kernel crash. This happens only when the firmware fallback mechanism (regular or custom) is used. The fallback mechanism exposes a sysfs interface for userspace to upload a file and notify the kernel when the file is loaded and ready, or to cancel an upload by echo'ing -1 into on the loading file: echo -n "-1" > /sys/$DEVPATH/loading This will call fw_load_abort(). Some distributions actually have a udev rule in place to *always* immediately cancel all firmware fallback mechanism requests (Debian), they have: $ cat /lib/udev/rules.d/50-firmware.rules # stub for immediately telling the kernel that userspace firmware loading # failed; necessary to avoid long timeouts with CONFIG_FW_LOADER_USER_HELPER=y SUBSYSTEM=="firmware", ACTION=="add", ATTR{loading}="-1 Distributions with this udev rule would run into this crash only if the fallback mechanism is used. Since most distributions disable by default using the fallback mechanism (CONFIG_FW_LOADER_USER_HELPER_FALLBACK), this would typicaly mean only 2 drivers which *require* the fallback mechanism could typically incur a crash: drivers/firmware/dell_rbu.c and the drivers/leds/leds-lp55xx-common.c driver. Distributions enabling CONFIG_FW_LOADER_USER_HELPER_FALLBACK by default are obviously more exposed to this crash. The crash happens because after commit 5b029624948d ("firmware: do not use fw_lock for fw_state protection") and subsequent fix commit 5d47ec02c37ea6 ("firmware: Correct handling of fw_state_wait() return value") a race can happen between this cancelation and the firmware fw_state_wait_timeout() being woken up after a state change with which fw_load_abort() as that calls swake_up(). Upon error fw_state_wait_timeout() will also again call fw_load_abort() and trigger a null reference. At first glance we could just fix this with a !buf check on fw_load_abort() before accessing buf->fw_st, however there is a logical issue in having a state machine used for the fallback mechanism and preventing access from it once we abort as its inside the buf (buf->fw_st). The firmware_class.c code is setting the buf to NULL to annotate an abort has occurred. Replace this mechanism by simply using the state check instead. All the other code in place already uses similar checks for aborting as well so no further changes are needed. An oops can be reproduced with the new fw_fallback.sh fallback mechanism cancellation test. Either cancelling the fallback mechanism or the custom fallback mechanism triggers a crash. mcgrof@piggy ~/linux-next/tools/testing/selftests/firmware (git::20170111-fw-fixes)$ sudo ./fw_fallback.sh ./fw_fallback.sh: timeout works ./fw_fallback.sh: firmware comparison works ./fw_fallback.sh: fallback mechanism works [ this then sits here when it is trying the cancellation test ] Kernel log: test_firmware: loading 'nope-test-firmware.bin' misc test_firmware: Direct firmware load for nope-test-firmware.bin failed with error -2 misc test_firmware: Falling back to user helper BUG: unable to handle kernel NULL pointer dereference at 0000000000000038 IP: _request_firmware+0xa27/0xad0 PGD 0 Oops: 0000 [#1] SMP Modules linked in: test_firmware(E) ... etc ... CPU: 1 PID: 1396 Comm: fw_fallback.sh Tainted: G W E 4.10.0-rc3-next-20170111+ #30 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.10.1-0-g8891697-prebuilt.qemu-project.org 04/01/2014 task: ffff9740b27f4340 task.stack: ffffbb15c0bc8000 RIP: 0010:_request_firmware+0xa27/0xad0 RSP: 0018:ffffbb15c0bcbd10 EFLAGS: 00010246 RAX: 00000000fffffffe RBX: ffff9740afe5aa80 RCX: 0000000000000000 RDX: ffff9740b27f4340 RSI: 0000000000000283 RDI: 0000000000000000 RBP: ffffbb15c0bcbd90 R08: ffffbb15c0bcbcd8 R09: 0000000000000000 R10: 0000000894a0d4b1 R11: 000000000000008c R12: ffffffffc0312480 R13: 0000000000000005 R14: ffff9740b1c32400 R15: 00000000000003e8 FS: 00007f8604422700(0000) GS:ffff9740bfc80000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000038 CR3: 000000012164c000 CR4: 00000000000006e0 Call Trace: request_firmware+0x37/0x50 trigger_request_store+0x79/0xd0 [test_firmware] dev_attr_store+0x18/0x30 sysfs_kf_write+0x37/0x40 kernfs_fop_write+0x110/0x1a0 __vfs_write+0x37/0x160 ? _cond_resched+0x1a/0x50 vfs_write+0xb5/0x1a0 SyS_write+0x55/0xc0 ? trace_do_page_fault+0x37/0xd0 entry_SYSCALL_64_fastpath+0x1e/0xad RIP: 0033:0x7f8603f49620 RSP: 002b:00007fff6287b788 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 000055c307b110a0 RCX: 00007f8603f49620 RDX: 0000000000000016 RSI: 000055c3084d8a90 RDI: 0000000000000001 RBP: 0000000000000016 R08: 000000000000c0ff R09: 000055c3084d6336 R10: 000055c307b108b0 R11: 0000000000000246 R12: 000055c307b13c80 R13: 000055c3084d6320 R14: 0000000000000000 R15: 00007fff6287b950 Code: 9f 64 84 e8 9c 61 fe ff b8 f4 ff ff ff e9 6b f9 ff ff 48 c7 c7 40 6b 8d 84 89 45 a8 e8 43 84 18 00 49 8b be 00 03 00 00 8b 45 a8 <83> 7f 38 02 74 08 e8 6e ec ff ff 8b 45 a8 49 c7 86 00 03 00 00 RIP: _request_firmware+0xa27/0xad0 RSP: ffffbb15c0bcbd10 CR2: 0000000000000038 ---[ end trace 6d94ac339c133e6f ]--- Fixes: 5d47ec02c37e ("firmware: Correct handling of fw_state_wait() return value") Reported-and-Tested-by: Jakub Kicinski Reported-and-Tested-by: Patrick Bruenn Reported-by: Chris Wilson CC: [3.10+] Signed-off-by: Luis R. Rodriguez Signed-off-by: Greg Kroah-Hartman --- drivers/base/firmware_class.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c index 4497d263209f..ac350c518e0c 100644 --- a/drivers/base/firmware_class.c +++ b/drivers/base/firmware_class.c @@ -558,9 +558,6 @@ static void fw_load_abort(struct firmware_priv *fw_priv) struct firmware_buf *buf = fw_priv->buf; __fw_load_abort(buf); - - /* avoid user action after loading abort */ - fw_priv->buf = NULL; } static LIST_HEAD(pending_fw_head); @@ -713,7 +710,7 @@ static ssize_t firmware_loading_store(struct device *dev, mutex_lock(&fw_lock); fw_buf = fw_priv->buf; - if (!fw_buf) + if (fw_state_is_aborted(&fw_buf->fw_st)) goto out; switch (loading) { -- cgit v1.2.3 From 606f42265d384b9149bfb953c5dfc6d4710fef4c Mon Sep 17 00:00:00 2001 From: Prashanth Prakash Date: Thu, 26 Jan 2017 11:08:32 -0700 Subject: arm64: skip register_cpufreq_notifier on ACPI-based systems On ACPI based systems where the topology is setup using the API store_cpu_topology, at the moment we do not have necessary code to parse cpu capacity and handle cpufreq notifier, thus resulting in a kernel panic. Stack: init_cpu_capacity_callback+0xb4/0x1c8 notifier_call_chain+0x5c/0xa0 __blocking_notifier_call_chain+0x58/0xa0 blocking_notifier_call_chain+0x3c/0x50 cpufreq_set_policy+0xe4/0x328 cpufreq_init_policy+0x80/0x100 cpufreq_online+0x418/0x710 cpufreq_add_dev+0x118/0x180 subsys_interface_register+0xa4/0xf8 cpufreq_register_driver+0x1c0/0x298 cppc_cpufreq_init+0xdc/0x1000 [cppc_cpufreq] do_one_initcall+0x5c/0x168 do_init_module+0x64/0x1e4 load_module+0x130c/0x14d0 SyS_finit_module+0x108/0x120 el0_svc_naked+0x24/0x28 Fixes: 7202bde8b7ae ("arm64: parse cpu capacity-dmips-mhz from DT") Acked-by: Will Deacon Signed-off-by: Prashanth Prakash Signed-off-by: Catalin Marinas --- arch/arm64/kernel/topology.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c index 23e9e13bd2aa..655e65f38f31 100644 --- a/arch/arm64/kernel/topology.c +++ b/arch/arm64/kernel/topology.c @@ -11,6 +11,7 @@ * for more details. */ +#include #include #include #include @@ -209,7 +210,12 @@ static struct notifier_block init_cpu_capacity_notifier = { static int __init register_cpufreq_notifier(void) { - if (cap_parsing_failed) + /* + * on ACPI-based systems we need to use the default cpu capacity + * until we have the necessary code to parse the cpu capacity, so + * skip registering cpufreq notifier. + */ + if (!acpi_disabled || cap_parsing_failed) return -EINVAL; if (!alloc_cpumask_var(&cpus_to_visit, GFP_KERNEL)) { -- cgit v1.2.3 From 92e55f412cffd016cc245a74278cb4d7b89bb3bc Mon Sep 17 00:00:00 2001 From: Pablo Neira Date: Thu, 26 Jan 2017 22:56:21 +0100 Subject: tcp: don't annotate mark on control socket from tcp_v6_send_response() Unlike ipv4, this control socket is shared by all cpus so we cannot use it as scratchpad area to annotate the mark that we pass to ip6_xmit(). Add a new parameter to ip6_xmit() to indicate the mark. The SCTP socket family caches the flowi6 structure in the sctp_transport structure, so we cannot use to carry the mark unless we later on reset it back, which I discarded since it looks ugly to me. Fixes: bf99b4ded5f8 ("tcp: fix mark propagation with fwmark_reflect enabled") Suggested-by: Eric Dumazet Signed-off-by: Pablo Neira Ayuso Signed-off-by: David S. Miller --- include/net/ipv6.h | 2 +- net/dccp/ipv6.c | 4 ++-- net/ipv6/inet6_connection_sock.c | 2 +- net/ipv6/ip6_output.c | 4 ++-- net/ipv6/tcp_ipv6.c | 5 ++--- net/sctp/ipv6.c | 3 ++- 6 files changed, 10 insertions(+), 10 deletions(-) diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 487e57391664..7afe991e900e 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -871,7 +871,7 @@ int ip6_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb); * upper-layer output functions */ int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, - struct ipv6_txoptions *opt, int tclass); + __u32 mark, struct ipv6_txoptions *opt, int tclass); int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr); diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index adfc790f7193..c4e879c02186 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -227,7 +227,7 @@ static int dccp_v6_send_response(const struct sock *sk, struct request_sock *req opt = ireq->ipv6_opt; if (!opt) opt = rcu_dereference(np->opt); - err = ip6_xmit(sk, skb, &fl6, opt, np->tclass); + err = ip6_xmit(sk, skb, &fl6, sk->sk_mark, opt, np->tclass); rcu_read_unlock(); err = net_xmit_eval(err); } @@ -281,7 +281,7 @@ static void dccp_v6_ctl_send_reset(const struct sock *sk, struct sk_buff *rxskb) dst = ip6_dst_lookup_flow(ctl_sk, &fl6, NULL); if (!IS_ERR(dst)) { skb_dst_set(skb, dst); - ip6_xmit(ctl_sk, skb, &fl6, NULL, 0); + ip6_xmit(ctl_sk, skb, &fl6, 0, NULL, 0); DCCP_INC_STATS(DCCP_MIB_OUTSEGS); DCCP_INC_STATS(DCCP_MIB_OUTRSTS); return; diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index 7396e75e161b..75c308239243 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -176,7 +176,7 @@ int inet6_csk_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl_unused /* Restore final destination back after routing done */ fl6.daddr = sk->sk_v6_daddr; - res = ip6_xmit(sk, skb, &fl6, rcu_dereference(np->opt), + res = ip6_xmit(sk, skb, &fl6, sk->sk_mark, rcu_dereference(np->opt), np->tclass); rcu_read_unlock(); return res; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 38122d04fadc..2c0df09e9036 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -172,7 +172,7 @@ int ip6_output(struct net *net, struct sock *sk, struct sk_buff *skb) * which are using proper atomic operations or spinlocks. */ int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, - struct ipv6_txoptions *opt, int tclass) + __u32 mark, struct ipv6_txoptions *opt, int tclass) { struct net *net = sock_net(sk); const struct ipv6_pinfo *np = inet6_sk(sk); @@ -240,7 +240,7 @@ int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, skb->protocol = htons(ETH_P_IPV6); skb->priority = sk->sk_priority; - skb->mark = sk->sk_mark; + skb->mark = mark; mtu = dst_mtu(dst); if ((skb->len <= mtu) || skb->ignore_df || skb_is_gso(skb)) { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 2b20622a5824..cb8929681dc7 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -469,7 +469,7 @@ static int tcp_v6_send_synack(const struct sock *sk, struct dst_entry *dst, opt = ireq->ipv6_opt; if (!opt) opt = rcu_dereference(np->opt); - err = ip6_xmit(sk, skb, fl6, opt, np->tclass); + err = ip6_xmit(sk, skb, fl6, sk->sk_mark, opt, np->tclass); rcu_read_unlock(); err = net_xmit_eval(err); } @@ -840,8 +840,7 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32 dst = ip6_dst_lookup_flow(ctl_sk, &fl6, NULL); if (!IS_ERR(dst)) { skb_dst_set(buff, dst); - ctl_sk->sk_mark = fl6.flowi6_mark; - ip6_xmit(ctl_sk, buff, &fl6, NULL, tclass); + ip6_xmit(ctl_sk, buff, &fl6, fl6.flowi6_mark, NULL, tclass); TCP_INC_STATS(net, TCP_MIB_OUTSEGS); if (rst) TCP_INC_STATS(net, TCP_MIB_OUTRSTS); diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 5ed8e79bf102..64dfd35ccdcc 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -222,7 +222,8 @@ static int sctp_v6_xmit(struct sk_buff *skb, struct sctp_transport *transport) SCTP_INC_STATS(sock_net(sk), SCTP_MIB_OUTSCTPPACKS); rcu_read_lock(); - res = ip6_xmit(sk, skb, fl6, rcu_dereference(np->opt), np->tclass); + res = ip6_xmit(sk, skb, fl6, sk->sk_mark, rcu_dereference(np->opt), + np->tclass); rcu_read_unlock(); return res; } -- cgit v1.2.3 From ab729823ec16aef384f09fd2cffe0b3d3f6e6cba Mon Sep 17 00:00:00 2001 From: Andreas Schultz Date: Fri, 27 Jan 2017 10:40:56 +0100 Subject: gtp: add genl family modules alias Auto-load the module when userspace asks for the gtp netlink family. Signed-off-by: Andreas Schultz Acked-by: Harald Welte Acked-by: Pablo Neira Ayuso Signed-off-by: David S. Miller --- drivers/net/gtp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c index 8b6810bad54b..7580ccc50e1d 100644 --- a/drivers/net/gtp.c +++ b/drivers/net/gtp.c @@ -1376,3 +1376,4 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Harald Welte "); MODULE_DESCRIPTION("Interface driver for GTP encapsulated traffic"); MODULE_ALIAS_RTNL_LINK("gtp"); +MODULE_ALIAS_GENL_FAMILY("gtp"); -- cgit v1.2.3 From c6ce1d08eede4c2968ed08aafa3165e8e183c5a1 Mon Sep 17 00:00:00 2001 From: Andreas Schultz Date: Fri, 27 Jan 2017 10:40:57 +0100 Subject: gtp: clear DF bit on GTP packet tx 3GPP TS 29.281 and 3GPP TS 29.060 imply that GTP-U packets should be sent with the DF bit cleared. For example 3GPP TS 29.060, Release 8, Section 13.2.2: > Backbone router: Any router in the backbone may fragment the GTP > packet if needed, according to IPv4. Signed-off-by: Andreas Schultz Acked-by: Harald Welte Acked-by: Pablo Neira Ayuso Signed-off-by: David S. Miller --- drivers/net/gtp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c index 7580ccc50e1d..1df54d626f20 100644 --- a/drivers/net/gtp.c +++ b/drivers/net/gtp.c @@ -612,7 +612,7 @@ static netdev_tx_t gtp_dev_xmit(struct sk_buff *skb, struct net_device *dev) pktinfo.fl4.saddr, pktinfo.fl4.daddr, pktinfo.iph->tos, ip4_dst_hoplimit(&pktinfo.rt->dst), - htons(IP_DF), + 0, pktinfo.gtph_port, pktinfo.gtph_port, true, false); break; -- cgit v1.2.3 From 3ab1b469e847ba425af3c5ad5068cc94b55b38d0 Mon Sep 17 00:00:00 2001 From: Andreas Schultz Date: Fri, 27 Jan 2017 10:40:58 +0100 Subject: gtp: fix cross netns recv on gtp socket The use of the passed through netlink src_net to check for a cross netns operation was wrong. Using the GTP socket and the GTP netdevice is always correct (even if the netdev has been moved to new netns after link creation). Remove the now obsolete net field from gtp_dev. Signed-off-by: Andreas Schultz Acked-by: Pablo Neira Ayuso Signed-off-by: David S. Miller --- drivers/net/gtp.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c index 1df54d626f20..99d3df788ce8 100644 --- a/drivers/net/gtp.c +++ b/drivers/net/gtp.c @@ -69,7 +69,6 @@ struct gtp_dev { struct socket *sock0; struct socket *sock1u; - struct net *net; struct net_device *dev; unsigned int hash_size; @@ -316,7 +315,7 @@ static int gtp_encap_recv(struct sock *sk, struct sk_buff *skb) netdev_dbg(gtp->dev, "encap_recv sk=%p\n", sk); - xnet = !net_eq(gtp->net, dev_net(gtp->dev)); + xnet = !net_eq(sock_net(sk), dev_net(gtp->dev)); switch (udp_sk(sk)->encap_type) { case UDP_ENCAP_GTP0: @@ -658,7 +657,7 @@ static void gtp_link_setup(struct net_device *dev) static int gtp_hashtable_new(struct gtp_dev *gtp, int hsize); static void gtp_hashtable_free(struct gtp_dev *gtp); static int gtp_encap_enable(struct net_device *dev, struct gtp_dev *gtp, - int fd_gtp0, int fd_gtp1, struct net *src_net); + int fd_gtp0, int fd_gtp1); static int gtp_newlink(struct net *src_net, struct net_device *dev, struct nlattr *tb[], struct nlattr *data[]) @@ -675,7 +674,7 @@ static int gtp_newlink(struct net *src_net, struct net_device *dev, fd0 = nla_get_u32(data[IFLA_GTP_FD0]); fd1 = nla_get_u32(data[IFLA_GTP_FD1]); - err = gtp_encap_enable(dev, gtp, fd0, fd1, src_net); + err = gtp_encap_enable(dev, gtp, fd0, fd1); if (err < 0) goto out_err; @@ -821,7 +820,7 @@ static void gtp_hashtable_free(struct gtp_dev *gtp) } static int gtp_encap_enable(struct net_device *dev, struct gtp_dev *gtp, - int fd_gtp0, int fd_gtp1, struct net *src_net) + int fd_gtp0, int fd_gtp1) { struct udp_tunnel_sock_cfg tuncfg = {NULL}; struct socket *sock0, *sock1u; @@ -858,7 +857,6 @@ static int gtp_encap_enable(struct net_device *dev, struct gtp_dev *gtp, gtp->sock0 = sock0; gtp->sock1u = sock1u; - gtp->net = src_net; tuncfg.sk_user_data = gtp; tuncfg.encap_rcv = gtp_encap_recv; -- cgit v1.2.3 From 9d162ed69f51cbd9ee5a0c7e82aba7acc96362ff Mon Sep 17 00:00:00 2001 From: Sean Nyekjaer Date: Fri, 27 Jan 2017 08:46:23 +0100 Subject: net: phy: micrel: add support for KSZ8795 This is adds support for the PHYs in the KSZ8795 5port managed switch. It will allow to detect the link between the switch and the soc and uses the same read_status functions as the KSZ8873MLL switch. Signed-off-by: Sean Nyekjaer Signed-off-by: David S. Miller --- drivers/net/phy/micrel.c | 14 ++++++++++++++ include/linux/micrel_phy.h | 2 ++ 2 files changed, 16 insertions(+) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 9a77289109b7..e55809c5beb7 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -1008,6 +1008,20 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, +}, { + .phy_id = PHY_ID_KSZ8795, + .phy_id_mask = MICREL_PHY_ID_MASK, + .name = "Micrel KSZ8795", + .features = (SUPPORTED_Pause | SUPPORTED_Asym_Pause), + .flags = PHY_HAS_MAGICANEG | PHY_HAS_INTERRUPT, + .config_init = kszphy_config_init, + .config_aneg = ksz8873mll_config_aneg, + .read_status = ksz8873mll_read_status, + .get_sset_count = kszphy_get_sset_count, + .get_strings = kszphy_get_strings, + .get_stats = kszphy_get_stats, + .suspend = genphy_suspend, + .resume = genphy_resume, } }; module_phy_driver(ksphy_driver); diff --git a/include/linux/micrel_phy.h b/include/linux/micrel_phy.h index 257173e0095e..f541da68d1e7 100644 --- a/include/linux/micrel_phy.h +++ b/include/linux/micrel_phy.h @@ -35,6 +35,8 @@ #define PHY_ID_KSZ886X 0x00221430 #define PHY_ID_KSZ8863 0x00221435 +#define PHY_ID_KSZ8795 0x00221550 + /* struct phy_device dev_flags definitions */ #define MICREL_PHY_50MHZ_CLK 0x00000001 #define MICREL_PHY_FXEN 0x00000002 -- cgit v1.2.3 From 950eabbd6ddedc1b08350b9169a6a51b130ebaaf Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 27 Jan 2017 13:32:14 +0100 Subject: ISDN: eicon: silence misleading array-bounds warning With some gcc versions, we get a warning about the eicon driver, and that currently shows up as the only remaining warning in one of the build bots: In file included from ../drivers/isdn/hardware/eicon/message.c:30:0: eicon/message.c: In function 'mixer_notify_update': eicon/platform.h:333:18: warning: array subscript is above array bounds [-Warray-bounds] The code is easily changed to open-code the unusual PUT_WORD() line causing this to avoid the warning. Cc: stable@vger.kernel.org Link: http://arm-soc.lixom.net/buildlogs/stable-rc/v4.4.45/ Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- drivers/isdn/hardware/eicon/message.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/isdn/hardware/eicon/message.c b/drivers/isdn/hardware/eicon/message.c index 1a1d99704fe6..296f1411fe84 100644 --- a/drivers/isdn/hardware/eicon/message.c +++ b/drivers/isdn/hardware/eicon/message.c @@ -11297,7 +11297,8 @@ static void mixer_notify_update(PLCI *plci, byte others) ((CAPI_MSG *) msg)->header.ncci = 0; ((CAPI_MSG *) msg)->info.facility_req.Selector = SELECTOR_LINE_INTERCONNECT; ((CAPI_MSG *) msg)->info.facility_req.structs[0] = 3; - PUT_WORD(&(((CAPI_MSG *) msg)->info.facility_req.structs[1]), LI_REQ_SILENT_UPDATE); + ((CAPI_MSG *) msg)->info.facility_req.structs[1] = LI_REQ_SILENT_UPDATE & 0xff; + ((CAPI_MSG *) msg)->info.facility_req.structs[2] = LI_REQ_SILENT_UPDATE >> 8; ((CAPI_MSG *) msg)->info.facility_req.structs[3] = 0; w = api_put(notify_plci->appl, (CAPI_MSG *) msg); if (w != _QUEUE_FULL) -- cgit v1.2.3 From e0d76fa4475ef2cf4b52d18588b8ce95153d021b Mon Sep 17 00:00:00 2001 From: Brian Foster Date: Thu, 26 Jan 2017 13:18:09 -0800 Subject: xfs: prevent quotacheck from overloading inode lru Quotacheck runs at mount time in situations where quota accounting must be recalculated. In doing so, it uses bulkstat to visit every inode in the filesystem. Historically, every inode processed during quotacheck was released and immediately tagged for reclaim because quotacheck runs before the superblock is marked active by the VFS. In other words, the final iput() lead to an immediate ->destroy_inode() call, which allowed the XFS background reclaim worker to start reclaiming inodes. Commit 17c12bcd3 ("xfs: when replaying bmap operations, don't let unlinked inodes get reaped") marks the XFS superblock active sooner as part of the mount process to support caching inodes processed during log recovery. This occurs before quotacheck and thus means all inodes processed by quotacheck are inserted to the LRU on release. The s_umount lock is held until the mount has completed and thus prevents the shrinkers from operating on the sb. This means that quotacheck can excessively populate the inode LRU and lead to OOM conditions on systems without sufficient RAM. Update the quotacheck bulkstat handler to set XFS_IGET_DONTCACHE on inodes processed by quotacheck. This causes ->drop_inode() to return 1 and in turn causes iput_final() to evict the inode. This preserves the original quotacheck behavior and prevents it from overloading the LRU and running out of memory. CC: stable@vger.kernel.org # v4.9 Reported-by: Martin Svec Signed-off-by: Brian Foster Reviewed-by: Eric Sandeen Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong --- fs/xfs/xfs_qm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c index 45e50ea90769..b669b123287b 100644 --- a/fs/xfs/xfs_qm.c +++ b/fs/xfs/xfs_qm.c @@ -1177,7 +1177,8 @@ xfs_qm_dqusage_adjust( * the case in all other instances. It's OK that we do this because * quotacheck is done only at mount time. */ - error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_EXCL, &ip); + error = xfs_iget(mp, NULL, ino, XFS_IGET_DONTCACHE, XFS_ILOCK_EXCL, + &ip); if (error) { *res = BULKSTAT_RV_NOTHING; return error; -- cgit v1.2.3 From 9aed02feae57bf7a40cb04ea0e3017cb7a998db4 Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Fri, 27 Jan 2017 10:45:27 -0800 Subject: ARC: [arcompact] handle unaligned access delay slot corner case After emulating an unaligned access in delay slot of a branch, we pretend as the delay slot never happened - so return back to actual branch target (or next PC if branch was not taken). Curently we did this by handling STATUS32.DE, we also need to clear the BTA.T bit, which is disregarded when returning from original misaligned exception, but could cause weirdness if it took the interrupt return path (in case interrupt was acive too) One ARC700 customer ran into this when enabling unaligned access fixup for kernel mode accesses as well Cc: stable@vger.kernel.org Signed-off-by: Vineet Gupta --- arch/arc/kernel/unaligned.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arc/kernel/unaligned.c b/arch/arc/kernel/unaligned.c index abd961f3e763..91ebe382147f 100644 --- a/arch/arc/kernel/unaligned.c +++ b/arch/arc/kernel/unaligned.c @@ -241,8 +241,9 @@ int misaligned_fixup(unsigned long address, struct pt_regs *regs, if (state.fault) goto fault; + /* clear any remanants of delay slot */ if (delay_mode(regs)) { - regs->ret = regs->bta; + regs->ret = regs->bta ~1U; regs->status32 &= ~STATUS_DE_MASK; } else { regs->ret += state.instr_len; -- cgit v1.2.3 From b4cfe3971f6eab542dd7ecc398bfa1aeec889934 Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Sun, 15 Jan 2017 20:15:00 +0200 Subject: RDMA/cma: Fix unknown symbol when CONFIG_IPV6 is not enabled If IPV6 has not been enabled in the underlying kernel, we must avoid calling IPV6 procedures in rdma_cm.ko. This requires using "IS_ENABLED(CONFIG_IPV6)" in "if" statements surrounding any code which calls external IPV6 procedures. In the instance fixed here, procedure cma_bind_addr() called ipv6_addr_type() -- which resulted in calling external procedure __ipv6_addr_type(). Fixes: 6c26a77124ff ("RDMA/cma: fix IPv6 address resolution") Cc: # v4.2+ Cc: Spencer Baugh Signed-off-by: Jack Morgenstein Reviewed-by: Moni Shoua Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/core/cma.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index e7dcfac877ca..3e70a9c5d79d 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -2811,7 +2811,8 @@ static int cma_bind_addr(struct rdma_cm_id *id, struct sockaddr *src_addr, if (!src_addr || !src_addr->sa_family) { src_addr = (struct sockaddr *) &id->route.addr.src_addr; src_addr->sa_family = dst_addr->sa_family; - if (dst_addr->sa_family == AF_INET6) { + if (IS_ENABLED(CONFIG_IPV6) && + dst_addr->sa_family == AF_INET6) { struct sockaddr_in6 *src_addr6 = (struct sockaddr_in6 *) src_addr; struct sockaddr_in6 *dst_addr6 = (struct sockaddr_in6 *) dst_addr; src_addr6->sin6_scope_id = dst_addr6->sin6_scope_id; -- cgit v1.2.3 From 030305d69fc6963c16003f50d7e8d74b02d0a143 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Fri, 27 Jan 2017 15:00:45 -0600 Subject: PCI/ASPM: Handle PCI-to-PCIe bridges as roots of PCIe hierarchies In a struct pcie_link_state, link->root points to the pcie_link_state of the root of the PCIe hierarchy. For the topmost link, this points to itself (link->root = link). For others, we copy the pointer from the parent (link->root = link->parent->root). Previously we recognized that Root Ports originated PCIe hierarchies, but we treated PCI/PCI-X to PCIe Bridges as being in the middle of the hierarchy, and when we tried to copy the pointer from link->parent->root, there was no parent, and we dereferenced a NULL pointer: BUG: unable to handle kernel NULL pointer dereference at 0000000000000090 IP: [] pcie_aspm_init_link_state+0x170/0x820 Recognize that PCI/PCI-X to PCIe Bridges originate PCIe hierarchies just like Root Ports do, so link->root for these devices should also point to itself. Fixes: 51ebfc92b72b ("PCI: Enumerate switches below PCI-to-PCIe bridges") Link: https://bugzilla.kernel.org/show_bug.cgi?id=193411 Link: https://bugzilla.opensuse.org/show_bug.cgi?id=1022181 Tested-by: lists@ssl-mail.com Tested-by: Jayachandran C. Signed-off-by: Bjorn Helgaas CC: stable@vger.kernel.org # v4.2+--- drivers/pci/pcie/aspm.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index 17ac1dce3286..3dd8bcbb3011 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -532,25 +532,32 @@ static struct pcie_link_state *alloc_pcie_link_state(struct pci_dev *pdev) link = kzalloc(sizeof(*link), GFP_KERNEL); if (!link) return NULL; + INIT_LIST_HEAD(&link->sibling); INIT_LIST_HEAD(&link->children); INIT_LIST_HEAD(&link->link); link->pdev = pdev; - if (pci_pcie_type(pdev) != PCI_EXP_TYPE_ROOT_PORT) { + + /* + * Root Ports and PCI/PCI-X to PCIe Bridges are roots of PCIe + * hierarchies. + */ + if (pci_pcie_type(pdev) == PCI_EXP_TYPE_ROOT_PORT || + pci_pcie_type(pdev) == PCI_EXP_TYPE_PCIE_BRIDGE) { + link->root = link; + } else { struct pcie_link_state *parent; + parent = pdev->bus->parent->self->link_state; if (!parent) { kfree(link); return NULL; } + link->parent = parent; + link->root = link->parent->root; list_add(&link->link, &parent->children); } - /* Setup a pointer to the root port link */ - if (!link->parent) - link->root = link; - else - link->root = link->parent->root; list_add(&link->sibling, &link_list); pdev->link_state = link; -- cgit v1.2.3 From 2b1d530cb3157f828fcaadd259613f59db3c6d1c Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Fri, 27 Jan 2017 14:19:25 +0200 Subject: MAINTAINERS: ath9k-devel is closed ath9k-devel list is now closed, only linux-wireless should be used. Reported-by: Michael Renzmann Signed-off-by: Kalle Valo --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index cfff2c9e3d94..d7699d5b5863 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -10169,7 +10169,6 @@ F: drivers/media/tuners/qt1010* QUALCOMM ATHEROS ATH9K WIRELESS DRIVER M: QCA ath9k Development L: linux-wireless@vger.kernel.org -L: ath9k-devel@lists.ath9k.org W: http://wireless.kernel.org/en/users/Drivers/ath9k S: Supported F: drivers/net/wireless/ath/ath9k/ -- cgit v1.2.3 From 966d2b04e070bc040319aaebfec09e0144dc3341 Mon Sep 17 00:00:00 2001 From: Douglas Miller Date: Sat, 28 Jan 2017 06:42:20 -0600 Subject: percpu-refcount: fix reference leak during percpu-atomic transition percpu_ref_tryget() and percpu_ref_tryget_live() should return "true" IFF they acquire a reference. But the return value from atomic_long_inc_not_zero() is a long and may have high bits set, e.g. PERCPU_COUNT_BIAS, and the return value of the tryget routines is bool so the reference may actually be acquired but the routines return "false" which results in a reference leak since the caller assumes it does not need to do a corresponding percpu_ref_put(). This was seen when performing CPU hotplug during I/O, as hangs in blk_mq_freeze_queue_wait where percpu_ref_kill (blk_mq_freeze_queue_start) raced with percpu_ref_tryget (blk_mq_timeout_work). Sample stack trace: __switch_to+0x2c0/0x450 __schedule+0x2f8/0x970 schedule+0x48/0xc0 blk_mq_freeze_queue_wait+0x94/0x120 blk_mq_queue_reinit_work+0xb8/0x180 blk_mq_queue_reinit_prepare+0x84/0xa0 cpuhp_invoke_callback+0x17c/0x600 cpuhp_up_callbacks+0x58/0x150 _cpu_up+0xf0/0x1c0 do_cpu_up+0x120/0x150 cpu_subsys_online+0x64/0xe0 device_online+0xb4/0x120 online_store+0xb4/0xc0 dev_attr_store+0x68/0xa0 sysfs_kf_write+0x80/0xb0 kernfs_fop_write+0x17c/0x250 __vfs_write+0x6c/0x1e0 vfs_write+0xd0/0x270 SyS_write+0x6c/0x110 system_call+0x38/0xe0 Examination of the queue showed a single reference (no PERCPU_COUNT_BIAS, and __PERCPU_REF_DEAD, __PERCPU_REF_ATOMIC set) and no requests. However, conditions at the time of the race are count of PERCPU_COUNT_BIAS + 0 and __PERCPU_REF_DEAD and __PERCPU_REF_ATOMIC set. The fix is to make the tryget routines use an actual boolean internally instead of the atomic long result truncated to a int. Fixes: e625305b3907 percpu-refcount: make percpu_ref based on longs instead of ints Link: https://bugzilla.kernel.org/show_bug.cgi?id=190751 Signed-off-by: Douglas Miller Reviewed-by: Jens Axboe Signed-off-by: Tejun Heo Fixes: e625305b3907 ("percpu-refcount: make percpu_ref based on longs instead of ints") Cc: stable@vger.kernel.org # v3.18+ --- include/linux/percpu-refcount.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/percpu-refcount.h b/include/linux/percpu-refcount.h index 1c7eec09e5eb..3a481a49546e 100644 --- a/include/linux/percpu-refcount.h +++ b/include/linux/percpu-refcount.h @@ -204,7 +204,7 @@ static inline void percpu_ref_get(struct percpu_ref *ref) static inline bool percpu_ref_tryget(struct percpu_ref *ref) { unsigned long __percpu *percpu_count; - int ret; + bool ret; rcu_read_lock_sched(); @@ -238,7 +238,7 @@ static inline bool percpu_ref_tryget(struct percpu_ref *ref) static inline bool percpu_ref_tryget_live(struct percpu_ref *ref) { unsigned long __percpu *percpu_count; - int ret = false; + bool ret = false; rcu_read_lock_sched(); -- cgit v1.2.3 From 83b5d1e3d3013dbf90645a5d07179d018c8243fa Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Tue, 3 Jan 2017 22:55:50 +0100 Subject: parisc, parport_gsc: Fixes for printk continuation lines Signed-off-by: Helge Deller --- drivers/parport/parport_gsc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/parport/parport_gsc.c b/drivers/parport/parport_gsc.c index dd6d4ccb41e4..3858b87fd0bb 100644 --- a/drivers/parport/parport_gsc.c +++ b/drivers/parport/parport_gsc.c @@ -293,7 +293,7 @@ struct parport *parport_gsc_probe_port(unsigned long base, p->irq = PARPORT_IRQ_NONE; } if (p->irq != PARPORT_IRQ_NONE) { - printk(", irq %d", p->irq); + pr_cont(", irq %d", p->irq); if (p->dma == PARPORT_DMA_AUTO) { p->dma = PARPORT_DMA_NONE; @@ -303,8 +303,8 @@ struct parport *parport_gsc_probe_port(unsigned long base, is mandatory (see above) */ p->dma = PARPORT_DMA_NONE; - printk(" ["); -#define printmode(x) {if(p->modes&PARPORT_MODE_##x){printk("%s%s",f?",":"",#x);f++;}} + pr_cont(" ["); +#define printmode(x) {if(p->modes&PARPORT_MODE_##x){pr_cont("%s%s",f?",":"",#x);f++;}} { int f = 0; printmode(PCSPP); @@ -315,7 +315,7 @@ struct parport *parport_gsc_probe_port(unsigned long base, // printmode(DMA); } #undef printmode - printk("]\n"); + pr_cont("]\n"); if (p->irq != PARPORT_IRQ_NONE) { if (request_irq (p->irq, parport_irq_handler, -- cgit v1.2.3 From 2ad5d52d42810bed95100a3d912679d8864421ec Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Sat, 28 Jan 2017 11:52:02 +0100 Subject: parisc: Don't use BITS_PER_LONG in userspace-exported swab.h header In swab.h the "#if BITS_PER_LONG > 32" breaks compiling userspace programs if BITS_PER_LONG is #defined by userspace with the sizeof() compiler builtin. Solve this problem by using __BITS_PER_LONG instead. Since we now #include asm/bitsperlong.h avoid further potential userspace pollution by moving the #define of SHIFT_PER_LONG to bitops.h which is not exported to userspace. This patch unbreaks compiling qemu on hppa/parisc. Signed-off-by: Helge Deller Cc: --- arch/parisc/include/asm/bitops.h | 8 +++++++- arch/parisc/include/uapi/asm/bitsperlong.h | 2 -- arch/parisc/include/uapi/asm/swab.h | 5 +++-- 3 files changed, 10 insertions(+), 5 deletions(-) diff --git a/arch/parisc/include/asm/bitops.h b/arch/parisc/include/asm/bitops.h index 3f9406d9b9d6..da87943328a5 100644 --- a/arch/parisc/include/asm/bitops.h +++ b/arch/parisc/include/asm/bitops.h @@ -6,7 +6,7 @@ #endif #include -#include /* for BITS_PER_LONG/SHIFT_PER_LONG */ +#include #include #include #include @@ -17,6 +17,12 @@ * to include/asm-i386/bitops.h or kerneldoc */ +#if __BITS_PER_LONG == 64 +#define SHIFT_PER_LONG 6 +#else +#define SHIFT_PER_LONG 5 +#endif + #define CHOP_SHIFTCOUNT(x) (((unsigned long) (x)) & (BITS_PER_LONG - 1)) diff --git a/arch/parisc/include/uapi/asm/bitsperlong.h b/arch/parisc/include/uapi/asm/bitsperlong.h index e0a23c7bdd43..07fa7e50bdc0 100644 --- a/arch/parisc/include/uapi/asm/bitsperlong.h +++ b/arch/parisc/include/uapi/asm/bitsperlong.h @@ -3,10 +3,8 @@ #if defined(__LP64__) #define __BITS_PER_LONG 64 -#define SHIFT_PER_LONG 6 #else #define __BITS_PER_LONG 32 -#define SHIFT_PER_LONG 5 #endif #include diff --git a/arch/parisc/include/uapi/asm/swab.h b/arch/parisc/include/uapi/asm/swab.h index e78403b129ef..928e1bbac98f 100644 --- a/arch/parisc/include/uapi/asm/swab.h +++ b/arch/parisc/include/uapi/asm/swab.h @@ -1,6 +1,7 @@ #ifndef _PARISC_SWAB_H #define _PARISC_SWAB_H +#include #include #include @@ -38,7 +39,7 @@ static inline __attribute_const__ __u32 __arch_swab32(__u32 x) } #define __arch_swab32 __arch_swab32 -#if BITS_PER_LONG > 32 +#if __BITS_PER_LONG > 32 /* ** From "PA-RISC 2.0 Architecture", HP Professional Books. ** See Appendix I page 8 , "Endian Byte Swapping". @@ -61,6 +62,6 @@ static inline __attribute_const__ __u64 __arch_swab64(__u64 x) return x; } #define __arch_swab64 __arch_swab64 -#endif /* BITS_PER_LONG > 32 */ +#endif /* __BITS_PER_LONG > 32 */ #endif /* _PARISC_SWAB_H */ -- cgit v1.2.3 From 9eb7892351a3a3b403d879b41c4e6efb2c96516f Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Wed, 11 Jan 2017 19:35:41 +0200 Subject: net/mlx5: Change ENOTSUPP to EOPNOTSUPP As ENOTSUPP is specific to NFS, change the return error value to EOPNOTSUPP in various places in the mlx5 driver. Signed-off-by: Or Gerlitz Suggested-by: Yotam Gigi Reviewed-by: Matan Barak Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 ++-- drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c | 6 +++--- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 10 +++++----- drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 4 ++-- drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 4 ++-- drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/main.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/port.c | 4 ++-- drivers/net/ethernet/mellanox/mlx5/core/vport.c | 2 +- 12 files changed, 22 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index 3797cc7c1288..caa837e5e2b9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -1728,7 +1728,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) if (cmd->cmdif_rev > CMD_IF_REV) { dev_err(&dev->pdev->dev, "driver does not support command interface version. driver %d, firmware %d\n", CMD_IF_REV, cmd->cmdif_rev); - err = -ENOTSUPP; + err = -EOPNOTSUPP; goto err_free_page; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 951dbd58594d..1619147a63e8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -863,12 +863,12 @@ static inline void mlx5e_arfs_destroy_tables(struct mlx5e_priv *priv) {} static inline int mlx5e_arfs_enable(struct mlx5e_priv *priv) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int mlx5e_arfs_disable(struct mlx5e_priv *priv) { - return -ENOTSUPP; + return -EOPNOTSUPP; } #else int mlx5e_arfs_create_tables(struct mlx5e_priv *priv); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index f0b460f47f29..35f9ae037ba0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -89,7 +89,7 @@ static int mlx5e_dcbnl_ieee_getets(struct net_device *netdev, int i; if (!MLX5_CAP_GEN(priv->mdev, ets)) - return -ENOTSUPP; + return -EOPNOTSUPP; ets->ets_cap = mlx5_max_tc(priv->mdev) + 1; for (i = 0; i < ets->ets_cap; i++) { @@ -236,7 +236,7 @@ static int mlx5e_dcbnl_ieee_setets(struct net_device *netdev, int err; if (!MLX5_CAP_GEN(priv->mdev, ets)) - return -ENOTSUPP; + return -EOPNOTSUPP; err = mlx5e_dbcnl_validate_ets(netdev, ets); if (err) @@ -402,7 +402,7 @@ static u8 mlx5e_dcbnl_setall(struct net_device *netdev) struct mlx5_core_dev *mdev = priv->mdev; struct ieee_ets ets; struct ieee_pfc pfc; - int err = -ENOTSUPP; + int err = -EOPNOTSUPP; int i; if (!MLX5_CAP_GEN(mdev, ets)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 5197817e4b2f..ffbdf9ee5a9b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -595,7 +595,7 @@ static int mlx5e_get_coalesce(struct net_device *netdev, struct mlx5e_priv *priv = netdev_priv(netdev); if (!MLX5_CAP_GEN(priv->mdev, cq_moderation)) - return -ENOTSUPP; + return -EOPNOTSUPP; coal->rx_coalesce_usecs = priv->params.rx_cq_moderation.usec; coal->rx_max_coalesced_frames = priv->params.rx_cq_moderation.pkts; @@ -620,7 +620,7 @@ static int mlx5e_set_coalesce(struct net_device *netdev, int i; if (!MLX5_CAP_GEN(mdev, cq_moderation)) - return -ENOTSUPP; + return -EOPNOTSUPP; mutex_lock(&priv->state_lock); @@ -1296,7 +1296,7 @@ static int mlx5e_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) u32 mlx5_wol_mode; if (!wol_supported) - return -ENOTSUPP; + return -EOPNOTSUPP; if (wol->wolopts & ~wol_supported) return -EINVAL; @@ -1426,7 +1426,7 @@ static int set_pflag_rx_cqe_based_moder(struct net_device *netdev, bool enable) if (rx_cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE && !MLX5_CAP_GEN(mdev, cq_period_start_from_cqe)) - return -ENOTSUPP; + return -EOPNOTSUPP; if (!rx_mode_changed) return 0; @@ -1452,7 +1452,7 @@ static int set_pflag_rx_cqe_compress(struct net_device *netdev, bool reset; if (!MLX5_CAP_GEN(mdev, cqe_compression)) - return -ENOTSUPP; + return -EOPNOTSUPP; if (enable && priv->tstamp.hwtstamp_config.rx_filter != HWTSTAMP_FILTER_NONE) { netdev_err(netdev, "Can't enable cqe compression while timestamping is enabled.\n"); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c index d088effd7160..f33f72d0237c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c @@ -92,7 +92,7 @@ static struct mlx5e_ethtool_table *get_flow_table(struct mlx5e_priv *priv, ns = mlx5_get_flow_namespace(priv->mdev, MLX5_FLOW_NAMESPACE_ETHTOOL); if (!ns) - return ERR_PTR(-ENOTSUPP); + return ERR_PTR(-EOPNOTSUPP); table_size = min_t(u32, BIT(MLX5_CAP_FLOWTABLE(priv->mdev, flow_table_properties_nic_receive.log_max_ft_size)), diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 2b7dd315020c..948351ae5bd2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -3331,7 +3331,7 @@ static const struct net_device_ops mlx5e_netdev_ops_sriov = { static int mlx5e_check_required_hca_cap(struct mlx5_core_dev *mdev) { if (MLX5_CAP_GEN(mdev, port_type) != MLX5_CAP_PORT_TYPE_ETH) - return -ENOTSUPP; + return -EOPNOTSUPP; if (!MLX5_CAP_GEN(mdev, eth_net_offloads) || !MLX5_CAP_GEN(mdev, nic_flow_table) || !MLX5_CAP_ETH(mdev, csum_cap) || @@ -3343,7 +3343,7 @@ static int mlx5e_check_required_hca_cap(struct mlx5_core_dev *mdev) < 3) { mlx5_core_warn(mdev, "Not creating net device, some required device capabilities are missing\n"); - return -ENOTSUPP; + return -EOPNOTSUPP; } if (!MLX5_CAP_ETH(mdev, self_lb_en_modifiable)) mlx5_core_warn(mdev, "Self loop back prevention is not supported\n"); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index f14d9c9ba773..bb712139b36e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -133,7 +133,7 @@ static int modify_esw_vport_cvlan(struct mlx5_core_dev *dev, u32 vport, if (!MLX5_CAP_ESW(dev, vport_cvlan_strip) || !MLX5_CAP_ESW(dev, vport_cvlan_insert_if_not_exist)) - return -ENOTSUPP; + return -EOPNOTSUPP; esw_debug(dev, "Set Vport[%d] VLAN %d qos %d set=%x\n", vport, vlan, qos, set_flags); @@ -1630,7 +1630,7 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode) if (!MLX5_CAP_GEN(esw->dev, eswitch_flow_table) || !MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ft_support)) { esw_warn(esw->dev, "E-Switch FDB is not supported, aborting ...\n"); - return -ENOTSUPP; + return -EOPNOTSUPP; } if (!MLX5_CAP_ESW_INGRESS_ACL(esw->dev, ft_support)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 03293ed1cc22..657d319fc4c6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -166,7 +166,7 @@ static int esw_add_vlan_action_check(struct mlx5_esw_flow_attr *attr, return 0; out_notsupp: - return -ENOTSUPP; + return -EOPNOTSUPP; } int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c index c4478ecd8056..b53fc85a2375 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c @@ -322,7 +322,7 @@ int mlx5_cmd_update_fte(struct mlx5_core_dev *dev, flow_table_properties_nic_receive. flow_modify_en); if (!atomic_mod_cap) - return -ENOTSUPP; + return -EOPNOTSUPP; opmod = 1; return mlx5_cmd_set_fte(dev, opmod, modify_mask, ft, group_id, fte); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index d01e9f21d469..3c315eb8d270 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -807,7 +807,7 @@ static int mlx5_core_set_issi(struct mlx5_core_dev *dev) return 0; } - return -ENOTSUPP; + return -EOPNOTSUPP; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index d2ec9d232a70..fd12e0a377a5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -620,7 +620,7 @@ static int mlx5_set_port_qetcr_reg(struct mlx5_core_dev *mdev, u32 *in, u32 out[MLX5_ST_SZ_DW(qtct_reg)]; if (!MLX5_CAP_GEN(mdev, ets)) - return -ENOTSUPP; + return -EOPNOTSUPP; return mlx5_core_access_reg(mdev, in, inlen, out, sizeof(out), MLX5_REG_QETCR, 0, 1); @@ -632,7 +632,7 @@ static int mlx5_query_port_qetcr_reg(struct mlx5_core_dev *mdev, u32 *out, u32 in[MLX5_ST_SZ_DW(qtct_reg)]; if (!MLX5_CAP_GEN(mdev, ets)) - return -ENOTSUPP; + return -EOPNOTSUPP; memset(in, 0, sizeof(in)); return mlx5_core_access_reg(mdev, in, sizeof(in), out, outlen, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index 269e4401c342..7129c30a2ab4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -532,7 +532,7 @@ int mlx5_modify_nic_vport_node_guid(struct mlx5_core_dev *mdev, if (!MLX5_CAP_GEN(mdev, vport_group_manager)) return -EACCES; if (!MLX5_CAP_ESW(mdev, nic_vport_node_guid_modify)) - return -ENOTSUPP; + return -EOPNOTSUPP; in = mlx5_vzalloc(inlen); if (!in) -- cgit v1.2.3 From eff596da48784316ccb83bef82bc1213b512d5e0 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 12 Jan 2017 13:04:01 +0200 Subject: net/mlx5: Return EOPNOTSUPP when failing to get steering name-space When we fail to retrieve a hardware steering name-space, the returned error code should say that this operation is not supported. Align the various places in the driver where this call is made to this convention. Also, make sure to warn when we fail to retrieve a SW (ANCHOR) name-space. Signed-off-by: Or Gerlitz Reviewed-by: Matan Barak Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_fs.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 6 +++--- drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 2 +- 4 files changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c index 1fe80de5d68f..a0e5a69402b3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c @@ -1089,7 +1089,7 @@ int mlx5e_create_flow_steering(struct mlx5e_priv *priv) MLX5_FLOW_NAMESPACE_KERNEL); if (!priv->fs.ns) - return -EINVAL; + return -EOPNOTSUPP; err = mlx5e_arfs_create_tables(priv); if (err) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index bb712139b36e..d0c8bf014453 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -353,7 +353,7 @@ static int esw_create_legacy_fdb_table(struct mlx5_eswitch *esw, int nvports) root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB); if (!root_ns) { esw_warn(dev, "Failed to get FDB flow namespace\n"); - return -ENOMEM; + return -EOPNOTSUPP; } flow_group_in = mlx5_vzalloc(inlen); @@ -962,7 +962,7 @@ static int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw, root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_EGRESS); if (!root_ns) { esw_warn(dev, "Failed to get E-Switch egress flow namespace\n"); - return -EIO; + return -EOPNOTSUPP; } flow_group_in = mlx5_vzalloc(inlen); @@ -1079,7 +1079,7 @@ static int esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw, root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_INGRESS); if (!root_ns) { esw_warn(dev, "Failed to get E-Switch ingress flow namespace\n"); - return -EIO; + return -EOPNOTSUPP; } flow_group_in = mlx5_vzalloc(inlen); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 657d319fc4c6..5803216157cf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -535,7 +535,7 @@ static int esw_create_offloads_table(struct mlx5_eswitch *esw) ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_OFFLOADS); if (!ns) { esw_warn(esw->dev, "Failed to get offloads flow namespace\n"); - return -ENOMEM; + return -EOPNOTSUPP; } ft_offloads = mlx5_create_flow_table(ns, 0, dev->priv.sriov.num_vfs + 2, 0, 0); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 0ac7a2fc916c..6346a8f5883b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -1822,7 +1822,7 @@ static int create_anchor_flow_table(struct mlx5_flow_steering *steering) struct mlx5_flow_table *ft; ns = mlx5_get_flow_namespace(steering->dev, MLX5_FLOW_NAMESPACE_ANCHOR); - if (!ns) + if (WARN_ON(!ns)) return -EINVAL; ft = mlx5_create_flow_table(ns, ANCHOR_PRIO, ANCHOR_SIZE, ANCHOR_LEVEL, 0); if (IS_ERR(ft)) { -- cgit v1.2.3 From 5403dc703ff277f8a2a12a83ac820750485f13b3 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Wed, 11 Jan 2017 19:39:42 +0200 Subject: net/mlx5: E-Switch, Err when retrieving steering name-space fails Make sure to return error when we failed retrieving the FDB steering name space. Also, while around, correctly print the error when mode change revert fails in the warning message. Signed-off-by: Or Gerlitz Reported-by: Leon Romanovsky Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 5803216157cf..c61bca138e65 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -424,6 +424,7 @@ static int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports) root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB); if (!root_ns) { esw_warn(dev, "Failed to get FDB flow namespace\n"); + err = -EOPNOTSUPP; goto ns_err; } @@ -655,7 +656,7 @@ static int esw_offloads_start(struct mlx5_eswitch *esw) esw_warn(esw->dev, "Failed setting eswitch to offloads, err %d\n", err); err1 = mlx5_eswitch_enable_sriov(esw, num_vfs, SRIOV_LEGACY); if (err1) - esw_warn(esw->dev, "Failed setting eswitch back to legacy, err %d\n", err); + esw_warn(esw->dev, "Failed setting eswitch back to legacy, err %d\n", err1); } if (esw->offloads.inline_mode == MLX5_INLINE_MODE_NONE) { if (mlx5_eswitch_inline_mode_get(esw, -- cgit v1.2.3 From 5bae8c031053c69b4aa74b7f1ba15d4ec8426208 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Sun, 15 Jan 2017 19:05:38 +0200 Subject: net/mlx5: E-Switch, Re-enable RoCE on mode change only after FDB destroy We must re-enable RoCE on the e-switch management port (PF) only after destroying the FDB in its switchdev/offloaded mode. Otherwise, when encapsulation is supported, this re-enablement will fail. Also, it's more natural and symmetric to disable RoCE on the PF before we create the FDB under switchdev mode, so do that as well and revert if getting into error during the mode change later. Fixes: 9da34cd34e85 ('net/mlx5: Disable RoCE on the e-switch management [..]') Signed-off-by: Or Gerlitz Reviewed-by: Hadar Hen Zion Signed-off-by: Saeed Mahameed --- .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 29 ++++++++++++++-------- 1 file changed, 18 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index c61bca138e65..595f7c7383b3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -675,9 +675,14 @@ int esw_offloads_init(struct mlx5_eswitch *esw, int nvports) int vport; int err; + /* disable PF RoCE so missed packets don't go through RoCE steering */ + mlx5_dev_list_lock(); + mlx5_remove_dev_by_protocol(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); + mlx5_dev_list_unlock(); + err = esw_create_offloads_fdb_table(esw, nvports); if (err) - return err; + goto create_fdb_err; err = esw_create_offloads_table(esw); if (err) @@ -697,11 +702,6 @@ int esw_offloads_init(struct mlx5_eswitch *esw, int nvports) goto err_reps; } - /* disable PF RoCE so missed packets don't go through RoCE steering */ - mlx5_dev_list_lock(); - mlx5_remove_dev_by_protocol(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); - mlx5_dev_list_unlock(); - return 0; err_reps: @@ -718,6 +718,13 @@ create_fg_err: create_ft_err: esw_destroy_offloads_fdb_table(esw); + +create_fdb_err: + /* enable back PF RoCE */ + mlx5_dev_list_lock(); + mlx5_add_dev_by_protocol(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); + mlx5_dev_list_unlock(); + return err; } @@ -725,11 +732,6 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw) { int err, err1, num_vfs = esw->dev->priv.sriov.num_vfs; - /* enable back PF RoCE */ - mlx5_dev_list_lock(); - mlx5_add_dev_by_protocol(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); - mlx5_dev_list_unlock(); - mlx5_eswitch_disable_sriov(esw); err = mlx5_eswitch_enable_sriov(esw, num_vfs, SRIOV_LEGACY); if (err) { @@ -739,6 +741,11 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw) esw_warn(esw->dev, "Failed setting eswitch back to offloads, err %d\n", err); } + /* enable back PF RoCE */ + mlx5_dev_list_lock(); + mlx5_add_dev_by_protocol(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); + mlx5_dev_list_unlock(); + return err; } -- cgit v1.2.3 From 3e621b19b0bb1f5bea34f1fbc5fb5629191eda2b Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Thu, 12 Jan 2017 11:07:40 +0200 Subject: net/mlx5e: Support TC encapsulation offloads with upper devices When tunneling is used, some virtualizations systems set the (mlx5e) uplink device to be stacked under upper devices such as bridge or ovs internal port, where the VTEP IP address used for the encapsulation is set on that upper device. In order to support such use-cases, we also deal with a setup where the egress mirred device isn't representing a port on the HW e-switch to where the ingress device belongs. We use eswitch service function which returns the uplink and set it as the egress device of the tc encap rule. Fixes: a54e20b4fcae ("net/mlx5e: Add basic TC tunnel set action for SRIOV offloads") Signed-off-by: Hadar Hen Zion Reviewed-by: Or Gerlitz Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 46bef6a26a8c..c5282b6aba8b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -663,6 +663,7 @@ static int mlx5e_route_lookup_ipv4(struct mlx5e_priv *priv, __be32 *saddr, int *out_ttl) { + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; struct rtable *rt; struct neighbour *n = NULL; int ttl; @@ -677,12 +678,11 @@ static int mlx5e_route_lookup_ipv4(struct mlx5e_priv *priv, #else return -EOPNOTSUPP; #endif - - if (!switchdev_port_same_parent_id(priv->netdev, rt->dst.dev)) { - pr_warn("%s: can't offload, devices not on same HW e-switch\n", __func__); - ip_rt_put(rt); - return -EOPNOTSUPP; - } + /* if the egress device isn't on the same HW e-switch, we use the uplink */ + if (!switchdev_port_same_parent_id(priv->netdev, rt->dst.dev)) + *out_dev = mlx5_eswitch_get_uplink_netdev(esw); + else + *out_dev = rt->dst.dev; ttl = ip4_dst_hoplimit(&rt->dst); n = dst_neigh_lookup(&rt->dst, &fl4->daddr); @@ -693,7 +693,6 @@ static int mlx5e_route_lookup_ipv4(struct mlx5e_priv *priv, *out_n = n; *saddr = fl4->saddr; *out_ttl = ttl; - *out_dev = rt->dst.dev; return 0; } -- cgit v1.2.3 From 1d3398facd08a7fd4202f269317a95668eb880b9 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Wed, 11 Jan 2017 14:32:26 +0200 Subject: net/mlx5e: Modify TIRs hash only when it's needed We don't need to modify our TIRs unless the user requested a change in the hash function/key, for example when changing indirection only. Tested: # Modify TIRs hash is needed ethtool -X ethX hkey ethtool -X ethX hfunc # Modify TIRs hash is not needed ethtool -X ethX equal All cases are verified with TCP Multi-Stream traffic over IPv4 & IPv6. Fixes: bdfc028de1b3 ("net/mlx5e: Fix ethtool RX hash func configuration change") Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index ffbdf9ee5a9b..6f4eb34259f0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -996,6 +996,7 @@ static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir, { struct mlx5e_priv *priv = netdev_priv(dev); int inlen = MLX5_ST_SZ_BYTES(modify_tir_in); + bool hash_changed = false; void *in; if ((hfunc != ETH_RSS_HASH_NO_CHANGE) && @@ -1017,14 +1018,21 @@ static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir, mlx5e_redirect_rqt(priv, rqtn, MLX5E_INDIR_RQT_SIZE, 0); } - if (key) + if (hfunc != ETH_RSS_HASH_NO_CHANGE && + hfunc != priv->params.rss_hfunc) { + priv->params.rss_hfunc = hfunc; + hash_changed = true; + } + + if (key) { memcpy(priv->params.toeplitz_hash_key, key, sizeof(priv->params.toeplitz_hash_key)); + hash_changed = hash_changed || + priv->params.rss_hfunc == ETH_RSS_HASH_TOP; + } - if (hfunc != ETH_RSS_HASH_NO_CHANGE) - priv->params.rss_hfunc = hfunc; - - mlx5e_modify_tirs_hash(priv, in, inlen); + if (hash_changed) + mlx5e_modify_tirs_hash(priv, in, inlen); mutex_unlock(&priv->state_lock); -- cgit v1.2.3 From a100ff3eef193d2d79daf98dcd97a54776ffeb78 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 12 Jan 2017 16:25:46 +0200 Subject: net/mlx5e: Fix update of hash function/key via ethtool Modifying TIR hash should change selected fields bitmask in addition to the function and key. Formerly, Only on ethool mlx5e_set_rxfh "ethtoo -X" we would not set this field resulting in zeroing of its value, which means no packet fields are used for RX RSS hash calculation thus causing all traffic to arrive in RQ[0]. On driver load out of the box we don't have this issue, since the TIR hash is fully created from scratch. Tested: ethtool -X ethX hkey ethtool -X ethX hfunc ethtool -X ethX equal All cases are verified with TCP Multi-Stream traffic over IPv4 & IPv6. Fixes: bdfc028de1b3 ("net/mlx5e: Fix ethtool RX hash func configuration change") Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 3 +- .../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 13 +- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 198 ++++++++++----------- 3 files changed, 109 insertions(+), 105 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 1619147a63e8..d5ecb8f53fd4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -791,7 +791,8 @@ void mlx5e_disable_vlan_filter(struct mlx5e_priv *priv); int mlx5e_modify_rqs_vsd(struct mlx5e_priv *priv, bool vsd); int mlx5e_redirect_rqt(struct mlx5e_priv *priv, u32 rqtn, int sz, int ix); -void mlx5e_build_tir_ctx_hash(void *tirc, struct mlx5e_priv *priv); +void mlx5e_build_indir_tir_ctx_hash(struct mlx5e_priv *priv, void *tirc, + enum mlx5e_traffic_types tt); int mlx5e_open_locked(struct net_device *netdev); int mlx5e_close_locked(struct net_device *netdev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 6f4eb34259f0..bb67863aa361 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -980,15 +980,18 @@ static int mlx5e_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, static void mlx5e_modify_tirs_hash(struct mlx5e_priv *priv, void *in, int inlen) { - struct mlx5_core_dev *mdev = priv->mdev; void *tirc = MLX5_ADDR_OF(modify_tir_in, in, ctx); - int i; + struct mlx5_core_dev *mdev = priv->mdev; + int ctxlen = MLX5_ST_SZ_BYTES(tirc); + int tt; MLX5_SET(modify_tir_in, in, bitmask.hash, 1); - mlx5e_build_tir_ctx_hash(tirc, priv); - for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) - mlx5_core_modify_tir(mdev, priv->indir_tir[i].tirn, in, inlen); + for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { + memset(tirc, 0, ctxlen); + mlx5e_build_indir_tir_ctx_hash(priv, tirc, tt); + mlx5_core_modify_tir(mdev, priv->indir_tir[tt].tirn, in, inlen); + } } static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 948351ae5bd2..f14ca3385fdd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -2022,8 +2022,23 @@ static void mlx5e_build_tir_ctx_lro(void *tirc, struct mlx5e_priv *priv) MLX5_SET(tirc, tirc, lro_timeout_period_usecs, priv->params.lro_timeout); } -void mlx5e_build_tir_ctx_hash(void *tirc, struct mlx5e_priv *priv) +void mlx5e_build_indir_tir_ctx_hash(struct mlx5e_priv *priv, void *tirc, + enum mlx5e_traffic_types tt) { + void *hfso = MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_outer); + +#define MLX5_HASH_IP (MLX5_HASH_FIELD_SEL_SRC_IP |\ + MLX5_HASH_FIELD_SEL_DST_IP) + +#define MLX5_HASH_IP_L4PORTS (MLX5_HASH_FIELD_SEL_SRC_IP |\ + MLX5_HASH_FIELD_SEL_DST_IP |\ + MLX5_HASH_FIELD_SEL_L4_SPORT |\ + MLX5_HASH_FIELD_SEL_L4_DPORT) + +#define MLX5_HASH_IP_IPSEC_SPI (MLX5_HASH_FIELD_SEL_SRC_IP |\ + MLX5_HASH_FIELD_SEL_DST_IP |\ + MLX5_HASH_FIELD_SEL_IPSEC_SPI) + MLX5_SET(tirc, tirc, rx_hash_fn, mlx5e_rx_hash_fn(priv->params.rss_hfunc)); if (priv->params.rss_hfunc == ETH_RSS_HASH_TOP) { @@ -2035,6 +2050,88 @@ void mlx5e_build_tir_ctx_hash(void *tirc, struct mlx5e_priv *priv) MLX5_SET(tirc, tirc, rx_hash_symmetric, 1); memcpy(rss_key, priv->params.toeplitz_hash_key, len); } + + switch (tt) { + case MLX5E_TT_IPV4_TCP: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV4); + MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, + MLX5_L4_PROT_TYPE_TCP); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP_L4PORTS); + break; + + case MLX5E_TT_IPV6_TCP: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV6); + MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, + MLX5_L4_PROT_TYPE_TCP); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP_L4PORTS); + break; + + case MLX5E_TT_IPV4_UDP: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV4); + MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, + MLX5_L4_PROT_TYPE_UDP); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP_L4PORTS); + break; + + case MLX5E_TT_IPV6_UDP: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV6); + MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, + MLX5_L4_PROT_TYPE_UDP); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP_L4PORTS); + break; + + case MLX5E_TT_IPV4_IPSEC_AH: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV4); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP_IPSEC_SPI); + break; + + case MLX5E_TT_IPV6_IPSEC_AH: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV6); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP_IPSEC_SPI); + break; + + case MLX5E_TT_IPV4_IPSEC_ESP: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV4); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP_IPSEC_SPI); + break; + + case MLX5E_TT_IPV6_IPSEC_ESP: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV6); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP_IPSEC_SPI); + break; + + case MLX5E_TT_IPV4: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV4); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP); + break; + + case MLX5E_TT_IPV6: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV6); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP); + break; + default: + WARN_ONCE(true, "%s: bad traffic type!\n", __func__); + } } static int mlx5e_modify_tirs_lro(struct mlx5e_priv *priv) @@ -2404,110 +2501,13 @@ void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv) static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, enum mlx5e_traffic_types tt) { - void *hfso = MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_outer); - MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn); -#define MLX5_HASH_IP (MLX5_HASH_FIELD_SEL_SRC_IP |\ - MLX5_HASH_FIELD_SEL_DST_IP) - -#define MLX5_HASH_IP_L4PORTS (MLX5_HASH_FIELD_SEL_SRC_IP |\ - MLX5_HASH_FIELD_SEL_DST_IP |\ - MLX5_HASH_FIELD_SEL_L4_SPORT |\ - MLX5_HASH_FIELD_SEL_L4_DPORT) - -#define MLX5_HASH_IP_IPSEC_SPI (MLX5_HASH_FIELD_SEL_SRC_IP |\ - MLX5_HASH_FIELD_SEL_DST_IP |\ - MLX5_HASH_FIELD_SEL_IPSEC_SPI) - mlx5e_build_tir_ctx_lro(tirc, priv); MLX5_SET(tirc, tirc, disp_type, MLX5_TIRC_DISP_TYPE_INDIRECT); MLX5_SET(tirc, tirc, indirect_table, priv->indir_rqt.rqtn); - mlx5e_build_tir_ctx_hash(tirc, priv); - - switch (tt) { - case MLX5E_TT_IPV4_TCP: - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - MLX5_L3_PROT_TYPE_IPV4); - MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, - MLX5_L4_PROT_TYPE_TCP); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - MLX5_HASH_IP_L4PORTS); - break; - - case MLX5E_TT_IPV6_TCP: - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - MLX5_L3_PROT_TYPE_IPV6); - MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, - MLX5_L4_PROT_TYPE_TCP); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - MLX5_HASH_IP_L4PORTS); - break; - - case MLX5E_TT_IPV4_UDP: - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - MLX5_L3_PROT_TYPE_IPV4); - MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, - MLX5_L4_PROT_TYPE_UDP); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - MLX5_HASH_IP_L4PORTS); - break; - - case MLX5E_TT_IPV6_UDP: - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - MLX5_L3_PROT_TYPE_IPV6); - MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, - MLX5_L4_PROT_TYPE_UDP); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - MLX5_HASH_IP_L4PORTS); - break; - - case MLX5E_TT_IPV4_IPSEC_AH: - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - MLX5_L3_PROT_TYPE_IPV4); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - MLX5_HASH_IP_IPSEC_SPI); - break; - - case MLX5E_TT_IPV6_IPSEC_AH: - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - MLX5_L3_PROT_TYPE_IPV6); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - MLX5_HASH_IP_IPSEC_SPI); - break; - - case MLX5E_TT_IPV4_IPSEC_ESP: - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - MLX5_L3_PROT_TYPE_IPV4); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - MLX5_HASH_IP_IPSEC_SPI); - break; - - case MLX5E_TT_IPV6_IPSEC_ESP: - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - MLX5_L3_PROT_TYPE_IPV6); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - MLX5_HASH_IP_IPSEC_SPI); - break; - - case MLX5E_TT_IPV4: - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - MLX5_L3_PROT_TYPE_IPV4); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - MLX5_HASH_IP); - break; - - case MLX5E_TT_IPV6: - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - MLX5_L3_PROT_TYPE_IPV6); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - MLX5_HASH_IP); - break; - default: - WARN_ONCE(true, - "mlx5e_build_indir_tir_ctx: bad traffic type!\n"); - } + mlx5e_build_indir_tir_ctx_hash(priv, tirc, tt); } static void mlx5e_build_direct_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, -- cgit v1.2.3 From d15118af268324ecfc968dd90396e966f4f9b3ff Mon Sep 17 00:00:00 2001 From: Moshe Shemesh Date: Thu, 19 Jan 2017 14:53:07 +0200 Subject: net/mlx5e: Check ets capability before ets query FW command On dcbnl callback getpgtccfgtx, the driver should check the ets capability before ets query command is sent to firmware. It is valid to return from this void function without changing in/out parameters, as these parameters are initialized to DCB_ATTR_VALUE_UNDEFINED. Fixes: 3a6a931dfb8e ("net/mlx5e: Support DCBX CEE API") Signed-off-by: Moshe Shemesh Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index 35f9ae037ba0..0523ed47f597 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -511,6 +511,11 @@ static void mlx5e_dcbnl_getpgtccfgtx(struct net_device *netdev, struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; + if (!MLX5_CAP_GEN(priv->mdev, ets)) { + netdev_err(netdev, "%s, ets is not supported\n", __func__); + return; + } + if (priority >= CEE_DCBX_MAX_PRIO) { netdev_err(netdev, "%s, priority is out of range\n", __func__); -- cgit v1.2.3 From 39cb2c9a316e77f6dfba96c543e55b6672d5a37e Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 29 Jan 2017 13:50:06 -0800 Subject: drm/i915: Check for NULL i915_vma in intel_unpin_fb_obj() I've seen this trigger twice now, where the i915_gem_object_to_ggtt() call in intel_unpin_fb_obj() returns NULL, resulting in an oops immediately afterwards as the (inlined) call to i915_vma_unpin_fence() tries to dereference it. It seems to be some race condition where the object is going away at shutdown time, since both times happened when shutting down the X server. The call chains were different: - VT ioctl(KDSETMODE, KD_TEXT): intel_cleanup_plane_fb+0x5b/0xa0 [i915] drm_atomic_helper_cleanup_planes+0x6f/0x90 [drm_kms_helper] intel_atomic_commit_tail+0x749/0xfe0 [i915] intel_atomic_commit+0x3cb/0x4f0 [i915] drm_atomic_commit+0x4b/0x50 [drm] restore_fbdev_mode+0x14c/0x2a0 [drm_kms_helper] drm_fb_helper_restore_fbdev_mode_unlocked+0x34/0x80 [drm_kms_helper] drm_fb_helper_set_par+0x2d/0x60 [drm_kms_helper] intel_fbdev_set_par+0x18/0x70 [i915] fb_set_var+0x236/0x460 fbcon_blank+0x30f/0x350 do_unblank_screen+0xd2/0x1a0 vt_ioctl+0x507/0x12a0 tty_ioctl+0x355/0xc30 do_vfs_ioctl+0xa3/0x5e0 SyS_ioctl+0x79/0x90 entry_SYSCALL_64_fastpath+0x13/0x94 - i915 unpin_work workqueue: intel_unpin_work_fn+0x58/0x140 [i915] process_one_work+0x1f1/0x480 worker_thread+0x48/0x4d0 kthread+0x101/0x140 and this patch purely papers over the issue by adding a NULL pointer check and a WARN_ON_ONCE() to avoid the oops that would then generally make the machine unresponsive. Other callers of i915_gem_object_to_ggtt() seem to also check for the returned pointer being NULL and warn about it, so this clearly has happened before in other places. [ Reported it originally to the i915 developers on Jan 8, applying the ugly workaround on my own now after triggering the problem for the second time with no feedback. This is likely to be the same bug reported as https://bugs.freedesktop.org/show_bug.cgi?id=98829 https://bugs.freedesktop.org/show_bug.cgi?id=99134 which has a patch for the underlying problem, but it hasn't gotten to me, so I'm applying the workaround. ] Cc: Daniel Vetter Cc: Jani Nikula Cc: Ville Syrjälä Cc: Chris Wilson Cc: Maarten Lankhorst Cc: Tvrtko Ursulin Cc: Imre Deak Signed-off-by: Linus Torvalds --- drivers/gpu/drm/i915/intel_display.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 77f7b1d849a4..f0b9aa7a0483 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2251,6 +2251,9 @@ void intel_unpin_fb_obj(struct drm_framebuffer *fb, unsigned int rotation) intel_fill_fb_ggtt_view(&view, fb, rotation); vma = i915_gem_object_to_ggtt(obj, &view); + if (WARN_ON_ONCE(!vma)) + return; + i915_vma_unpin_fence(vma); i915_gem_object_unpin_from_display_plane(vma); } -- cgit v1.2.3 From 566cf877a1fcb6d6dc0126b076aad062054c2637 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 29 Jan 2017 14:25:17 -0800 Subject: Linux 4.10-rc6 --- Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Makefile b/Makefile index 098840012b9b..96b27a888285 100644 --- a/Makefile +++ b/Makefile @@ -1,8 +1,8 @@ VERSION = 4 PATCHLEVEL = 10 SUBLEVEL = 0 -EXTRAVERSION = -rc5 -NAME = Anniversary Edition +EXTRAVERSION = -rc6 +NAME = Fearless Coyote # *DOCUMENTATION* # To see a list of typical targets execute "make help" -- cgit v1.2.3 From 0a764db103376cf69d04449b10688f3516cc0b88 Mon Sep 17 00:00:00 2001 From: Alexey Brodkin Date: Fri, 27 Jan 2017 15:24:43 +0300 Subject: stmmac: Discard masked flags in interrupt status register DW GMAC databook says the following about bits in "Register 15 (Interrupt Mask Register)": --------------------------->8------------------------- When set, this bit __disables_the_assertion_of_the_interrupt_signal__ because of the setting of XXX bit in Register 14 (Interrupt Status Register). --------------------------->8------------------------- In fact even if we mask one bit in the mask register it doesn't prevent corresponding bit to appear in the status register, it only disables interrupt generation for corresponding event. But currently we expect a bit different behavior: status bits to be in sync with their masks, i.e. if mask for bit A is set in the mask register then bit A won't appear in the interrupt status register. This was proven to be incorrect assumption, see discussion here [1]. That misunderstanding causes unexpected behaviour of the GMAC, for example we were happy enough to just see bogus messages about link state changes. So from now on we'll be only checking bits that really may trigger an interrupt. [1] https://lkml.org/lkml/2016/11/3/413 Signed-off-by: Alexey Brodkin Cc: Giuseppe Cavallaro Cc: Fabrice Gasnier Cc: Joachim Eastwood Cc: Phil Reid Cc: David Miller Cc: Alexandre Torgue Cc: Vineet Gupta Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index be3c91c7f211..5484fd726d5a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -305,8 +305,12 @@ static int dwmac1000_irq_status(struct mac_device_info *hw, { void __iomem *ioaddr = hw->pcsr; u32 intr_status = readl(ioaddr + GMAC_INT_STATUS); + u32 intr_mask = readl(ioaddr + GMAC_INT_MASK); int ret = 0; + /* Discard masked bits */ + intr_status &= ~intr_mask; + /* Not used events (e.g. MMC interrupts) are not handled. */ if ((intr_status & GMAC_INT_STATUS_MMCTIS)) x->mmc_tx_irq_n++; -- cgit v1.2.3 From dc97a89e726c4e1830320d1db8215ef77ecebae0 Mon Sep 17 00:00:00 2001 From: Rafal Ozieblo Date: Fri, 27 Jan 2017 15:08:20 +0000 Subject: net: macb: Fix 64 bit addressing support for GEM This patch adds support for 32 bit GEM in 64 bit system. It checks capability at runtime and uses appropriate buffer descriptor. Signed-off-by: Rafal Ozieblo Signed-off-by: David S. Miller --- drivers/net/ethernet/cadence/macb.c | 188 +++++++++++++++++++++++++----------- drivers/net/ethernet/cadence/macb.h | 20 +++- 2 files changed, 147 insertions(+), 61 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index c0fb80acc2da..baba2db9d9c2 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -43,13 +43,13 @@ #define DEFAULT_RX_RING_SIZE 512 /* must be power of 2 */ #define MIN_RX_RING_SIZE 64 #define MAX_RX_RING_SIZE 8192 -#define RX_RING_BYTES(bp) (sizeof(struct macb_dma_desc) \ +#define RX_RING_BYTES(bp) (macb_dma_desc_get_size(bp) \ * (bp)->rx_ring_size) #define DEFAULT_TX_RING_SIZE 512 /* must be power of 2 */ #define MIN_TX_RING_SIZE 64 #define MAX_TX_RING_SIZE 4096 -#define TX_RING_BYTES(bp) (sizeof(struct macb_dma_desc) \ +#define TX_RING_BYTES(bp) (macb_dma_desc_get_size(bp) \ * (bp)->tx_ring_size) /* level of occupied TX descriptors under which we wake up TX process */ @@ -78,6 +78,37 @@ */ #define MACB_HALT_TIMEOUT 1230 +/* DMA buffer descriptor might be different size + * depends on hardware configuration. + */ +static unsigned int macb_dma_desc_get_size(struct macb *bp) +{ +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + if (bp->hw_dma_cap == HW_DMA_CAP_64B) + return sizeof(struct macb_dma_desc) + sizeof(struct macb_dma_desc_64); +#endif + return sizeof(struct macb_dma_desc); +} + +static unsigned int macb_adj_dma_desc_idx(struct macb *bp, unsigned int idx) +{ +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + /* Dma buffer descriptor is 4 words length (instead of 2 words) + * for 64b GEM. + */ + if (bp->hw_dma_cap == HW_DMA_CAP_64B) + idx <<= 1; +#endif + return idx; +} + +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT +static struct macb_dma_desc_64 *macb_64b_desc(struct macb *bp, struct macb_dma_desc *desc) +{ + return (struct macb_dma_desc_64 *)((void *)desc + sizeof(struct macb_dma_desc)); +} +#endif + /* Ring buffer accessors */ static unsigned int macb_tx_ring_wrap(struct macb *bp, unsigned int index) { @@ -87,7 +118,9 @@ static unsigned int macb_tx_ring_wrap(struct macb *bp, unsigned int index) static struct macb_dma_desc *macb_tx_desc(struct macb_queue *queue, unsigned int index) { - return &queue->tx_ring[macb_tx_ring_wrap(queue->bp, index)]; + index = macb_tx_ring_wrap(queue->bp, index); + index = macb_adj_dma_desc_idx(queue->bp, index); + return &queue->tx_ring[index]; } static struct macb_tx_skb *macb_tx_skb(struct macb_queue *queue, @@ -101,7 +134,7 @@ static dma_addr_t macb_tx_dma(struct macb_queue *queue, unsigned int index) dma_addr_t offset; offset = macb_tx_ring_wrap(queue->bp, index) * - sizeof(struct macb_dma_desc); + macb_dma_desc_get_size(queue->bp); return queue->tx_ring_dma + offset; } @@ -113,7 +146,9 @@ static unsigned int macb_rx_ring_wrap(struct macb *bp, unsigned int index) static struct macb_dma_desc *macb_rx_desc(struct macb *bp, unsigned int index) { - return &bp->rx_ring[macb_rx_ring_wrap(bp, index)]; + index = macb_rx_ring_wrap(bp, index); + index = macb_adj_dma_desc_idx(bp, index); + return &bp->rx_ring[index]; } static void *macb_rx_buffer(struct macb *bp, unsigned int index) @@ -560,12 +595,32 @@ static void macb_tx_unmap(struct macb *bp, struct macb_tx_skb *tx_skb) } } -static inline void macb_set_addr(struct macb_dma_desc *desc, dma_addr_t addr) +static void macb_set_addr(struct macb *bp, struct macb_dma_desc *desc, dma_addr_t addr) { - desc->addr = (u32)addr; #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - desc->addrh = (u32)(addr >> 32); + struct macb_dma_desc_64 *desc_64; + + if (bp->hw_dma_cap == HW_DMA_CAP_64B) { + desc_64 = macb_64b_desc(bp, desc); + desc_64->addrh = upper_32_bits(addr); + } #endif + desc->addr = lower_32_bits(addr); +} + +static dma_addr_t macb_get_addr(struct macb *bp, struct macb_dma_desc *desc) +{ + dma_addr_t addr = 0; +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + struct macb_dma_desc_64 *desc_64; + + if (bp->hw_dma_cap == HW_DMA_CAP_64B) { + desc_64 = macb_64b_desc(bp, desc); + addr = ((u64)(desc_64->addrh) << 32); + } +#endif + addr |= MACB_BF(RX_WADDR, MACB_BFEXT(RX_WADDR, desc->addr)); + return addr; } static void macb_tx_error_task(struct work_struct *work) @@ -649,16 +704,17 @@ static void macb_tx_error_task(struct work_struct *work) /* Set end of TX queue */ desc = macb_tx_desc(queue, 0); - macb_set_addr(desc, 0); + macb_set_addr(bp, desc, 0); desc->ctrl = MACB_BIT(TX_USED); /* Make descriptor updates visible to hardware */ wmb(); /* Reinitialize the TX desc queue */ - queue_writel(queue, TBQP, (u32)(queue->tx_ring_dma)); + queue_writel(queue, TBQP, lower_32_bits(queue->tx_ring_dma)); #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - queue_writel(queue, TBQPH, (u32)(queue->tx_ring_dma >> 32)); + if (bp->hw_dma_cap == HW_DMA_CAP_64B) + queue_writel(queue, TBQPH, upper_32_bits(queue->tx_ring_dma)); #endif /* Make TX ring reflect state of hardware */ queue->tx_head = 0; @@ -750,6 +806,7 @@ static void gem_rx_refill(struct macb *bp) unsigned int entry; struct sk_buff *skb; dma_addr_t paddr; + struct macb_dma_desc *desc; while (CIRC_SPACE(bp->rx_prepared_head, bp->rx_tail, bp->rx_ring_size) > 0) { @@ -759,6 +816,7 @@ static void gem_rx_refill(struct macb *bp) rmb(); bp->rx_prepared_head++; + desc = macb_rx_desc(bp, entry); if (!bp->rx_skbuff[entry]) { /* allocate sk_buff for this free entry in ring */ @@ -782,14 +840,14 @@ static void gem_rx_refill(struct macb *bp) if (entry == bp->rx_ring_size - 1) paddr |= MACB_BIT(RX_WRAP); - macb_set_addr(&(bp->rx_ring[entry]), paddr); - bp->rx_ring[entry].ctrl = 0; + macb_set_addr(bp, desc, paddr); + desc->ctrl = 0; /* properly align Ethernet header */ skb_reserve(skb, NET_IP_ALIGN); } else { - bp->rx_ring[entry].addr &= ~MACB_BIT(RX_USED); - bp->rx_ring[entry].ctrl = 0; + desc->addr &= ~MACB_BIT(RX_USED); + desc->ctrl = 0; } } @@ -835,16 +893,13 @@ static int gem_rx(struct macb *bp, int budget) bool rxused; entry = macb_rx_ring_wrap(bp, bp->rx_tail); - desc = &bp->rx_ring[entry]; + desc = macb_rx_desc(bp, entry); /* Make hw descriptor updates visible to CPU */ rmb(); rxused = (desc->addr & MACB_BIT(RX_USED)) ? true : false; - addr = MACB_BF(RX_WADDR, MACB_BFEXT(RX_WADDR, desc->addr)); -#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - addr |= ((u64)(desc->addrh) << 32); -#endif + addr = macb_get_addr(bp, desc); ctrl = desc->ctrl; if (!rxused) @@ -987,15 +1042,17 @@ static int macb_rx_frame(struct macb *bp, unsigned int first_frag, static inline void macb_init_rx_ring(struct macb *bp) { dma_addr_t addr; + struct macb_dma_desc *desc = NULL; int i; addr = bp->rx_buffers_dma; for (i = 0; i < bp->rx_ring_size; i++) { - bp->rx_ring[i].addr = addr; - bp->rx_ring[i].ctrl = 0; + desc = macb_rx_desc(bp, i); + macb_set_addr(bp, desc, addr); + desc->ctrl = 0; addr += bp->rx_buffer_size; } - bp->rx_ring[bp->rx_ring_size - 1].addr |= MACB_BIT(RX_WRAP); + desc->addr |= MACB_BIT(RX_WRAP); bp->rx_tail = 0; } @@ -1008,15 +1065,14 @@ static int macb_rx(struct macb *bp, int budget) for (tail = bp->rx_tail; budget > 0; tail++) { struct macb_dma_desc *desc = macb_rx_desc(bp, tail); - u32 addr, ctrl; + u32 ctrl; /* Make hw descriptor updates visible to CPU */ rmb(); - addr = desc->addr; ctrl = desc->ctrl; - if (!(addr & MACB_BIT(RX_USED))) + if (!(desc->addr & MACB_BIT(RX_USED))) break; if (ctrl & MACB_BIT(RX_SOF)) { @@ -1336,7 +1392,7 @@ static unsigned int macb_tx_map(struct macb *bp, i = tx_head; entry = macb_tx_ring_wrap(bp, i); ctrl = MACB_BIT(TX_USED); - desc = &queue->tx_ring[entry]; + desc = macb_tx_desc(queue, entry); desc->ctrl = ctrl; if (lso_ctrl) { @@ -1358,7 +1414,7 @@ static unsigned int macb_tx_map(struct macb *bp, i--; entry = macb_tx_ring_wrap(bp, i); tx_skb = &queue->tx_skb[entry]; - desc = &queue->tx_ring[entry]; + desc = macb_tx_desc(queue, entry); ctrl = (u32)tx_skb->size; if (eof) { @@ -1379,7 +1435,7 @@ static unsigned int macb_tx_map(struct macb *bp, ctrl |= MACB_BF(MSS_MFS, mss_mfs); /* Set TX buffer descriptor */ - macb_set_addr(desc, tx_skb->mapping); + macb_set_addr(bp, desc, tx_skb->mapping); /* desc->addr must be visible to hardware before clearing * 'TX_USED' bit in desc->ctrl. */ @@ -1586,11 +1642,9 @@ static void gem_free_rx_buffers(struct macb *bp) if (!skb) continue; - desc = &bp->rx_ring[i]; - addr = MACB_BF(RX_WADDR, MACB_BFEXT(RX_WADDR, desc->addr)); -#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - addr |= ((u64)(desc->addrh) << 32); -#endif + desc = macb_rx_desc(bp, i); + addr = macb_get_addr(bp, desc); + dma_unmap_single(&bp->pdev->dev, addr, bp->rx_buffer_size, DMA_FROM_DEVICE); dev_kfree_skb_any(skb); @@ -1711,15 +1765,17 @@ out_err: static void gem_init_rings(struct macb *bp) { struct macb_queue *queue; + struct macb_dma_desc *desc = NULL; unsigned int q; int i; for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) { for (i = 0; i < bp->tx_ring_size; i++) { - queue->tx_ring[i].addr = 0; - queue->tx_ring[i].ctrl = MACB_BIT(TX_USED); + desc = macb_tx_desc(queue, i); + macb_set_addr(bp, desc, 0); + desc->ctrl = MACB_BIT(TX_USED); } - queue->tx_ring[bp->tx_ring_size - 1].ctrl |= MACB_BIT(TX_WRAP); + desc->ctrl |= MACB_BIT(TX_WRAP); queue->tx_head = 0; queue->tx_tail = 0; } @@ -1733,16 +1789,18 @@ static void gem_init_rings(struct macb *bp) static void macb_init_rings(struct macb *bp) { int i; + struct macb_dma_desc *desc = NULL; macb_init_rx_ring(bp); for (i = 0; i < bp->tx_ring_size; i++) { - bp->queues[0].tx_ring[i].addr = 0; - bp->queues[0].tx_ring[i].ctrl = MACB_BIT(TX_USED); + desc = macb_tx_desc(&bp->queues[0], i); + macb_set_addr(bp, desc, 0); + desc->ctrl = MACB_BIT(TX_USED); } bp->queues[0].tx_head = 0; bp->queues[0].tx_tail = 0; - bp->queues[0].tx_ring[bp->tx_ring_size - 1].ctrl |= MACB_BIT(TX_WRAP); + desc->ctrl |= MACB_BIT(TX_WRAP); } static void macb_reset_hw(struct macb *bp) @@ -1863,7 +1921,8 @@ static void macb_configure_dma(struct macb *bp) dmacfg &= ~GEM_BIT(TXCOEN); #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - dmacfg |= GEM_BIT(ADDR64); + if (bp->hw_dma_cap == HW_DMA_CAP_64B) + dmacfg |= GEM_BIT(ADDR64); #endif netdev_dbg(bp->dev, "Cadence configure DMA with 0x%08x\n", dmacfg); @@ -1910,14 +1969,16 @@ static void macb_init_hw(struct macb *bp) macb_configure_dma(bp); /* Initialize TX and RX buffers */ - macb_writel(bp, RBQP, (u32)(bp->rx_ring_dma)); + macb_writel(bp, RBQP, lower_32_bits(bp->rx_ring_dma)); #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - macb_writel(bp, RBQPH, (u32)(bp->rx_ring_dma >> 32)); + if (bp->hw_dma_cap == HW_DMA_CAP_64B) + macb_writel(bp, RBQPH, upper_32_bits(bp->rx_ring_dma)); #endif for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) { - queue_writel(queue, TBQP, (u32)(queue->tx_ring_dma)); + queue_writel(queue, TBQP, lower_32_bits(queue->tx_ring_dma)); #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - queue_writel(queue, TBQPH, (u32)(queue->tx_ring_dma >> 32)); + if (bp->hw_dma_cap == HW_DMA_CAP_64B) + queue_writel(queue, TBQPH, upper_32_bits(queue->tx_ring_dma)); #endif /* Enable interrupts */ @@ -2627,7 +2688,8 @@ static int macb_init(struct platform_device *pdev) queue->IMR = GEM_IMR(hw_q - 1); queue->TBQP = GEM_TBQP(hw_q - 1); #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - queue->TBQPH = GEM_TBQPH(hw_q -1); + if (bp->hw_dma_cap == HW_DMA_CAP_64B) + queue->TBQPH = GEM_TBQPH(hw_q - 1); #endif } else { /* queue0 uses legacy registers */ @@ -2637,7 +2699,8 @@ static int macb_init(struct platform_device *pdev) queue->IMR = MACB_IMR; queue->TBQP = MACB_TBQP; #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - queue->TBQPH = MACB_TBQPH; + if (bp->hw_dma_cap == HW_DMA_CAP_64B) + queue->TBQPH = MACB_TBQPH; #endif } @@ -2730,13 +2793,14 @@ static int macb_init(struct platform_device *pdev) static int at91ether_start(struct net_device *dev) { struct macb *lp = netdev_priv(dev); + struct macb_dma_desc *desc; dma_addr_t addr; u32 ctl; int i; lp->rx_ring = dma_alloc_coherent(&lp->pdev->dev, (AT91ETHER_MAX_RX_DESCR * - sizeof(struct macb_dma_desc)), + macb_dma_desc_get_size(lp)), &lp->rx_ring_dma, GFP_KERNEL); if (!lp->rx_ring) return -ENOMEM; @@ -2748,7 +2812,7 @@ static int at91ether_start(struct net_device *dev) if (!lp->rx_buffers) { dma_free_coherent(&lp->pdev->dev, AT91ETHER_MAX_RX_DESCR * - sizeof(struct macb_dma_desc), + macb_dma_desc_get_size(lp), lp->rx_ring, lp->rx_ring_dma); lp->rx_ring = NULL; return -ENOMEM; @@ -2756,13 +2820,14 @@ static int at91ether_start(struct net_device *dev) addr = lp->rx_buffers_dma; for (i = 0; i < AT91ETHER_MAX_RX_DESCR; i++) { - lp->rx_ring[i].addr = addr; - lp->rx_ring[i].ctrl = 0; + desc = macb_rx_desc(lp, i); + macb_set_addr(lp, desc, addr); + desc->ctrl = 0; addr += AT91ETHER_MAX_RBUFF_SZ; } /* Set the Wrap bit on the last descriptor */ - lp->rx_ring[AT91ETHER_MAX_RX_DESCR - 1].addr |= MACB_BIT(RX_WRAP); + desc->addr |= MACB_BIT(RX_WRAP); /* Reset buffer index */ lp->rx_tail = 0; @@ -2834,7 +2899,7 @@ static int at91ether_close(struct net_device *dev) dma_free_coherent(&lp->pdev->dev, AT91ETHER_MAX_RX_DESCR * - sizeof(struct macb_dma_desc), + macb_dma_desc_get_size(lp), lp->rx_ring, lp->rx_ring_dma); lp->rx_ring = NULL; @@ -2885,13 +2950,15 @@ static int at91ether_start_xmit(struct sk_buff *skb, struct net_device *dev) static void at91ether_rx(struct net_device *dev) { struct macb *lp = netdev_priv(dev); + struct macb_dma_desc *desc; unsigned char *p_recv; struct sk_buff *skb; unsigned int pktlen; - while (lp->rx_ring[lp->rx_tail].addr & MACB_BIT(RX_USED)) { + desc = macb_rx_desc(lp, lp->rx_tail); + while (desc->addr & MACB_BIT(RX_USED)) { p_recv = lp->rx_buffers + lp->rx_tail * AT91ETHER_MAX_RBUFF_SZ; - pktlen = MACB_BF(RX_FRMLEN, lp->rx_ring[lp->rx_tail].ctrl); + pktlen = MACB_BF(RX_FRMLEN, desc->ctrl); skb = netdev_alloc_skb(dev, pktlen + 2); if (skb) { skb_reserve(skb, 2); @@ -2905,17 +2972,19 @@ static void at91ether_rx(struct net_device *dev) lp->stats.rx_dropped++; } - if (lp->rx_ring[lp->rx_tail].ctrl & MACB_BIT(RX_MHASH_MATCH)) + if (desc->ctrl & MACB_BIT(RX_MHASH_MATCH)) lp->stats.multicast++; /* reset ownership bit */ - lp->rx_ring[lp->rx_tail].addr &= ~MACB_BIT(RX_USED); + desc->addr &= ~MACB_BIT(RX_USED); /* wrap after last buffer */ if (lp->rx_tail == AT91ETHER_MAX_RX_DESCR - 1) lp->rx_tail = 0; else lp->rx_tail++; + + desc = macb_rx_desc(lp, lp->rx_tail); } } @@ -3211,8 +3280,11 @@ static int macb_probe(struct platform_device *pdev) device_init_wakeup(&pdev->dev, bp->wol & MACB_WOL_HAS_MAGIC_PACKET); #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - if (GEM_BFEXT(DBWDEF, gem_readl(bp, DCFG1)) > GEM_DBW32) + if (GEM_BFEXT(DAW64, gem_readl(bp, DCFG6))) { dma_set_mask(&pdev->dev, DMA_BIT_MASK(44)); + bp->hw_dma_cap = HW_DMA_CAP_64B; + } else + bp->hw_dma_cap = HW_DMA_CAP_32B; #endif spin_lock_init(&bp->lock); diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index d67adad67be1..fc8550a5d47f 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -385,6 +385,8 @@ /* Bitfields in DCFG6. */ #define GEM_PBUF_LSO_OFFSET 27 #define GEM_PBUF_LSO_SIZE 1 +#define GEM_DAW64_OFFSET 23 +#define GEM_DAW64_SIZE 1 /* Constants for CLK */ #define MACB_CLK_DIV8 0 @@ -487,12 +489,20 @@ struct macb_dma_desc { u32 addr; u32 ctrl; +}; + #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - u32 addrh; - u32 resvd; -#endif +enum macb_hw_dma_cap { + HW_DMA_CAP_32B, + HW_DMA_CAP_64B, }; +struct macb_dma_desc_64 { + u32 addrh; + u32 resvd; +}; +#endif + /* DMA descriptor bitfields */ #define MACB_RX_USED_OFFSET 0 #define MACB_RX_USED_SIZE 1 @@ -874,6 +884,10 @@ struct macb { unsigned int jumbo_max_len; u32 wol; + +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + enum macb_hw_dma_cap hw_dma_cap; +#endif }; static inline bool macb_is_gem(struct macb *bp) -- cgit v1.2.3 From f1712c73714088a7252d276a57126d56c7d37e64 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Jan 2017 08:11:44 -0800 Subject: can: Fix kernel panic at security_sock_rcv_skb Zhang Yanmin reported crashes [1] and provided a patch adding a synchronize_rcu() call in can_rx_unregister() The main problem seems that the sockets themselves are not RCU protected. If CAN uses RCU for delivery, then sockets should be freed only after one RCU grace period. Recent kernels could use sock_set_flag(sk, SOCK_RCU_FREE), but let's ease stable backports with the following fix instead. [1] BUG: unable to handle kernel NULL pointer dereference at (null) IP: [] selinux_socket_sock_rcv_skb+0x65/0x2a0 Call Trace: [] security_sock_rcv_skb+0x4c/0x60 [] sk_filter+0x41/0x210 [] sock_queue_rcv_skb+0x53/0x3a0 [] raw_rcv+0x2a3/0x3c0 [] can_rcv_filter+0x12b/0x370 [] can_receive+0xd9/0x120 [] can_rcv+0xab/0x100 [] __netif_receive_skb_core+0xd8c/0x11f0 [] __netif_receive_skb+0x24/0xb0 [] process_backlog+0x127/0x280 [] net_rx_action+0x33b/0x4f0 [] __do_softirq+0x184/0x440 [] do_softirq_own_stack+0x1c/0x30 [] do_softirq.part.18+0x3b/0x40 [] do_softirq+0x1d/0x20 [] netif_rx_ni+0xe5/0x110 [] slcan_receive_buf+0x507/0x520 [] flush_to_ldisc+0x21c/0x230 [] process_one_work+0x24f/0x670 [] worker_thread+0x9d/0x6f0 [] ? rescuer_thread+0x480/0x480 [] kthread+0x12c/0x150 [] ret_from_fork+0x3f/0x70 Reported-by: Zhang Yanmin Signed-off-by: Eric Dumazet Acked-by: Oliver Hartkopp Signed-off-by: David S. Miller --- include/linux/can/core.h | 7 +++---- net/can/af_can.c | 12 ++++++++++-- net/can/af_can.h | 3 ++- net/can/bcm.c | 4 ++-- net/can/gw.c | 2 +- net/can/raw.c | 4 ++-- 6 files changed, 20 insertions(+), 12 deletions(-) diff --git a/include/linux/can/core.h b/include/linux/can/core.h index a0875001b13c..df08a41d5be5 100644 --- a/include/linux/can/core.h +++ b/include/linux/can/core.h @@ -45,10 +45,9 @@ struct can_proto { extern int can_proto_register(const struct can_proto *cp); extern void can_proto_unregister(const struct can_proto *cp); -extern int can_rx_register(struct net_device *dev, canid_t can_id, - canid_t mask, - void (*func)(struct sk_buff *, void *), - void *data, char *ident); +int can_rx_register(struct net_device *dev, canid_t can_id, canid_t mask, + void (*func)(struct sk_buff *, void *), + void *data, char *ident, struct sock *sk); extern void can_rx_unregister(struct net_device *dev, canid_t can_id, canid_t mask, diff --git a/net/can/af_can.c b/net/can/af_can.c index 1108079d934f..5488e4a6ccd0 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -445,6 +445,7 @@ static struct hlist_head *find_rcv_list(canid_t *can_id, canid_t *mask, * @func: callback function on filter match * @data: returned parameter for callback function * @ident: string for calling module identification + * @sk: socket pointer (might be NULL) * * Description: * Invokes the callback function with the received sk_buff and the given @@ -468,7 +469,7 @@ static struct hlist_head *find_rcv_list(canid_t *can_id, canid_t *mask, */ int can_rx_register(struct net_device *dev, canid_t can_id, canid_t mask, void (*func)(struct sk_buff *, void *), void *data, - char *ident) + char *ident, struct sock *sk) { struct receiver *r; struct hlist_head *rl; @@ -496,6 +497,7 @@ int can_rx_register(struct net_device *dev, canid_t can_id, canid_t mask, r->func = func; r->data = data; r->ident = ident; + r->sk = sk; hlist_add_head_rcu(&r->list, rl); d->entries++; @@ -520,8 +522,11 @@ EXPORT_SYMBOL(can_rx_register); static void can_rx_delete_receiver(struct rcu_head *rp) { struct receiver *r = container_of(rp, struct receiver, rcu); + struct sock *sk = r->sk; kmem_cache_free(rcv_cache, r); + if (sk) + sock_put(sk); } /** @@ -596,8 +601,11 @@ void can_rx_unregister(struct net_device *dev, canid_t can_id, canid_t mask, spin_unlock(&can_rcvlists_lock); /* schedule the receiver item for deletion */ - if (r) + if (r) { + if (r->sk) + sock_hold(r->sk); call_rcu(&r->rcu, can_rx_delete_receiver); + } } EXPORT_SYMBOL(can_rx_unregister); diff --git a/net/can/af_can.h b/net/can/af_can.h index fca0fe9fc45a..b86f5129e838 100644 --- a/net/can/af_can.h +++ b/net/can/af_can.h @@ -50,13 +50,14 @@ struct receiver { struct hlist_node list; - struct rcu_head rcu; canid_t can_id; canid_t mask; unsigned long matches; void (*func)(struct sk_buff *, void *); void *data; char *ident; + struct sock *sk; + struct rcu_head rcu; }; #define CAN_SFF_RCV_ARRAY_SZ (1 << CAN_SFF_ID_BITS) diff --git a/net/can/bcm.c b/net/can/bcm.c index 21ac75390e3d..5c9407181918 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -1216,7 +1216,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, err = can_rx_register(dev, op->can_id, REGMASK(op->can_id), bcm_rx_handler, op, - "bcm"); + "bcm", sk); op->rx_reg_dev = dev; dev_put(dev); @@ -1225,7 +1225,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, } else err = can_rx_register(NULL, op->can_id, REGMASK(op->can_id), - bcm_rx_handler, op, "bcm"); + bcm_rx_handler, op, "bcm", sk); if (err) { /* this bcm rx op is broken -> remove it */ list_del(&op->list); diff --git a/net/can/gw.c b/net/can/gw.c index a54ab0c82104..7056a1a2bb70 100644 --- a/net/can/gw.c +++ b/net/can/gw.c @@ -442,7 +442,7 @@ static inline int cgw_register_filter(struct cgw_job *gwj) { return can_rx_register(gwj->src.dev, gwj->ccgw.filter.can_id, gwj->ccgw.filter.can_mask, can_can_gw_rcv, - gwj, "gw"); + gwj, "gw", NULL); } static inline void cgw_unregister_filter(struct cgw_job *gwj) diff --git a/net/can/raw.c b/net/can/raw.c index b075f028d7e2..6dc546a06673 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -190,7 +190,7 @@ static int raw_enable_filters(struct net_device *dev, struct sock *sk, for (i = 0; i < count; i++) { err = can_rx_register(dev, filter[i].can_id, filter[i].can_mask, - raw_rcv, sk, "raw"); + raw_rcv, sk, "raw", sk); if (err) { /* clean up successfully registered filters */ while (--i >= 0) @@ -211,7 +211,7 @@ static int raw_enable_errfilter(struct net_device *dev, struct sock *sk, if (err_mask) err = can_rx_register(dev, 0, err_mask | CAN_ERR_FLAG, - raw_rcv, sk, "raw"); + raw_rcv, sk, "raw", sk); return err; } -- cgit v1.2.3 From cf626c3b252b2c9d131be0dd66096ec3bf729e54 Mon Sep 17 00:00:00 2001 From: Sean Nyekjaer Date: Fri, 27 Jan 2017 21:39:03 +0100 Subject: net: phy: micrel: KSZ8795 do not set SUPPORTED_[Asym_]Pause As pr commit "net: phy: phy drivers should not set SUPPORTED_[Asym_]Pause" this phy driver should not set these feature bits. Signed-off-by: Sean Nyekjaer Fixes: 9d162ed69f51 ("net: phy: micrel: add support for KSZ8795") Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/micrel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index e55809c5beb7..6742070ca676 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -1012,7 +1012,7 @@ static struct phy_driver ksphy_driver[] = { .phy_id = PHY_ID_KSZ8795, .phy_id_mask = MICREL_PHY_ID_MASK, .name = "Micrel KSZ8795", - .features = (SUPPORTED_Pause | SUPPORTED_Asym_Pause), + .features = PHY_BASIC_FEATURES, .flags = PHY_HAS_MAGICANEG | PHY_HAS_INTERRUPT, .config_init = kszphy_config_init, .config_aneg = ksz8873mll_config_aneg, -- cgit v1.2.3 From d1156b489fa734d1af763d6a07b1637c01bb0aed Mon Sep 17 00:00:00 2001 From: Alexey Khoroshilov Date: Sat, 28 Jan 2017 01:07:30 +0300 Subject: net: adaptec: starfire: add checks for dma mapping errors init_ring(), refill_rx_ring() and start_tx() don't check if mapping dma memory succeed. The patch adds the checks and failure handling. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov Signed-off-by: David S. Miller --- drivers/net/ethernet/adaptec/starfire.c | 45 +++++++++++++++++++++++++++++++-- 1 file changed, 43 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/adaptec/starfire.c b/drivers/net/ethernet/adaptec/starfire.c index c12d2618eebf..3872ab96b80a 100644 --- a/drivers/net/ethernet/adaptec/starfire.c +++ b/drivers/net/ethernet/adaptec/starfire.c @@ -1152,6 +1152,12 @@ static void init_ring(struct net_device *dev) if (skb == NULL) break; np->rx_info[i].mapping = pci_map_single(np->pci_dev, skb->data, np->rx_buf_sz, PCI_DMA_FROMDEVICE); + if (pci_dma_mapping_error(np->pci_dev, + np->rx_info[i].mapping)) { + dev_kfree_skb(skb); + np->rx_info[i].skb = NULL; + break; + } /* Grrr, we cannot offset to correctly align the IP header. */ np->rx_ring[i].rxaddr = cpu_to_dma(np->rx_info[i].mapping | RxDescValid); } @@ -1182,8 +1188,9 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev) { struct netdev_private *np = netdev_priv(dev); unsigned int entry; + unsigned int prev_tx; u32 status; - int i; + int i, j; /* * be cautious here, wrapping the queue has weird semantics @@ -1201,6 +1208,7 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev) } #endif /* ZEROCOPY && HAS_BROKEN_FIRMWARE */ + prev_tx = np->cur_tx; entry = np->cur_tx % TX_RING_SIZE; for (i = 0; i < skb_num_frags(skb); i++) { int wrap_ring = 0; @@ -1234,6 +1242,11 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev) skb_frag_size(this_frag), PCI_DMA_TODEVICE); } + if (pci_dma_mapping_error(np->pci_dev, + np->tx_info[entry].mapping)) { + dev->stats.tx_dropped++; + goto err_out; + } np->tx_ring[entry].addr = cpu_to_dma(np->tx_info[entry].mapping); np->tx_ring[entry].status = cpu_to_le32(status); @@ -1268,8 +1281,30 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev) netif_stop_queue(dev); return NETDEV_TX_OK; -} +err_out: + entry = prev_tx % TX_RING_SIZE; + np->tx_info[entry].skb = NULL; + if (i > 0) { + pci_unmap_single(np->pci_dev, + np->tx_info[entry].mapping, + skb_first_frag_len(skb), + PCI_DMA_TODEVICE); + np->tx_info[entry].mapping = 0; + entry = (entry + np->tx_info[entry].used_slots) % TX_RING_SIZE; + for (j = 1; j < i; j++) { + pci_unmap_single(np->pci_dev, + np->tx_info[entry].mapping, + skb_frag_size( + &skb_shinfo(skb)->frags[j-1]), + PCI_DMA_TODEVICE); + entry++; + } + } + dev_kfree_skb_any(skb); + np->cur_tx = prev_tx; + return NETDEV_TX_OK; +} /* The interrupt handler does all of the Rx thread work and cleans up after the Tx thread. */ @@ -1569,6 +1604,12 @@ static void refill_rx_ring(struct net_device *dev) break; /* Better luck next round. */ np->rx_info[entry].mapping = pci_map_single(np->pci_dev, skb->data, np->rx_buf_sz, PCI_DMA_FROMDEVICE); + if (pci_dma_mapping_error(np->pci_dev, + np->rx_info[entry].mapping)) { + dev_kfree_skb(skb); + np->rx_info[entry].skb = NULL; + break; + } np->rx_ring[entry].rxaddr = cpu_to_dma(np->rx_info[entry].mapping | RxDescValid); } -- cgit v1.2.3 From a0615a16f7d0ceb5804d295203c302d496d8ee91 Mon Sep 17 00:00:00 2001 From: Reza Arbab Date: Wed, 25 Jan 2017 09:54:33 -0600 Subject: powerpc/mm: Use the correct pointer when setting a 2MB pte When setting a 2MB pte, radix__map_kernel_page() is using the address ptep = (pte_t *)pudp; Fix this conversion to use pmdp instead. Use pmdp_ptep() to do this instead of casting the pointer. Fixes: 2bfd65e45e87 ("powerpc/mm/radix: Add radix callbacks for early init routines") Cc: stable@vger.kernel.org # v4.7+ Reviewed-by: Aneesh Kumar K.V Signed-off-by: Reza Arbab Signed-off-by: Michael Ellerman --- arch/powerpc/mm/pgtable-radix.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/mm/pgtable-radix.c b/arch/powerpc/mm/pgtable-radix.c index cfa53ccc8baf..34f1a0dbc898 100644 --- a/arch/powerpc/mm/pgtable-radix.c +++ b/arch/powerpc/mm/pgtable-radix.c @@ -65,7 +65,7 @@ int radix__map_kernel_page(unsigned long ea, unsigned long pa, if (!pmdp) return -ENOMEM; if (map_page_size == PMD_SIZE) { - ptep = (pte_t *)pudp; + ptep = pmdp_ptep(pmdp); goto set_the_pte; } ptep = pte_alloc_kernel(pmdp, ea); @@ -90,7 +90,7 @@ int radix__map_kernel_page(unsigned long ea, unsigned long pa, } pmdp = pmd_offset(pudp, ea); if (map_page_size == PMD_SIZE) { - ptep = (pte_t *)pudp; + ptep = pmdp_ptep(pmdp); goto set_the_pte; } if (!pmd_present(*pmdp)) { -- cgit v1.2.3 From e82d02580af45663fad6d3596e4344c606e81e10 Mon Sep 17 00:00:00 2001 From: Jisheng Zhang Date: Mon, 23 Jan 2017 15:15:32 +0800 Subject: pinctrl: berlin-bg4ct: fix the value for "sd1a" of pin SCRD0_CRD_PRES This should be a typo. Signed-off-by: Jisheng Zhang Signed-off-by: Linus Walleij --- drivers/pinctrl/berlin/berlin-bg4ct.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pinctrl/berlin/berlin-bg4ct.c b/drivers/pinctrl/berlin/berlin-bg4ct.c index 09172043d589..c617ec49e9ed 100644 --- a/drivers/pinctrl/berlin/berlin-bg4ct.c +++ b/drivers/pinctrl/berlin/berlin-bg4ct.c @@ -217,7 +217,7 @@ static const struct berlin_desc_group berlin4ct_soc_pinctrl_groups[] = { BERLIN_PINCTRL_GROUP("SCRD0_CRD_PRES", 0xc, 0x3, 0x15, BERLIN_PINCTRL_FUNCTION(0x0, "gpio"), /* GPIO20 */ BERLIN_PINCTRL_FUNCTION(0x1, "scrd0"), /* crd pres */ - BERLIN_PINCTRL_FUNCTION(0x1, "sd1a")), /* DAT3 */ + BERLIN_PINCTRL_FUNCTION(0x3, "sd1a")), /* DAT3 */ BERLIN_PINCTRL_GROUP("SPI1_SS0n", 0xc, 0x3, 0x18, BERLIN_PINCTRL_FUNCTION(0x0, "spi1"), /* SS0n */ BERLIN_PINCTRL_FUNCTION(0x1, "gpio"), /* GPIO37 */ -- cgit v1.2.3 From 2154d94b40ea2a5de05245521371d0461bb0d669 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Mon, 23 Jan 2017 09:21:30 +0100 Subject: pinctrl: sunxi: Don't enforce bias disable (for now) Commit 07fe64ba213f ("pinctrl: sunxi: Handle bias disable") actually enforced enforced the disabling of the pull up/down resistors instead of ignoring it like it was done before. This was part of a wider rework to switch to the generic pinconf bindings, and was meant to be merged together with DT patches that were switching to it, and removing what was considered default values by both the binding and the boards. This included no bias on a pin. However, those DT patches were delayed to 4.11, which would be fine only for a significant number boards having the bias setup wrong, which in turns break the MMC on those boards (and possibly other devices too). In order to avoid conflicts as much as possible, bring back the old behaviour for 4.10, and we'll revert that commit once all the DT bits will have landed. Tested-by: Priit Laes Signed-off-by: Maxime Ripard Acked-by: Chen-Yu Tsai Signed-off-by: Linus Walleij --- drivers/pinctrl/sunxi/pinctrl-sunxi.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/pinctrl/sunxi/pinctrl-sunxi.c b/drivers/pinctrl/sunxi/pinctrl-sunxi.c index 0eb51e33cb1b..207a8de4e1ed 100644 --- a/drivers/pinctrl/sunxi/pinctrl-sunxi.c +++ b/drivers/pinctrl/sunxi/pinctrl-sunxi.c @@ -564,8 +564,7 @@ static int sunxi_pconf_group_set(struct pinctrl_dev *pctldev, val = arg / 10 - 1; break; case PIN_CONFIG_BIAS_DISABLE: - val = 0; - break; + continue; case PIN_CONFIG_BIAS_PULL_UP: if (arg == 0) return -EINVAL; -- cgit v1.2.3 From 19b26d92dfb70f56440c187a20c49102ab648b97 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 24 Jan 2017 17:28:22 +0200 Subject: pinctrl: intel: merrifield: Add missed check in mrfld_config_set() Not every pin can be configured. Add missed check to prevent access violation. Fixes: 4e80c8f50574 ("pinctrl: intel: Add Intel Merrifield pin controller support") Acked-by: Mika Westerberg Signed-off-by: Andy Shevchenko Signed-off-by: Linus Walleij --- drivers/pinctrl/intel/pinctrl-merrifield.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/pinctrl/intel/pinctrl-merrifield.c b/drivers/pinctrl/intel/pinctrl-merrifield.c index b21896126f76..4d4ef42a39b5 100644 --- a/drivers/pinctrl/intel/pinctrl-merrifield.c +++ b/drivers/pinctrl/intel/pinctrl-merrifield.c @@ -794,6 +794,9 @@ static int mrfld_config_set(struct pinctrl_dev *pctldev, unsigned int pin, unsigned int i; int ret; + if (!mrfld_buf_available(mp, pin)) + return -ENOTSUPP; + for (i = 0; i < nconfigs; i++) { switch (pinconf_to_config_param(configs[i])) { case PIN_CONFIG_BIAS_DISABLE: -- cgit v1.2.3 From 24c2503255d35c269b67162c397a1a1c1e02f6ce Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Wed, 25 Jan 2017 21:00:48 +0100 Subject: x86/microcode: Do not access the initrd after it has been freed When we look for microcode blobs, we first try builtin and if that doesn't succeed, we fallback to the initrd supplied to the kernel. However, at some point doing boot, that initrd gets jettisoned and we shouldn't access it anymore. But we do, as the below KASAN report shows. That's because find_microcode_in_initrd() doesn't check whether the initrd is still valid or not. So do that. ================================================================== BUG: KASAN: use-after-free in find_cpio_data Read of size 1 by task swapper/1/0 page:ffffea0000db9d40 count:0 mapcount:0 mapping: (null) index:0x1 flags: 0x100000000000000() raw: 0100000000000000 0000000000000000 0000000000000001 00000000ffffffff raw: dead000000000100 dead000000000200 0000000000000000 0000000000000000 page dumped because: kasan: bad access detected CPU: 1 PID: 0 Comm: swapper/1 Tainted: G W 4.10.0-rc5-debug-00075-g2dbde22 #3 Hardware name: Dell Inc. XPS 13 9360/0839Y6, BIOS 1.2.3 12/01/2016 Call Trace: dump_stack ? _atomic_dec_and_lock ? __dump_page kasan_report_error ? pointer ? find_cpio_data __asan_report_load1_noabort ? find_cpio_data find_cpio_data ? vsprintf ? dump_stack ? get_ucode_user ? print_usage_bug find_microcode_in_initrd __load_ucode_intel ? collect_cpu_info_early ? debug_check_no_locks_freed load_ucode_intel_ap ? collect_cpu_info ? trace_hardirqs_on ? flat_send_IPI_mask_allbutself load_ucode_ap ? get_builtin_firmware ? flush_tlb_func ? do_raw_spin_trylock ? cpumask_weight cpu_init ? trace_hardirqs_off ? play_dead_common ? native_play_dead ? hlt_play_dead ? syscall_init ? arch_cpu_idle_dead ? do_idle start_secondary start_cpu Memory state around the buggy address: ffff880036e74f00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ffff880036e74f80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff >ffff880036e75000: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ^ ffff880036e75080: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ffff880036e75100: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ================================================================== Reported-by: Andrey Ryabinin Tested-by: Andrey Ryabinin Signed-off-by: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20170126165833.evjemhbqzaepirxo@pd.tnic Signed-off-by: Ingo Molnar --- arch/x86/include/asm/microcode.h | 1 + arch/x86/kernel/cpu/microcode/amd.c | 5 +++-- arch/x86/kernel/cpu/microcode/core.c | 22 +++++++++++++++++----- 3 files changed, 21 insertions(+), 7 deletions(-) diff --git a/arch/x86/include/asm/microcode.h b/arch/x86/include/asm/microcode.h index 38711df3bcb5..2266f864b747 100644 --- a/arch/x86/include/asm/microcode.h +++ b/arch/x86/include/asm/microcode.h @@ -140,6 +140,7 @@ extern void __init load_ucode_bsp(void); extern void load_ucode_ap(void); void reload_early_microcode(void); extern bool get_builtin_firmware(struct cpio_data *cd, const char *name); +extern bool initrd_gone; #else static inline int __init microcode_init(void) { return 0; }; static inline void __init load_ucode_bsp(void) { } diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c index 6a31e2691f3a..079e81733a58 100644 --- a/arch/x86/kernel/cpu/microcode/amd.c +++ b/arch/x86/kernel/cpu/microcode/amd.c @@ -384,8 +384,9 @@ void load_ucode_amd_ap(unsigned int family) reget: if (!get_builtin_microcode(&cp, family)) { #ifdef CONFIG_BLK_DEV_INITRD - cp = find_cpio_data(ucode_path, (void *)initrd_start, - initrd_end - initrd_start, NULL); + if (!initrd_gone) + cp = find_cpio_data(ucode_path, (void *)initrd_start, + initrd_end - initrd_start, NULL); #endif if (!(cp.data && cp.size)) { /* diff --git a/arch/x86/kernel/cpu/microcode/core.c b/arch/x86/kernel/cpu/microcode/core.c index 2af69d27da62..73102d932760 100644 --- a/arch/x86/kernel/cpu/microcode/core.c +++ b/arch/x86/kernel/cpu/microcode/core.c @@ -46,6 +46,8 @@ static struct microcode_ops *microcode_ops; static bool dis_ucode_ldr = true; +bool initrd_gone; + LIST_HEAD(microcode_cache); /* @@ -190,21 +192,24 @@ void load_ucode_ap(void) static int __init save_microcode_in_initrd(void) { struct cpuinfo_x86 *c = &boot_cpu_data; + int ret = -EINVAL; switch (c->x86_vendor) { case X86_VENDOR_INTEL: if (c->x86 >= 6) - return save_microcode_in_initrd_intel(); + ret = save_microcode_in_initrd_intel(); break; case X86_VENDOR_AMD: if (c->x86 >= 0x10) - return save_microcode_in_initrd_amd(c->x86); + ret = save_microcode_in_initrd_amd(c->x86); break; default: break; } - return -EINVAL; + initrd_gone = true; + + return ret; } struct cpio_data find_microcode_in_initrd(const char *path, bool use_pa) @@ -247,9 +252,16 @@ struct cpio_data find_microcode_in_initrd(const char *path, bool use_pa) * has the virtual address of the beginning of the initrd. It also * possibly relocates the ramdisk. In either case, initrd_start contains * the updated address so use that instead. + * + * initrd_gone is for the hotplug case where we've thrown out initrd + * already. */ - if (!use_pa && initrd_start) - start = initrd_start; + if (!use_pa) { + if (initrd_gone) + return (struct cpio_data){ NULL, 0, "" }; + if (initrd_start) + start = initrd_start; + } return find_cpio_data(path, (void *)start, size, NULL); #else /* !CONFIG_BLK_DEV_INITRD */ -- cgit v1.2.3 From 4e5b54f127426c82dc2816340c26d951a5bb3429 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Sun, 18 Dec 2016 14:35:45 +0100 Subject: drm: prevent double-(un)registration for connectors If we're unlucky then the registration from a hotplugged connector might race with the final registration step on driver load. And since MST topology discover is asynchronous that's even somewhat likely. v2: Also update the kerneldoc for @registered! v3: Review from Chris: - Improve kerneldoc for late_register/early_unregister callbacks. - Use mutex_destroy. Reviewed-by: Chris Wilson Cc: Chris Wilson Reviewed-by: Sean Paul Reported-by: Chris Wilson Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/20161218133545.2106-1-daniel.vetter@ffwll.ch (cherry picked from commit e73ab00e9a0f1731f34d0620a9c55f5c30c4ad4e) --- drivers/gpu/drm/drm_connector.c | 20 +++++++++++++++----- include/drm/drm_connector.h | 16 +++++++++++++++- 2 files changed, 30 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c index 5a4526289392..8be60a4576b3 100644 --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -225,6 +225,7 @@ int drm_connector_init(struct drm_device *dev, INIT_LIST_HEAD(&connector->probed_modes); INIT_LIST_HEAD(&connector->modes); + mutex_init(&connector->mutex); connector->edid_blob_ptr = NULL; connector->status = connector_status_unknown; @@ -359,6 +360,8 @@ void drm_connector_cleanup(struct drm_connector *connector) connector->funcs->atomic_destroy_state(connector, connector->state); + mutex_destroy(&connector->mutex); + memset(connector, 0, sizeof(*connector)); } EXPORT_SYMBOL(drm_connector_cleanup); @@ -374,14 +377,15 @@ EXPORT_SYMBOL(drm_connector_cleanup); */ int drm_connector_register(struct drm_connector *connector) { - int ret; + int ret = 0; + mutex_lock(&connector->mutex); if (connector->registered) - return 0; + goto unlock; ret = drm_sysfs_connector_add(connector); if (ret) - return ret; + goto unlock; ret = drm_debugfs_connector_add(connector); if (ret) { @@ -397,12 +401,14 @@ int drm_connector_register(struct drm_connector *connector) drm_mode_object_register(connector->dev, &connector->base); connector->registered = true; - return 0; + goto unlock; err_debugfs: drm_debugfs_connector_remove(connector); err_sysfs: drm_sysfs_connector_remove(connector); +unlock: + mutex_unlock(&connector->mutex); return ret; } EXPORT_SYMBOL(drm_connector_register); @@ -415,8 +421,11 @@ EXPORT_SYMBOL(drm_connector_register); */ void drm_connector_unregister(struct drm_connector *connector) { - if (!connector->registered) + mutex_lock(&connector->mutex); + if (!connector->registered) { + mutex_unlock(&connector->mutex); return; + } if (connector->funcs->early_unregister) connector->funcs->early_unregister(connector); @@ -425,6 +434,7 @@ void drm_connector_unregister(struct drm_connector *connector) drm_debugfs_connector_remove(connector); connector->registered = false; + mutex_unlock(&connector->mutex); } EXPORT_SYMBOL(drm_connector_unregister); diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h index a9b95246e26e..045a97cbeba2 100644 --- a/include/drm/drm_connector.h +++ b/include/drm/drm_connector.h @@ -381,6 +381,8 @@ struct drm_connector_funcs { * core drm connector interfaces. Everything added from this callback * should be unregistered in the early_unregister callback. * + * This is called while holding drm_connector->mutex. + * * Returns: * * 0 on success, or a negative error code on failure. @@ -395,6 +397,8 @@ struct drm_connector_funcs { * late_register(). It is called from drm_connector_unregister(), * early in the driver unload sequence to disable userspace access * before data structures are torndown. + * + * This is called while holding drm_connector->mutex. */ void (*early_unregister)(struct drm_connector *connector); @@ -559,7 +563,6 @@ struct drm_cmdline_mode { * @interlace_allowed: can this connector handle interlaced modes? * @doublescan_allowed: can this connector handle doublescan? * @stereo_allowed: can this connector handle stereo modes? - * @registered: is this connector exposed (registered) with userspace? * @modes: modes available on this connector (from fill_modes() + user) * @status: one of the drm_connector_status enums (connected, not, or unknown) * @probed_modes: list of modes derived directly from the display @@ -607,6 +610,13 @@ struct drm_connector { char *name; + /** + * @mutex: Lock for general connector state, but currently only protects + * @registered. Most of the connector state is still protected by the + * mutex in &drm_mode_config. + */ + struct mutex mutex; + /** * @index: Compacted connector index, which matches the position inside * the mode_config.list for drivers not supporting hot-add/removing. Can @@ -620,6 +630,10 @@ struct drm_connector { bool interlace_allowed; bool doublescan_allowed; bool stereo_allowed; + /** + * @registered: Is this connector exposed (registered) with userspace? + * Protected by @mutex. + */ bool registered; struct list_head modes; /* list of modes on this connector */ -- cgit v1.2.3 From e6e7b48b295afa5a5ab440de0a94d9ad8b3ce2d0 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Thu, 12 Jan 2017 17:15:56 +0100 Subject: drm: Don't race connector registration I was under the misconception that the sysfs dev stuff can be fully set up, and then registered all in one step with device_add. That's true for properties and property groups, but not for parents and child devices. Those must be fully registered before you can register a child. Add a bit of tracking to make sure that asynchronous mst connector hotplugging gets this right. For consistency we rely upon the implicit barriers of the connector->mutex, which is taken anyway, to ensure that at least either the connector or device registration call will work out. Mildly tested since I can't reliably reproduce this on my mst box here. Reported-by: Dave Hansen Cc: Dave Hansen Acked-by: Chris Wilson Cc: Chris Wilson Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/1484237756-2720-1-git-send-email-daniel.vetter@ffwll.ch --- drivers/gpu/drm/drm_connector.c | 3 +++ drivers/gpu/drm/drm_drv.c | 4 ++++ include/drm/drmP.h | 1 + 3 files changed, 8 insertions(+) diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c index 8be60a4576b3..7a7019ac9388 100644 --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -379,6 +379,9 @@ int drm_connector_register(struct drm_connector *connector) { int ret = 0; + if (!connector->dev->registered) + return 0; + mutex_lock(&connector->mutex); if (connector->registered) goto unlock; diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c index a525751b4559..6594b4088f11 100644 --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -745,6 +745,8 @@ int drm_dev_register(struct drm_device *dev, unsigned long flags) if (ret) goto err_minors; + dev->registered = true; + if (dev->driver->load) { ret = dev->driver->load(dev, flags); if (ret) @@ -785,6 +787,8 @@ void drm_dev_unregister(struct drm_device *dev) drm_lastclose(dev); + dev->registered = false; + if (drm_core_check_feature(dev, DRIVER_MODESET)) drm_modeset_unregister_all(dev); diff --git a/include/drm/drmP.h b/include/drm/drmP.h index 192016e2b518..9c4ee144b5f6 100644 --- a/include/drm/drmP.h +++ b/include/drm/drmP.h @@ -517,6 +517,7 @@ struct drm_device { struct drm_minor *control; /**< Control node */ struct drm_minor *primary; /**< Primary node */ struct drm_minor *render; /**< Render node */ + bool registered; /* currently active master for this device. Protected by master_mutex */ struct drm_master *master; -- cgit v1.2.3 From a06393ed03167771246c4c43192d9c264bc48412 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Wed, 18 Jan 2017 21:30:51 +0100 Subject: can: bcm: fix hrtimer/tasklet termination in bcm op removal When removing a bcm tx operation either a hrtimer or a tasklet might run. As the hrtimer triggers its associated tasklet and vice versa we need to take care to mutually terminate both handlers. Reported-by: Michael Josenhans Signed-off-by: Oliver Hartkopp Tested-by: Michael Josenhans Cc: linux-stable Signed-off-by: Marc Kleine-Budde --- net/can/bcm.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) diff --git a/net/can/bcm.c b/net/can/bcm.c index 5c9407181918..95d13b233c65 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -734,14 +734,23 @@ static struct bcm_op *bcm_find_op(struct list_head *ops, static void bcm_remove_op(struct bcm_op *op) { - hrtimer_cancel(&op->timer); - hrtimer_cancel(&op->thrtimer); - - if (op->tsklet.func) - tasklet_kill(&op->tsklet); + if (op->tsklet.func) { + while (test_bit(TASKLET_STATE_SCHED, &op->tsklet.state) || + test_bit(TASKLET_STATE_RUN, &op->tsklet.state) || + hrtimer_active(&op->timer)) { + hrtimer_cancel(&op->timer); + tasklet_kill(&op->tsklet); + } + } - if (op->thrtsklet.func) - tasklet_kill(&op->thrtsklet); + if (op->thrtsklet.func) { + while (test_bit(TASKLET_STATE_SCHED, &op->thrtsklet.state) || + test_bit(TASKLET_STATE_RUN, &op->thrtsklet.state) || + hrtimer_active(&op->thrtimer)) { + hrtimer_cancel(&op->thrtimer); + tasklet_kill(&op->thrtsklet); + } + } if ((op->frames) && (op->frames != &op->sframe)) kfree(op->frames); -- cgit v1.2.3 From a76a82a3e38c8d3fb6499e3dfaeb0949241ab588 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 26 Jan 2017 16:39:55 +0100 Subject: perf/core: Fix use-after-free bug Dmitry reported a KASAN use-after-free on event->group_leader. It turns out there's a hole in perf_remove_from_context() due to event_function_call() not calling its function when the task associated with the event is already dead. In this case the event will have been detached from the task, but the grouping will have been retained, such that group operations might still work properly while there are live child events etc. This does however mean that we can miss a perf_group_detach() call when the group decomposes, this in turn can then lead to use-after-free. Fix it by explicitly doing the group detach if its still required. Reported-by: Dmitry Vyukov Tested-by: Dmitry Vyukov Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Mathieu Desnoyers Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: stable@vger.kernel.org # v4.5+ Cc: syzkaller Fixes: 63b6da39bb38 ("perf: Fix perf_event_exit_task() race") Link: http://lkml.kernel.org/r/20170126153955.GD6515@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- kernel/events/core.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 110b38a58493..4e1f4c0070ce 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1469,7 +1469,6 @@ ctx_group_list(struct perf_event *event, struct perf_event_context *ctx) static void list_add_event(struct perf_event *event, struct perf_event_context *ctx) { - lockdep_assert_held(&ctx->lock); WARN_ON_ONCE(event->attach_state & PERF_ATTACH_CONTEXT); @@ -1624,6 +1623,8 @@ static void perf_group_attach(struct perf_event *event) { struct perf_event *group_leader = event->group_leader, *pos; + lockdep_assert_held(&event->ctx->lock); + /* * We can have double attach due to group movement in perf_event_open. */ @@ -1697,6 +1698,8 @@ static void perf_group_detach(struct perf_event *event) struct perf_event *sibling, *tmp; struct list_head *list = NULL; + lockdep_assert_held(&event->ctx->lock); + /* * We can have double detach due to exit/hot-unplug + close. */ @@ -1895,9 +1898,29 @@ __perf_remove_from_context(struct perf_event *event, */ static void perf_remove_from_context(struct perf_event *event, unsigned long flags) { - lockdep_assert_held(&event->ctx->mutex); + struct perf_event_context *ctx = event->ctx; + + lockdep_assert_held(&ctx->mutex); event_function_call(event, __perf_remove_from_context, (void *)flags); + + /* + * The above event_function_call() can NO-OP when it hits + * TASK_TOMBSTONE. In that case we must already have been detached + * from the context (by perf_event_exit_event()) but the grouping + * might still be in-tact. + */ + WARN_ON_ONCE(event->attach_state & PERF_ATTACH_CONTEXT); + if ((flags & DETACH_GROUP) && + (event->attach_state & PERF_ATTACH_GROUP)) { + /* + * Since in that case we cannot possibly be scheduled, simply + * detach now. + */ + raw_spin_lock_irq(&ctx->lock); + perf_group_detach(event); + raw_spin_unlock_irq(&ctx->lock); + } } /* -- cgit v1.2.3 From 0b3589be9b98994ce3d5aeca52445d1f5627c4ba Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 26 Jan 2017 23:15:08 +0100 Subject: perf/core: Fix PERF_RECORD_MMAP2 prot/flags for anonymous memory Andres reported that MMAP2 records for anonymous memory always have their protection field 0. Turns out, someone daft put the prot/flags generation code in the file branch, leaving them unset for anonymous memory. Reported-by: Andres Freund Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Don Zickus Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Stephane Eranian Cc: Thomas Gleixner Cc: acme@kernel.org Cc: anton@ozlabs.org Cc: namhyung@kernel.org Cc: stable@vger.kernel.org # v3.16+ Fixes: f972eb63b100 ("perf: Pass protection and flags bits through mmap2 interface") Link: http://lkml.kernel.org/r/20170126221508.GF6536@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- kernel/events/core.c | 42 +++++++++++++++++++++--------------------- 1 file changed, 21 insertions(+), 21 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 4e1f4c0070ce..e5aaa806702d 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -6632,6 +6632,27 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event) char *buf = NULL; char *name; + if (vma->vm_flags & VM_READ) + prot |= PROT_READ; + if (vma->vm_flags & VM_WRITE) + prot |= PROT_WRITE; + if (vma->vm_flags & VM_EXEC) + prot |= PROT_EXEC; + + if (vma->vm_flags & VM_MAYSHARE) + flags = MAP_SHARED; + else + flags = MAP_PRIVATE; + + if (vma->vm_flags & VM_DENYWRITE) + flags |= MAP_DENYWRITE; + if (vma->vm_flags & VM_MAYEXEC) + flags |= MAP_EXECUTABLE; + if (vma->vm_flags & VM_LOCKED) + flags |= MAP_LOCKED; + if (vma->vm_flags & VM_HUGETLB) + flags |= MAP_HUGETLB; + if (file) { struct inode *inode; dev_t dev; @@ -6658,27 +6679,6 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event) maj = MAJOR(dev); min = MINOR(dev); - if (vma->vm_flags & VM_READ) - prot |= PROT_READ; - if (vma->vm_flags & VM_WRITE) - prot |= PROT_WRITE; - if (vma->vm_flags & VM_EXEC) - prot |= PROT_EXEC; - - if (vma->vm_flags & VM_MAYSHARE) - flags = MAP_SHARED; - else - flags = MAP_PRIVATE; - - if (vma->vm_flags & VM_DENYWRITE) - flags |= MAP_DENYWRITE; - if (vma->vm_flags & VM_MAYEXEC) - flags |= MAP_EXECUTABLE; - if (vma->vm_flags & VM_LOCKED) - flags |= MAP_LOCKED; - if (vma->vm_flags & VM_HUGETLB) - flags |= MAP_HUGETLB; - goto got_name; } else { if (vma->vm_ops && vma->vm_ops->name) { -- cgit v1.2.3 From 08d85f3ea99f1eeafc4e8507936190e86a16ee8c Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 17 Jan 2017 16:00:48 +0000 Subject: irqdomain: Avoid activating interrupts more than once Since commit f3b0946d629c ("genirq/msi: Make sure PCI MSIs are activated early"), we can end-up activating a PCI/MSI twice (once at allocation time, and once at startup time). This is normally of no consequences, except that there is some HW out there that may misbehave if activate is used more than once (the GICv3 ITS, for example, uses the activate callback to issue the MAPVI command, and the architecture spec says that "If there is an existing mapping for the EventID-DeviceID combination, behavior is UNPREDICTABLE"). While this could be worked around in each individual driver, it may make more sense to tackle the issue at the core level. In order to avoid getting in that situation, let's have a per-interrupt flag to remember if we have already activated that interrupt or not. Fixes: f3b0946d629c ("genirq/msi: Make sure PCI MSIs are activated early") Reported-and-tested-by: Andre Przywara Signed-off-by: Marc Zyngier Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1484668848-24361-1-git-send-email-marc.zyngier@arm.com Signed-off-by: Thomas Gleixner --- include/linux/irq.h | 17 +++++++++++++++++ kernel/irq/irqdomain.c | 44 ++++++++++++++++++++++++++++++-------------- 2 files changed, 47 insertions(+), 14 deletions(-) diff --git a/include/linux/irq.h b/include/linux/irq.h index e79875574b39..39e3254e5769 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -184,6 +184,7 @@ struct irq_data { * * IRQD_TRIGGER_MASK - Mask for the trigger type bits * IRQD_SETAFFINITY_PENDING - Affinity setting is pending + * IRQD_ACTIVATED - Interrupt has already been activated * IRQD_NO_BALANCING - Balancing disabled for this IRQ * IRQD_PER_CPU - Interrupt is per cpu * IRQD_AFFINITY_SET - Interrupt affinity was set @@ -202,6 +203,7 @@ struct irq_data { enum { IRQD_TRIGGER_MASK = 0xf, IRQD_SETAFFINITY_PENDING = (1 << 8), + IRQD_ACTIVATED = (1 << 9), IRQD_NO_BALANCING = (1 << 10), IRQD_PER_CPU = (1 << 11), IRQD_AFFINITY_SET = (1 << 12), @@ -312,6 +314,21 @@ static inline bool irqd_affinity_is_managed(struct irq_data *d) return __irqd_to_state(d) & IRQD_AFFINITY_MANAGED; } +static inline bool irqd_is_activated(struct irq_data *d) +{ + return __irqd_to_state(d) & IRQD_ACTIVATED; +} + +static inline void irqd_set_activated(struct irq_data *d) +{ + __irqd_to_state(d) |= IRQD_ACTIVATED; +} + +static inline void irqd_clr_activated(struct irq_data *d) +{ + __irqd_to_state(d) &= ~IRQD_ACTIVATED; +} + #undef __irqd_to_state static inline irq_hw_number_t irqd_to_hwirq(struct irq_data *d) diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 8c0a0ae43521..b59e6768c5e9 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -1346,6 +1346,30 @@ void irq_domain_free_irqs_parent(struct irq_domain *domain, } EXPORT_SYMBOL_GPL(irq_domain_free_irqs_parent); +static void __irq_domain_activate_irq(struct irq_data *irq_data) +{ + if (irq_data && irq_data->domain) { + struct irq_domain *domain = irq_data->domain; + + if (irq_data->parent_data) + __irq_domain_activate_irq(irq_data->parent_data); + if (domain->ops->activate) + domain->ops->activate(domain, irq_data); + } +} + +static void __irq_domain_deactivate_irq(struct irq_data *irq_data) +{ + if (irq_data && irq_data->domain) { + struct irq_domain *domain = irq_data->domain; + + if (domain->ops->deactivate) + domain->ops->deactivate(domain, irq_data); + if (irq_data->parent_data) + __irq_domain_deactivate_irq(irq_data->parent_data); + } +} + /** * irq_domain_activate_irq - Call domain_ops->activate recursively to activate * interrupt @@ -1356,13 +1380,9 @@ EXPORT_SYMBOL_GPL(irq_domain_free_irqs_parent); */ void irq_domain_activate_irq(struct irq_data *irq_data) { - if (irq_data && irq_data->domain) { - struct irq_domain *domain = irq_data->domain; - - if (irq_data->parent_data) - irq_domain_activate_irq(irq_data->parent_data); - if (domain->ops->activate) - domain->ops->activate(domain, irq_data); + if (!irqd_is_activated(irq_data)) { + __irq_domain_activate_irq(irq_data); + irqd_set_activated(irq_data); } } @@ -1376,13 +1396,9 @@ void irq_domain_activate_irq(struct irq_data *irq_data) */ void irq_domain_deactivate_irq(struct irq_data *irq_data) { - if (irq_data && irq_data->domain) { - struct irq_domain *domain = irq_data->domain; - - if (domain->ops->deactivate) - domain->ops->deactivate(domain, irq_data); - if (irq_data->parent_data) - irq_domain_deactivate_irq(irq_data->parent_data); + if (irqd_is_activated(irq_data)) { + __irq_domain_deactivate_irq(irq_data); + irqd_clr_activated(irq_data); } } -- cgit v1.2.3 From 827e1579e1d5cb66e340e7be1944b825b542bbdf Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Thu, 26 Jan 2017 19:24:07 +0200 Subject: pinctrl: baytrail: Rectify debounce support (part 2) The commit 04ff5a095d66 ("pinctrl: baytrail: Rectify debounce support") almost fixes the logic of debuonce but missed couple of things, i.e. typo in mask when disabling debounce and lack of enabling it back. This patch addresses above issues. Reported-by: Jean Delvare Fixes: 04ff5a095d66 ("pinctrl: baytrail: Rectify debounce support") Signed-off-by: Andy Shevchenko Reviewed-by: Jean Delvare Acked-by: Mika Westerberg Signed-off-by: Linus Walleij --- drivers/pinctrl/intel/pinctrl-baytrail.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/pinctrl/intel/pinctrl-baytrail.c b/drivers/pinctrl/intel/pinctrl-baytrail.c index c123488266ce..3eb7c2bde87e 100644 --- a/drivers/pinctrl/intel/pinctrl-baytrail.c +++ b/drivers/pinctrl/intel/pinctrl-baytrail.c @@ -1243,10 +1243,12 @@ static int byt_pin_config_set(struct pinctrl_dev *pctl_dev, debounce = readl(db_reg); debounce &= ~BYT_DEBOUNCE_PULSE_MASK; + if (arg) + conf |= BYT_DEBOUNCE_EN; + else + conf &= ~BYT_DEBOUNCE_EN; + switch (arg) { - case 0: - conf &= BYT_DEBOUNCE_EN; - break; case 375: debounce |= BYT_DEBOUNCE_PULSE_375US; break; @@ -1269,7 +1271,9 @@ static int byt_pin_config_set(struct pinctrl_dev *pctl_dev, debounce |= BYT_DEBOUNCE_PULSE_24MS; break; default: - ret = -EINVAL; + if (arg) + ret = -EINVAL; + break; } if (!ret) -- cgit v1.2.3 From 1b89970d81bbd52720fc64a3fe9572ee33588363 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Thu, 26 Jan 2017 19:24:08 +0200 Subject: pinctrl: baytrail: Debounce register is one per community Debounce value is set globally per community. Otherwise user will easily get a kernel crash when they start using the feature: BUG: unable to handle kernel paging request at ffffc900003be000 IP: byt_gpio_dbg_show+0xa9/0x430 Make it clear in byt_gpio_reg(). Note that this fix just prevents kernel to crash, but doesn't make any difference to the existing logic. It means the last caller will win the trade and debounce value will be configured accordingly. The actual logic fix needs to be thought about and it's not as important as crash fix. That's why the latter goes separately and right now. Fixes: 658b476c742f ("pinctrl: baytrail: Add debounce configuration") Cc: Cristina Ciocan Signed-off-by: Andy Shevchenko Reviewed-by: Jean Delvare Acked-by: Mika Westerberg Signed-off-by: Linus Walleij --- drivers/pinctrl/intel/pinctrl-baytrail.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/drivers/pinctrl/intel/pinctrl-baytrail.c b/drivers/pinctrl/intel/pinctrl-baytrail.c index 3eb7c2bde87e..05ba052ff6b9 100644 --- a/drivers/pinctrl/intel/pinctrl-baytrail.c +++ b/drivers/pinctrl/intel/pinctrl-baytrail.c @@ -731,16 +731,23 @@ static void __iomem *byt_gpio_reg(struct byt_gpio *vg, unsigned int offset, int reg) { struct byt_community *comm = byt_get_community(vg, offset); - u32 reg_offset = 0; + u32 reg_offset; if (!comm) return NULL; offset -= comm->pin_base; - if (reg == BYT_INT_STAT_REG) + switch (reg) { + case BYT_INT_STAT_REG: reg_offset = (offset / 32) * 4; - else + break; + case BYT_DEBOUNCE_REG: + reg_offset = 0; + break; + default: reg_offset = comm->pad_map[offset] * 16; + break; + } return comm->reg_base + reg_offset + reg; } -- cgit v1.2.3 From cdca06e4e85974d8a3503ab15709dbbaf90d3dd1 Mon Sep 17 00:00:00 2001 From: Alexander Stein Date: Mon, 30 Jan 2017 12:35:28 +0100 Subject: pinctrl: baytrail: Add missing spinlock usage in byt_gpio_irq_handler According to VLI64 Intel Atom E3800 Specification Update (#329901) concurrent read accesses may result in returning 0xffffffff and write accesses may be dropped silently. To workaround all accesses must be protected by locks. Cc: stable@vger.kernel.org Signed-off-by: Alexander Stein Acked-by: Mika Westerberg Signed-off-by: Linus Walleij --- drivers/pinctrl/intel/pinctrl-baytrail.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/pinctrl/intel/pinctrl-baytrail.c b/drivers/pinctrl/intel/pinctrl-baytrail.c index 05ba052ff6b9..d94aef17348b 100644 --- a/drivers/pinctrl/intel/pinctrl-baytrail.c +++ b/drivers/pinctrl/intel/pinctrl-baytrail.c @@ -1623,7 +1623,9 @@ static void byt_gpio_irq_handler(struct irq_desc *desc) continue; } + raw_spin_lock(&vg->lock); pending = readl(reg); + raw_spin_unlock(&vg->lock); for_each_set_bit(pin, &pending, 32) { virq = irq_find_mapping(vg->chip.irqdomain, base + pin); generic_handle_irq(virq); -- cgit v1.2.3 From 94842b4fc4d6b1691cfc86c6f5251f299d27f4ba Mon Sep 17 00:00:00 2001 From: Pavel Belous Date: Sat, 28 Jan 2017 22:53:28 +0300 Subject: net: ethtool: add support for 2500BaseT and 5000BaseT link modes This patch introduce support for 2500BaseT and 5000BaseT link modes. These modes are included in the new IEEE 802.3bz standard. Signed-off-by: Pavel Belous Signed-off-by: David S. Miller --- include/uapi/linux/ethtool.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index f0db7788f887..3dc91a46e8b8 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -1384,6 +1384,8 @@ enum ethtool_link_mode_bit_indices { ETHTOOL_LINK_MODE_10000baseLR_Full_BIT = 44, ETHTOOL_LINK_MODE_10000baseLRM_Full_BIT = 45, ETHTOOL_LINK_MODE_10000baseER_Full_BIT = 46, + ETHTOOL_LINK_MODE_2500baseT_Full_BIT = 47, + ETHTOOL_LINK_MODE_5000baseT_Full_BIT = 48, /* Last allowed bit for __ETHTOOL_LINK_MODE_LEGACY_MASK is bit @@ -1393,7 +1395,7 @@ enum ethtool_link_mode_bit_indices { */ __ETHTOOL_LINK_MODE_LAST - = ETHTOOL_LINK_MODE_10000baseER_Full_BIT, + = ETHTOOL_LINK_MODE_5000baseT_Full_BIT, }; #define __ETHTOOL_LINK_MODE_LEGACY_MASK(base_name) \ -- cgit v1.2.3 From 4af0e5bb95ee3ba5ea4bd7dbb94e1648a5279cc9 Mon Sep 17 00:00:00 2001 From: Arseny Solokha Date: Sun, 29 Jan 2017 19:52:20 +0700 Subject: gianfar: synchronize DMA API usage by free_skb_rx_queue w/ gfar_new_page In spite of switching to paged allocation of Rx buffers, the driver still called dma_unmap_single() in the Rx queues tear-down path. The DMA region unmapping code in free_skb_rx_queue() basically predates the introduction of paged allocation to the driver. While being refactored, it apparently hasn't reflected the change in the DMA API usage by its counterpart gfar_new_page(). As a result, setting an interface to the DOWN state now yields the following: # ip link set eth2 down fsl-gianfar ffe24000.ethernet: DMA-API: device driver frees DMA memory with wrong function [device address=0x000000001ecd0000] [size=40] ------------[ cut here ]------------ WARNING: CPU: 1 PID: 189 at lib/dma-debug.c:1123 check_unmap+0x8e0/0xa28 CPU: 1 PID: 189 Comm: ip Tainted: G O 4.9.5 #1 task: dee73400 task.stack: dede2000 NIP: c02101e8 LR: c02101e8 CTR: c0260d74 REGS: dede3bb0 TRAP: 0700 Tainted: G O (4.9.5) MSR: 00021000 CR: 28002222 XER: 00000000 GPR00: c02101e8 dede3c60 dee73400 000000b6 dfbd033c dfbd36c4 1f622000 dede2000 GPR08: 00000007 c05b1634 1f622000 00000000 22002484 100a9904 00000000 00000000 GPR16: 00000000 db4c849c 00000002 db4c8480 00000001 df142240 db4c84bc 00000000 GPR24: c0706148 c0700000 00029000 c07552e8 c07323b4 dede3cb8 c07605e0 db535540 NIP [c02101e8] check_unmap+0x8e0/0xa28 LR [c02101e8] check_unmap+0x8e0/0xa28 Call Trace: [dede3c60] [c02101e8] check_unmap+0x8e0/0xa28 (unreliable) [dede3cb0] [c02103b8] debug_dma_unmap_page+0x88/0x9c [dede3d30] [c02dffbc] free_skb_resources+0x2c4/0x404 [dede3d80] [c02e39b4] gfar_close+0x24/0xc8 [dede3da0] [c0361550] __dev_close_many+0xa0/0xf8 [dede3dd0] [c03616f0] __dev_close+0x2c/0x4c [dede3df0] [c036b1b8] __dev_change_flags+0xa0/0x174 [dede3e10] [c036b2ac] dev_change_flags+0x20/0x60 [dede3e30] [c03e130c] devinet_ioctl+0x540/0x824 [dede3e90] [c0347dcc] sock_ioctl+0x134/0x298 [dede3eb0] [c0111814] do_vfs_ioctl+0xac/0x854 [dede3f20] [c0111ffc] SyS_ioctl+0x40/0x74 [dede3f40] [c000f290] ret_from_syscall+0x0/0x3c --- interrupt: c01 at 0xff45da0 LR = 0xff45cd0 Instruction dump: 811d001c 7c66482e 813d0020 9061000c 807f000c 5463103a 7cc6182e 3c60c052 386309ac 90c10008 4cc63182 4826b845 <0fe00000> 4bfffa60 3c80c052 388402c4 ---[ end trace 695ae6d7ac1d0c47 ]--- Mapped at: [] gfar_alloc_rx_buffs+0x178/0x248 [] startup_gfar+0x368/0x570 [] __dev_open+0xdc/0x150 [] __dev_change_flags+0xa0/0x174 [] dev_change_flags+0x20/0x60 Even though the issue was discovered in 4.9 kernel, the code in question is identical in the current net and net-next trees. Fixes: 75354148ce69 ("gianfar: Add paged allocation and Rx S/G") Signed-off-by: Arseny Solokha Acked-by: Claudiu Manoil Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/gianfar.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index c1b671667920..957bfc220978 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -2010,8 +2010,8 @@ static void free_skb_rx_queue(struct gfar_priv_rx_q *rx_queue) if (!rxb->page) continue; - dma_unmap_single(rx_queue->dev, rxb->dma, - PAGE_SIZE, DMA_FROM_DEVICE); + dma_unmap_page(rx_queue->dev, rxb->dma, + PAGE_SIZE, DMA_FROM_DEVICE); __free_page(rxb->page); rxb->page = NULL; -- cgit v1.2.3 From d585df1c5ccf995fcee910705ad7a9cdd11d4152 Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Mon, 30 Jan 2017 15:11:45 +0200 Subject: net/mlx4_core: Avoid command timeouts during VF driver device shutdown Some Hypervisors detach VFs from VMs by instantly causing an FLR event to be generated for a VF. In the mlx4 case, this will cause that VF's comm channel to be disabled before the VM has an opportunity to invoke the VF device's "shutdown" method. The result is that the VF driver on the VM will experience a command timeout during the shutdown process when the Hypervisor does not deliver a command-completion event to the VM. To avoid FW command timeouts on the VM when the driver's shutdown method is invoked, we detect the absence of the VF's comm channel at the very start of the shutdown process. If the comm-channel has already been disabled, we cause all FW commands during the device shutdown process to immediately return success (and thus avoid all command timeouts). Signed-off-by: Jack Morgenstein Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/catas.c | 2 +- drivers/net/ethernet/mellanox/mlx4/intf.c | 12 ++++++++++++ drivers/net/ethernet/mellanox/mlx4/mlx4.h | 1 + 3 files changed, 14 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/catas.c b/drivers/net/ethernet/mellanox/mlx4/catas.c index c7e939945259..53daa6ca5d83 100644 --- a/drivers/net/ethernet/mellanox/mlx4/catas.c +++ b/drivers/net/ethernet/mellanox/mlx4/catas.c @@ -158,7 +158,7 @@ static int mlx4_reset_slave(struct mlx4_dev *dev) return -ETIMEDOUT; } -static int mlx4_comm_internal_err(u32 slave_read) +int mlx4_comm_internal_err(u32 slave_read) { return (u32)COMM_CHAN_EVENT_INTERNAL_ERR == (slave_read & (u32)COMM_CHAN_EVENT_INTERNAL_ERR) ? 1 : 0; diff --git a/drivers/net/ethernet/mellanox/mlx4/intf.c b/drivers/net/ethernet/mellanox/mlx4/intf.c index 0e8b7c44931f..8258d08acd8c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/intf.c +++ b/drivers/net/ethernet/mellanox/mlx4/intf.c @@ -222,6 +222,18 @@ void mlx4_unregister_device(struct mlx4_dev *dev) return; mlx4_stop_catas_poll(dev); + if (dev->persist->interface_state & MLX4_INTERFACE_STATE_DELETION && + mlx4_is_slave(dev)) { + /* In mlx4_remove_one on a VF */ + u32 slave_read = + swab32(readl(&mlx4_priv(dev)->mfunc.comm->slave_read)); + + if (mlx4_comm_internal_err(slave_read)) { + mlx4_dbg(dev, "%s: comm channel is down, entering error state.\n", + __func__); + mlx4_enter_error_state(dev->persist); + } + } mutex_lock(&intf_mutex); list_for_each_entry(intf, &intf_list, list) diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h index 88ee7d8a5923..086920b615af 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h @@ -1220,6 +1220,7 @@ void mlx4_qp_event(struct mlx4_dev *dev, u32 qpn, int event_type); void mlx4_srq_event(struct mlx4_dev *dev, u32 srqn, int event_type); void mlx4_enter_error_state(struct mlx4_dev_persistent *persist); +int mlx4_comm_internal_err(u32 slave_read); int mlx4_SENSE_PORT(struct mlx4_dev *dev, int port, enum mlx4_port_type *type); -- cgit v1.2.3 From 2b89ed65a6f201a6a4f0450ad289aa4bf491608c Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Sun, 29 Jan 2017 22:52:53 -0500 Subject: ipv6: Paritially checksum full MTU frames IPv6 will mark data that is smaller that mtu - headersize as CHECKSUM_PARTIAL, but if the data will completely fill the mtu, the packet checksum will be computed in software instead. Extend the conditional to include the data that fills the mtu as well. Signed-off-by: Vladislav Yasevich Signed-off-by: David S. Miller --- net/ipv6/ip6_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 2c0df09e9036..b6a94ff0bbd0 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -1344,7 +1344,7 @@ emsgsize: */ if (transhdrlen && sk->sk_protocol == IPPROTO_UDP && headersize == sizeof(struct ipv6hdr) && - length < mtu - headersize && + length <= mtu - headersize && !(flags & MSG_MORE) && rt->dst.dev->features & (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM)) csummode = CHECKSUM_PARTIAL; -- cgit v1.2.3 From 040587af31228d82c52267f717c9fcdb65f36335 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Mon, 30 Jan 2017 16:19:02 +0100 Subject: net/sched: cls_flower: Correct matching on ICMPv6 code When matching on the ICMPv6 code ICMPV6_CODE rather than ICMPV4_CODE attributes should be used. This corrects what appears to be a typo. Sample usage: tc qdisc add dev eth0 ingress tc filter add dev eth0 protocol ipv6 parent ffff: flower \ indev eth0 ip_proto icmpv6 type 128 code 0 action drop Without this change the code parameter above is effectively ignored. Fixes: 7b684884fbfa ("net/sched: cls_flower: Support matching on ICMP type and code") Signed-off-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/cls_flower.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 970db7a41684..5752789acc13 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -568,9 +568,9 @@ static int fl_set_key(struct net *net, struct nlattr **tb, &mask->icmp.type, TCA_FLOWER_KEY_ICMPV6_TYPE_MASK, sizeof(key->icmp.type)); - fl_set_key_val(tb, &key->icmp.code, TCA_FLOWER_KEY_ICMPV4_CODE, + fl_set_key_val(tb, &key->icmp.code, TCA_FLOWER_KEY_ICMPV6_CODE, &mask->icmp.code, - TCA_FLOWER_KEY_ICMPV4_CODE_MASK, + TCA_FLOWER_KEY_ICMPV6_CODE_MASK, sizeof(key->icmp.code)); } -- cgit v1.2.3 From 7a7dc961a28b965a0d0303c2e989df17b411708b Mon Sep 17 00:00:00 2001 From: Liam R. Howlett Date: Tue, 17 Jan 2017 10:59:02 -0500 Subject: sparc64: Zero pages on allocation for mondo and error queues. Error queues use a non-zero first word to detect if the queues are full. Using pages that have not been zeroed may result in false positive overflow events. These queues are set up once during boot so zeroing all mondo and error queue pages is safe. Note that the false positive overflow does not always occur because the page allocation for these queues is so early in the boot cycle that higher number CPUs get fresh pages. It is only when traps are serviced with lower number CPUs who were given already used pages that this issue is exposed. Signed-off-by: Liam R. Howlett Signed-off-by: David S. Miller --- arch/sparc/kernel/irq_64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/sparc/kernel/irq_64.c b/arch/sparc/kernel/irq_64.c index 34a7930b76ef..baed4cdeda75 100644 --- a/arch/sparc/kernel/irq_64.c +++ b/arch/sparc/kernel/irq_64.c @@ -1021,7 +1021,7 @@ static void __init alloc_one_queue(unsigned long *pa_ptr, unsigned long qmask) unsigned long order = get_order(size); unsigned long p; - p = __get_free_pages(GFP_KERNEL, order); + p = __get_free_pages(GFP_KERNEL | __GFP_ZERO, order); if (!p) { prom_printf("SUN4V: Error, cannot allocate queue.\n"); prom_halt(); -- cgit v1.2.3 From 047487241ff59374fded8c477f21453681f5995c Mon Sep 17 00:00:00 2001 From: Liam R. Howlett Date: Tue, 17 Jan 2017 10:59:03 -0500 Subject: sparc64: Handle PIO & MEM non-resumable errors. User processes trying to access an invalid memory address via PIO will receive a SIGBUS signal instead of causing a panic. Memory errors will receive a SIGKILL since a SIGBUS may result in a coredump which may attempt to repeat the faulting access. Signed-off-by: Liam R. Howlett Signed-off-by: David S. Miller --- arch/sparc/kernel/traps_64.c | 73 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 73 insertions(+) diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c index 496fa926e1e0..d44fb806bbd7 100644 --- a/arch/sparc/kernel/traps_64.c +++ b/arch/sparc/kernel/traps_64.c @@ -2051,6 +2051,73 @@ void sun4v_resum_overflow(struct pt_regs *regs) atomic_inc(&sun4v_resum_oflow_cnt); } +/* Given a set of registers, get the virtual addressi that was being accessed + * by the faulting instructions at tpc. + */ +static unsigned long sun4v_get_vaddr(struct pt_regs *regs) +{ + unsigned int insn; + + if (!copy_from_user(&insn, (void __user *)regs->tpc, 4)) { + return compute_effective_address(regs, insn, + (insn >> 25) & 0x1f); + } + return 0; +} + +/* Attempt to handle non-resumable errors generated from userspace. + * Returns true if the signal was handled, false otherwise. + */ +bool sun4v_nonresum_error_user_handled(struct pt_regs *regs, + struct sun4v_error_entry *ent) { + + unsigned int attrs = ent->err_attrs; + + if (attrs & SUN4V_ERR_ATTRS_MEMORY) { + unsigned long addr = ent->err_raddr; + siginfo_t info; + + if (addr == ~(u64)0) { + /* This seems highly unlikely to ever occur */ + pr_emerg("SUN4V NON-RECOVERABLE ERROR: Memory error detected in unknown location!\n"); + } else { + unsigned long page_cnt = DIV_ROUND_UP(ent->err_size, + PAGE_SIZE); + + /* Break the unfortunate news. */ + pr_emerg("SUN4V NON-RECOVERABLE ERROR: Memory failed at %016lX\n", + addr); + pr_emerg("SUN4V NON-RECOVERABLE ERROR: Claiming %lu ages.\n", + page_cnt); + + while (page_cnt-- > 0) { + if (pfn_valid(addr >> PAGE_SHIFT)) + get_page(pfn_to_page(addr >> PAGE_SHIFT)); + addr += PAGE_SIZE; + } + } + info.si_signo = SIGKILL; + info.si_errno = 0; + info.si_trapno = 0; + force_sig_info(info.si_signo, &info, current); + + return true; + } + if (attrs & SUN4V_ERR_ATTRS_PIO) { + siginfo_t info; + + info.si_signo = SIGBUS; + info.si_code = BUS_ADRERR; + info.si_addr = (void __user *)sun4v_get_vaddr(regs); + force_sig_info(info.si_signo, &info, current); + + return true; + } + + /* Default to doing nothing */ + return false; +} + /* We run with %pil set to PIL_NORMAL_MAX and PSTATE_IE enabled in %pstate. * Log the event, clear the first word of the entry, and die. */ @@ -2075,6 +2142,12 @@ void sun4v_nonresum_error(struct pt_regs *regs, unsigned long offset) put_cpu(); + if (!(regs->tstate & TSTATE_PRIV) && + sun4v_nonresum_error_user_handled(regs, &local_copy)) { + /* DON'T PANIC: This userspace error was handled. */ + return; + } + #ifdef CONFIG_PCI /* Check for the special PCI poke sequence. */ if (pci_poke_in_progress && pci_poke_cpu == cpu) { -- cgit v1.2.3 From 05e0be7c900797e9164976a6014d534ce3035909 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Tue, 31 Jan 2017 00:47:30 -0800 Subject: Input: synaptics-rmi4 - fix reversed conditions in enable/disable_irq_wake These tests are reversed. A warning should be displayed if an error is returned, not on success. Signed-off-by: Christophe JAILLET Reviewed-by: Benjamin Tissoires Signed-off-by: Dmitry Torokhov --- drivers/input/rmi4/rmi_driver.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/input/rmi4/rmi_driver.c b/drivers/input/rmi4/rmi_driver.c index 11447ab1055c..bf5c36e229ba 100644 --- a/drivers/input/rmi4/rmi_driver.c +++ b/drivers/input/rmi4/rmi_driver.c @@ -901,7 +901,7 @@ void rmi_enable_irq(struct rmi_device *rmi_dev, bool clear_wake) data->enabled = true; if (clear_wake && device_may_wakeup(rmi_dev->xport->dev)) { retval = disable_irq_wake(irq); - if (!retval) + if (retval) dev_warn(&rmi_dev->dev, "Failed to disable irq for wake: %d\n", retval); @@ -936,7 +936,7 @@ void rmi_disable_irq(struct rmi_device *rmi_dev, bool enable_wake) disable_irq(irq); if (enable_wake && device_may_wakeup(rmi_dev->xport->dev)) { retval = enable_irq_wake(irq); - if (!retval) + if (retval) dev_warn(&rmi_dev->dev, "Failed to enable irq for wake: %d\n", retval); -- cgit v1.2.3 From 433e19cf33d34bb6751c874a9c00980552fe508c Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Sat, 28 Jan 2017 11:46:02 -0700 Subject: Drivers: hv: vmbus: finally fix hv_need_to_signal_on_read() Commit a389fcfd2cb5 ("Drivers: hv: vmbus: Fix signaling logic in hv_need_to_signal_on_read()") added the proper mb(), but removed the test "prev_write_sz < pending_sz" when making the signal decision. As a result, the guest can signal the host unnecessarily, and then the host can throttle the guest because the host thinks the guest is buggy or malicious; finally the user running stress test can perceive intermittent freeze of the guest. This patch brings back the test, and properly handles the in-place consumption APIs used by NetVSC (see get_next_pkt_raw(), put_pkt_raw() and commit_rd_index()). Fixes: a389fcfd2cb5 ("Drivers: hv: vmbus: Fix signaling logic in hv_need_to_signal_on_read()") Signed-off-by: Dexuan Cui Reported-by: Rolf Neugebauer Tested-by: Rolf Neugebauer Cc: "K. Y. Srinivasan" Cc: Haiyang Zhang Cc: Stephen Hemminger Cc: Signed-off-by: K. Y. Srinivasan Signed-off-by: Greg Kroah-Hartman --- drivers/hv/ring_buffer.c | 1 + drivers/net/hyperv/netvsc.c | 6 ++++++ include/linux/hyperv.h | 32 ++++++++++++++++++++++++++++++-- 3 files changed, 37 insertions(+), 2 deletions(-) diff --git a/drivers/hv/ring_buffer.c b/drivers/hv/ring_buffer.c index cd49cb17eb7f..308dbda700eb 100644 --- a/drivers/hv/ring_buffer.c +++ b/drivers/hv/ring_buffer.c @@ -383,6 +383,7 @@ int hv_ringbuffer_read(struct vmbus_channel *channel, return ret; } + init_cached_read_index(channel); next_read_location = hv_get_next_read_location(inring_info); next_read_location = hv_copyfrom_ringbuffer(inring_info, &desc, sizeof(desc), diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index 5a1cc089acb7..86e5749226ef 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -1295,6 +1295,9 @@ void netvsc_channel_cb(void *context) ndev = hv_get_drvdata(device); buffer = get_per_channel_state(channel); + /* commit_rd_index() -> hv_signal_on_read() needs this. */ + init_cached_read_index(channel); + do { desc = get_next_pkt_raw(channel); if (desc != NULL) { @@ -1347,6 +1350,9 @@ void netvsc_channel_cb(void *context) bufferlen = bytes_recvd; } + + init_cached_read_index(channel); + } while (1); if (bufferlen > NETVSC_PACKET_SIZE) diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h index 42fe43fb0c80..183efde54269 100644 --- a/include/linux/hyperv.h +++ b/include/linux/hyperv.h @@ -128,6 +128,7 @@ struct hv_ring_buffer_info { u32 ring_data_startoffset; u32 priv_write_index; u32 priv_read_index; + u32 cached_read_index; }; /* @@ -180,6 +181,19 @@ static inline u32 hv_get_bytes_to_write(struct hv_ring_buffer_info *rbi) return write; } +static inline u32 hv_get_cached_bytes_to_write( + const struct hv_ring_buffer_info *rbi) +{ + u32 read_loc, write_loc, dsize, write; + + dsize = rbi->ring_datasize; + read_loc = rbi->cached_read_index; + write_loc = rbi->ring_buffer->write_index; + + write = write_loc >= read_loc ? dsize - (write_loc - read_loc) : + read_loc - write_loc; + return write; +} /* * VMBUS version is 32 bit entity broken up into * two 16 bit quantities: major_number. minor_number. @@ -1488,7 +1502,7 @@ hv_get_ring_buffer(struct hv_ring_buffer_info *ring_info) static inline void hv_signal_on_read(struct vmbus_channel *channel) { - u32 cur_write_sz; + u32 cur_write_sz, cached_write_sz; u32 pending_sz; struct hv_ring_buffer_info *rbi = &channel->inbound; @@ -1512,12 +1526,24 @@ static inline void hv_signal_on_read(struct vmbus_channel *channel) cur_write_sz = hv_get_bytes_to_write(rbi); - if (cur_write_sz >= pending_sz) + if (cur_write_sz < pending_sz) + return; + + cached_write_sz = hv_get_cached_bytes_to_write(rbi); + if (cached_write_sz < pending_sz) vmbus_setevent(channel); return; } +static inline void +init_cached_read_index(struct vmbus_channel *channel) +{ + struct hv_ring_buffer_info *rbi = &channel->inbound; + + rbi->cached_read_index = rbi->ring_buffer->read_index; +} + /* * An API to support in-place processing of incoming VMBUS packets. */ @@ -1569,6 +1595,8 @@ static inline void put_pkt_raw(struct vmbus_channel *channel, * This call commits the read index and potentially signals the host. * Here is the pattern for using the "in-place" consumption APIs: * + * init_cached_read_index(); + * * while (get_next_pkt_raw() { * process the packet "in-place"; * put_pkt_raw(); -- cgit v1.2.3 From 96692b097ba76d0c637ae8af47b29c73da33c9d0 Mon Sep 17 00:00:00 2001 From: Ben Skeggs Date: Wed, 14 Dec 2016 09:52:39 +1000 Subject: drm/nouveau/fence/g84-: protect against concurrent access to semaphore buffers Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_fence.h | 1 + drivers/gpu/drm/nouveau/nv84_fence.c | 6 ++++++ 2 files changed, 7 insertions(+) diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.h b/drivers/gpu/drm/nouveau/nouveau_fence.h index ccdce1b4eec4..d5e58a38f160 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.h +++ b/drivers/gpu/drm/nouveau/nouveau_fence.h @@ -99,6 +99,7 @@ struct nv84_fence_priv { struct nouveau_bo *bo; struct nouveau_bo *bo_gart; u32 *suspend; + struct mutex mutex; }; int nv84_fence_context_new(struct nouveau_channel *); diff --git a/drivers/gpu/drm/nouveau/nv84_fence.c b/drivers/gpu/drm/nouveau/nv84_fence.c index 52b87ae83e7b..f0b322bec7df 100644 --- a/drivers/gpu/drm/nouveau/nv84_fence.c +++ b/drivers/gpu/drm/nouveau/nv84_fence.c @@ -107,8 +107,10 @@ nv84_fence_context_del(struct nouveau_channel *chan) struct nv84_fence_chan *fctx = chan->fence; nouveau_bo_wr32(priv->bo, chan->chid * 16 / 4, fctx->base.sequence); + mutex_lock(&priv->mutex); nouveau_bo_vma_del(priv->bo, &fctx->vma_gart); nouveau_bo_vma_del(priv->bo, &fctx->vma); + mutex_unlock(&priv->mutex); nouveau_fence_context_del(&fctx->base); chan->fence = NULL; nouveau_fence_context_free(&fctx->base); @@ -134,11 +136,13 @@ nv84_fence_context_new(struct nouveau_channel *chan) fctx->base.sync32 = nv84_fence_sync32; fctx->base.sequence = nv84_fence_read(chan); + mutex_lock(&priv->mutex); ret = nouveau_bo_vma_add(priv->bo, cli->vm, &fctx->vma); if (ret == 0) { ret = nouveau_bo_vma_add(priv->bo_gart, cli->vm, &fctx->vma_gart); } + mutex_unlock(&priv->mutex); if (ret) nv84_fence_context_del(chan); @@ -212,6 +216,8 @@ nv84_fence_create(struct nouveau_drm *drm) priv->base.context_base = dma_fence_context_alloc(priv->base.contexts); priv->base.uevent = true; + mutex_init(&priv->mutex); + /* Use VRAM if there is any ; otherwise fallback to system memory */ domain = drm->device.info.ram_size != 0 ? TTM_PL_FLAG_VRAM : /* -- cgit v1.2.3 From c966b6279f610a24ac1d42dcbe30e10fa61220b2 Mon Sep 17 00:00:00 2001 From: Ben Skeggs Date: Wed, 25 May 2016 17:11:40 +1000 Subject: drm/nouveau: prevent userspace from deleting client object Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_usif.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_usif.c b/drivers/gpu/drm/nouveau/nouveau_usif.c index 08f9c6fa0f7f..1fba38622744 100644 --- a/drivers/gpu/drm/nouveau/nouveau_usif.c +++ b/drivers/gpu/drm/nouveau/nouveau_usif.c @@ -313,7 +313,8 @@ usif_ioctl(struct drm_file *filp, void __user *user, u32 argc) if (!(ret = nvif_unpack(-ENOSYS, &data, &size, argv->v0, 0, 0, true))) { /* block access to objects not created via this interface */ owner = argv->v0.owner; - if (argv->v0.object == 0ULL) + if (argv->v0.object == 0ULL && + argv->v0.type != NVIF_IOCTL_V0_DEL) argv->v0.owner = NVDRM_OBJECT_ANY; /* except client */ else argv->v0.owner = NVDRM_OBJECT_USIF; -- cgit v1.2.3 From 7dfee6827780d4228148263545af936d0cae8930 Mon Sep 17 00:00:00 2001 From: Ben Skeggs Date: Mon, 9 Jan 2017 10:22:15 +1000 Subject: drm/nouveau/disp/mcp7x: disable dptmds workaround The workaround appears to cause regressions on these boards, and from inspection of RM traces, NVIDIA don't appear to do it on them either. Signed-off-by: Ben Skeggs Tested-by: Roy Spliet --- drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c index 567466f93cd5..0db8efbf1c2e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c @@ -433,8 +433,6 @@ nv50_disp_dptmds_war(struct nvkm_device *device) case 0x94: case 0x96: case 0x98: - case 0xaa: - case 0xac: return true; default: break; -- cgit v1.2.3 From d72498ca2cbcf15e5038b184a95f061bca21e820 Mon Sep 17 00:00:00 2001 From: Martin Peres Date: Wed, 7 Dec 2016 06:30:15 +0200 Subject: drm/nouveau/nouveau/led: prevent compiling the led-code if nouveau=y and leds=m The proper fix would have been to select LEDS_CLASS but this can lead to a circular dependency, as found out by Arnd. This patch implements Arnd's suggestion instead, at the cost of some auto-magic for a fringe feature. Reported-by: Arnd Bergmann Reported-by: Intel's 0-DAY Fixes: 8d021d71b324 ("drm/nouveau/drm/nouveau: add a LED driver for the NVIDIA logo") Signed-off-by: Martin Peres Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_led.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_led.h b/drivers/gpu/drm/nouveau/nouveau_led.h index 187ecdb82002..21a5775028cc 100644 --- a/drivers/gpu/drm/nouveau/nouveau_led.h +++ b/drivers/gpu/drm/nouveau/nouveau_led.h @@ -42,7 +42,7 @@ nouveau_led(struct drm_device *dev) } /* nouveau_led.c */ -#if IS_ENABLED(CONFIG_LEDS_CLASS) +#if IS_REACHABLE(CONFIG_LEDS_CLASS) int nouveau_led_init(struct drm_device *dev); void nouveau_led_suspend(struct drm_device *dev); void nouveau_led_resume(struct drm_device *dev); -- cgit v1.2.3 From d347583a39e2df609a9e40c835f72d3614665b53 Mon Sep 17 00:00:00 2001 From: Alastair Bridgewater Date: Wed, 11 Jan 2017 15:47:18 -0500 Subject: drm/nouveau/disp/gt215: Fix HDA ELD handling (thus, HDMI audio) on gt215 Store the ELD correctly, not just enough copies of the first byte to pad out the given ELD size. Signed-off-by: Alastair Bridgewater Fixes: 120b0c39c756 ("drm/nv50-/disp: audit and version SOR_HDA_ELD method") Cc: stable@vger.kernel.org # v3.17+ Reviewed-by: Ilia Mirkin Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nvkm/engine/disp/hdagt215.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/hdagt215.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/hdagt215.c index 6f0436df0219..f8f2f16c22a2 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/hdagt215.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/hdagt215.c @@ -59,7 +59,7 @@ gt215_hda_eld(NV50_DISP_MTHD_V1) ); } for (i = 0; i < size; i++) - nvkm_wr32(device, 0x61c440 + soff, (i << 8) | args->v0.data[0]); + nvkm_wr32(device, 0x61c440 + soff, (i << 8) | args->v0.data[i]); for (; i < 0x60; i++) nvkm_wr32(device, 0x61c440 + soff, (i << 8)); nvkm_mask(device, 0x61c448 + soff, 0x80000003, 0x80000003); -- cgit v1.2.3 From 24bf7ae359b8cca165bb30742d2b1c03a1eb23af Mon Sep 17 00:00:00 2001 From: Ilia Mirkin Date: Thu, 19 Jan 2017 22:56:30 -0500 Subject: drm/nouveau/nv1a,nv1f/disp: fix memory clock rate retrieval Based on the xf86-video-nv code, NFORCE (NV1A) and NFORCE2 (NV1F) have a different way of retrieving clocks. See the nv_hw.c:nForceUpdateArbitrationSettings function in the original code for how these clocks were accessed. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=54587 Cc: stable@vger.kernel.org Signed-off-by: Ilia Mirkin Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/dispnv04/hw.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/dispnv04/hw.c b/drivers/gpu/drm/nouveau/dispnv04/hw.c index 74856a8b8f35..e64f52464ecf 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/hw.c +++ b/drivers/gpu/drm/nouveau/dispnv04/hw.c @@ -222,6 +222,7 @@ nouveau_hw_get_clock(struct drm_device *dev, enum nvbios_pll_type plltype) uint32_t mpllP; pci_read_config_dword(pci_get_bus_and_slot(0, 3), 0x6c, &mpllP); + mpllP = (mpllP >> 8) & 0xf; if (!mpllP) mpllP = 4; @@ -232,7 +233,7 @@ nouveau_hw_get_clock(struct drm_device *dev, enum nvbios_pll_type plltype) uint32_t clock; pci_read_config_dword(pci_get_bus_and_slot(0, 5), 0x4c, &clock); - return clock; + return clock / 1000; } ret = nouveau_hw_get_pllvals(dev, plltype, &pllvals); -- cgit v1.2.3 From 2b5078937355c0d662ecef54b7d4d04f48da4fa9 Mon Sep 17 00:00:00 2001 From: Ben Skeggs Date: Tue, 24 Jan 2017 09:32:26 +1000 Subject: drm/nouveau/kms/nv50: request vblank events for commits that send completion events This somehow fixes an issue where sync-to-vblank longer works correctly after resume from suspend. From a HW perspective, we don't need the IRQs turned on to be able to detect flip completion, so it's assumed that this is required for the voodoo in the core DRM vblank core. Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nv50_display.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c index 2c2c64507661..32097fd615fd 100644 --- a/drivers/gpu/drm/nouveau/nv50_display.c +++ b/drivers/gpu/drm/nouveau/nv50_display.c @@ -4052,6 +4052,11 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state) } } + for_each_crtc_in_state(state, crtc, crtc_state, i) { + if (crtc->state->event) + drm_crtc_vblank_get(crtc); + } + /* Update plane(s). */ for_each_plane_in_state(state, plane, plane_state, i) { struct nv50_wndw_atom *asyw = nv50_wndw_atom(plane->state); @@ -4101,6 +4106,7 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state) drm_crtc_send_vblank_event(crtc, crtc->state->event); spin_unlock_irqrestore(&crtc->dev->event_lock, flags); crtc->state->event = NULL; + drm_crtc_vblank_put(crtc); } } -- cgit v1.2.3 From 161e6d44a5e2d3f85365cb717d60e363171b39e6 Mon Sep 17 00:00:00 2001 From: Gabriel Krisman Bertazi Date: Mon, 16 Jan 2017 12:23:42 -0200 Subject: mmc: sdhci: Ignore unexpected CARD_INT interrupts One of our kernelCI boxes hanged at boot because a faulty eSDHC device was triggering spurious CARD_INT interrupts for SD cards, causing CMD52 reads, which are not allowed for SD devices. This adds a sanity check to the interruption path, preventing that illegal command from getting sent if the CARD_INT interruption should be disabled. This quirk allows that particular machine to resume boot despite the faulty hardware, instead of getting hung dealing with thousands of mishandled interrupts. Suggested-by: Adrian Hunter Signed-off-by: Gabriel Krisman Bertazi Acked-by: Adrian Hunter Signed-off-by: Ulf Hansson Cc: --- drivers/mmc/host/sdhci.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c index 23909804ffb8..0def99590d16 100644 --- a/drivers/mmc/host/sdhci.c +++ b/drivers/mmc/host/sdhci.c @@ -2733,7 +2733,8 @@ static irqreturn_t sdhci_irq(int irq, void *dev_id) if (intmask & SDHCI_INT_RETUNE) mmc_retune_needed(host->mmc); - if (intmask & SDHCI_INT_CARD_INT) { + if ((intmask & SDHCI_INT_CARD_INT) && + (host->ier & SDHCI_INT_CARD_INT)) { sdhci_enable_sdio_irq_nolock(host, false); host->thread_isr |= SDHCI_INT_CARD_INT; result = IRQ_WAKE_THREAD; -- cgit v1.2.3 From 7a7b5df84b6b4e5d599c7289526eed96541a0654 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Mon, 30 Jan 2017 11:26:38 +0100 Subject: HID: cp2112: fix sleep-while-atomic A recent commit fixing DMA-buffers on stack added a shared transfer buffer protected by a spinlock. This is broken as the USB HID request callbacks can sleep. Fix this up by replacing the spinlock with a mutex. Fixes: 1ffb3c40ffb5 ("HID: cp2112: make transfer buffers DMA capable") Cc: stable # 4.9 Signed-off-by: Johan Hovold Reviewed-by: Benjamin Tissoires Signed-off-by: Jiri Kosina --- drivers/hid/hid-cp2112.c | 26 +++++++++++--------------- 1 file changed, 11 insertions(+), 15 deletions(-) diff --git a/drivers/hid/hid-cp2112.c b/drivers/hid/hid-cp2112.c index f31a778b0851..3e0b6bad29f2 100644 --- a/drivers/hid/hid-cp2112.c +++ b/drivers/hid/hid-cp2112.c @@ -168,7 +168,7 @@ struct cp2112_device { atomic_t xfer_avail; struct gpio_chip gc; u8 *in_out_buffer; - spinlock_t lock; + struct mutex lock; struct gpio_desc *desc[8]; bool gpio_poll; @@ -186,10 +186,9 @@ static int cp2112_gpio_direction_input(struct gpio_chip *chip, unsigned offset) struct cp2112_device *dev = gpiochip_get_data(chip); struct hid_device *hdev = dev->hdev; u8 *buf = dev->in_out_buffer; - unsigned long flags; int ret; - spin_lock_irqsave(&dev->lock, flags); + mutex_lock(&dev->lock); ret = hid_hw_raw_request(hdev, CP2112_GPIO_CONFIG, buf, CP2112_GPIO_CONFIG_LENGTH, HID_FEATURE_REPORT, @@ -213,7 +212,7 @@ static int cp2112_gpio_direction_input(struct gpio_chip *chip, unsigned offset) ret = 0; exit: - spin_unlock_irqrestore(&dev->lock, flags); + mutex_unlock(&dev->lock); return ret <= 0 ? ret : -EIO; } @@ -222,10 +221,9 @@ static void cp2112_gpio_set(struct gpio_chip *chip, unsigned offset, int value) struct cp2112_device *dev = gpiochip_get_data(chip); struct hid_device *hdev = dev->hdev; u8 *buf = dev->in_out_buffer; - unsigned long flags; int ret; - spin_lock_irqsave(&dev->lock, flags); + mutex_lock(&dev->lock); buf[0] = CP2112_GPIO_SET; buf[1] = value ? 0xff : 0; @@ -237,7 +235,7 @@ static void cp2112_gpio_set(struct gpio_chip *chip, unsigned offset, int value) if (ret < 0) hid_err(hdev, "error setting GPIO values: %d\n", ret); - spin_unlock_irqrestore(&dev->lock, flags); + mutex_unlock(&dev->lock); } static int cp2112_gpio_get_all(struct gpio_chip *chip) @@ -245,10 +243,9 @@ static int cp2112_gpio_get_all(struct gpio_chip *chip) struct cp2112_device *dev = gpiochip_get_data(chip); struct hid_device *hdev = dev->hdev; u8 *buf = dev->in_out_buffer; - unsigned long flags; int ret; - spin_lock_irqsave(&dev->lock, flags); + mutex_lock(&dev->lock); ret = hid_hw_raw_request(hdev, CP2112_GPIO_GET, buf, CP2112_GPIO_GET_LENGTH, HID_FEATURE_REPORT, @@ -262,7 +259,7 @@ static int cp2112_gpio_get_all(struct gpio_chip *chip) ret = buf[1]; exit: - spin_unlock_irqrestore(&dev->lock, flags); + mutex_unlock(&dev->lock); return ret; } @@ -284,10 +281,9 @@ static int cp2112_gpio_direction_output(struct gpio_chip *chip, struct cp2112_device *dev = gpiochip_get_data(chip); struct hid_device *hdev = dev->hdev; u8 *buf = dev->in_out_buffer; - unsigned long flags; int ret; - spin_lock_irqsave(&dev->lock, flags); + mutex_lock(&dev->lock); ret = hid_hw_raw_request(hdev, CP2112_GPIO_CONFIG, buf, CP2112_GPIO_CONFIG_LENGTH, HID_FEATURE_REPORT, @@ -308,7 +304,7 @@ static int cp2112_gpio_direction_output(struct gpio_chip *chip, goto fail; } - spin_unlock_irqrestore(&dev->lock, flags); + mutex_unlock(&dev->lock); /* * Set gpio value when output direction is already set, @@ -319,7 +315,7 @@ static int cp2112_gpio_direction_output(struct gpio_chip *chip, return 0; fail: - spin_unlock_irqrestore(&dev->lock, flags); + mutex_unlock(&dev->lock); return ret < 0 ? ret : -EIO; } @@ -1235,7 +1231,7 @@ static int cp2112_probe(struct hid_device *hdev, const struct hid_device_id *id) if (!dev->in_out_buffer) return -ENOMEM; - spin_lock_init(&dev->lock); + mutex_init(&dev->lock); ret = hid_parse(hdev); if (ret) { -- cgit v1.2.3 From 8e9faa15469ed7c7467423db4c62aeed3ff4cae3 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Mon, 30 Jan 2017 11:26:39 +0100 Subject: HID: cp2112: fix gpio-callback error handling In case of a zero-length report, the gpio direction_input callback would currently return success instead of an errno. Fixes: 1ffb3c40ffb5 ("HID: cp2112: make transfer buffers DMA capable") Cc: stable # 4.9 Signed-off-by: Johan Hovold Reviewed-by: Benjamin Tissoires Signed-off-by: Jiri Kosina --- drivers/hid/hid-cp2112.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hid/hid-cp2112.c b/drivers/hid/hid-cp2112.c index 3e0b6bad29f2..b22d0f83f8e3 100644 --- a/drivers/hid/hid-cp2112.c +++ b/drivers/hid/hid-cp2112.c @@ -213,7 +213,7 @@ static int cp2112_gpio_direction_input(struct gpio_chip *chip, unsigned offset) exit: mutex_unlock(&dev->lock); - return ret <= 0 ? ret : -EIO; + return ret < 0 ? ret : -EIO; } static void cp2112_gpio_set(struct gpio_chip *chip, unsigned offset, int value) -- cgit v1.2.3 From 92c715fca907686f5298220ece53423e38ba3aed Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Tue, 31 Jan 2017 10:25:25 +0100 Subject: drm/atomic: Fix double free in drm_atomic_state_default_clear drm_atomic_helper_page_flip and drm_atomic_ioctl set their own events in crtc_state->event. But when it's set the event is freed in 2 places. Solve this by only freeing the event in the atomic ioctl when it allocated its own event. This has been broken twice. The first time when the code was introduced, but only in the corner case when an event is allocated, but more crtc's were included by atomic check and then failing. This can mostly happen when you do an atomic modeset in i915 and the display clock is changed, which forces all crtc's to be included to the state. This has been broken worse by adding in-fences support, which caused the double free to be done unconditionally. [IGT] kms_rotation_crc: starting subtest primary-rotation-180 ============================================================================= BUG kmalloc-128 (Tainted: G U ): Object already free ----------------------------------------------------------------------------- Disabling lock debugging due to kernel taint INFO: Allocated in drm_atomic_helper_setup_commit+0x285/0x2f0 [drm_kms_helper] age=0 cpu=3 pid=1529 ___slab_alloc+0x308/0x3b0 __slab_alloc+0xd/0x20 kmem_cache_alloc_trace+0x92/0x1c0 drm_atomic_helper_setup_commit+0x285/0x2f0 [drm_kms_helper] intel_atomic_commit+0x35/0x4f0 [i915] drm_atomic_commit+0x46/0x50 [drm] drm_mode_atomic_ioctl+0x7d4/0xab0 [drm] drm_ioctl+0x2b3/0x490 [drm] do_vfs_ioctl+0x69c/0x700 SyS_ioctl+0x4e/0x80 entry_SYSCALL_64_fastpath+0x13/0x94 INFO: Freed in drm_event_cancel_free+0xa3/0xb0 [drm] age=0 cpu=3 pid=1529 __slab_free+0x48/0x2e0 kfree+0x159/0x1a0 drm_event_cancel_free+0xa3/0xb0 [drm] drm_mode_atomic_ioctl+0x86d/0xab0 [drm] drm_ioctl+0x2b3/0x490 [drm] do_vfs_ioctl+0x69c/0x700 SyS_ioctl+0x4e/0x80 entry_SYSCALL_64_fastpath+0x13/0x94 INFO: Slab 0xffffde1f0997b080 objects=17 used=2 fp=0xffff92fb65ec2578 flags=0x200000000008101 INFO: Object 0xffff92fb65ec2578 @offset=1400 fp=0xffff92fb65ec2ae8 Redzone ffff92fb65ec2570: bb bb bb bb bb bb bb bb ........ Object ffff92fb65ec2578: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk Object ffff92fb65ec2588: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk Object ffff92fb65ec2598: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk Object ffff92fb65ec25a8: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk Object ffff92fb65ec25b8: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk Object ffff92fb65ec25c8: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk Object ffff92fb65ec25d8: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk Object ffff92fb65ec25e8: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b a5 kkkkkkkkkkkkkkk. Redzone ffff92fb65ec25f8: bb bb bb bb bb bb bb bb ........ Padding ffff92fb65ec2738: 5a 5a 5a 5a 5a 5a 5a 5a ZZZZZZZZ CPU: 3 PID: 180 Comm: kworker/3:2 Tainted: G BU 4.10.0-rc6-patser+ #5039 Hardware name: /NUC5PPYB, BIOS PYBSWCEL.86A.0031.2015.0601.1712 06/01/2015 Workqueue: events intel_atomic_helper_free_state [i915] Call Trace: dump_stack+0x4d/0x6d print_trailer+0x20c/0x220 free_debug_processing+0x1c6/0x330 ? drm_atomic_state_default_clear+0xf7/0x1c0 [drm] __slab_free+0x48/0x2e0 ? drm_atomic_state_default_clear+0xf7/0x1c0 [drm] kfree+0x159/0x1a0 drm_atomic_state_default_clear+0xf7/0x1c0 [drm] ? drm_atomic_state_clear+0x30/0x30 [drm] intel_atomic_state_clear+0xd/0x20 [i915] drm_atomic_state_clear+0x1a/0x30 [drm] __drm_atomic_state_free+0x13/0x60 [drm] intel_atomic_helper_free_state+0x5d/0x70 [i915] process_one_work+0x260/0x4a0 worker_thread+0x2d1/0x4f0 kthread+0x127/0x130 ? process_one_work+0x4a0/0x4a0 ? kthread_stop+0x120/0x120 ret_from_fork+0x29/0x40 FIX kmalloc-128: Object at 0xffff92fb65ec2578 not freed Fixes: 3b24f7d67581 ("drm/atomic: Add struct drm_crtc_commit to track async updates") Fixes: 9626014258a5 ("drm/fence: add in-fences support") Cc: # v4.8+ Cc: Daniel Vetter Signed-off-by: Maarten Lankhorst Reviewed-by: Daniel Vetter Reviewed-by: Gustavo Padovan Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/1485854725-27640-1-git-send-email-maarten.lankhorst@linux.intel.com --- drivers/gpu/drm/drm_atomic.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index 50f5cf7b69d1..fdfb1ec17e66 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -2032,13 +2032,16 @@ static void complete_crtc_signaling(struct drm_device *dev, } for_each_crtc_in_state(state, crtc, crtc_state, i) { + struct drm_pending_vblank_event *event = crtc_state->event; /* - * TEST_ONLY and PAGE_FLIP_EVENT are mutually - * exclusive, if they weren't, this code should be - * called on success for TEST_ONLY too. + * Free the allocated event. drm_atomic_helper_setup_commit + * can allocate an event too, so only free it if it's ours + * to prevent a double free in drm_atomic_state_clear. */ - if (crtc_state->event) - drm_event_cancel_free(dev, &crtc_state->event->base); + if (event && (event->base.fence || event->base.file_priv)) { + drm_event_cancel_free(dev, &event->base); + crtc_state->event = NULL; + } } if (!fence_state) -- cgit v1.2.3 From 79c6f448c8b79c321e4a1f31f98194e4f6b6cae7 Mon Sep 17 00:00:00 2001 From: Steven Rostedt (VMware) Date: Mon, 30 Jan 2017 19:27:10 -0500 Subject: tracing: Fix hwlat kthread migration The hwlat tracer creates a kernel thread at start of the tracer. It is pinned to a single CPU and will move to the next CPU after each period of running. If the user modifies the migration thread's affinity, it will not change after that happens. The original code created the thread at the first instance it was called, but later was changed to destroy the thread after the tracer was finished, and would not be created until the next instance of the tracer was established. The code that initialized the affinity was only called on the initial instantiation of the tracer. After that, it was not initialized, and the previous affinity did not match the current newly created one, making it appear that the user modified the thread's affinity when it did not, and the thread failed to migrate again. Cc: stable@vger.kernel.org Fixes: 0330f7aa8ee6 ("tracing: Have hwlat trace migrate across tracing_cpumask CPUs") Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_hwlat.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/kernel/trace/trace_hwlat.c b/kernel/trace/trace_hwlat.c index 775569ec50d0..af344a1bf0d0 100644 --- a/kernel/trace/trace_hwlat.c +++ b/kernel/trace/trace_hwlat.c @@ -266,7 +266,7 @@ out: static struct cpumask save_cpumask; static bool disable_migrate; -static void move_to_next_cpu(void) +static void move_to_next_cpu(bool initmask) { static struct cpumask *current_mask; int next_cpu; @@ -275,7 +275,7 @@ static void move_to_next_cpu(void) return; /* Just pick the first CPU on first iteration */ - if (!current_mask) { + if (initmask) { current_mask = &save_cpumask; get_online_cpus(); cpumask_and(current_mask, cpu_online_mask, tracing_buffer_mask); @@ -330,10 +330,12 @@ static void move_to_next_cpu(void) static int kthread_fn(void *data) { u64 interval; + bool initmask = true; while (!kthread_should_stop()) { - move_to_next_cpu(); + move_to_next_cpu(initmask); + initmask = false; local_irq_disable(); get_sample(); -- cgit v1.2.3 From d07830db1bdb254e4b50d366010b219286b8c937 Mon Sep 17 00:00:00 2001 From: Marcel J.E. Mol Date: Mon, 30 Jan 2017 19:26:40 +0100 Subject: USB: serial: pl2303: add ATEN device ID Seems that ATEN serial-to-usb devices using pl2303 exist with different device ids. This patch adds a missing device ID so it is recognised by the driver. Signed-off-by: Marcel J.E. Mol Cc: stable Signed-off-by: Johan Hovold --- drivers/usb/serial/pl2303.c | 1 + drivers/usb/serial/pl2303.h | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c index 46fca6b75846..1db4b61bdf7b 100644 --- a/drivers/usb/serial/pl2303.c +++ b/drivers/usb/serial/pl2303.c @@ -49,6 +49,7 @@ static const struct usb_device_id id_table[] = { { USB_DEVICE(IODATA_VENDOR_ID, IODATA_PRODUCT_ID) }, { USB_DEVICE(IODATA_VENDOR_ID, IODATA_PRODUCT_ID_RSAQ5) }, { USB_DEVICE(ATEN_VENDOR_ID, ATEN_PRODUCT_ID) }, + { USB_DEVICE(ATEN_VENDOR_ID, ATEN_PRODUCT_ID2) }, { USB_DEVICE(ATEN_VENDOR_ID2, ATEN_PRODUCT_ID) }, { USB_DEVICE(ELCOM_VENDOR_ID, ELCOM_PRODUCT_ID) }, { USB_DEVICE(ELCOM_VENDOR_ID, ELCOM_PRODUCT_ID_UCSGT) }, diff --git a/drivers/usb/serial/pl2303.h b/drivers/usb/serial/pl2303.h index e3b7af8adfb7..09d9be88209e 100644 --- a/drivers/usb/serial/pl2303.h +++ b/drivers/usb/serial/pl2303.h @@ -27,6 +27,7 @@ #define ATEN_VENDOR_ID 0x0557 #define ATEN_VENDOR_ID2 0x0547 #define ATEN_PRODUCT_ID 0x2008 +#define ATEN_PRODUCT_ID2 0x2118 #define IODATA_VENDOR_ID 0x04bb #define IODATA_PRODUCT_ID 0x0a03 -- cgit v1.2.3 From d19fb70dd68c4e960e2ac09b0b9c79dfdeefa726 Mon Sep 17 00:00:00 2001 From: Kinglong Mee Date: Wed, 18 Jan 2017 19:04:42 +0800 Subject: NFSD: Fix a null reference case in find_or_create_lock_stateid() nfsd assigns the nfs4_free_lock_stateid to .sc_free in init_lock_stateid(). If nfsd doesn't go through init_lock_stateid() and put stateid at end, there is a NULL reference to .sc_free when calling nfs4_put_stid(ns). This patch let the nfs4_stid.sc_free assignment to nfs4_alloc_stid(). Cc: stable@vger.kernel.org Fixes: 356a95ece7aa "nfsd: clean up races in lock stateid searching..." Signed-off-by: Kinglong Mee Reviewed-by: Jeff Layton Signed-off-by: J. Bruce Fields --- fs/nfsd/nfs4layouts.c | 5 +++-- fs/nfsd/nfs4state.c | 19 ++++++++----------- fs/nfsd/state.h | 4 ++-- 3 files changed, 13 insertions(+), 15 deletions(-) diff --git a/fs/nfsd/nfs4layouts.c b/fs/nfsd/nfs4layouts.c index 596205d939a1..1fc07a9c70e9 100644 --- a/fs/nfsd/nfs4layouts.c +++ b/fs/nfsd/nfs4layouts.c @@ -223,10 +223,11 @@ nfsd4_alloc_layout_stateid(struct nfsd4_compound_state *cstate, struct nfs4_layout_stateid *ls; struct nfs4_stid *stp; - stp = nfs4_alloc_stid(cstate->clp, nfs4_layout_stateid_cache); + stp = nfs4_alloc_stid(cstate->clp, nfs4_layout_stateid_cache, + nfsd4_free_layout_stateid); if (!stp) return NULL; - stp->sc_free = nfsd4_free_layout_stateid; + get_nfs4_file(fp); stp->sc_file = fp; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 4b4beaaa4eaa..a0dee8ae9f97 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -633,8 +633,8 @@ out: return co; } -struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, - struct kmem_cache *slab) +struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, struct kmem_cache *slab, + void (*sc_free)(struct nfs4_stid *)) { struct nfs4_stid *stid; int new_id; @@ -650,6 +650,8 @@ struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, idr_preload_end(); if (new_id < 0) goto out_free; + + stid->sc_free = sc_free; stid->sc_client = cl; stid->sc_stateid.si_opaque.so_id = new_id; stid->sc_stateid.si_opaque.so_clid = cl->cl_clientid; @@ -675,15 +677,12 @@ out_free: static struct nfs4_ol_stateid * nfs4_alloc_open_stateid(struct nfs4_client *clp) { struct nfs4_stid *stid; - struct nfs4_ol_stateid *stp; - stid = nfs4_alloc_stid(clp, stateid_slab); + stid = nfs4_alloc_stid(clp, stateid_slab, nfs4_free_ol_stateid); if (!stid) return NULL; - stp = openlockstateid(stid); - stp->st_stid.sc_free = nfs4_free_ol_stateid; - return stp; + return openlockstateid(stid); } static void nfs4_free_deleg(struct nfs4_stid *stid) @@ -781,11 +780,10 @@ alloc_init_deleg(struct nfs4_client *clp, struct svc_fh *current_fh, goto out_dec; if (delegation_blocked(¤t_fh->fh_handle)) goto out_dec; - dp = delegstateid(nfs4_alloc_stid(clp, deleg_slab)); + dp = delegstateid(nfs4_alloc_stid(clp, deleg_slab, nfs4_free_deleg)); if (dp == NULL) goto out_dec; - dp->dl_stid.sc_free = nfs4_free_deleg; /* * delegation seqid's are never incremented. The 4.1 special * meaning of seqid 0 isn't meaningful, really, but let's avoid @@ -5580,7 +5578,6 @@ init_lock_stateid(struct nfs4_ol_stateid *stp, struct nfs4_lockowner *lo, stp->st_stateowner = nfs4_get_stateowner(&lo->lo_owner); get_nfs4_file(fp); stp->st_stid.sc_file = fp; - stp->st_stid.sc_free = nfs4_free_lock_stateid; stp->st_access_bmap = 0; stp->st_deny_bmap = open_stp->st_deny_bmap; stp->st_openstp = open_stp; @@ -5623,7 +5620,7 @@ find_or_create_lock_stateid(struct nfs4_lockowner *lo, struct nfs4_file *fi, lst = find_lock_stateid(lo, fi); if (lst == NULL) { spin_unlock(&clp->cl_lock); - ns = nfs4_alloc_stid(clp, stateid_slab); + ns = nfs4_alloc_stid(clp, stateid_slab, nfs4_free_lock_stateid); if (ns == NULL) return NULL; diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index c9399366f9df..4516e8b7d776 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -603,8 +603,8 @@ extern __be32 nfs4_preprocess_stateid_op(struct svc_rqst *rqstp, __be32 nfsd4_lookup_stateid(struct nfsd4_compound_state *cstate, stateid_t *stateid, unsigned char typemask, struct nfs4_stid **s, struct nfsd_net *nn); -struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, - struct kmem_cache *slab); +struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, struct kmem_cache *slab, + void (*sc_free)(struct nfs4_stid *)); void nfs4_unhash_stid(struct nfs4_stid *s); void nfs4_put_stid(struct nfs4_stid *s); void nfs4_inc_and_copy_stateid(stateid_t *dst, struct nfs4_stid *stid); -- cgit v1.2.3 From 41f53350a0f36a7b8e31bec0d0ca907e028ab4cd Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 24 Jan 2017 09:22:41 +0100 Subject: nfsd: special case truncates some more Both the NFS protocols and the Linux VFS use a setattr operation with a bitmap of attributs to set to set various file attributes including the file size and the uid/gid. The Linux syscalls never mixes size updates with unrelated updates like the uid/gid, and some file systems like XFS and GFS2 rely on the fact that truncates might not update random other attributes, and many other file systems handle the case but do not update the different attributes in the same transaction. NFSD on the other hand passes the attributes it gets on the wire more or less directly through to the VFS, leading to updates the file systems don't expect. XFS at least has an assert on the allowed attributes, which caught an unusual NFS client setting the size and group at the same time. To handle this issue properly this switches nfsd to call vfs_truncate for size changes, and then handle all other attributes through notify_change. As a side effect this also means less boilerplace code around the size change as we can now reuse the VFS code. Signed-off-by: Christoph Hellwig Signed-off-by: J. Bruce Fields --- fs/nfsd/vfs.c | 97 +++++++++++++++++++++++------------------------------------ 1 file changed, 37 insertions(+), 60 deletions(-) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 26c6fdb4bf67..ca13236dbb1f 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -332,37 +332,6 @@ nfsd_sanitize_attrs(struct inode *inode, struct iattr *iap) } } -static __be32 -nfsd_get_write_access(struct svc_rqst *rqstp, struct svc_fh *fhp, - struct iattr *iap) -{ - struct inode *inode = d_inode(fhp->fh_dentry); - int host_err; - - if (iap->ia_size < inode->i_size) { - __be32 err; - - err = nfsd_permission(rqstp, fhp->fh_export, fhp->fh_dentry, - NFSD_MAY_TRUNC | NFSD_MAY_OWNER_OVERRIDE); - if (err) - return err; - } - - host_err = get_write_access(inode); - if (host_err) - goto out_nfserrno; - - host_err = locks_verify_truncate(inode, NULL, iap->ia_size); - if (host_err) - goto out_put_write_access; - return 0; - -out_put_write_access: - put_write_access(inode); -out_nfserrno: - return nfserrno(host_err); -} - /* * Set various file attributes. After this call fhp needs an fh_put. */ @@ -377,7 +346,6 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, __be32 err; int host_err; bool get_write_count; - int size_change = 0; if (iap->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_SIZE)) accmode |= NFSD_MAY_WRITE|NFSD_MAY_OWNER_OVERRIDE; @@ -390,11 +358,11 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, /* Get inode */ err = fh_verify(rqstp, fhp, ftype, accmode); if (err) - goto out; + return err; if (get_write_count) { host_err = fh_want_write(fhp); if (host_err) - return nfserrno(host_err); + goto out_host_err; } dentry = fhp->fh_dentry; @@ -405,50 +373,59 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, iap->ia_valid &= ~ATTR_MODE; if (!iap->ia_valid) - goto out; + return 0; nfsd_sanitize_attrs(inode, iap); + if (check_guard && guardtime != inode->i_ctime.tv_sec) + return nfserr_notsync; + /* * The size case is special, it changes the file in addition to the - * attributes. + * attributes, and file systems don't expect it to be mixed with + * "random" attribute changes. We thus split out the size change + * into a separate call for vfs_truncate, and do the rest as a + * a separate setattr call. */ if (iap->ia_valid & ATTR_SIZE) { - err = nfsd_get_write_access(rqstp, fhp, iap); - if (err) - goto out; - size_change = 1; + struct path path = { + .mnt = fhp->fh_export->ex_path.mnt, + .dentry = dentry, + }; + bool implicit_mtime = false; /* - * RFC5661, Section 18.30.4: - * Changing the size of a file with SETATTR indirectly - * changes the time_modify and change attributes. - * - * (and similar for the older RFCs) + * vfs_truncate implicity updates the mtime IFF the file size + * actually changes. Avoid the additional seattr call below if + * the only other attribute that the client sends is the mtime. */ - if (iap->ia_size != i_size_read(inode)) - iap->ia_valid |= ATTR_MTIME; - } + if (iap->ia_size != i_size_read(inode) && + ((iap->ia_valid & ~(ATTR_SIZE | ATTR_MTIME)) == 0)) + implicit_mtime = true; - iap->ia_valid |= ATTR_CTIME; + host_err = vfs_truncate(&path, iap->ia_size); + if (host_err) + goto out_host_err; - if (check_guard && guardtime != inode->i_ctime.tv_sec) { - err = nfserr_notsync; - goto out_put_write_access; + iap->ia_valid &= ~ATTR_SIZE; + if (implicit_mtime) + iap->ia_valid &= ~ATTR_MTIME; + if (!iap->ia_valid) + goto done; } + iap->ia_valid |= ATTR_CTIME; + fh_lock(fhp); host_err = notify_change(dentry, iap, NULL); fh_unlock(fhp); - err = nfserrno(host_err); + if (host_err) + goto out_host_err; -out_put_write_access: - if (size_change) - put_write_access(inode); - if (!err) - err = nfserrno(commit_metadata(fhp)); -out: - return err; +done: + host_err = commit_metadata(fhp); +out_host_err: + return nfserrno(host_err); } #if defined(CONFIG_NFSD_V4) -- cgit v1.2.3 From 034dd34ff4916ec1f8f74e39ca3efb04eab2f791 Mon Sep 17 00:00:00 2001 From: J. Bruce Fields Date: Tue, 31 Jan 2017 11:37:50 -0500 Subject: svcrpc: fix oops in absence of krb5 module Olga Kornievskaia says: "I ran into this oops in the nfsd (below) (4.10-rc3 kernel). To trigger this I had a client (unsuccessfully) try to mount the server with krb5 where the server doesn't have the rpcsec_gss_krb5 module built." The problem is that rsci.cred is copied from a svc_cred structure that gss_proxy didn't properly initialize. Fix that. [120408.542387] general protection fault: 0000 [#1] SMP ... [120408.565724] CPU: 0 PID: 3601 Comm: nfsd Not tainted 4.10.0-rc3+ #16 [120408.567037] Hardware name: VMware, Inc. VMware Virtual = Platform/440BX Desktop Reference Platform, BIOS 6.00 07/02/2015 [120408.569225] task: ffff8800776f95c0 task.stack: ffffc90003d58000 [120408.570483] RIP: 0010:gss_mech_put+0xb/0x20 [auth_rpcgss] ... [120408.584946] ? rsc_free+0x55/0x90 [auth_rpcgss] [120408.585901] gss_proxy_save_rsc+0xb2/0x2a0 [auth_rpcgss] [120408.587017] svcauth_gss_proxy_init+0x3cc/0x520 [auth_rpcgss] [120408.588257] ? __enqueue_entity+0x6c/0x70 [120408.589101] svcauth_gss_accept+0x391/0xb90 [auth_rpcgss] [120408.590212] ? try_to_wake_up+0x4a/0x360 [120408.591036] ? wake_up_process+0x15/0x20 [120408.592093] ? svc_xprt_do_enqueue+0x12e/0x2d0 [sunrpc] [120408.593177] svc_authenticate+0xe1/0x100 [sunrpc] [120408.594168] svc_process_common+0x203/0x710 [sunrpc] [120408.595220] svc_process+0x105/0x1c0 [sunrpc] [120408.596278] nfsd+0xe9/0x160 [nfsd] [120408.597060] kthread+0x101/0x140 [120408.597734] ? nfsd_destroy+0x60/0x60 [nfsd] [120408.598626] ? kthread_park+0x90/0x90 [120408.599448] ret_from_fork+0x22/0x30 Fixes: 1d658336b05f "SUNRPC: Add RPC based upcall mechanism for RPCGSS auth" Cc: stable@vger.kernel.org Cc: Simo Sorce Reported-by: Olga Kornievskaia Tested-by: Olga Kornievskaia Signed-off-by: J. Bruce Fields --- net/sunrpc/auth_gss/gss_rpc_xdr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sunrpc/auth_gss/gss_rpc_xdr.c b/net/sunrpc/auth_gss/gss_rpc_xdr.c index dc6fb79a361f..25d9a9cf7b66 100644 --- a/net/sunrpc/auth_gss/gss_rpc_xdr.c +++ b/net/sunrpc/auth_gss/gss_rpc_xdr.c @@ -260,7 +260,7 @@ static int gssx_dec_option_array(struct xdr_stream *xdr, if (!oa->data) return -ENOMEM; - creds = kmalloc(sizeof(struct svc_cred), GFP_KERNEL); + creds = kzalloc(sizeof(struct svc_cred), GFP_KERNEL); if (!creds) { kfree(oa->data); return -ENOMEM; -- cgit v1.2.3 From c73e44269369e936165f0f9b61f1f09a11dae01c Mon Sep 17 00:00:00 2001 From: Vincent Date: Mon, 30 Jan 2017 15:06:43 +0100 Subject: net: thunderx: avoid dereferencing xcv when NULL This fixes the following smatch and coccinelle warnings: drivers/net/ethernet/cavium/thunder/thunder_xcv.c:119 xcv_setup_link() error: we previously assumed 'xcv' could be null (see line 118) [smatch] drivers/net/ethernet/cavium/thunder/thunder_xcv.c:119:16-20: ERROR: xcv is NULL but dereferenced. [coccinelle] Fixes: 6465859aba1e66a5 ("net: thunderx: Add RGMII interface type support") Signed-off-by: Vincent Stehlé Cc: Sunil Goutham Signed-off-by: David S. Miller --- drivers/net/ethernet/cavium/thunder/thunder_xcv.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/ethernet/cavium/thunder/thunder_xcv.c b/drivers/net/ethernet/cavium/thunder/thunder_xcv.c index 67befedef709..578c7f8f11bf 100644 --- a/drivers/net/ethernet/cavium/thunder/thunder_xcv.c +++ b/drivers/net/ethernet/cavium/thunder/thunder_xcv.c @@ -116,8 +116,7 @@ void xcv_setup_link(bool link_up, int link_speed) int speed = 2; if (!xcv) { - dev_err(&xcv->pdev->dev, - "XCV init not done, probe may have failed\n"); + pr_err("XCV init not done, probe may have failed\n"); return; } -- cgit v1.2.3 From 90427ef5d2a4b9a24079889bf16afdcdaebc4240 Mon Sep 17 00:00:00 2001 From: Dimitris Michailidis Date: Mon, 30 Jan 2017 14:09:42 -0800 Subject: ipv6: fix flow labels when the traffic class is non-0 ip6_make_flowlabel() determines the flow label for IPv6 packets. It's supposed to be passed a flow label, which it returns as is if non-0 and in some other cases, otherwise it calculates a new value. The problem is callers often pass a flowi6.flowlabel, which may also contain traffic class bits. If the traffic class is non-0 ip6_make_flowlabel() mistakes the non-0 it gets as a flow label and returns the whole thing. Thus it can return a 'flow label' longer than 20b and the low 20b of that is typically 0 resulting in packets with 0 label. Moreover, different packets of a flow may be labeled differently. For a TCP flow with ECN non-payload and payload packets get different labels as exemplified by this pair of consecutive packets: (pure ACK) Internet Protocol Version 6, Src: 2002:af5:11a3::, Dst: 2002:af5:11a2:: 0110 .... = Version: 6 .... 0000 0000 .... .... .... .... .... = Traffic Class: 0x00 (DSCP: CS0, ECN: Not-ECT) .... 0000 00.. .... .... .... .... .... = Differentiated Services Codepoint: Default (0) .... .... ..00 .... .... .... .... .... = Explicit Congestion Notification: Not ECN-Capable Transport (0) .... .... .... 0001 1100 1110 0100 1001 = Flow Label: 0x1ce49 Payload Length: 32 Next Header: TCP (6) (payload) Internet Protocol Version 6, Src: 2002:af5:11a3::, Dst: 2002:af5:11a2:: 0110 .... = Version: 6 .... 0000 0010 .... .... .... .... .... = Traffic Class: 0x02 (DSCP: CS0, ECN: ECT(0)) .... 0000 00.. .... .... .... .... .... = Differentiated Services Codepoint: Default (0) .... .... ..10 .... .... .... .... .... = Explicit Congestion Notification: ECN-Capable Transport codepoint '10' (2) .... .... .... 0000 0000 0000 0000 0000 = Flow Label: 0x00000 Payload Length: 688 Next Header: TCP (6) This patch allows ip6_make_flowlabel() to be passed more than just a flow label and has it extract the part it really wants. This was simpler than modifying the callers. With this patch packets like the above become Internet Protocol Version 6, Src: 2002:af5:11a3::, Dst: 2002:af5:11a2:: 0110 .... = Version: 6 .... 0000 0000 .... .... .... .... .... = Traffic Class: 0x00 (DSCP: CS0, ECN: Not-ECT) .... 0000 00.. .... .... .... .... .... = Differentiated Services Codepoint: Default (0) .... .... ..00 .... .... .... .... .... = Explicit Congestion Notification: Not ECN-Capable Transport (0) .... .... .... 1010 1111 1010 0101 1110 = Flow Label: 0xafa5e Payload Length: 32 Next Header: TCP (6) Internet Protocol Version 6, Src: 2002:af5:11a3::, Dst: 2002:af5:11a2:: 0110 .... = Version: 6 .... 0000 0010 .... .... .... .... .... = Traffic Class: 0x02 (DSCP: CS0, ECN: ECT(0)) .... 0000 00.. .... .... .... .... .... = Differentiated Services Codepoint: Default (0) .... .... ..10 .... .... .... .... .... = Explicit Congestion Notification: ECN-Capable Transport codepoint '10' (2) .... .... .... 1010 1111 1010 0101 1110 = Flow Label: 0xafa5e Payload Length: 688 Next Header: TCP (6) Signed-off-by: Dimitris Michailidis Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ipv6.h | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 7afe991e900e..dbf0abba33b8 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -776,6 +776,11 @@ static inline __be32 ip6_make_flowlabel(struct net *net, struct sk_buff *skb, { u32 hash; + /* @flowlabel may include more than a flow label, eg, the traffic class. + * Here we want only the flow label value. + */ + flowlabel &= IPV6_FLOWLABEL_MASK; + if (flowlabel || net->ipv6.sysctl.auto_flowlabels == IP6_AUTO_FLOW_LABEL_OFF || (!autolabel && -- cgit v1.2.3 From 62deb8187d116581c88c69a2dd9b5c16588545d4 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 18 Jan 2017 14:29:17 +0000 Subject: FS-Cache: Initialise stores_lock in netfs cookie Initialise the stores_lock in fscache netfs cookies. Technically, it shouldn't be necessary, since the netfs cookie is an index and stores no data, but initialising it anyway adds insignificant overhead. Signed-off-by: David Howells Reviewed-by: Jeff Layton Acked-by: Steve Dickson Signed-off-by: Al Viro --- fs/fscache/netfs.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/fscache/netfs.c b/fs/fscache/netfs.c index 9b28649df3a1..a8aa00be4444 100644 --- a/fs/fscache/netfs.c +++ b/fs/fscache/netfs.c @@ -48,6 +48,7 @@ int __fscache_register_netfs(struct fscache_netfs *netfs) cookie->flags = 1 << FSCACHE_COOKIE_ENABLED; spin_lock_init(&cookie->lock); + spin_lock_init(&cookie->stores_lock); INIT_HLIST_HEAD(&cookie->backing_objects); /* check the netfs type is not already present */ -- cgit v1.2.3 From 6bdded59c8933940ac7e5b416448276ac89d1144 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 18 Jan 2017 14:29:25 +0000 Subject: fscache: Clear outstanding writes when disabling a cookie fscache_disable_cookie() needs to clear the outstanding writes on the cookie it's disabling because they cannot be completed after. Without this, fscache_nfs_open_file() gets stuck because it disables the cookie when the file is opened for writing but can't uncache the pages till afterwards - otherwise there's a race between the open routine and anyone who already has it open R/O and is still reading from it. Looking in /proc/pid/stack of the offending process shows: [] __fscache_wait_on_page_write+0x82/0x9b [fscache] [] __fscache_uncache_all_inode_pages+0x91/0xe1 [fscache] [] nfs_fscache_open_file+0x59/0x9e [nfs] [] nfs4_file_open+0x17f/0x1b8 [nfsv4] [] do_dentry_open+0x16d/0x2b7 [] vfs_open+0x5c/0x65 [] path_openat+0x785/0x8fb [] do_filp_open+0x48/0x9e [] do_sys_open+0x13b/0x1cb [] SyS_open+0x19/0x1b [] do_syscall_64+0x80/0x17a [] return_from_SYSCALL_64+0x0/0x7a [] 0xffffffffffffffff Reported-by: Jianhong Yin Signed-off-by: David Howells Acked-by: Jeff Layton Acked-by: Steve Dickson Signed-off-by: Al Viro --- fs/fscache/cookie.c | 5 +++++ fs/fscache/object.c | 6 ++++++ 2 files changed, 11 insertions(+) diff --git a/fs/fscache/cookie.c b/fs/fscache/cookie.c index 4304072161aa..40d61077bead 100644 --- a/fs/fscache/cookie.c +++ b/fs/fscache/cookie.c @@ -542,6 +542,7 @@ void __fscache_disable_cookie(struct fscache_cookie *cookie, bool invalidate) hlist_for_each_entry(object, &cookie->backing_objects, cookie_link) { if (invalidate) set_bit(FSCACHE_OBJECT_RETIRED, &object->flags); + clear_bit(FSCACHE_OBJECT_PENDING_WRITE, &object->flags); fscache_raise_event(object, FSCACHE_OBJECT_EV_KILL); } } else { @@ -560,6 +561,10 @@ void __fscache_disable_cookie(struct fscache_cookie *cookie, bool invalidate) wait_on_atomic_t(&cookie->n_active, fscache_wait_atomic_t, TASK_UNINTERRUPTIBLE); + /* Make sure any pending writes are cancelled. */ + if (cookie->def->type != FSCACHE_COOKIE_TYPE_INDEX) + fscache_invalidate_writes(cookie); + /* Reset the cookie state if it wasn't relinquished */ if (!test_bit(FSCACHE_COOKIE_RELINQUISHED, &cookie->flags)) { atomic_inc(&cookie->n_active); diff --git a/fs/fscache/object.c b/fs/fscache/object.c index 9e792e30f4db..be02a086ed9b 100644 --- a/fs/fscache/object.c +++ b/fs/fscache/object.c @@ -645,6 +645,12 @@ static const struct fscache_state *fscache_kill_object(struct fscache_object *ob fscache_mark_object_dead(object); object->oob_event_mask = 0; + if (test_bit(FSCACHE_OBJECT_RETIRED, &object->flags)) { + /* Reject any new read/write ops and abort any that are pending. */ + clear_bit(FSCACHE_OBJECT_PENDING_WRITE, &object->flags); + fscache_cancel_all_ops(object); + } + if (list_empty(&object->dependents) && object->n_ops == 0 && object->n_children == 0) -- cgit v1.2.3 From e26bfebdfc0d212d366de9990a096665d5c0209a Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 31 Jan 2017 09:45:28 +0000 Subject: fscache: Fix dead object requeue Under some circumstances, an fscache object can become queued such that it fscache_object_work_func() can be called once the object is in the OBJECT_DEAD state. This results in the kernel oopsing when it tries to invoke the handler for the state (which is hard coded to 0x2). The way this comes about is something like the following: (1) The object dispatcher is processing a work state for an object. This is done in workqueue context. (2) An out-of-band event comes in that isn't masked, causing the object to be queued, say EV_KILL. (3) The object dispatcher finishes processing the current work state on that object and then sees there's another event to process, so, without returning to the workqueue core, it processes that event too. It then follows the chain of events that initiates until we reach OBJECT_DEAD without going through a wait state (such as WAIT_FOR_CLEARANCE). At this point, object->events may be 0, object->event_mask will be 0 and oob_event_mask will be 0. (4) The object dispatcher returns to the workqueue processor, and in due course, this sees that the object's work item is still queued and invokes it again. (5) The current state is a work state (OBJECT_DEAD), so the dispatcher jumps to it - resulting in an OOPS. When I'm seeing this, the work state in (1) appears to have been either LOOK_UP_OBJECT or CREATE_OBJECT (object->oob_table is fscache_osm_lookup_oob). The window for (2) is very small: (A) object->event_mask is cleared whilst the event dispatch process is underway - though there's no memory barrier to force this to the top of the function. The window, therefore is from the time the object was selected by the workqueue processor and made requeueable to the time the mask was cleared. (B) fscache_raise_event() will only queue the object if it manages to set the event bit and the corresponding event_mask bit was set. The enqueuement is then deferred slightly whilst we get a ref on the object and get the per-CPU variable for workqueue congestion. This slight deferral slightly increases the probability by allowing extra time for the workqueue to make the item requeueable. Handle this by giving the dead state a processor function and checking the for the dead state address rather than seeing if the processor function is address 0x2. The dead state processor function can then set a flag to indicate that it's occurred and give a warning if it occurs more than once per object. If this race occurs, an oops similar to the following is seen (note the RIP value): BUG: unable to handle kernel NULL pointer dereference at 0000000000000002 IP: [<0000000000000002>] 0x1 PGD 0 Oops: 0010 [#1] SMP Modules linked in: ... CPU: 17 PID: 16077 Comm: kworker/u48:9 Not tainted 3.10.0-327.18.2.el7.x86_64 #1 Hardware name: HP ProLiant DL380 Gen9/ProLiant DL380 Gen9, BIOS P89 12/27/2015 Workqueue: fscache_object fscache_object_work_func [fscache] task: ffff880302b63980 ti: ffff880717544000 task.ti: ffff880717544000 RIP: 0010:[<0000000000000002>] [<0000000000000002>] 0x1 RSP: 0018:ffff880717547df8 EFLAGS: 00010202 RAX: ffffffffa0368640 RBX: ffff880edf7a4480 RCX: dead000000200200 RDX: 0000000000000002 RSI: 00000000ffffffff RDI: ffff880edf7a4480 RBP: ffff880717547e18 R08: 0000000000000000 R09: dfc40a25cb3a4510 R10: dfc40a25cb3a4510 R11: 0000000000000400 R12: 0000000000000000 R13: ffff880edf7a4510 R14: ffff8817f6153400 R15: 0000000000000600 FS: 0000000000000000(0000) GS:ffff88181f420000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000002 CR3: 000000000194a000 CR4: 00000000001407e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Stack: ffffffffa0363695 ffff880edf7a4510 ffff88093f16f900 ffff8817faa4ec00 ffff880717547e60 ffffffff8109d5db 00000000faa4ec18 0000000000000000 ffff8817faa4ec18 ffff88093f16f930 ffff880302b63980 ffff88093f16f900 Call Trace: [] ? fscache_object_work_func+0xa5/0x200 [fscache] [] process_one_work+0x17b/0x470 [] worker_thread+0x21c/0x400 [] ? rescuer_thread+0x400/0x400 [] kthread+0xcf/0xe0 [] ? kthread_create_on_node+0x140/0x140 [] ret_from_fork+0x58/0x90 [] ? kthread_create_on_node+0x140/0x140 Signed-off-by: David Howells Acked-by: Jeremy McNicoll Tested-by: Frank Sorenson Tested-by: Benjamin Coddington Reviewed-by: Benjamin Coddington Signed-off-by: Al Viro --- fs/fscache/object.c | 26 ++++++++++++++++++++++++-- include/linux/fscache-cache.h | 1 + 2 files changed, 25 insertions(+), 2 deletions(-) diff --git a/fs/fscache/object.c b/fs/fscache/object.c index be02a086ed9b..7a182c87f378 100644 --- a/fs/fscache/object.c +++ b/fs/fscache/object.c @@ -30,6 +30,7 @@ static const struct fscache_state *fscache_look_up_object(struct fscache_object static const struct fscache_state *fscache_object_available(struct fscache_object *, int); static const struct fscache_state *fscache_parent_ready(struct fscache_object *, int); static const struct fscache_state *fscache_update_object(struct fscache_object *, int); +static const struct fscache_state *fscache_object_dead(struct fscache_object *, int); #define __STATE_NAME(n) fscache_osm_##n #define STATE(n) (&__STATE_NAME(n)) @@ -91,7 +92,7 @@ static WORK_STATE(LOOKUP_FAILURE, "LCFL", fscache_lookup_failure); static WORK_STATE(KILL_OBJECT, "KILL", fscache_kill_object); static WORK_STATE(KILL_DEPENDENTS, "KDEP", fscache_kill_dependents); static WORK_STATE(DROP_OBJECT, "DROP", fscache_drop_object); -static WORK_STATE(OBJECT_DEAD, "DEAD", (void*)2UL); +static WORK_STATE(OBJECT_DEAD, "DEAD", fscache_object_dead); static WAIT_STATE(WAIT_FOR_INIT, "?INI", TRANSIT_TO(INIT_OBJECT, 1 << FSCACHE_OBJECT_EV_NEW_CHILD)); @@ -229,6 +230,10 @@ execute_work_state: event = -1; if (new_state == NO_TRANSIT) { _debug("{OBJ%x} %s notrans", object->debug_id, state->name); + if (unlikely(state == STATE(OBJECT_DEAD))) { + _leave(" [dead]"); + return; + } fscache_enqueue_object(object); event_mask = object->oob_event_mask; goto unmask_events; @@ -239,7 +244,7 @@ execute_work_state: object->state = state = new_state; if (state->work) { - if (unlikely(state->work == ((void *)2UL))) { + if (unlikely(state == STATE(OBJECT_DEAD))) { _leave(" [dead]"); return; } @@ -1083,3 +1088,20 @@ void fscache_object_mark_killed(struct fscache_object *object, } } EXPORT_SYMBOL(fscache_object_mark_killed); + +/* + * The object is dead. We can get here if an object gets queued by an event + * that would lead to its death (such as EV_KILL) when the dispatcher is + * already running (and so can be requeued) but hasn't yet cleared the event + * mask. + */ +static const struct fscache_state *fscache_object_dead(struct fscache_object *object, + int event) +{ + if (!test_and_set_bit(FSCACHE_OBJECT_RUN_AFTER_DEAD, + &object->flags)) + return NO_TRANSIT; + + WARN(true, "FS-Cache object redispatched after death"); + return NO_TRANSIT; +} diff --git a/include/linux/fscache-cache.h b/include/linux/fscache-cache.h index 13ba552e6c09..4c467ef50159 100644 --- a/include/linux/fscache-cache.h +++ b/include/linux/fscache-cache.h @@ -360,6 +360,7 @@ struct fscache_object { #define FSCACHE_OBJECT_IS_AVAILABLE 5 /* T if object has become active */ #define FSCACHE_OBJECT_RETIRED 6 /* T if object was retired on relinquishment */ #define FSCACHE_OBJECT_KILLED_BY_CACHE 7 /* T if object was killed by the cache */ +#define FSCACHE_OBJECT_RUN_AFTER_DEAD 8 /* T if object has been dispatched after death */ struct list_head cache_link; /* link in cache->object_list */ struct hlist_node cookie_link; /* link in cookie->backing_objects */ -- cgit v1.2.3 From aaaec6fc755447a1d056765b11b24d8ff2b81366 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 31 Jan 2017 19:03:21 +0100 Subject: x86/irq: Make irq activate operations symmetric The recent commit which prevents double activation of interrupts unearthed interesting code in x86. The code (ab)uses irq_domain_activate_irq() to reconfigure an already activated interrupt. That trips over the prevention code now. Fix it by deactivating the interrupt before activating the new configuration. Fixes: 08d85f3ea99f1 "irqdomain: Avoid activating interrupts more than once" Reported-and-tested-by: Mike Galbraith Reported-and-tested-by: Borislav Petkov Signed-off-by: Thomas Gleixner Cc: Andrey Ryabinin Cc: Marc Zyngier Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/alpine.DEB.2.20.1701311901580.3457@nanos --- arch/x86/kernel/apic/io_apic.c | 2 ++ arch/x86/kernel/hpet.c | 1 + 2 files changed, 3 insertions(+) diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 1e35dd06b090..52f352b063fd 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -2117,6 +2117,7 @@ static inline void __init check_timer(void) if (idx != -1 && irq_trigger(idx)) unmask_ioapic_irq(irq_get_chip_data(0)); } + irq_domain_deactivate_irq(irq_data); irq_domain_activate_irq(irq_data); if (timer_irq_works()) { if (disable_timer_pin_1 > 0) @@ -2138,6 +2139,7 @@ static inline void __init check_timer(void) * legacy devices should be connected to IO APIC #0 */ replace_pin_at_irq_node(data, node, apic1, pin1, apic2, pin2); + irq_domain_deactivate_irq(irq_data); irq_domain_activate_irq(irq_data); legacy_pic->unmask(0); if (timer_irq_works()) { diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 85e87b46c318..dc6ba5bda9fc 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -352,6 +352,7 @@ static int hpet_resume(struct clock_event_device *evt, int timer) } else { struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt); + irq_domain_deactivate_irq(irq_get_irq_data(hdev->irq)); irq_domain_activate_irq(irq_get_irq_data(hdev->irq)); disable_irq(hdev->irq); irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu)); -- cgit v1.2.3 From 0becc0ae5b42828785b589f686725ff5bc3b9b25 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 31 Jan 2017 09:37:34 +0100 Subject: x86/mce: Make timer handling more robust Erik reported that on a preproduction hardware a CMCI storm triggers the BUG_ON in add_timer_on(). The reason is that the per CPU MCE timer is started by the CMCI logic before the MCE CPU hotplug callback starts the timer with add_timer_on(). So the timer is already queued which triggers the BUG. Using add_timer_on() is pretty pointless in this code because the timer is strictlty per CPU, initialized as pinned and all operations which arm the timer happen on the CPU to which the timer belongs. Simplify the whole machinery by using mod_timer() instead of add_timer_on() which avoids the problem because mod_timer() can handle already queued timers. Use __start_timer() everywhere so the earliest armed expiry time is preserved. Reported-by: Erik Veijola Tested-by: Borislav Petkov Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Cc: Tony Luck Link: http://lkml.kernel.org/r/alpine.DEB.2.20.1701310936080.3457@nanos Signed-off-by: Thomas Gleixner --- arch/x86/kernel/cpu/mcheck/mce.c | 31 ++++++++++++------------------- 1 file changed, 12 insertions(+), 19 deletions(-) diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 00ef43233e03..537c6647d84c 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1373,20 +1373,15 @@ static unsigned long mce_adjust_timer_default(unsigned long interval) static unsigned long (*mce_adjust_timer)(unsigned long interval) = mce_adjust_timer_default; -static void __restart_timer(struct timer_list *t, unsigned long interval) +static void __start_timer(struct timer_list *t, unsigned long interval) { unsigned long when = jiffies + interval; unsigned long flags; local_irq_save(flags); - if (timer_pending(t)) { - if (time_before(when, t->expires)) - mod_timer(t, when); - } else { - t->expires = round_jiffies(when); - add_timer_on(t, smp_processor_id()); - } + if (!timer_pending(t) || time_before(when, t->expires)) + mod_timer(t, round_jiffies(when)); local_irq_restore(flags); } @@ -1421,7 +1416,7 @@ static void mce_timer_fn(unsigned long data) done: __this_cpu_write(mce_next_interval, iv); - __restart_timer(t, iv); + __start_timer(t, iv); } /* @@ -1432,7 +1427,7 @@ void mce_timer_kick(unsigned long interval) struct timer_list *t = this_cpu_ptr(&mce_timer); unsigned long iv = __this_cpu_read(mce_next_interval); - __restart_timer(t, interval); + __start_timer(t, interval); if (interval < iv) __this_cpu_write(mce_next_interval, interval); @@ -1779,17 +1774,15 @@ static void __mcheck_cpu_clear_vendor(struct cpuinfo_x86 *c) } } -static void mce_start_timer(unsigned int cpu, struct timer_list *t) +static void mce_start_timer(struct timer_list *t) { unsigned long iv = check_interval * HZ; if (mca_cfg.ignore_ce || !iv) return; - per_cpu(mce_next_interval, cpu) = iv; - - t->expires = round_jiffies(jiffies + iv); - add_timer_on(t, cpu); + this_cpu_write(mce_next_interval, iv); + __start_timer(t, iv); } static void __mcheck_cpu_setup_timer(void) @@ -1806,7 +1799,7 @@ static void __mcheck_cpu_init_timer(void) unsigned int cpu = smp_processor_id(); setup_pinned_timer(t, mce_timer_fn, cpu); - mce_start_timer(cpu, t); + mce_start_timer(t); } /* Handle unconfigured int18 (should never happen) */ @@ -2566,7 +2559,7 @@ static int mce_cpu_dead(unsigned int cpu) static int mce_cpu_online(unsigned int cpu) { - struct timer_list *t = &per_cpu(mce_timer, cpu); + struct timer_list *t = this_cpu_ptr(&mce_timer); int ret; mce_device_create(cpu); @@ -2577,13 +2570,13 @@ static int mce_cpu_online(unsigned int cpu) return ret; } mce_reenable_cpu(); - mce_start_timer(cpu, t); + mce_start_timer(t); return 0; } static int mce_cpu_pre_down(unsigned int cpu) { - struct timer_list *t = &per_cpu(mce_timer, cpu); + struct timer_list *t = this_cpu_ptr(&mce_timer); mce_disable_cpu(); del_timer_sync(t); -- cgit v1.2.3 From 4b3e6f2ef3722f1a6a97b6034ed492c1a21fd4ae Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Tue, 31 Jan 2017 18:35:37 -0800 Subject: xtensa: fix noMMU build on cores with MMU Commit bf15f86b343ed8 ("xtensa: initialize MMU before jumping to reset vector") calls MMU management functions even when CONFIG_MMU is not selected. That breaks noMMU build on cores with MMU. Don't manage MMU when CONFIG_MMU is not selected. Cc: stable@vger.kernel.org Signed-off-by: Max Filippov --- arch/xtensa/kernel/setup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/xtensa/kernel/setup.c b/arch/xtensa/kernel/setup.c index 88a044af7504..32cdc2c52e98 100644 --- a/arch/xtensa/kernel/setup.c +++ b/arch/xtensa/kernel/setup.c @@ -540,7 +540,7 @@ subsys_initcall(topology_init); void cpu_reset(void) { -#if XCHAL_HAVE_PTP_MMU +#if XCHAL_HAVE_PTP_MMU && IS_ENABLED(CONFIG_MMU) local_irq_disable(); /* * We have full MMU: all autoload ways, ways 7, 8 and 9 of DTLB must -- cgit v1.2.3 From dd86e373e09fb16b83e8adf5c48c421a4ca76468 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 31 Jan 2017 23:58:38 +0100 Subject: perf/x86/intel/rapl: Make package handling more robust The package management code in RAPL relies on package mapping being available before a CPU is started. This changed with: 9d85eb9119f4 ("x86/smpboot: Make logical package management more robust") because the ACPI/BIOS information turned out to be unreliable, but that left RAPL in broken state. This was not noticed because on a regular boot all CPUs are online before RAPL is initialized. A possible fix would be to reintroduce the mess which allocates a package data structure in CPU prepare and when it turns out to already exist in starting throw it away later in the CPU online callback. But that's a horrible hack and not required at all because RAPL becomes functional for perf only in the CPU online callback. That's correct because user space is not yet informed about the CPU being onlined, so nothing caan rely on RAPL being available on that particular CPU. Move the allocation to the CPU online callback and simplify the hotplug handling. At this point the package mapping is established and correct. This also adds a missing check for available package data in the event_init() function. Reported-by: Yasuaki Ishimatsu Signed-off-by: Thomas Gleixner Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Sebastian Siewior Cc: Stephane Eranian Cc: Vince Weaver Fixes: 9d85eb9119f4 ("x86/smpboot: Make logical package management more robust") Link: http://lkml.kernel.org/r/20170131230141.212593966@linutronix.de Signed-off-by: Ingo Molnar --- arch/x86/events/intel/rapl.c | 60 +++++++++++++++++++------------------------- include/linux/cpuhotplug.h | 1 - 2 files changed, 26 insertions(+), 35 deletions(-) diff --git a/arch/x86/events/intel/rapl.c b/arch/x86/events/intel/rapl.c index 17c3564d087a..22ef4f72cf32 100644 --- a/arch/x86/events/intel/rapl.c +++ b/arch/x86/events/intel/rapl.c @@ -161,7 +161,13 @@ static u64 rapl_timer_ms; static inline struct rapl_pmu *cpu_to_rapl_pmu(unsigned int cpu) { - return rapl_pmus->pmus[topology_logical_package_id(cpu)]; + unsigned int pkgid = topology_logical_package_id(cpu); + + /* + * The unsigned check also catches the '-1' return value for non + * existent mappings in the topology map. + */ + return pkgid < rapl_pmus->maxpkg ? rapl_pmus->pmus[pkgid] : NULL; } static inline u64 rapl_read_counter(struct perf_event *event) @@ -402,6 +408,8 @@ static int rapl_pmu_event_init(struct perf_event *event) /* must be done before validate_group */ pmu = cpu_to_rapl_pmu(event->cpu); + if (!pmu) + return -EINVAL; event->cpu = pmu->cpu; event->pmu_private = pmu; event->hw.event_base = msr; @@ -585,6 +593,20 @@ static int rapl_cpu_online(unsigned int cpu) struct rapl_pmu *pmu = cpu_to_rapl_pmu(cpu); int target; + if (!pmu) { + pmu = kzalloc_node(sizeof(*pmu), GFP_KERNEL, cpu_to_node(cpu)); + if (!pmu) + return -ENOMEM; + + raw_spin_lock_init(&pmu->lock); + INIT_LIST_HEAD(&pmu->active_list); + pmu->pmu = &rapl_pmus->pmu; + pmu->timer_interval = ms_to_ktime(rapl_timer_ms); + rapl_hrtimer_init(pmu); + + rapl_pmus->pmus[topology_logical_package_id(cpu)] = pmu; + } + /* * Check if there is an online cpu in the package which collects rapl * events already. @@ -598,27 +620,6 @@ static int rapl_cpu_online(unsigned int cpu) return 0; } -static int rapl_cpu_prepare(unsigned int cpu) -{ - struct rapl_pmu *pmu = cpu_to_rapl_pmu(cpu); - - if (pmu) - return 0; - - pmu = kzalloc_node(sizeof(*pmu), GFP_KERNEL, cpu_to_node(cpu)); - if (!pmu) - return -ENOMEM; - - raw_spin_lock_init(&pmu->lock); - INIT_LIST_HEAD(&pmu->active_list); - pmu->pmu = &rapl_pmus->pmu; - pmu->timer_interval = ms_to_ktime(rapl_timer_ms); - pmu->cpu = -1; - rapl_hrtimer_init(pmu); - rapl_pmus->pmus[topology_logical_package_id(cpu)] = pmu; - return 0; -} - static int rapl_check_hw_unit(bool apply_quirk) { u64 msr_rapl_power_unit_bits; @@ -803,29 +804,21 @@ static int __init rapl_pmu_init(void) /* * Install callbacks. Core will call them for each online cpu. */ - - ret = cpuhp_setup_state(CPUHP_PERF_X86_RAPL_PREP, "perf/x86/rapl:prepare", - rapl_cpu_prepare, NULL); - if (ret) - goto out; - ret = cpuhp_setup_state(CPUHP_AP_PERF_X86_RAPL_ONLINE, "perf/x86/rapl:online", rapl_cpu_online, rapl_cpu_offline); if (ret) - goto out1; + goto out; ret = perf_pmu_register(&rapl_pmus->pmu, "power", -1); if (ret) - goto out2; + goto out1; rapl_advertise(); return 0; -out2: - cpuhp_remove_state(CPUHP_AP_PERF_X86_RAPL_ONLINE); out1: - cpuhp_remove_state(CPUHP_PERF_X86_RAPL_PREP); + cpuhp_remove_state(CPUHP_AP_PERF_X86_RAPL_ONLINE); out: pr_warn("Initialization failed (%d), disabled\n", ret); cleanup_rapl_pmus(); @@ -836,7 +829,6 @@ module_init(rapl_pmu_init); static void __exit intel_rapl_exit(void) { cpuhp_remove_state_nocalls(CPUHP_AP_PERF_X86_RAPL_ONLINE); - cpuhp_remove_state_nocalls(CPUHP_PERF_X86_RAPL_PREP); perf_pmu_unregister(&rapl_pmus->pmu); cleanup_rapl_pmus(); } diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h index d936a0021839..8329f3dc592c 100644 --- a/include/linux/cpuhotplug.h +++ b/include/linux/cpuhotplug.h @@ -10,7 +10,6 @@ enum cpuhp_state { CPUHP_PERF_X86_PREPARE, CPUHP_PERF_X86_UNCORE_PREP, CPUHP_PERF_X86_AMD_UNCORE_PREP, - CPUHP_PERF_X86_RAPL_PREP, CPUHP_PERF_BFIN, CPUHP_PERF_POWER, CPUHP_PERF_SUPERH, -- cgit v1.2.3 From 1aa6cfd33df492939b0be15ebdbcff1f8ae5ddb6 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 31 Jan 2017 23:58:39 +0100 Subject: perf/x86/intel/uncore: Clean up hotplug conversion fallout The recent conversion to the hotplug state machine kept two mechanisms from the original code: 1) The first_init logic which adds the number of online CPUs in a package to the refcount. That's wrong because the callbacks are executed for all online CPUs. Remove it so the refcounting is correct. 2) The on_each_cpu() call to undo box->init() in the error handling path. That's bogus because when the prepare callback fails no box has been initialized yet. Remove it. Signed-off-by: Thomas Gleixner Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Sebastian Siewior Cc: Stephane Eranian Cc: Vince Weaver Cc: Yasuaki Ishimatsu Fixes: 1a246b9f58c6 ("perf/x86/intel/uncore: Convert to hotplug state machine") Link: http://lkml.kernel.org/r/20170131230141.298032324@linutronix.de Signed-off-by: Ingo Molnar --- arch/x86/events/intel/uncore.c | 44 ++++-------------------------------------- 1 file changed, 4 insertions(+), 40 deletions(-) diff --git a/arch/x86/events/intel/uncore.c b/arch/x86/events/intel/uncore.c index 8c4ccdc3a3f3..56c5235dcc29 100644 --- a/arch/x86/events/intel/uncore.c +++ b/arch/x86/events/intel/uncore.c @@ -764,30 +764,6 @@ static void uncore_pmu_unregister(struct intel_uncore_pmu *pmu) pmu->registered = false; } -static void __uncore_exit_boxes(struct intel_uncore_type *type, int cpu) -{ - struct intel_uncore_pmu *pmu = type->pmus; - struct intel_uncore_box *box; - int i, pkg; - - if (pmu) { - pkg = topology_physical_package_id(cpu); - for (i = 0; i < type->num_boxes; i++, pmu++) { - box = pmu->boxes[pkg]; - if (box) - uncore_box_exit(box); - } - } -} - -static void uncore_exit_boxes(void *dummy) -{ - struct intel_uncore_type **types; - - for (types = uncore_msr_uncores; *types; types++) - __uncore_exit_boxes(*types++, smp_processor_id()); -} - static void uncore_free_boxes(struct intel_uncore_pmu *pmu) { int pkg; @@ -1078,22 +1054,12 @@ static int uncore_cpu_dying(unsigned int cpu) return 0; } -static int first_init; - static int uncore_cpu_starting(unsigned int cpu) { struct intel_uncore_type *type, **types = uncore_msr_uncores; struct intel_uncore_pmu *pmu; struct intel_uncore_box *box; - int i, pkg, ncpus = 1; - - if (first_init) { - /* - * On init we get the number of online cpus in the package - * and set refcount for all of them. - */ - ncpus = cpumask_weight(topology_core_cpumask(cpu)); - } + int i, pkg; pkg = topology_logical_package_id(cpu); for (; *types; types++) { @@ -1104,7 +1070,7 @@ static int uncore_cpu_starting(unsigned int cpu) if (!box) continue; /* The first cpu on a package activates the box */ - if (atomic_add_return(ncpus, &box->refcnt) == ncpus) + if (atomic_inc_return(&box->refcnt) == 1) uncore_box_init(box); } } @@ -1408,19 +1374,17 @@ static int __init intel_uncore_init(void) "perf/x86/intel/uncore:prepare", uncore_cpu_prepare, NULL); } - first_init = 1; + cpuhp_setup_state(CPUHP_AP_PERF_X86_UNCORE_STARTING, "perf/x86/uncore:starting", uncore_cpu_starting, uncore_cpu_dying); - first_init = 0; + cpuhp_setup_state(CPUHP_AP_PERF_X86_UNCORE_ONLINE, "perf/x86/uncore:online", uncore_event_cpu_online, uncore_event_cpu_offline); return 0; err: - /* Undo box->init_box() */ - on_each_cpu_mask(&uncore_cpu_mask, uncore_exit_boxes, NULL, 1); uncore_types_exit(uncore_msr_uncores); uncore_pci_exit(); return ret; -- cgit v1.2.3 From fff4b87e594ad3d2e4f51e8d3d86a6f9d3d8b654 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 31 Jan 2017 23:58:40 +0100 Subject: perf/x86/intel/uncore: Make package handling more robust The package management code in uncore relies on package mapping being available before a CPU is started. This changed with: 9d85eb9119f4 ("x86/smpboot: Make logical package management more robust") because the ACPI/BIOS information turned out to be unreliable, but that left uncore in broken state. This was not noticed because on a regular boot all CPUs are online before uncore is initialized. Move the allocation to the CPU online callback and simplify the hotplug handling. At this point the package mapping is established and correct. Signed-off-by: Thomas Gleixner Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Sebastian Siewior Cc: Stephane Eranian Cc: Vince Weaver Cc: Yasuaki Ishimatsu Fixes: 9d85eb9119f4 ("x86/smpboot: Make logical package management more robust") Link: http://lkml.kernel.org/r/20170131230141.377156255@linutronix.de Signed-off-by: Ingo Molnar --- arch/x86/events/intel/uncore.c | 196 +++++++++++++++++++---------------------- include/linux/cpuhotplug.h | 2 - 2 files changed, 91 insertions(+), 107 deletions(-) diff --git a/arch/x86/events/intel/uncore.c b/arch/x86/events/intel/uncore.c index 56c5235dcc29..1ab45976474d 100644 --- a/arch/x86/events/intel/uncore.c +++ b/arch/x86/events/intel/uncore.c @@ -100,7 +100,13 @@ ssize_t uncore_event_show(struct kobject *kobj, struct intel_uncore_box *uncore_pmu_to_box(struct intel_uncore_pmu *pmu, int cpu) { - return pmu->boxes[topology_logical_package_id(cpu)]; + unsigned int pkgid = topology_logical_package_id(cpu); + + /* + * The unsigned check also catches the '-1' return value for non + * existent mappings in the topology map. + */ + return pkgid < max_packages ? pmu->boxes[pkgid] : NULL; } u64 uncore_msr_read_counter(struct intel_uncore_box *box, struct perf_event *event) @@ -1034,76 +1040,6 @@ static void uncore_pci_exit(void) } } -static int uncore_cpu_dying(unsigned int cpu) -{ - struct intel_uncore_type *type, **types = uncore_msr_uncores; - struct intel_uncore_pmu *pmu; - struct intel_uncore_box *box; - int i, pkg; - - pkg = topology_logical_package_id(cpu); - for (; *types; types++) { - type = *types; - pmu = type->pmus; - for (i = 0; i < type->num_boxes; i++, pmu++) { - box = pmu->boxes[pkg]; - if (box && atomic_dec_return(&box->refcnt) == 0) - uncore_box_exit(box); - } - } - return 0; -} - -static int uncore_cpu_starting(unsigned int cpu) -{ - struct intel_uncore_type *type, **types = uncore_msr_uncores; - struct intel_uncore_pmu *pmu; - struct intel_uncore_box *box; - int i, pkg; - - pkg = topology_logical_package_id(cpu); - for (; *types; types++) { - type = *types; - pmu = type->pmus; - for (i = 0; i < type->num_boxes; i++, pmu++) { - box = pmu->boxes[pkg]; - if (!box) - continue; - /* The first cpu on a package activates the box */ - if (atomic_inc_return(&box->refcnt) == 1) - uncore_box_init(box); - } - } - - return 0; -} - -static int uncore_cpu_prepare(unsigned int cpu) -{ - struct intel_uncore_type *type, **types = uncore_msr_uncores; - struct intel_uncore_pmu *pmu; - struct intel_uncore_box *box; - int i, pkg; - - pkg = topology_logical_package_id(cpu); - for (; *types; types++) { - type = *types; - pmu = type->pmus; - for (i = 0; i < type->num_boxes; i++, pmu++) { - if (pmu->boxes[pkg]) - continue; - /* First cpu of a package allocates the box */ - box = uncore_alloc_box(type, cpu_to_node(cpu)); - if (!box) - return -ENOMEM; - box->pmu = pmu; - box->pkgid = pkg; - pmu->boxes[pkg] = box; - } - } - return 0; -} - static void uncore_change_type_ctx(struct intel_uncore_type *type, int old_cpu, int new_cpu) { @@ -1143,12 +1079,14 @@ static void uncore_change_context(struct intel_uncore_type **uncores, static int uncore_event_cpu_offline(unsigned int cpu) { - int target; + struct intel_uncore_type *type, **types = uncore_msr_uncores; + struct intel_uncore_pmu *pmu; + struct intel_uncore_box *box; + int i, pkg, target; /* Check if exiting cpu is used for collecting uncore events */ if (!cpumask_test_and_clear_cpu(cpu, &uncore_cpu_mask)) - return 0; - + goto unref; /* Find a new cpu to collect uncore events */ target = cpumask_any_but(topology_core_cpumask(cpu), cpu); @@ -1160,12 +1098,82 @@ static int uncore_event_cpu_offline(unsigned int cpu) uncore_change_context(uncore_msr_uncores, cpu, target); uncore_change_context(uncore_pci_uncores, cpu, target); + +unref: + /* Clear the references */ + pkg = topology_logical_package_id(cpu); + for (; *types; types++) { + type = *types; + pmu = type->pmus; + for (i = 0; i < type->num_boxes; i++, pmu++) { + box = pmu->boxes[pkg]; + if (box && atomic_dec_return(&box->refcnt) == 0) + uncore_box_exit(box); + } + } return 0; } +static int allocate_boxes(struct intel_uncore_type **types, + unsigned int pkg, unsigned int cpu) +{ + struct intel_uncore_box *box, *tmp; + struct intel_uncore_type *type; + struct intel_uncore_pmu *pmu; + LIST_HEAD(allocated); + int i; + + /* Try to allocate all required boxes */ + for (; *types; types++) { + type = *types; + pmu = type->pmus; + for (i = 0; i < type->num_boxes; i++, pmu++) { + if (pmu->boxes[pkg]) + continue; + box = uncore_alloc_box(type, cpu_to_node(cpu)); + if (!box) + goto cleanup; + box->pmu = pmu; + box->pkgid = pkg; + list_add(&box->active_list, &allocated); + } + } + /* Install them in the pmus */ + list_for_each_entry_safe(box, tmp, &allocated, active_list) { + list_del_init(&box->active_list); + box->pmu->boxes[pkg] = box; + } + return 0; + +cleanup: + list_for_each_entry_safe(box, tmp, &allocated, active_list) { + list_del_init(&box->active_list); + kfree(box); + } + return -ENOMEM; +} + static int uncore_event_cpu_online(unsigned int cpu) { - int target; + struct intel_uncore_type *type, **types = uncore_msr_uncores; + struct intel_uncore_pmu *pmu; + struct intel_uncore_box *box; + int i, ret, pkg, target; + + pkg = topology_logical_package_id(cpu); + ret = allocate_boxes(types, pkg, cpu); + if (ret) + return ret; + + for (; *types; types++) { + type = *types; + pmu = type->pmus; + for (i = 0; i < type->num_boxes; i++, pmu++) { + box = pmu->boxes[pkg]; + if (!box && atomic_inc_return(&box->refcnt) == 1) + uncore_box_init(box); + } + } /* * Check if there is an online cpu in the package @@ -1355,33 +1363,13 @@ static int __init intel_uncore_init(void) if (cret && pret) return -ENODEV; - /* - * Install callbacks. Core will call them for each online cpu. - * - * The first online cpu of each package allocates and takes - * the refcounts for all other online cpus in that package. - * If msrs are not enabled no allocation is required and - * uncore_cpu_prepare() is not called for each online cpu. - */ - if (!cret) { - ret = cpuhp_setup_state(CPUHP_PERF_X86_UNCORE_PREP, - "perf/x86/intel/uncore:prepare", - uncore_cpu_prepare, NULL); - if (ret) - goto err; - } else { - cpuhp_setup_state_nocalls(CPUHP_PERF_X86_UNCORE_PREP, - "perf/x86/intel/uncore:prepare", - uncore_cpu_prepare, NULL); - } - - cpuhp_setup_state(CPUHP_AP_PERF_X86_UNCORE_STARTING, - "perf/x86/uncore:starting", - uncore_cpu_starting, uncore_cpu_dying); - - cpuhp_setup_state(CPUHP_AP_PERF_X86_UNCORE_ONLINE, - "perf/x86/uncore:online", - uncore_event_cpu_online, uncore_event_cpu_offline); + /* Install hotplug callbacks to setup the targets for each package */ + ret = cpuhp_setup_state(CPUHP_AP_PERF_X86_UNCORE_ONLINE, + "perf/x86/intel/uncore:online", + uncore_event_cpu_online, + uncore_event_cpu_offline); + if (ret) + goto err; return 0; err: @@ -1393,9 +1381,7 @@ module_init(intel_uncore_init); static void __exit intel_uncore_exit(void) { - cpuhp_remove_state_nocalls(CPUHP_AP_PERF_X86_UNCORE_ONLINE); - cpuhp_remove_state_nocalls(CPUHP_AP_PERF_X86_UNCORE_STARTING); - cpuhp_remove_state_nocalls(CPUHP_PERF_X86_UNCORE_PREP); + cpuhp_remove_state(CPUHP_AP_PERF_X86_UNCORE_ONLINE); uncore_types_exit(uncore_msr_uncores); uncore_pci_exit(); } diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h index 8329f3dc592c..921acaaa1601 100644 --- a/include/linux/cpuhotplug.h +++ b/include/linux/cpuhotplug.h @@ -8,7 +8,6 @@ enum cpuhp_state { CPUHP_CREATE_THREADS, CPUHP_PERF_PREPARE, CPUHP_PERF_X86_PREPARE, - CPUHP_PERF_X86_UNCORE_PREP, CPUHP_PERF_X86_AMD_UNCORE_PREP, CPUHP_PERF_BFIN, CPUHP_PERF_POWER, @@ -85,7 +84,6 @@ enum cpuhp_state { CPUHP_AP_IRQ_ARMADA_XP_STARTING, CPUHP_AP_IRQ_BCM2836_STARTING, CPUHP_AP_ARM_MVEBU_COHERENCY, - CPUHP_AP_PERF_X86_UNCORE_STARTING, CPUHP_AP_PERF_X86_AMD_UNCORE_STARTING, CPUHP_AP_PERF_X86_STARTING, CPUHP_AP_PERF_X86_AMD_IBS_STARTING, -- cgit v1.2.3 From eeee74a4f6625b77c3e8db0693c2d4546507ba0d Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Tue, 31 Jan 2017 10:21:30 +0100 Subject: drm/atomic: Unconditionally call prepare_fb. Atomic drivers may set properties like rotation on the same fb, which may require a call to prepare_fb even when framebuffer stays identical. Instead of handling all the special cases in the core, let the driver decide when prepare_fb and cleanup_fb are noops. This is a revert of: commit fcc60b413d14dd06ddbd79ec50e83c4fb2a097ba Author: Keith Packard Date: Sat Jun 4 01:16:22 2016 -0700 drm: Don't prepare or cleanup unchanging frame buffers [v3] The original commit mentions that this prevents waiting in i915 on all previous rendering during cursor updates, but there are better ways to fix this. Signed-off-by: Maarten Lankhorst Acked-by: Laurent Pinchart Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/6d82f9b6-9d16-91d1-d176-4a37b09afc44@linux.intel.com (cherry picked from commit 0532be078a207d7dd6ad26ebd0834e258acc4ee7) Signed-off-by: Jani Nikula Link: http://patchwork.freedesktop.org/patch/msgid/1485854491-27389-2-git-send-email-maarten.lankhorst@linux.intel.com --- drivers/gpu/drm/drm_atomic_helper.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 34f757bcabae..4594477dee00 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -1666,9 +1666,6 @@ int drm_atomic_helper_prepare_planes(struct drm_device *dev, funcs = plane->helper_private; - if (!drm_atomic_helper_framebuffer_changed(dev, state, plane_state->crtc)) - continue; - if (funcs->prepare_fb) { ret = funcs->prepare_fb(plane, plane_state); if (ret) @@ -1685,9 +1682,6 @@ fail: if (j >= i) continue; - if (!drm_atomic_helper_framebuffer_changed(dev, state, plane_state->crtc)) - continue; - funcs = plane->helper_private; if (funcs->cleanup_fb) @@ -1954,9 +1948,6 @@ void drm_atomic_helper_cleanup_planes(struct drm_device *dev, for_each_plane_in_state(old_state, plane, plane_state, i) { const struct drm_plane_helper_funcs *funcs; - if (!drm_atomic_helper_framebuffer_changed(dev, old_state, plane_state->crtc)) - continue; - funcs = plane->helper_private; if (funcs->cleanup_fb) -- cgit v1.2.3 From e8fe4f4b2b7b93048729538321c681c0cff33b39 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Tue, 31 Jan 2017 10:21:31 +0100 Subject: drm/i915: Track pinned vma in intel_plane_state With atomic plane states we are able to track an allocation right from preparation, during use and through to the final free after being swapped out for a new plane. We can couple the VMA we pin for the framebuffer (and its rotation) to this lifetime and avoid all the clumsy lookups in between. v2: Remove residual vma on plane cleanup (Chris) v3: Add a description for the vma destruction in intel_plane_destroy_state (Maarten) References: https://bugs.freedesktop.org/show_bug.cgi?id=98829 Signed-off-by: Chris Wilson Signed-off-by: Maarten Lankhorst Link: http://patchwork.freedesktop.org/patch/msgid/20170116152131.18089-1-chris@chris-wilson.co.uk Acked-by: Joonas Lahtinen (cherry picked from commit be1e341513ca23b0668b7b0f26fa6e2ffc46ba20) Signed-off-by: Jani Nikula Link: http://patchwork.freedesktop.org/patch/msgid/1485854491-27389-3-git-send-email-maarten.lankhorst@linux.intel.com --- drivers/gpu/drm/i915/i915_drv.h | 16 +--- drivers/gpu/drm/i915/intel_atomic_plane.c | 20 +++++ drivers/gpu/drm/i915/intel_display.c | 125 ++++++++++-------------------- drivers/gpu/drm/i915/intel_drv.h | 9 ++- drivers/gpu/drm/i915/intel_fbc.c | 52 +++++-------- drivers/gpu/drm/i915/intel_fbdev.c | 4 +- drivers/gpu/drm/i915/intel_sprite.c | 8 +- 7 files changed, 99 insertions(+), 135 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 69bc3b0c4390..8493e19b563a 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1012,6 +1012,8 @@ struct intel_fbc { struct work_struct underrun_work; struct intel_fbc_state_cache { + struct i915_vma *vma; + struct { unsigned int mode_flags; uint32_t hsw_bdw_pixel_rate; @@ -1025,15 +1027,14 @@ struct intel_fbc { } plane; struct { - u64 ilk_ggtt_offset; uint32_t pixel_format; unsigned int stride; - int fence_reg; - unsigned int tiling_mode; } fb; } state_cache; struct intel_fbc_reg_params { + struct i915_vma *vma; + struct { enum pipe pipe; enum plane plane; @@ -1041,10 +1042,8 @@ struct intel_fbc { } crtc; struct { - u64 ggtt_offset; uint32_t pixel_format; unsigned int stride; - int fence_reg; } fb; int cfb_size; @@ -3168,13 +3167,6 @@ i915_gem_object_to_ggtt(struct drm_i915_gem_object *obj, return i915_gem_obj_to_vma(obj, &to_i915(obj->base.dev)->ggtt.base, view); } -static inline unsigned long -i915_gem_object_ggtt_offset(struct drm_i915_gem_object *o, - const struct i915_ggtt_view *view) -{ - return i915_ggtt_offset(i915_gem_object_to_ggtt(o, view)); -} - /* i915_gem_fence_reg.c */ int __must_check i915_vma_get_fence(struct i915_vma *vma); int __must_check i915_vma_put_fence(struct i915_vma *vma); diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.c b/drivers/gpu/drm/i915/intel_atomic_plane.c index dbe9fb41ae53..8d3e515f27ba 100644 --- a/drivers/gpu/drm/i915/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/intel_atomic_plane.c @@ -85,6 +85,8 @@ intel_plane_duplicate_state(struct drm_plane *plane) __drm_atomic_helper_plane_duplicate_state(plane, state); + intel_state->vma = NULL; + return state; } @@ -100,6 +102,24 @@ void intel_plane_destroy_state(struct drm_plane *plane, struct drm_plane_state *state) { + struct i915_vma *vma; + + vma = fetch_and_zero(&to_intel_plane_state(state)->vma); + + /* + * FIXME: Normally intel_cleanup_plane_fb handles destruction of vma. + * We currently don't clear all planes during driver unload, so we have + * to be able to unpin vma here for now. + * + * Normally this can only happen during unload when kmscon is disabled + * and userspace doesn't attempt to set a framebuffer at all. + */ + if (vma) { + mutex_lock(&plane->dev->struct_mutex); + intel_unpin_fb_vma(vma); + mutex_unlock(&plane->dev->struct_mutex); + } + drm_atomic_helper_plane_destroy_state(plane, state); } diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index f0b9aa7a0483..f1e4a21d4664 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2235,27 +2235,22 @@ intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb, unsigned int rotation) i915_vma_pin_fence(vma); } + i915_vma_get(vma); err: intel_runtime_pm_put(dev_priv); return vma; } -void intel_unpin_fb_obj(struct drm_framebuffer *fb, unsigned int rotation) +void intel_unpin_fb_vma(struct i915_vma *vma) { - struct drm_i915_gem_object *obj = intel_fb_obj(fb); - struct i915_ggtt_view view; - struct i915_vma *vma; - - WARN_ON(!mutex_is_locked(&obj->base.dev->struct_mutex)); - - intel_fill_fb_ggtt_view(&view, fb, rotation); - vma = i915_gem_object_to_ggtt(obj, &view); + lockdep_assert_held(&vma->vm->dev->struct_mutex); if (WARN_ON_ONCE(!vma)) return; i915_vma_unpin_fence(vma); i915_gem_object_unpin_from_display_plane(vma); + i915_vma_put(vma); } static int intel_fb_pitch(const struct drm_framebuffer *fb, int plane, @@ -2750,7 +2745,6 @@ intel_find_initial_plane_obj(struct intel_crtc *intel_crtc, struct drm_device *dev = intel_crtc->base.dev; struct drm_i915_private *dev_priv = to_i915(dev); struct drm_crtc *c; - struct intel_crtc *i; struct drm_i915_gem_object *obj; struct drm_plane *primary = intel_crtc->base.primary; struct drm_plane_state *plane_state = primary->state; @@ -2775,20 +2769,20 @@ intel_find_initial_plane_obj(struct intel_crtc *intel_crtc, * an fb with another CRTC instead */ for_each_crtc(dev, c) { - i = to_intel_crtc(c); + struct intel_plane_state *state; if (c == &intel_crtc->base) continue; - if (!i->active) + if (!to_intel_crtc(c)->active) continue; - fb = c->primary->fb; - if (!fb) + state = to_intel_plane_state(c->primary->state); + if (!state->vma) continue; - obj = intel_fb_obj(fb); - if (i915_gem_object_ggtt_offset(obj, NULL) == plane_config->base) { + if (intel_plane_ggtt_offset(state) == plane_config->base) { + fb = c->primary->fb; drm_framebuffer_reference(fb); goto valid_fb; } @@ -2809,6 +2803,19 @@ intel_find_initial_plane_obj(struct intel_crtc *intel_crtc, return; valid_fb: + mutex_lock(&dev->struct_mutex); + intel_state->vma = + intel_pin_and_fence_fb_obj(fb, primary->state->rotation); + mutex_unlock(&dev->struct_mutex); + if (IS_ERR(intel_state->vma)) { + DRM_ERROR("failed to pin boot fb on pipe %d: %li\n", + intel_crtc->pipe, PTR_ERR(intel_state->vma)); + + intel_state->vma = NULL; + drm_framebuffer_unreference(fb); + return; + } + plane_state->src_x = 0; plane_state->src_y = 0; plane_state->src_w = fb->width << 16; @@ -3104,13 +3111,13 @@ static void i9xx_update_primary_plane(struct drm_plane *primary, I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]); if (INTEL_GEN(dev_priv) >= 4) { I915_WRITE(DSPSURF(plane), - intel_fb_gtt_offset(fb, rotation) + + intel_plane_ggtt_offset(plane_state) + intel_crtc->dspaddr_offset); I915_WRITE(DSPTILEOFF(plane), (y << 16) | x); I915_WRITE(DSPLINOFF(plane), linear_offset); } else { I915_WRITE(DSPADDR(plane), - intel_fb_gtt_offset(fb, rotation) + + intel_plane_ggtt_offset(plane_state) + intel_crtc->dspaddr_offset); } POSTING_READ(reg); @@ -3207,7 +3214,7 @@ static void ironlake_update_primary_plane(struct drm_plane *primary, I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]); I915_WRITE(DSPSURF(plane), - intel_fb_gtt_offset(fb, rotation) + + intel_plane_ggtt_offset(plane_state) + intel_crtc->dspaddr_offset); if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) { I915_WRITE(DSPOFFSET(plane), (y << 16) | x); @@ -3230,23 +3237,6 @@ u32 intel_fb_stride_alignment(const struct drm_i915_private *dev_priv, } } -u32 intel_fb_gtt_offset(struct drm_framebuffer *fb, - unsigned int rotation) -{ - struct drm_i915_gem_object *obj = intel_fb_obj(fb); - struct i915_ggtt_view view; - struct i915_vma *vma; - - intel_fill_fb_ggtt_view(&view, fb, rotation); - - vma = i915_gem_object_to_ggtt(obj, &view); - if (WARN(!vma, "ggtt vma for display object not found! (view=%u)\n", - view.type)) - return -1; - - return i915_ggtt_offset(vma); -} - static void skl_detach_scaler(struct intel_crtc *intel_crtc, int id) { struct drm_device *dev = intel_crtc->base.dev; @@ -3441,7 +3431,7 @@ static void skylake_update_primary_plane(struct drm_plane *plane, } I915_WRITE(PLANE_SURF(pipe, 0), - intel_fb_gtt_offset(fb, rotation) + surf_addr); + intel_plane_ggtt_offset(plane_state) + surf_addr); POSTING_READ(PLANE_SURF(pipe, 0)); } @@ -11536,7 +11526,7 @@ static void intel_unpin_work_fn(struct work_struct *__work) flush_work(&work->mmio_work); mutex_lock(&dev->struct_mutex); - intel_unpin_fb_obj(work->old_fb, primary->state->rotation); + intel_unpin_fb_vma(work->old_vma); i915_gem_object_put(work->pending_flip_obj); mutex_unlock(&dev->struct_mutex); @@ -12246,8 +12236,10 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc, goto cleanup_pending; } - work->gtt_offset = intel_fb_gtt_offset(fb, primary->state->rotation); - work->gtt_offset += intel_crtc->dspaddr_offset; + work->old_vma = to_intel_plane_state(primary->state)->vma; + to_intel_plane_state(primary->state)->vma = vma; + + work->gtt_offset = i915_ggtt_offset(vma) + intel_crtc->dspaddr_offset; work->rotation = crtc->primary->state->rotation; /* @@ -12301,7 +12293,8 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc, cleanup_request: i915_add_request_no_flush(request); cleanup_unpin: - intel_unpin_fb_obj(fb, crtc->primary->state->rotation); + to_intel_plane_state(primary->state)->vma = work->old_vma; + intel_unpin_fb_vma(vma); cleanup_pending: atomic_dec(&intel_crtc->unpin_work_count); unlock: @@ -14794,6 +14787,8 @@ intel_prepare_plane_fb(struct drm_plane *plane, DRM_DEBUG_KMS("failed to pin object\n"); return PTR_ERR(vma); } + + to_intel_plane_state(new_state)->vma = vma; } return 0; @@ -14812,19 +14807,12 @@ void intel_cleanup_plane_fb(struct drm_plane *plane, struct drm_plane_state *old_state) { - struct drm_i915_private *dev_priv = to_i915(plane->dev); - struct intel_plane_state *old_intel_state; - struct drm_i915_gem_object *old_obj = intel_fb_obj(old_state->fb); - struct drm_i915_gem_object *obj = intel_fb_obj(plane->state->fb); - - old_intel_state = to_intel_plane_state(old_state); - - if (!obj && !old_obj) - return; + struct i915_vma *vma; - if (old_obj && (plane->type != DRM_PLANE_TYPE_CURSOR || - !INTEL_INFO(dev_priv)->cursor_needs_physical)) - intel_unpin_fb_obj(old_state->fb, old_state->rotation); + /* Should only be called after a successful intel_prepare_plane_fb()! */ + vma = fetch_and_zero(&to_intel_plane_state(old_state)->vma); + if (vma) + intel_unpin_fb_vma(vma); } int @@ -15166,7 +15154,7 @@ intel_update_cursor_plane(struct drm_plane *plane, if (!obj) addr = 0; else if (!INTEL_INFO(dev_priv)->cursor_needs_physical) - addr = i915_gem_object_ggtt_offset(obj, NULL); + addr = intel_plane_ggtt_offset(state); else addr = obj->phys_handle->busaddr; @@ -17066,41 +17054,12 @@ void intel_display_resume(struct drm_device *dev) void intel_modeset_gem_init(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); - struct drm_crtc *c; - struct drm_i915_gem_object *obj; intel_init_gt_powersave(dev_priv); intel_modeset_init_hw(dev); intel_setup_overlay(dev_priv); - - /* - * Make sure any fbs we allocated at startup are properly - * pinned & fenced. When we do the allocation it's too early - * for this. - */ - for_each_crtc(dev, c) { - struct i915_vma *vma; - - obj = intel_fb_obj(c->primary->fb); - if (obj == NULL) - continue; - - mutex_lock(&dev->struct_mutex); - vma = intel_pin_and_fence_fb_obj(c->primary->fb, - c->primary->state->rotation); - mutex_unlock(&dev->struct_mutex); - if (IS_ERR(vma)) { - DRM_ERROR("failed to pin boot fb on pipe %d\n", - to_intel_crtc(c)->pipe); - drm_framebuffer_unreference(c->primary->fb); - c->primary->fb = NULL; - c->primary->crtc = c->primary->state->crtc = NULL; - update_state_fb(c->primary); - c->state->plane_mask &= ~(1 << drm_plane_index(c->primary)); - } - } } int intel_connector_register(struct drm_connector *connector) diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index cd72ae171eeb..03a2112004f9 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -377,6 +377,7 @@ struct intel_atomic_state { struct intel_plane_state { struct drm_plane_state base; struct drm_rect clip; + struct i915_vma *vma; struct { u32 offset; @@ -1046,6 +1047,7 @@ struct intel_flip_work { struct work_struct mmio_work; struct drm_crtc *crtc; + struct i915_vma *old_vma; struct drm_framebuffer *old_fb; struct drm_i915_gem_object *pending_flip_obj; struct drm_pending_vblank_event *event; @@ -1273,7 +1275,7 @@ void intel_release_load_detect_pipe(struct drm_connector *connector, struct drm_modeset_acquire_ctx *ctx); struct i915_vma * intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb, unsigned int rotation); -void intel_unpin_fb_obj(struct drm_framebuffer *fb, unsigned int rotation); +void intel_unpin_fb_vma(struct i915_vma *vma); struct drm_framebuffer * __intel_framebuffer_create(struct drm_device *dev, struct drm_mode_fb_cmd2 *mode_cmd, @@ -1362,7 +1364,10 @@ void intel_mode_from_pipe_config(struct drm_display_mode *mode, int skl_update_scaler_crtc(struct intel_crtc_state *crtc_state); int skl_max_scale(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state); -u32 intel_fb_gtt_offset(struct drm_framebuffer *fb, unsigned int rotation); +static inline u32 intel_plane_ggtt_offset(const struct intel_plane_state *state) +{ + return i915_ggtt_offset(state->vma); +} u32 skl_plane_ctl_format(uint32_t pixel_format); u32 skl_plane_ctl_tiling(uint64_t fb_modifier); diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c index 62f215b12eb5..f3a1d6a5cabe 100644 --- a/drivers/gpu/drm/i915/intel_fbc.c +++ b/drivers/gpu/drm/i915/intel_fbc.c @@ -173,7 +173,7 @@ static void i8xx_fbc_activate(struct drm_i915_private *dev_priv) if (IS_I945GM(dev_priv)) fbc_ctl |= FBC_CTL_C3_IDLE; /* 945 needs special SR handling */ fbc_ctl |= (cfb_pitch & 0xff) << FBC_CTL_STRIDE_SHIFT; - fbc_ctl |= params->fb.fence_reg; + fbc_ctl |= params->vma->fence->id; I915_WRITE(FBC_CONTROL, fbc_ctl); } @@ -193,8 +193,8 @@ static void g4x_fbc_activate(struct drm_i915_private *dev_priv) else dpfc_ctl |= DPFC_CTL_LIMIT_1X; - if (params->fb.fence_reg != I915_FENCE_REG_NONE) { - dpfc_ctl |= DPFC_CTL_FENCE_EN | params->fb.fence_reg; + if (params->vma->fence) { + dpfc_ctl |= DPFC_CTL_FENCE_EN | params->vma->fence->id; I915_WRITE(DPFC_FENCE_YOFF, params->crtc.fence_y_offset); } else { I915_WRITE(DPFC_FENCE_YOFF, 0); @@ -251,13 +251,14 @@ static void ilk_fbc_activate(struct drm_i915_private *dev_priv) break; } - if (params->fb.fence_reg != I915_FENCE_REG_NONE) { + if (params->vma->fence) { dpfc_ctl |= DPFC_CTL_FENCE_EN; if (IS_GEN5(dev_priv)) - dpfc_ctl |= params->fb.fence_reg; + dpfc_ctl |= params->vma->fence->id; if (IS_GEN6(dev_priv)) { I915_WRITE(SNB_DPFC_CTL_SA, - SNB_CPU_FENCE_ENABLE | params->fb.fence_reg); + SNB_CPU_FENCE_ENABLE | + params->vma->fence->id); I915_WRITE(DPFC_CPU_FENCE_OFFSET, params->crtc.fence_y_offset); } @@ -269,7 +270,8 @@ static void ilk_fbc_activate(struct drm_i915_private *dev_priv) } I915_WRITE(ILK_DPFC_FENCE_YOFF, params->crtc.fence_y_offset); - I915_WRITE(ILK_FBC_RT_BASE, params->fb.ggtt_offset | ILK_FBC_RT_VALID); + I915_WRITE(ILK_FBC_RT_BASE, + i915_ggtt_offset(params->vma) | ILK_FBC_RT_VALID); /* enable it... */ I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN); @@ -319,10 +321,11 @@ static void gen7_fbc_activate(struct drm_i915_private *dev_priv) break; } - if (params->fb.fence_reg != I915_FENCE_REG_NONE) { + if (params->vma->fence) { dpfc_ctl |= IVB_DPFC_CTL_FENCE_EN; I915_WRITE(SNB_DPFC_CTL_SA, - SNB_CPU_FENCE_ENABLE | params->fb.fence_reg); + SNB_CPU_FENCE_ENABLE | + params->vma->fence->id); I915_WRITE(DPFC_CPU_FENCE_OFFSET, params->crtc.fence_y_offset); } else { I915_WRITE(SNB_DPFC_CTL_SA,0); @@ -727,14 +730,6 @@ static bool intel_fbc_hw_tracking_covers_screen(struct intel_crtc *crtc) return effective_w <= max_w && effective_h <= max_h; } -/* XXX replace me when we have VMA tracking for intel_plane_state */ -static int get_fence_id(struct drm_framebuffer *fb) -{ - struct i915_vma *vma = i915_gem_object_to_ggtt(intel_fb_obj(fb), NULL); - - return vma && vma->fence ? vma->fence->id : I915_FENCE_REG_NONE; -} - static void intel_fbc_update_state_cache(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state, struct intel_plane_state *plane_state) @@ -743,7 +738,8 @@ static void intel_fbc_update_state_cache(struct intel_crtc *crtc, struct intel_fbc *fbc = &dev_priv->fbc; struct intel_fbc_state_cache *cache = &fbc->state_cache; struct drm_framebuffer *fb = plane_state->base.fb; - struct drm_i915_gem_object *obj; + + cache->vma = NULL; cache->crtc.mode_flags = crtc_state->base.adjusted_mode.flags; if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) @@ -758,16 +754,10 @@ static void intel_fbc_update_state_cache(struct intel_crtc *crtc, if (!cache->plane.visible) return; - obj = intel_fb_obj(fb); - - /* FIXME: We lack the proper locking here, so only run this on the - * platforms that need. */ - if (IS_GEN(dev_priv, 5, 6)) - cache->fb.ilk_ggtt_offset = i915_gem_object_ggtt_offset(obj, NULL); cache->fb.pixel_format = fb->pixel_format; cache->fb.stride = fb->pitches[0]; - cache->fb.fence_reg = get_fence_id(fb); - cache->fb.tiling_mode = i915_gem_object_get_tiling(obj); + + cache->vma = plane_state->vma; } static bool intel_fbc_can_activate(struct intel_crtc *crtc) @@ -784,7 +774,7 @@ static bool intel_fbc_can_activate(struct intel_crtc *crtc) return false; } - if (!cache->plane.visible) { + if (!cache->vma) { fbc->no_fbc_reason = "primary plane not visible"; return false; } @@ -807,8 +797,7 @@ static bool intel_fbc_can_activate(struct intel_crtc *crtc) * so have no fence associated with it) due to aperture constaints * at the time of pinning. */ - if (cache->fb.tiling_mode != I915_TILING_X || - cache->fb.fence_reg == I915_FENCE_REG_NONE) { + if (!cache->vma->fence) { fbc->no_fbc_reason = "framebuffer not tiled or fenced"; return false; } @@ -888,17 +877,16 @@ static void intel_fbc_get_reg_params(struct intel_crtc *crtc, * zero. */ memset(params, 0, sizeof(*params)); + params->vma = cache->vma; + params->crtc.pipe = crtc->pipe; params->crtc.plane = crtc->plane; params->crtc.fence_y_offset = get_crtc_fence_y_offset(crtc); params->fb.pixel_format = cache->fb.pixel_format; params->fb.stride = cache->fb.stride; - params->fb.fence_reg = cache->fb.fence_reg; params->cfb_size = intel_fbc_calculate_cfb_size(dev_priv, cache); - - params->fb.ggtt_offset = cache->fb.ilk_ggtt_offset; } static bool intel_fbc_reg_params_equal(struct intel_fbc_reg_params *params1, diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c index 8cf2d80f2254..f4a8c4fc57c4 100644 --- a/drivers/gpu/drm/i915/intel_fbdev.c +++ b/drivers/gpu/drm/i915/intel_fbdev.c @@ -284,7 +284,7 @@ static int intelfb_create(struct drm_fb_helper *helper, out_destroy_fbi: drm_fb_helper_release_fbi(helper); out_unpin: - intel_unpin_fb_obj(&ifbdev->fb->base, DRM_ROTATE_0); + intel_unpin_fb_vma(vma); out_unlock: mutex_unlock(&dev->struct_mutex); return ret; @@ -549,7 +549,7 @@ static void intel_fbdev_destroy(struct intel_fbdev *ifbdev) if (ifbdev->fb) { mutex_lock(&ifbdev->helper.dev->struct_mutex); - intel_unpin_fb_obj(&ifbdev->fb->base, DRM_ROTATE_0); + intel_unpin_fb_vma(ifbdev->vma); mutex_unlock(&ifbdev->helper.dev->struct_mutex); drm_framebuffer_remove(&ifbdev->fb->base); diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c index 8f131a08d440..242a73e66d82 100644 --- a/drivers/gpu/drm/i915/intel_sprite.c +++ b/drivers/gpu/drm/i915/intel_sprite.c @@ -273,7 +273,7 @@ skl_update_plane(struct drm_plane *drm_plane, I915_WRITE(PLANE_CTL(pipe, plane), plane_ctl); I915_WRITE(PLANE_SURF(pipe, plane), - intel_fb_gtt_offset(fb, rotation) + surf_addr); + intel_plane_ggtt_offset(plane_state) + surf_addr); POSTING_READ(PLANE_SURF(pipe, plane)); } @@ -458,7 +458,7 @@ vlv_update_plane(struct drm_plane *dplane, I915_WRITE(SPSIZE(pipe, plane), (crtc_h << 16) | crtc_w); I915_WRITE(SPCNTR(pipe, plane), sprctl); I915_WRITE(SPSURF(pipe, plane), - intel_fb_gtt_offset(fb, rotation) + sprsurf_offset); + intel_plane_ggtt_offset(plane_state) + sprsurf_offset); POSTING_READ(SPSURF(pipe, plane)); } @@ -594,7 +594,7 @@ ivb_update_plane(struct drm_plane *plane, I915_WRITE(SPRSCALE(pipe), sprscale); I915_WRITE(SPRCTL(pipe), sprctl); I915_WRITE(SPRSURF(pipe), - intel_fb_gtt_offset(fb, rotation) + sprsurf_offset); + intel_plane_ggtt_offset(plane_state) + sprsurf_offset); POSTING_READ(SPRSURF(pipe)); } @@ -721,7 +721,7 @@ ilk_update_plane(struct drm_plane *plane, I915_WRITE(DVSSCALE(pipe), dvsscale); I915_WRITE(DVSCNTR(pipe), dvscntr); I915_WRITE(DVSSURF(pipe), - intel_fb_gtt_offset(fb, rotation) + dvssurf_offset); + intel_plane_ggtt_offset(plane_state) + dvssurf_offset); POSTING_READ(DVSSURF(pipe)); } -- cgit v1.2.3 From 4993b39ab04b083ff6ee1147e7e7f120feb6bf7f Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Tue, 31 Jan 2017 20:01:31 +0100 Subject: be2net: fix initial MAC setting Recent commit 34393529163a ("be2net: fix MAC addr setting on privileged BE3 VFs") allows privileged BE3 VFs to set its MAC address during initialization. Although the initial MAC for such VFs is already programmed by parent PF the subsequent setting performed by VF is OK, but in certain cases (after fresh boot) this command in VF can fail. The MAC should be initialized only when: 1) no MAC is programmed (always except BE3 VFs during first init) 2) programmed MAC is different from requested (e.g. MAC is set when interface is down). In this case the initial MAC programmed by PF needs to be deleted. The adapter->dev_mac contains MAC address currently programmed in HW so it should be zeroed when the MAC is deleted from HW and should not be filled when MAC is set when interface is down in be_mac_addr_set() as no programming is performed in this case. Example of failure without the fix (immediately after fresh boot): # ip link set eth0 up <- eth0 is BE3 PF be2net 0000:01:00.0 eth0: Link is Up # echo 1 > /sys/class/net/eth0/device/sriov_numvfs <- Create 1 VF ... be2net 0000:01:04.0: Emulex OneConnect(be3): VF port 0 # ip link set eth8 up <- eth8 is created privileged VF be2net 0000:01:04.0: opcode 59-1 failed:status 1-76 RTNETLINK answers: Input/output error # echo 0 > /sys/class/net/eth0/device/sriov_numvfs <- Delete VF iommu: Removing device 0000:01:04.0 from group 33 ... # echo 1 > /sys/class/net/eth0/device/sriov_numvfs <- Create it again iommu: Removing device 0000:01:04.0 from group 33 ... # ip link set eth8 up be2net 0000:01:04.0 eth8: Link is Up Initialization is now OK. v2 - Corrected the comment and condition check suggested by Suresh & Harsha Fixes: 34393529163a ("be2net: fix MAC addr setting on privileged BE3 VFs") Cc: Sathya Perla Cc: Ajit Khaparde Cc: Sriharsha Basavapatna Cc: Somnath Kotur Signed-off-by: Ivan Vecera Acked-by: Sriharsha Basavapatna Signed-off-by: David S. Miller --- drivers/net/ethernet/emulex/benet/be_main.c | 33 ++++++++++++++++++++++++----- 1 file changed, 28 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 1a7f8ad7b9c6..cd49a54c538d 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -362,8 +362,10 @@ static int be_mac_addr_set(struct net_device *netdev, void *p) status = -EPERM; goto err; } -done: + + /* Remember currently programmed MAC */ ether_addr_copy(adapter->dev_mac, addr->sa_data); +done: ether_addr_copy(netdev->dev_addr, addr->sa_data); dev_info(dev, "MAC address changed to %pM\n", addr->sa_data); return 0; @@ -3618,8 +3620,10 @@ static void be_disable_if_filters(struct be_adapter *adapter) { /* Don't delete MAC on BE3 VFs without FILTMGMT privilege */ if (!BEx_chip(adapter) || !be_virtfn(adapter) || - check_privilege(adapter, BE_PRIV_FILTMGMT)) + check_privilege(adapter, BE_PRIV_FILTMGMT)) { be_dev_mac_del(adapter, adapter->pmac_id[0]); + eth_zero_addr(adapter->dev_mac); + } be_clear_uc_list(adapter); be_clear_mc_list(adapter); @@ -3773,12 +3777,27 @@ static int be_enable_if_filters(struct be_adapter *adapter) if (status) return status; - /* Don't add MAC on BE3 VFs without FILTMGMT privilege */ - if (!BEx_chip(adapter) || !be_virtfn(adapter) || - check_privilege(adapter, BE_PRIV_FILTMGMT)) { + /* Normally this condition usually true as the ->dev_mac is zeroed. + * But on BE3 VFs the initial MAC is pre-programmed by PF and + * subsequent be_dev_mac_add() can fail (after fresh boot) + */ + if (!ether_addr_equal(adapter->dev_mac, adapter->netdev->dev_addr)) { + int old_pmac_id = -1; + + /* Remember old programmed MAC if any - can happen on BE3 VF */ + if (!is_zero_ether_addr(adapter->dev_mac)) + old_pmac_id = adapter->pmac_id[0]; + status = be_dev_mac_add(adapter, adapter->netdev->dev_addr); if (status) return status; + + /* Delete the old programmed MAC as we successfully programmed + * a new MAC + */ + if (old_pmac_id >= 0 && old_pmac_id != adapter->pmac_id[0]) + be_dev_mac_del(adapter, old_pmac_id); + ether_addr_copy(adapter->dev_mac, adapter->netdev->dev_addr); } @@ -4552,6 +4571,10 @@ static int be_mac_setup(struct be_adapter *adapter) memcpy(adapter->netdev->dev_addr, mac, ETH_ALEN); memcpy(adapter->netdev->perm_addr, mac, ETH_ALEN); + + /* Initial MAC for BE3 VFs is already programmed by PF */ + if (BEx_chip(adapter) && be_virtfn(adapter)) + memcpy(adapter->dev_mac, mac, ETH_ALEN); } return 0; -- cgit v1.2.3 From 2da64d20a0b20046d688e44f4033efd09157e29d Mon Sep 17 00:00:00 2001 From: Alexey Kardashevskiy Date: Wed, 1 Feb 2017 14:26:16 +1100 Subject: vfio/spapr: Fix missing mutex unlock when creating a window Commit d9c728949ddc ("vfio/spapr: Postpone default window creation") added an additional exit to the VFIO_IOMMU_SPAPR_TCE_CREATE case and made it possible to return from tce_iommu_ioctl() without unlocking container->lock; this fixes the issue. Fixes: d9c728949ddc ("vfio/spapr: Postpone default window creation") Signed-off-by: Alexey Kardashevskiy Reviewed-by: David Gibson Signed-off-by: Alex Williamson --- drivers/vfio/vfio_iommu_spapr_tce.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/drivers/vfio/vfio_iommu_spapr_tce.c b/drivers/vfio/vfio_iommu_spapr_tce.c index 128d10282d16..7690e5bf3cf1 100644 --- a/drivers/vfio/vfio_iommu_spapr_tce.c +++ b/drivers/vfio/vfio_iommu_spapr_tce.c @@ -1123,12 +1123,11 @@ static long tce_iommu_ioctl(void *iommu_data, mutex_lock(&container->lock); ret = tce_iommu_create_default_window(container); - if (ret) - return ret; - - ret = tce_iommu_create_window(container, create.page_shift, - create.window_size, create.levels, - &create.start_addr); + if (!ret) + ret = tce_iommu_create_window(container, + create.page_shift, + create.window_size, create.levels, + &create.start_addr); mutex_unlock(&container->lock); -- cgit v1.2.3 From fd62d9f5c575f0792f150109f1fd24a0d4b3f854 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Tue, 31 Jan 2017 15:14:29 +0200 Subject: net/sched: matchall: Fix configuration race In the current version, the matchall internal state is split into two structs: cls_matchall_head and cls_matchall_filter. This makes little sense, as matchall instance supports only one filter, and there is no situation where one exists and the other does not. In addition, that led to some races when filter was deleted while packet was processed. Unify that two structs into one, thus simplifying the process of matchall creation and deletion. As a result, the new, delete and get callbacks have a dummy implementation where all the work is done in destroy and change callbacks, as was done in cls_cgroup. Fixes: bf3994d2ed31 ("net/sched: introduce Match-all classifier") Reported-by: Daniel Borkmann Signed-off-by: Yotam Gigi Acked-by: Jiri Pirko Signed-off-by: David S. Miller --- net/sched/cls_matchall.c | 127 +++++++++++++++++------------------------------ 1 file changed, 45 insertions(+), 82 deletions(-) diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c index f935429bd5ef..b12bc2abea93 100644 --- a/net/sched/cls_matchall.c +++ b/net/sched/cls_matchall.c @@ -16,16 +16,11 @@ #include #include -struct cls_mall_filter { +struct cls_mall_head { struct tcf_exts exts; struct tcf_result res; u32 handle; - struct rcu_head rcu; u32 flags; -}; - -struct cls_mall_head { - struct cls_mall_filter *filter; struct rcu_head rcu; }; @@ -33,38 +28,29 @@ static int mall_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct tcf_result *res) { struct cls_mall_head *head = rcu_dereference_bh(tp->root); - struct cls_mall_filter *f = head->filter; - if (tc_skip_sw(f->flags)) + if (tc_skip_sw(head->flags)) return -1; - return tcf_exts_exec(skb, &f->exts, res); + return tcf_exts_exec(skb, &head->exts, res); } static int mall_init(struct tcf_proto *tp) { - struct cls_mall_head *head; - - head = kzalloc(sizeof(*head), GFP_KERNEL); - if (!head) - return -ENOBUFS; - - rcu_assign_pointer(tp->root, head); - return 0; } -static void mall_destroy_filter(struct rcu_head *head) +static void mall_destroy_rcu(struct rcu_head *rcu) { - struct cls_mall_filter *f = container_of(head, struct cls_mall_filter, rcu); + struct cls_mall_head *head = container_of(rcu, struct cls_mall_head, + rcu); - tcf_exts_destroy(&f->exts); - - kfree(f); + tcf_exts_destroy(&head->exts); + kfree(head); } static int mall_replace_hw_filter(struct tcf_proto *tp, - struct cls_mall_filter *f, + struct cls_mall_head *head, unsigned long cookie) { struct net_device *dev = tp->q->dev_queue->dev; @@ -74,7 +60,7 @@ static int mall_replace_hw_filter(struct tcf_proto *tp, offload.type = TC_SETUP_MATCHALL; offload.cls_mall = &mall_offload; offload.cls_mall->command = TC_CLSMATCHALL_REPLACE; - offload.cls_mall->exts = &f->exts; + offload.cls_mall->exts = &head->exts; offload.cls_mall->cookie = cookie; return dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, @@ -82,7 +68,7 @@ static int mall_replace_hw_filter(struct tcf_proto *tp, } static void mall_destroy_hw_filter(struct tcf_proto *tp, - struct cls_mall_filter *f, + struct cls_mall_head *head, unsigned long cookie) { struct net_device *dev = tp->q->dev_queue->dev; @@ -103,29 +89,20 @@ static bool mall_destroy(struct tcf_proto *tp, bool force) { struct cls_mall_head *head = rtnl_dereference(tp->root); struct net_device *dev = tp->q->dev_queue->dev; - struct cls_mall_filter *f = head->filter; - if (!force && f) - return false; + if (!head) + return true; - if (f) { - if (tc_should_offload(dev, tp, f->flags)) - mall_destroy_hw_filter(tp, f, (unsigned long) f); + if (tc_should_offload(dev, tp, head->flags)) + mall_destroy_hw_filter(tp, head, (unsigned long) head); - call_rcu(&f->rcu, mall_destroy_filter); - } - kfree_rcu(head, rcu); + call_rcu(&head->rcu, mall_destroy_rcu); return true; } static unsigned long mall_get(struct tcf_proto *tp, u32 handle) { - struct cls_mall_head *head = rtnl_dereference(tp->root); - struct cls_mall_filter *f = head->filter; - - if (f && f->handle == handle) - return (unsigned long) f; - return 0; + return 0UL; } static const struct nla_policy mall_policy[TCA_MATCHALL_MAX + 1] = { @@ -134,7 +111,7 @@ static const struct nla_policy mall_policy[TCA_MATCHALL_MAX + 1] = { }; static int mall_set_parms(struct net *net, struct tcf_proto *tp, - struct cls_mall_filter *f, + struct cls_mall_head *head, unsigned long base, struct nlattr **tb, struct nlattr *est, bool ovr) { @@ -147,11 +124,11 @@ static int mall_set_parms(struct net *net, struct tcf_proto *tp, return err; if (tb[TCA_MATCHALL_CLASSID]) { - f->res.classid = nla_get_u32(tb[TCA_MATCHALL_CLASSID]); - tcf_bind_filter(tp, &f->res, base); + head->res.classid = nla_get_u32(tb[TCA_MATCHALL_CLASSID]); + tcf_bind_filter(tp, &head->res, base); } - tcf_exts_change(tp, &f->exts, &e); + tcf_exts_change(tp, &head->exts, &e); return 0; } @@ -162,21 +139,17 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, unsigned long *arg, bool ovr) { struct cls_mall_head *head = rtnl_dereference(tp->root); - struct cls_mall_filter *fold = (struct cls_mall_filter *) *arg; struct net_device *dev = tp->q->dev_queue->dev; - struct cls_mall_filter *f; struct nlattr *tb[TCA_MATCHALL_MAX + 1]; + struct cls_mall_head *new; u32 flags = 0; int err; if (!tca[TCA_OPTIONS]) return -EINVAL; - if (head->filter) - return -EBUSY; - - if (fold) - return -EINVAL; + if (head) + return -EEXIST; err = nla_parse_nested(tb, TCA_MATCHALL_MAX, tca[TCA_OPTIONS], mall_policy); @@ -189,23 +162,23 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, return -EINVAL; } - f = kzalloc(sizeof(*f), GFP_KERNEL); - if (!f) + new = kzalloc(sizeof(*new), GFP_KERNEL); + if (!new) return -ENOBUFS; - tcf_exts_init(&f->exts, TCA_MATCHALL_ACT, 0); + tcf_exts_init(&new->exts, TCA_MATCHALL_ACT, 0); if (!handle) handle = 1; - f->handle = handle; - f->flags = flags; + new->handle = handle; + new->flags = flags; - err = mall_set_parms(net, tp, f, base, tb, tca[TCA_RATE], ovr); + err = mall_set_parms(net, tp, new, base, tb, tca[TCA_RATE], ovr); if (err) goto errout; if (tc_should_offload(dev, tp, flags)) { - err = mall_replace_hw_filter(tp, f, (unsigned long) f); + err = mall_replace_hw_filter(tp, new, (unsigned long) new); if (err) { if (tc_skip_sw(flags)) goto errout; @@ -214,39 +187,29 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, } } - *arg = (unsigned long) f; - rcu_assign_pointer(head->filter, f); - + *arg = (unsigned long) head; + rcu_assign_pointer(tp->root, new); + if (head) + call_rcu(&head->rcu, mall_destroy_rcu); return 0; errout: - kfree(f); + kfree(new); return err; } static int mall_delete(struct tcf_proto *tp, unsigned long arg) { - struct cls_mall_head *head = rtnl_dereference(tp->root); - struct cls_mall_filter *f = (struct cls_mall_filter *) arg; - struct net_device *dev = tp->q->dev_queue->dev; - - if (tc_should_offload(dev, tp, f->flags)) - mall_destroy_hw_filter(tp, f, (unsigned long) f); - - RCU_INIT_POINTER(head->filter, NULL); - tcf_unbind_filter(tp, &f->res); - call_rcu(&f->rcu, mall_destroy_filter); - return 0; + return -EOPNOTSUPP; } static void mall_walk(struct tcf_proto *tp, struct tcf_walker *arg) { struct cls_mall_head *head = rtnl_dereference(tp->root); - struct cls_mall_filter *f = head->filter; if (arg->count < arg->skip) goto skip; - if (arg->fn(tp, (unsigned long) f, arg) < 0) + if (arg->fn(tp, (unsigned long) head, arg) < 0) arg->stop = 1; skip: arg->count++; @@ -255,28 +218,28 @@ skip: static int mall_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, struct sk_buff *skb, struct tcmsg *t) { - struct cls_mall_filter *f = (struct cls_mall_filter *) fh; + struct cls_mall_head *head = (struct cls_mall_head *) fh; struct nlattr *nest; - if (!f) + if (!head) return skb->len; - t->tcm_handle = f->handle; + t->tcm_handle = head->handle; nest = nla_nest_start(skb, TCA_OPTIONS); if (!nest) goto nla_put_failure; - if (f->res.classid && - nla_put_u32(skb, TCA_MATCHALL_CLASSID, f->res.classid)) + if (head->res.classid && + nla_put_u32(skb, TCA_MATCHALL_CLASSID, head->res.classid)) goto nla_put_failure; - if (tcf_exts_dump(skb, &f->exts)) + if (tcf_exts_dump(skb, &head->exts)) goto nla_put_failure; nla_nest_end(skb, nest); - if (tcf_exts_dump_stats(skb, &f->exts) < 0) + if (tcf_exts_dump_stats(skb, &head->exts) < 0) goto nla_put_failure; return skb->len; -- cgit v1.2.3 From 1a2a14444d32b89b28116daea86f63ced1716668 Mon Sep 17 00:00:00 2001 From: Dimitris Michailidis Date: Tue, 31 Jan 2017 16:03:13 -0800 Subject: net: fix ndo_features_check/ndo_fix_features comment ordering Commit cdba756f5803a2 ("net: move ndo_features_check() close to ndo_start_xmit()") inadvertently moved the doc comment for .ndo_fix_features instead of .ndo_features_check. Fix the comment ordering. Fixes: cdba756f5803a2 ("net: move ndo_features_check() close to ndo_start_xmit()") Signed-off-by: Dimitris Michailidis Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 29 +++++++++++++++-------------- 1 file changed, 15 insertions(+), 14 deletions(-) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 9bde9558b596..70ad0291d517 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -866,11 +866,15 @@ struct netdev_xdp { * of useless work if you return NETDEV_TX_BUSY. * Required; cannot be NULL. * - * netdev_features_t (*ndo_fix_features)(struct net_device *dev, - * netdev_features_t features); - * Adjusts the requested feature flags according to device-specific - * constraints, and returns the resulting flags. Must not modify - * the device state. + * netdev_features_t (*ndo_features_check)(struct sk_buff *skb, + * struct net_device *dev + * netdev_features_t features); + * Called by core transmit path to determine if device is capable of + * performing offload operations on a given packet. This is to give + * the device an opportunity to implement any restrictions that cannot + * be otherwise expressed by feature flags. The check is called with + * the set of features that the stack has calculated and it returns + * those the driver believes to be appropriate. * * u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb, * void *accel_priv, select_queue_fallback_t fallback); @@ -1028,6 +1032,12 @@ struct netdev_xdp { * Called to release previously enslaved netdev. * * Feature/offload setting functions. + * netdev_features_t (*ndo_fix_features)(struct net_device *dev, + * netdev_features_t features); + * Adjusts the requested feature flags according to device-specific + * constraints, and returns the resulting flags. Must not modify + * the device state. + * * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features); * Called to update device configuration to new features. Passed * feature set might be less than what was returned by ndo_fix_features()). @@ -1100,15 +1110,6 @@ struct netdev_xdp { * Callback to use for xmit over the accelerated station. This * is used in place of ndo_start_xmit on accelerated net * devices. - * netdev_features_t (*ndo_features_check)(struct sk_buff *skb, - * struct net_device *dev - * netdev_features_t features); - * Called by core transmit path to determine if device is capable of - * performing offload operations on a given packet. This is to give - * the device an opportunity to implement any restrictions that cannot - * be otherwise expressed by feature flags. The check is called with - * the set of features that the stack has calculated and it returns - * those the driver believes to be appropriate. * int (*ndo_set_tx_maxrate)(struct net_device *dev, * int queue_index, u32 maxrate); * Called when a user wants to set a max-rate limitation of specific -- cgit v1.2.3 From 63117f09c768be05a0bf465911297dc76394f686 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 1 Feb 2017 11:46:32 +0300 Subject: ipv6: pointer math error in ip6_tnl_parse_tlv_enc_lim() Casting is a high precedence operation but "off" and "i" are in terms of bytes so we need to have some parenthesis here. Fixes: fbfa743a9d2a ("ipv6: fix ip6_tnl_parse_tlv_enc_lim()") Signed-off-by: Dan Carpenter Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/ip6_tunnel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index ff8ee06491c3..75fac933c209 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -441,7 +441,7 @@ __u16 ip6_tnl_parse_tlv_enc_lim(struct sk_buff *skb, __u8 *raw) if (i + sizeof(*tel) > optlen) break; - tel = (struct ipv6_tlv_tnl_enc_lim *) skb->data + off + i; + tel = (struct ipv6_tlv_tnl_enc_lim *)(skb->data + off + i); /* return index of option if found and valid */ if (tel->type == IPV6_TLV_TNL_ENCAP_LIMIT && tel->length == 1) -- cgit v1.2.3 From 06425c308b92eaf60767bc71d359f4cbc7a561f8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 1 Feb 2017 08:33:53 -0800 Subject: tcp: fix 0 divide in __tcp_select_window() syszkaller fuzzer was able to trigger a divide by zero, when TCP window scaling is not enabled. SO_RCVBUF can be used not only to increase sk_rcvbuf, also to decrease it below current receive buffers utilization. If mss is negative or 0, just return a zero TCP window. Signed-off-by: Eric Dumazet Reported-by: Dmitry Vyukov Acked-by: Neal Cardwell Signed-off-by: David S. Miller --- net/ipv4/tcp_output.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 1d5331a1b1dc..8ce50dc3ab8c 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2518,9 +2518,11 @@ u32 __tcp_select_window(struct sock *sk) int full_space = min_t(int, tp->window_clamp, allowed_space); int window; - if (mss > full_space) + if (unlikely(mss > full_space)) { mss = full_space; - + if (mss <= 0) + return 0; + } if (free_space < (full_space >> 1)) { icsk->icsk_ack.quick = 0; -- cgit v1.2.3 From c8f325a59cfc718d13a50fbc746ed9b415c25e92 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Wed, 1 Feb 2017 17:45:02 +0000 Subject: efi/fdt: Avoid FDT manipulation after ExitBootServices() Some AArch64 UEFI implementations disable the MMU in ExitBootServices(), after which unaligned accesses to RAM are no longer supported. Commit: abfb7b686a3e ("efi/libstub/arm*: Pass latest memory map to the kernel") fixed an issue in the memory map handling of the stub FDT code, but inadvertently created an issue with such firmware, by moving some of the FDT manipulation to after the invocation of ExitBootServices(). Given that the stub's libfdt implementation uses the ordinary, accelerated string functions, which rely on hardware handling of unaligned accesses, manipulating the FDT with the MMU off may result in alignment faults. So fix the situation by moving the update_fdt_memmap() call into the callback function invoked by efi_exit_boot_services() right before it calls the ExitBootServices() UEFI service (which is arguably a better place for it anyway) Note that disabling the MMU in ExitBootServices() is not compliant with the UEFI spec, and carries great risk due to the fact that switching from cached to uncached memory accesses halfway through compiler generated code (i.e., involving a stack) can never be done in a way that is architecturally safe. Fixes: abfb7b686a3e ("efi/libstub/arm*: Pass latest memory map to the kernel") Signed-off-by: Ard Biesheuvel Tested-by: Riku Voipio Cc: Cc: mark.rutland@arm.com Cc: linux-efi@vger.kernel.org Cc: matt@codeblueprint.co.uk Cc: leif.lindholm@linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/1485971102-23330-2-git-send-email-ard.biesheuvel@linaro.org Signed-off-by: Ingo Molnar --- drivers/firmware/efi/libstub/fdt.c | 14 +++----------- 1 file changed, 3 insertions(+), 11 deletions(-) diff --git a/drivers/firmware/efi/libstub/fdt.c b/drivers/firmware/efi/libstub/fdt.c index 921dfa047202..260c4b4b492e 100644 --- a/drivers/firmware/efi/libstub/fdt.c +++ b/drivers/firmware/efi/libstub/fdt.c @@ -187,6 +187,7 @@ static efi_status_t update_fdt_memmap(void *fdt, struct efi_boot_memmap *map) struct exit_boot_struct { efi_memory_desc_t *runtime_map; int *runtime_entry_count; + void *new_fdt_addr; }; static efi_status_t exit_boot_func(efi_system_table_t *sys_table_arg, @@ -202,7 +203,7 @@ static efi_status_t exit_boot_func(efi_system_table_t *sys_table_arg, efi_get_virtmap(*map->map, *map->map_size, *map->desc_size, p->runtime_map, p->runtime_entry_count); - return EFI_SUCCESS; + return update_fdt_memmap(p->new_fdt_addr, map); } /* @@ -300,22 +301,13 @@ efi_status_t allocate_new_fdt_and_exit_boot(efi_system_table_t *sys_table, priv.runtime_map = runtime_map; priv.runtime_entry_count = &runtime_entry_count; + priv.new_fdt_addr = (void *)*new_fdt_addr; status = efi_exit_boot_services(sys_table, handle, &map, &priv, exit_boot_func); if (status == EFI_SUCCESS) { efi_set_virtual_address_map_t *svam; - status = update_fdt_memmap((void *)*new_fdt_addr, &map); - if (status != EFI_SUCCESS) { - /* - * The kernel won't get far without the memory map, but - * may still be able to print something meaningful so - * return success here. - */ - return EFI_SUCCESS; - } - /* Install the new virtual address map */ svam = sys_table->runtime->set_virtual_address_map; status = svam(runtime_entry_count * desc_size, desc_size, -- cgit v1.2.3 From 26a346f23c5291d1d9521e72763103daf2c6f0d1 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 1 Feb 2017 17:57:56 +0100 Subject: tracing/kprobes: Fix __init annotation clang complains about "__init" being attached to a struct name: kernel/trace/trace_kprobe.c:1375:15: error: '__section__' attribute only applies to functions and global variables The intention must have been to mark the function as __init instead of the type, so move the attribute there. Link: http://lkml.kernel.org/r/20170201165826.2625888-1-arnd@arndb.de Fixes: f18f97ac43d7 ("tracing/kprobes: Add a helper method to return number of probe hits") Signed-off-by: Arnd Bergmann Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_kprobe.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index a133ecd741e4..7ad9e53ad174 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1372,7 +1372,7 @@ kprobe_trace_selftest_target(int a1, int a2, int a3, int a4, int a5, int a6) return a1 + a2 + a3 + a4 + a5 + a6; } -static struct __init trace_event_file * +static __init struct trace_event_file * find_trace_probe_file(struct trace_kprobe *tk, struct trace_array *tr) { struct trace_event_file *file; -- cgit v1.2.3 From 57bcd0a6364cd4eaa362d7ff1777e88ddf501602 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 27 Jan 2017 10:31:52 -0500 Subject: drm/amdgpu/si: fix crash on headless asics Missing check for crtcs present. Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=193341 https://bugs.freedesktop.org/show_bug.cgi?id=99387 Reviewed-by: Christian König Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index e2b0b1646f99..0635829b18cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -254,6 +254,9 @@ static void gmc_v6_0_mc_program(struct amdgpu_device *adev) } WREG32(mmHDP_REG_COHERENCY_FLUSH_CNTL, 0); + if (adev->mode_info.num_crtc) + amdgpu_display_set_vga_render_state(adev, false); + gmc_v6_0_mc_stop(adev, &save); if (gmc_v6_0_wait_for_idle((void *)adev)) { @@ -283,7 +286,6 @@ static void gmc_v6_0_mc_program(struct amdgpu_device *adev) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); } gmc_v6_0_mc_resume(adev, &save); - amdgpu_display_set_vga_render_state(adev, false); } static int gmc_v6_0_mc_init(struct amdgpu_device *adev) -- cgit v1.2.3 From 51964e9e12d0a054002a1a0d1dec4f661c7aaf28 Mon Sep 17 00:00:00 2001 From: Michel Dänzer Date: Mon, 30 Jan 2017 12:06:35 +0900 Subject: drm/radeon: Fix vram_size/visible values in DRM_RADEON_GEM_INFO ioctl vram_size is supposed to be the total amount of VRAM that can be used by userspace, which corresponds to the TTM VRAM manager size (which is normally the full amount of VRAM, but can be just the visible VRAM when DMA can't be used for BO migration for some reason). The above was incorrectly used for vram_visible before, resulting in generally too large values being reported. Reviewed-by: Christian König Reviewed-by: Nicolai Hähnle Reviewed-by: Alex Deucher Signed-off-by: Michel Dänzer Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/radeon_drv.c | 3 ++- drivers/gpu/drm/radeon/radeon_gem.c | 4 ++-- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index e0c143b865f3..30bd4a6a9d46 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -97,9 +97,10 @@ * 2.46.0 - Add PFP_SYNC_ME support on evergreen * 2.47.0 - Add UVD_NO_OP register support * 2.48.0 - TA_CS_BC_BASE_ADDR allowed on SI + * 2.49.0 - DRM_RADEON_GEM_INFO ioctl returns correct vram_size/visible values */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 48 +#define KMS_DRIVER_MINOR 49 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index 0bcffd8a7bd3..96683f5b2b1b 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -220,8 +220,8 @@ int radeon_gem_info_ioctl(struct drm_device *dev, void *data, man = &rdev->mman.bdev.man[TTM_PL_VRAM]; - args->vram_size = rdev->mc.real_vram_size; - args->vram_visible = (u64)man->size << PAGE_SHIFT; + args->vram_size = (u64)man->size << PAGE_SHIFT; + args->vram_visible = rdev->mc.visible_vram_size; args->vram_visible -= rdev->vram_pin_size; args->gart_size = rdev->mc.gtt_size; args->gart_size -= rdev->gart_pin_size; -- cgit v1.2.3 From 56067812d5b0e737ac2063e94a50f76b810d6ca3 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Fri, 3 Feb 2017 09:54:05 +0000 Subject: kbuild: modversions: add infrastructure for emitting relative CRCs This add the kbuild infrastructure that will allow architectures to emit vmlinux symbol CRCs as 32-bit offsets to another location in the kernel where the actual value is stored. This works around problems with CRCs being mistaken for relocatable symbols on kernels that self relocate at runtime (i.e., powerpc with CONFIG_RELOCATABLE=y) For the kbuild side of things, this comes down to the following: - introducing a Kconfig symbol MODULE_REL_CRCS - adding a -R switch to genksyms to instruct it to emit the CRC symbols as references into the .rodata section - making modpost distinguish such references from absolute CRC symbols by the section index (SHN_ABS) - making kallsyms disregard non-absolute symbols with a __crc_ prefix Signed-off-by: Ard Biesheuvel Signed-off-by: Linus Torvalds --- init/Kconfig | 4 ++++ scripts/Makefile.build | 2 ++ scripts/genksyms/genksyms.c | 19 ++++++++++++++----- scripts/kallsyms.c | 12 ++++++++++++ scripts/mod/modpost.c | 10 ++++++++++ 5 files changed, 42 insertions(+), 5 deletions(-) diff --git a/init/Kconfig b/init/Kconfig index e1a937348a3e..4dd8bd232a1d 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -1987,6 +1987,10 @@ config MODVERSIONS make them incompatible with the kernel you are running. If unsure, say N. +config MODULE_REL_CRCS + bool + depends on MODVERSIONS + config MODULE_SRCVERSION_ALL bool "Source checksum for all modules" help diff --git a/scripts/Makefile.build b/scripts/Makefile.build index eadcd4d359d9..d883116ebaa4 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -164,6 +164,7 @@ cmd_gensymtypes_c = \ $(CPP) -D__GENKSYMS__ $(c_flags) $< | \ $(GENKSYMS) $(if $(1), -T $(2)) \ $(patsubst y,-s _,$(CONFIG_HAVE_UNDERSCORE_SYMBOL_PREFIX)) \ + $(patsubst y,-R,$(CONFIG_MODULE_REL_CRCS)) \ $(if $(KBUILD_PRESERVE),-p) \ -r $(firstword $(wildcard $(2:.symtypes=.symref) /dev/null)) @@ -337,6 +338,7 @@ cmd_gensymtypes_S = \ $(CPP) -D__GENKSYMS__ $(c_flags) -xc - | \ $(GENKSYMS) $(if $(1), -T $(2)) \ $(patsubst y,-s _,$(CONFIG_HAVE_UNDERSCORE_SYMBOL_PREFIX)) \ + $(patsubst y,-R,$(CONFIG_MODULE_REL_CRCS)) \ $(if $(KBUILD_PRESERVE),-p) \ -r $(firstword $(wildcard $(2:.symtypes=.symref) /dev/null)) diff --git a/scripts/genksyms/genksyms.c b/scripts/genksyms/genksyms.c index 06121ce524a7..c9235d8340f1 100644 --- a/scripts/genksyms/genksyms.c +++ b/scripts/genksyms/genksyms.c @@ -44,7 +44,7 @@ char *cur_filename, *source_file; int in_source_file; static int flag_debug, flag_dump_defs, flag_reference, flag_dump_types, - flag_preserve, flag_warnings; + flag_preserve, flag_warnings, flag_rel_crcs; static const char *mod_prefix = ""; static int errors; @@ -693,7 +693,10 @@ void export_symbol(const char *name) fputs(">\n", debugfile); /* Used as a linker script. */ - printf("%s__crc_%s = 0x%08lx ;\n", mod_prefix, name, crc); + printf(!flag_rel_crcs ? "%s__crc_%s = 0x%08lx;\n" : + "SECTIONS { .rodata : ALIGN(4) { " + "%s__crc_%s = .; LONG(0x%08lx); } }\n", + mod_prefix, name, crc); } } @@ -730,7 +733,7 @@ void error_with_pos(const char *fmt, ...) static void genksyms_usage(void) { - fputs("Usage:\n" "genksyms [-adDTwqhV] > /path/to/.tmp_obj.ver\n" "\n" + fputs("Usage:\n" "genksyms [-adDTwqhVR] > /path/to/.tmp_obj.ver\n" "\n" #ifdef __GNU_LIBRARY__ " -s, --symbol-prefix Select symbol prefix\n" " -d, --debug Increment the debug level (repeatable)\n" @@ -742,6 +745,7 @@ static void genksyms_usage(void) " -q, --quiet Disable warnings (default)\n" " -h, --help Print this message\n" " -V, --version Print the release version\n" + " -R, --relative-crc Emit section relative symbol CRCs\n" #else /* __GNU_LIBRARY__ */ " -s Select symbol prefix\n" " -d Increment the debug level (repeatable)\n" @@ -753,6 +757,7 @@ static void genksyms_usage(void) " -q Disable warnings (default)\n" " -h Print this message\n" " -V Print the release version\n" + " -R Emit section relative symbol CRCs\n" #endif /* __GNU_LIBRARY__ */ , stderr); } @@ -774,13 +779,14 @@ int main(int argc, char **argv) {"preserve", 0, 0, 'p'}, {"version", 0, 0, 'V'}, {"help", 0, 0, 'h'}, + {"relative-crc", 0, 0, 'R'}, {0, 0, 0, 0} }; - while ((o = getopt_long(argc, argv, "s:dwqVDr:T:ph", + while ((o = getopt_long(argc, argv, "s:dwqVDr:T:phR", &long_opts[0], NULL)) != EOF) #else /* __GNU_LIBRARY__ */ - while ((o = getopt(argc, argv, "s:dwqVDr:T:ph")) != EOF) + while ((o = getopt(argc, argv, "s:dwqVDr:T:phR")) != EOF) #endif /* __GNU_LIBRARY__ */ switch (o) { case 's': @@ -823,6 +829,9 @@ int main(int argc, char **argv) case 'h': genksyms_usage(); return 0; + case 'R': + flag_rel_crcs = 1; + break; default: genksyms_usage(); return 1; diff --git a/scripts/kallsyms.c b/scripts/kallsyms.c index 299b92ca1ae0..5d554419170b 100644 --- a/scripts/kallsyms.c +++ b/scripts/kallsyms.c @@ -219,6 +219,10 @@ static int symbol_valid(struct sym_entry *s) "_SDA2_BASE_", /* ppc */ NULL }; + static char *special_prefixes[] = { + "__crc_", /* modversions */ + NULL }; + static char *special_suffixes[] = { "_veneer", /* arm */ "_from_arm", /* arm */ @@ -259,6 +263,14 @@ static int symbol_valid(struct sym_entry *s) if (strcmp(sym_name, special_symbols[i]) == 0) return 0; + for (i = 0; special_prefixes[i]; i++) { + int l = strlen(special_prefixes[i]); + + if (l <= strlen(sym_name) && + strncmp(sym_name, special_prefixes[i], l) == 0) + return 0; + } + for (i = 0; special_suffixes[i]; i++) { int l = strlen(sym_name) - strlen(special_suffixes[i]); diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c index 29c89a6bad3d..4dedd0d3d3a7 100644 --- a/scripts/mod/modpost.c +++ b/scripts/mod/modpost.c @@ -621,6 +621,16 @@ static void handle_modversions(struct module *mod, struct elf_info *info, if (strncmp(symname, CRC_PFX, strlen(CRC_PFX)) == 0) { is_crc = true; crc = (unsigned int) sym->st_value; + if (sym->st_shndx != SHN_UNDEF && sym->st_shndx != SHN_ABS) { + unsigned int *crcp; + + /* symbol points to the CRC in the ELF object */ + crcp = (void *)info->hdr + sym->st_value + + info->sechdrs[sym->st_shndx].sh_offset - + (info->hdr->e_type != ET_REL ? + info->sechdrs[sym->st_shndx].sh_addr : 0); + crc = *crcp; + } sym_update_crc(symname + strlen(CRC_PFX), mod, crc, export); } -- cgit v1.2.3 From 71810db27c1c853b335675bee335d893bc3d324b Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Fri, 3 Feb 2017 09:54:06 +0000 Subject: modversions: treat symbol CRCs as 32 bit quantities The modversion symbol CRCs are emitted as ELF symbols, which allows us to easily populate the kcrctab sections by relying on the linker to associate each kcrctab slot with the correct value. This has a couple of downsides: - Given that the CRCs are treated as memory addresses, we waste 4 bytes for each CRC on 64 bit architectures, - On architectures that support runtime relocation, a R__RELATIVE relocation entry is emitted for each CRC value, which identifies it as a quantity that requires fixing up based on the actual runtime load offset of the kernel. This results in corrupted CRCs unless we explicitly undo the fixup (and this is currently being handled in the core module code) - Such runtime relocation entries take up 24 bytes of __init space each, resulting in a x8 overhead in [uncompressed] kernel size for CRCs. Switching to explicit 32 bit values on 64 bit architectures fixes most of these issues, given that 32 bit values are not treated as quantities that require fixing up based on the actual runtime load offset. Note that on some ELF64 architectures [such as PPC64], these 32-bit values are still emitted as [absolute] runtime relocatable quantities, even if the value resolves to a build time constant. Since relative relocations are always resolved at build time, this patch enables MODULE_REL_CRCS on powerpc when CONFIG_RELOCATABLE=y, which turns the absolute CRC references into relative references into .rodata where the actual CRC value is stored. So redefine all CRC fields and variables as u32, and redefine the __CRC_SYMBOL() macro for 64 bit builds to emit the CRC reference using inline assembler (which is necessary since 64-bit C code cannot use 32-bit types to hold memory addresses, even if they are ultimately resolved using values that do not exceed 0xffffffff). To avoid potential problems with legacy 32-bit architectures using legacy toolchains, the equivalent C definition of the kcrctab entry is retained for 32-bit architectures. Note that this mostly reverts commit d4703aefdbc8 ("module: handle ppc64 relocating kcrctabs when CONFIG_RELOCATABLE=y") Acked-by: Rusty Russell Signed-off-by: Ard Biesheuvel Signed-off-by: Linus Torvalds --- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/module.h | 4 --- arch/powerpc/kernel/module_64.c | 8 ------ include/asm-generic/export.h | 11 ++++---- include/linux/export.h | 14 +++++++++++ include/linux/module.h | 14 +++++------ kernel/module.c | 53 ++++++++++++++++++--------------------- 7 files changed, 53 insertions(+), 52 deletions(-) diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index a8ee573fe610..db8a1ef6bfaf 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -484,6 +484,7 @@ config RELOCATABLE bool "Build a relocatable kernel" depends on (PPC64 && !COMPILE_TEST) || (FLATMEM && (44x || FSL_BOOKE)) select NONSTATIC_KERNEL + select MODULE_REL_CRCS if MODVERSIONS help This builds a kernel image that is capable of running at the location the kernel is loaded at. For ppc32, there is no any diff --git a/arch/powerpc/include/asm/module.h b/arch/powerpc/include/asm/module.h index cc12c61ef315..53885512b8d3 100644 --- a/arch/powerpc/include/asm/module.h +++ b/arch/powerpc/include/asm/module.h @@ -90,9 +90,5 @@ static inline int module_finalize_ftrace(struct module *mod, const Elf_Shdr *sec } #endif -#if defined(CONFIG_MODVERSIONS) && defined(CONFIG_PPC64) -#define ARCH_RELOCATES_KCRCTAB -#define reloc_start PHYSICAL_START -#endif #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_MODULE_H */ diff --git a/arch/powerpc/kernel/module_64.c b/arch/powerpc/kernel/module_64.c index bb1807184bad..0b0f89685b67 100644 --- a/arch/powerpc/kernel/module_64.c +++ b/arch/powerpc/kernel/module_64.c @@ -286,14 +286,6 @@ static void dedotify_versions(struct modversion_info *vers, for (end = (void *)vers + size; vers < end; vers++) if (vers->name[0] == '.') { memmove(vers->name, vers->name+1, strlen(vers->name)); -#ifdef ARCH_RELOCATES_KCRCTAB - /* The TOC symbol has no CRC computed. To avoid CRC - * check failing, we must force it to the expected - * value (see CRC check in module.c). - */ - if (!strcmp(vers->name, "TOC.")) - vers->crc = -(unsigned long)reloc_start; -#endif } } diff --git a/include/asm-generic/export.h b/include/asm-generic/export.h index 63554e9f6e0c..719db1968d81 100644 --- a/include/asm-generic/export.h +++ b/include/asm-generic/export.h @@ -9,18 +9,15 @@ #ifndef KSYM_ALIGN #define KSYM_ALIGN 8 #endif -#ifndef KCRC_ALIGN -#define KCRC_ALIGN 8 -#endif #else #define __put .long #ifndef KSYM_ALIGN #define KSYM_ALIGN 4 #endif +#endif #ifndef KCRC_ALIGN #define KCRC_ALIGN 4 #endif -#endif #ifdef CONFIG_HAVE_UNDERSCORE_SYMBOL_PREFIX #define KSYM(name) _##name @@ -52,7 +49,11 @@ KSYM(__kstrtab_\name): .section ___kcrctab\sec+\name,"a" .balign KCRC_ALIGN KSYM(__kcrctab_\name): - __put KSYM(__crc_\name) +#if defined(CONFIG_MODULE_REL_CRCS) + .long KSYM(__crc_\name) - . +#else + .long KSYM(__crc_\name) +#endif .weak KSYM(__crc_\name) .previous #endif diff --git a/include/linux/export.h b/include/linux/export.h index 2a0f61fbc731..7473fba6a60c 100644 --- a/include/linux/export.h +++ b/include/linux/export.h @@ -43,6 +43,13 @@ extern struct module __this_module; #ifdef CONFIG_MODVERSIONS /* Mark the CRC weak since genksyms apparently decides not to * generate a checksums for some symbols */ +#if defined(CONFIG_MODULE_REL_CRCS) +#define __CRC_SYMBOL(sym, sec) \ + asm(" .section \"___kcrctab" sec "+" #sym "\", \"a\" \n" \ + " .weak " VMLINUX_SYMBOL_STR(__crc_##sym) " \n" \ + " .long " VMLINUX_SYMBOL_STR(__crc_##sym) " - . \n" \ + " .previous \n"); +#elif !defined(CONFIG_64BIT) #define __CRC_SYMBOL(sym, sec) \ extern __visible void *__crc_##sym __attribute__((weak)); \ static const unsigned long __kcrctab_##sym \ @@ -50,6 +57,13 @@ extern struct module __this_module; __attribute__((section("___kcrctab" sec "+" #sym), used)) \ = (unsigned long) &__crc_##sym; #else +#define __CRC_SYMBOL(sym, sec) \ + asm(" .section \"___kcrctab" sec "+" #sym "\", \"a\" \n" \ + " .weak " VMLINUX_SYMBOL_STR(__crc_##sym) " \n" \ + " .long " VMLINUX_SYMBOL_STR(__crc_##sym) " \n" \ + " .previous \n"); +#endif +#else #define __CRC_SYMBOL(sym, sec) #endif diff --git a/include/linux/module.h b/include/linux/module.h index 7c84273d60b9..cc7cba219b20 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -346,7 +346,7 @@ struct module { /* Exported symbols */ const struct kernel_symbol *syms; - const unsigned long *crcs; + const s32 *crcs; unsigned int num_syms; /* Kernel parameters. */ @@ -359,18 +359,18 @@ struct module { /* GPL-only exported symbols. */ unsigned int num_gpl_syms; const struct kernel_symbol *gpl_syms; - const unsigned long *gpl_crcs; + const s32 *gpl_crcs; #ifdef CONFIG_UNUSED_SYMBOLS /* unused exported symbols. */ const struct kernel_symbol *unused_syms; - const unsigned long *unused_crcs; + const s32 *unused_crcs; unsigned int num_unused_syms; /* GPL-only, unused exported symbols. */ unsigned int num_unused_gpl_syms; const struct kernel_symbol *unused_gpl_syms; - const unsigned long *unused_gpl_crcs; + const s32 *unused_gpl_crcs; #endif #ifdef CONFIG_MODULE_SIG @@ -382,7 +382,7 @@ struct module { /* symbols that will be GPL-only in the near future. */ const struct kernel_symbol *gpl_future_syms; - const unsigned long *gpl_future_crcs; + const s32 *gpl_future_crcs; unsigned int num_gpl_future_syms; /* Exception table */ @@ -523,7 +523,7 @@ struct module *find_module(const char *name); struct symsearch { const struct kernel_symbol *start, *stop; - const unsigned long *crcs; + const s32 *crcs; enum { NOT_GPL_ONLY, GPL_ONLY, @@ -539,7 +539,7 @@ struct symsearch { */ const struct kernel_symbol *find_symbol(const char *name, struct module **owner, - const unsigned long **crc, + const s32 **crc, bool gplok, bool warn); diff --git a/kernel/module.c b/kernel/module.c index 38d4270925d4..3d8f126208e3 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -389,16 +389,16 @@ extern const struct kernel_symbol __start___ksymtab_gpl[]; extern const struct kernel_symbol __stop___ksymtab_gpl[]; extern const struct kernel_symbol __start___ksymtab_gpl_future[]; extern const struct kernel_symbol __stop___ksymtab_gpl_future[]; -extern const unsigned long __start___kcrctab[]; -extern const unsigned long __start___kcrctab_gpl[]; -extern const unsigned long __start___kcrctab_gpl_future[]; +extern const s32 __start___kcrctab[]; +extern const s32 __start___kcrctab_gpl[]; +extern const s32 __start___kcrctab_gpl_future[]; #ifdef CONFIG_UNUSED_SYMBOLS extern const struct kernel_symbol __start___ksymtab_unused[]; extern const struct kernel_symbol __stop___ksymtab_unused[]; extern const struct kernel_symbol __start___ksymtab_unused_gpl[]; extern const struct kernel_symbol __stop___ksymtab_unused_gpl[]; -extern const unsigned long __start___kcrctab_unused[]; -extern const unsigned long __start___kcrctab_unused_gpl[]; +extern const s32 __start___kcrctab_unused[]; +extern const s32 __start___kcrctab_unused_gpl[]; #endif #ifndef CONFIG_MODVERSIONS @@ -497,7 +497,7 @@ struct find_symbol_arg { /* Output */ struct module *owner; - const unsigned long *crc; + const s32 *crc; const struct kernel_symbol *sym; }; @@ -563,7 +563,7 @@ static bool find_symbol_in_section(const struct symsearch *syms, * (optional) module which owns it. Needs preempt disabled or module_mutex. */ const struct kernel_symbol *find_symbol(const char *name, struct module **owner, - const unsigned long **crc, + const s32 **crc, bool gplok, bool warn) { @@ -1249,23 +1249,17 @@ static int try_to_force_load(struct module *mod, const char *reason) } #ifdef CONFIG_MODVERSIONS -/* If the arch applies (non-zero) relocations to kernel kcrctab, unapply it. */ -static unsigned long maybe_relocated(unsigned long crc, - const struct module *crc_owner) + +static u32 resolve_rel_crc(const s32 *crc) { -#ifdef ARCH_RELOCATES_KCRCTAB - if (crc_owner == NULL) - return crc - (unsigned long)reloc_start; -#endif - return crc; + return *(u32 *)((void *)crc + *crc); } static int check_version(Elf_Shdr *sechdrs, unsigned int versindex, const char *symname, struct module *mod, - const unsigned long *crc, - const struct module *crc_owner) + const s32 *crc) { unsigned int i, num_versions; struct modversion_info *versions; @@ -1283,13 +1277,19 @@ static int check_version(Elf_Shdr *sechdrs, / sizeof(struct modversion_info); for (i = 0; i < num_versions; i++) { + u32 crcval; + if (strcmp(versions[i].name, symname) != 0) continue; - if (versions[i].crc == maybe_relocated(*crc, crc_owner)) + if (IS_ENABLED(CONFIG_MODULE_REL_CRCS)) + crcval = resolve_rel_crc(crc); + else + crcval = *crc; + if (versions[i].crc == crcval) return 1; - pr_debug("Found checksum %lX vs module %lX\n", - maybe_relocated(*crc, crc_owner), versions[i].crc); + pr_debug("Found checksum %X vs module %lX\n", + crcval, versions[i].crc); goto bad_version; } @@ -1307,7 +1307,7 @@ static inline int check_modstruct_version(Elf_Shdr *sechdrs, unsigned int versindex, struct module *mod) { - const unsigned long *crc; + const s32 *crc; /* * Since this should be found in kernel (which can't be removed), no @@ -1321,8 +1321,7 @@ static inline int check_modstruct_version(Elf_Shdr *sechdrs, } preempt_enable(); return check_version(sechdrs, versindex, - VMLINUX_SYMBOL_STR(module_layout), mod, crc, - NULL); + VMLINUX_SYMBOL_STR(module_layout), mod, crc); } /* First part is kernel version, which we ignore if module has crcs. */ @@ -1340,8 +1339,7 @@ static inline int check_version(Elf_Shdr *sechdrs, unsigned int versindex, const char *symname, struct module *mod, - const unsigned long *crc, - const struct module *crc_owner) + const s32 *crc) { return 1; } @@ -1368,7 +1366,7 @@ static const struct kernel_symbol *resolve_symbol(struct module *mod, { struct module *owner; const struct kernel_symbol *sym; - const unsigned long *crc; + const s32 *crc; int err; /* @@ -1383,8 +1381,7 @@ static const struct kernel_symbol *resolve_symbol(struct module *mod, if (!sym) goto unlock; - if (!check_version(info->sechdrs, info->index.vers, name, mod, crc, - owner)) { + if (!check_version(info->sechdrs, info->index.vers, name, mod, crc)) { sym = ERR_PTR(-EINVAL); goto getname; } -- cgit v1.2.3 From 4b9eee96fcb361a5e16a8d2619825e8a048f81f7 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Fri, 3 Feb 2017 09:54:07 +0000 Subject: module: unify absolute krctab definitions for 32-bit and 64-bit The previous patch introduced a separate inline asm version of the krcrctab declaration template for use with 64-bit architectures, which cannot refer to ELF symbols using 32-bit quantities. This declaration should be equivalent to the C one for 32-bit architectures, but just in case - unify them in a separate patch, which can simply be dropped if it turns out to break anything. Signed-off-by: Ard Biesheuvel Signed-off-by: Linus Torvalds --- include/linux/export.h | 7 ------- 1 file changed, 7 deletions(-) diff --git a/include/linux/export.h b/include/linux/export.h index 7473fba6a60c..1a1dfdb2a5c6 100644 --- a/include/linux/export.h +++ b/include/linux/export.h @@ -49,13 +49,6 @@ extern struct module __this_module; " .weak " VMLINUX_SYMBOL_STR(__crc_##sym) " \n" \ " .long " VMLINUX_SYMBOL_STR(__crc_##sym) " - . \n" \ " .previous \n"); -#elif !defined(CONFIG_64BIT) -#define __CRC_SYMBOL(sym, sec) \ - extern __visible void *__crc_##sym __attribute__((weak)); \ - static const unsigned long __kcrctab_##sym \ - __used \ - __attribute__((section("___kcrctab" sec "+" #sym), used)) \ - = (unsigned long) &__crc_##sym; #else #define __CRC_SYMBOL(sym, sec) \ asm(" .section \"___kcrctab" sec "+" #sym "\", \"a\" \n" \ -- cgit v1.2.3 From 00c87e9a70a17b355b81c36adedf05e84f54e10d Mon Sep 17 00:00:00 2001 From: Radim Krčmář Date: Wed, 1 Feb 2017 14:19:53 +0100 Subject: KVM: x86: do not save guest-unsupported XSAVE state Saving unsupported state prevents migration when the new host does not support a XSAVE feature of the original host, even if the feature is not exposed to the guest. We've masked host features with guest-visible features before, with 4344ee981e21 ("KVM: x86: only copy XSAVE state for the supported features") and dropped it when implementing XSAVES. Do it again. Fixes: df1daba7d1cb ("KVM: x86: support XSAVES usage in the host") Cc: stable@vger.kernel.org Reviewed-by: Paolo Bonzini Signed-off-by: Radim Krčmář --- arch/x86/kvm/x86.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index d153be8929a6..e52c9088660f 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3182,6 +3182,7 @@ static void fill_xsave(u8 *dest, struct kvm_vcpu *vcpu) memcpy(dest, xsave, XSAVE_HDR_OFFSET); /* Set XSTATE_BV */ + xstate_bv &= vcpu->arch.guest_supported_xcr0 | XFEATURE_MASK_FPSSE; *(u64 *)(dest + XSAVE_HDR_OFFSET) = xstate_bv; /* -- cgit v1.2.3 From 29905b52fad0854351f57bab867647e4982285bf Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Thu, 2 Feb 2017 18:05:26 +0000 Subject: log2: make order_base_2() behave correctly on const input value zero The function order_base_2() is defined (according to the comment block) as returning zero on input zero, but subsequently passes the input into roundup_pow_of_two(), which is explicitly undefined for input zero. This has gone unnoticed until now, but optimization passes in GCC 7 may produce constant folded function instances where a constant value of zero is passed into order_base_2(), resulting in link errors against the deliberately undefined '____ilog2_NaN'. So update order_base_2() to adhere to its own documented interface. [ See http://marc.info/?l=linux-kernel&m=147672952517795&w=2 and follow-up discussion for more background. The gcc "optimization pass" is really just broken, but now the GCC trunk problem seems to have escaped out of just specially built daily images, so we need to work around it in mainline. - Linus ] Signed-off-by: Ard Biesheuvel Signed-off-by: Linus Torvalds --- include/linux/log2.h | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/include/linux/log2.h b/include/linux/log2.h index fd7ff3d91e6a..ef3d4f67118c 100644 --- a/include/linux/log2.h +++ b/include/linux/log2.h @@ -203,6 +203,17 @@ unsigned long __rounddown_pow_of_two(unsigned long n) * ... and so on. */ -#define order_base_2(n) ilog2(roundup_pow_of_two(n)) +static inline __attribute_const__ +int __order_base_2(unsigned long n) +{ + return n > 1 ? ilog2(n - 1) + 1 : 0; +} +#define order_base_2(n) \ +( \ + __builtin_constant_p(n) ? ( \ + ((n) == 0 || (n) == 1) ? 0 : \ + ilog2((n) - 1) + 1) : \ + __order_base_2(n) \ +) #endif /* _LINUX_LOG2_H */ -- cgit v1.2.3 From 0d5415b489f68b58e1983a53793d25d53098ed4b Mon Sep 17 00:00:00 2001 From: Michael S. Tsirkin Date: Fri, 3 Feb 2017 05:43:52 +0200 Subject: Revert "vring: Force use of DMA API for ARM-based systems with legacy devices" This reverts commit c7070619f3408d9a0dffbed9149e6f00479cf43b. This has been shown to regress on some ARM systems: by forcing on DMA API usage for ARM systems, we have inadvertently kicked open a hornets' nest in terms of cache-coherency. Namely that unless the virtio device is explicitly described as capable of coherent DMA by firmware, the DMA APIs on ARM and other DT-based platforms will assume it is non-coherent. This turns out to cause a big problem for the likes of QEMU and kvmtool, which generate virtio-mmio devices in their guest DTs but neglect to add the often-overlooked "dma-coherent" property; as a result, we end up with the guest making non-cacheable accesses to the vring, the host doing so cacheably, both talking past each other and things going horribly wrong. We are working on a safer work-around. Fixes: c7070619f340 ("vring: Force use of DMA API for ARM-based systems with legacy devices") Reported-by: Robin Murphy Cc: Signed-off-by: Will Deacon Signed-off-by: Michael S. Tsirkin Acked-by: Marc Zyngier --- drivers/virtio/virtio_ring.c | 7 ------- 1 file changed, 7 deletions(-) diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 7e38ed79c3fc..409aeaa49246 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -159,13 +159,6 @@ static bool vring_use_dma_api(struct virtio_device *vdev) if (xen_domain()) return true; - /* - * On ARM-based machines, the DMA ops will do the right thing, - * so always use them with legacy devices. - */ - if (IS_ENABLED(CONFIG_ARM) || IS_ENABLED(CONFIG_ARM64)) - return !virtio_has_feature(vdev, VIRTIO_F_VERSION_1); - return false; } -- cgit v1.2.3 From cda8bba0f99d25d2061c531113c14fa41effc3ae Mon Sep 17 00:00:00 2001 From: Halil Pasic Date: Mon, 30 Jan 2017 11:09:36 +0100 Subject: vhost: fix initialization for vq->is_le Currently, under certain circumstances vhost_init_is_le does just a part of the initialization job, and depends on vhost_reset_is_le being called too. For this reason vhost_vq_init_access used to call vhost_reset_is_le when vq->private_data is NULL. This is not only counter intuitive, but also real a problem because it breaks vhost_net. The bug was introduced to vhost_net with commit 2751c9882b94 ("vhost: cross-endian support for legacy devices"). The symptom is corruption of the vq's used.idx field (virtio) after VHOST_NET_SET_BACKEND was issued as a part of the vhost shutdown on a vq with pending descriptors. Let us make sure the outcome of vhost_init_is_le never depend on the state it is actually supposed to initialize, and fix virtio_net by removing the reset from vhost_vq_init_access. With the above, there is no reason for vhost_reset_is_le to do just half of the job. Let us make vhost_reset_is_le reinitialize is_le. Signed-off-by: Halil Pasic Reported-by: Michael A. Tebolt Reported-by: Dr. David Alan Gilbert Fixes: commit 2751c9882b94 ("vhost: cross-endian support for legacy devices") Cc: Signed-off-by: Michael S. Tsirkin Reviewed-by: Greg Kurz Tested-by: Michael A. Tebolt --- drivers/vhost/vhost.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index d6432603880c..8f99fe08de02 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -130,14 +130,14 @@ static long vhost_get_vring_endian(struct vhost_virtqueue *vq, u32 idx, static void vhost_init_is_le(struct vhost_virtqueue *vq) { - if (vhost_has_feature(vq, VIRTIO_F_VERSION_1)) - vq->is_le = true; + vq->is_le = vhost_has_feature(vq, VIRTIO_F_VERSION_1) + || virtio_legacy_is_little_endian(); } #endif /* CONFIG_VHOST_CROSS_ENDIAN_LEGACY */ static void vhost_reset_is_le(struct vhost_virtqueue *vq) { - vq->is_le = virtio_legacy_is_little_endian(); + vhost_init_is_le(vq); } struct vhost_flush_struct { @@ -1714,10 +1714,8 @@ int vhost_vq_init_access(struct vhost_virtqueue *vq) int r; bool is_le = vq->is_le; - if (!vq->private_data) { - vhost_reset_is_le(vq); + if (!vq->private_data) return 0; - } vhost_init_is_le(vq); -- cgit v1.2.3 From 79134d11d030b886106bf45a5638c1ccb1f0856c Mon Sep 17 00:00:00 2001 From: Amit Shah Date: Fri, 3 Feb 2017 16:48:14 +0530 Subject: MAINTAINERS: update email address for Amit Shah I'm leaving my job at Red Hat, this email address will stop working next week. Update it to one that I will have access to later. Signed-off-by: Amit Shah Signed-off-by: Michael S. Tsirkin --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 5f10c28b2e15..e0f979294e23 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -13066,7 +13066,7 @@ F: drivers/input/serio/userio.c F: include/uapi/linux/userio.h VIRTIO CONSOLE DRIVER -M: Amit Shah +M: Amit Shah L: virtualization@lists.linux-foundation.org S: Maintained F: drivers/char/virtio_console.c -- cgit v1.2.3 From d7b028f56a971a2e4d8d7887540a144eeefcd4ab Mon Sep 17 00:00:00 2001 From: Dan Streetman Date: Fri, 3 Feb 2017 13:13:09 -0800 Subject: zswap: disable changing params if init fails Add zswap_init_failed bool that prevents changing any of the module params, if init_zswap() fails, and set zswap_enabled to false. Change 'enabled' param to a callback, and check zswap_init_failed before allowing any change to 'enabled', 'zpool', or 'compressor' params. Any driver that is built-in to the kernel will not be unloaded if its init function returns error, and its module params remain accessible for users to change via sysfs. Since zswap uses param callbacks, which assume that zswap has been initialized, changing the zswap params after a failed initialization will result in WARNING due to the param callbacks expecting a pool to already exist. This prevents that by immediately exiting any of the param callbacks if initialization failed. This was reported here: https://marc.info/?l=linux-mm&m=147004228125528&w=4 And fixes this WARNING: [ 429.723476] WARNING: CPU: 0 PID: 5140 at mm/zswap.c:503 __zswap_pool_current+0x56/0x60 The warning is just noise, and not serious. However, when init fails, zswap frees all its percpu dstmem pages and its kmem cache. The kmem cache might be serious, if kmem_cache_alloc(NULL, gfp) has problems; but the percpu dstmem pages are definitely a problem, as they're used as temporary buffer for compressed pages before copying into place in the zpool. If the user does get zswap enabled after an init failure, then zswap will likely Oops on the first page it tries to compress (or worse, start corrupting memory). Fixes: 90b0fc26d5db ("zswap: change zpool/compressor at runtime") Link: http://lkml.kernel.org/r/20170124200259.16191-2-ddstreet@ieee.org Signed-off-by: Dan Streetman Reported-by: Marcin Miroslaw Cc: Seth Jennings Cc: Michal Hocko Cc: Sergey Senozhatsky Cc: Minchan Kim Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/zswap.c | 30 +++++++++++++++++++++++++++++- 1 file changed, 29 insertions(+), 1 deletion(-) diff --git a/mm/zswap.c b/mm/zswap.c index 067a0d62f318..cabf09e0128b 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -78,7 +78,13 @@ static u64 zswap_duplicate_entry; /* Enable/disable zswap (disabled by default) */ static bool zswap_enabled; -module_param_named(enabled, zswap_enabled, bool, 0644); +static int zswap_enabled_param_set(const char *, + const struct kernel_param *); +static struct kernel_param_ops zswap_enabled_param_ops = { + .set = zswap_enabled_param_set, + .get = param_get_bool, +}; +module_param_cb(enabled, &zswap_enabled_param_ops, &zswap_enabled, 0644); /* Crypto compressor to use */ #define ZSWAP_COMPRESSOR_DEFAULT "lzo" @@ -176,6 +182,9 @@ static atomic_t zswap_pools_count = ATOMIC_INIT(0); /* used by param callback function */ static bool zswap_init_started; +/* fatal error during init */ +static bool zswap_init_failed; + /********************************* * helpers and fwd declarations **********************************/ @@ -624,6 +633,11 @@ static int __zswap_param_set(const char *val, const struct kernel_param *kp, char *s = strstrip((char *)val); int ret; + if (zswap_init_failed) { + pr_err("can't set param, initialization failed\n"); + return -ENODEV; + } + /* no change required */ if (!strcmp(s, *(char **)kp->arg)) return 0; @@ -703,6 +717,17 @@ static int zswap_zpool_param_set(const char *val, return __zswap_param_set(val, kp, NULL, zswap_compressor); } +static int zswap_enabled_param_set(const char *val, + const struct kernel_param *kp) +{ + if (zswap_init_failed) { + pr_err("can't enable, initialization failed\n"); + return -ENODEV; + } + + return param_set_bool(val, kp); +} + /********************************* * writeback code **********************************/ @@ -1201,6 +1226,9 @@ hp_fail: dstmem_fail: zswap_entry_cache_destroy(); cache_fail: + /* if built-in, we aren't unloaded on failure; don't allow use */ + zswap_init_failed = true; + zswap_enabled = false; return -ENOMEM; } /* must be late so crypto has time to come up */ -- cgit v1.2.3 From 4f40c6e5627ea73b4e7c615c59631f38cc880885 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 3 Feb 2017 13:13:12 -0800 Subject: kasan: respect /proc/sys/kernel/traceoff_on_warning After much waiting I finally reproduced a KASAN issue, only to find my trace-buffer empty of useful information because it got spooled out :/ Make kasan_report honour the /proc/sys/kernel/traceoff_on_warning interface. Link: http://lkml.kernel.org/r/20170125164106.3514-1-aryabinin@virtuozzo.com Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Andrey Ryabinin Acked-by: Alexander Potapenko Cc: Dmitry Vyukov Cc: Steven Rostedt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/kasan/report.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/mm/kasan/report.c b/mm/kasan/report.c index b82b3e215157..f479365530b6 100644 --- a/mm/kasan/report.c +++ b/mm/kasan/report.c @@ -13,6 +13,7 @@ * */ +#include #include #include #include @@ -300,6 +301,8 @@ void kasan_report(unsigned long addr, size_t size, if (likely(!kasan_report_enabled())) return; + disable_trace_on_warning(); + info.access_addr = (void *)addr; info.access_size = size; info.is_write = is_write; -- cgit v1.2.3 From 253fd0f02040a19c6fe80e4171659fa3482a422d Mon Sep 17 00:00:00 2001 From: Kirill A. Shutemov Date: Fri, 3 Feb 2017 13:13:15 -0800 Subject: shmem: fix sleeping from atomic context Syzkaller fuzzer managed to trigger this: BUG: sleeping function called from invalid context at mm/shmem.c:852 in_atomic(): 1, irqs_disabled(): 0, pid: 529, name: khugepaged 3 locks held by khugepaged/529: #0: (shrinker_rwsem){++++..}, at: [] shrink_slab.part.59+0x121/0xd30 mm/vmscan.c:451 #1: (&type->s_umount_key#29){++++..}, at: [] trylock_super+0x20/0x100 fs/super.c:392 #2: (&(&sbinfo->shrinklist_lock)->rlock){+.+.-.}, at: [] spin_lock include/linux/spinlock.h:302 [inline] #2: (&(&sbinfo->shrinklist_lock)->rlock){+.+.-.}, at: [] shmem_unused_huge_shrink+0x28e/0x1490 mm/shmem.c:427 CPU: 2 PID: 529 Comm: khugepaged Not tainted 4.10.0-rc5+ #201 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011 Call Trace: shmem_undo_range+0xb20/0x2710 mm/shmem.c:852 shmem_truncate_range+0x27/0xa0 mm/shmem.c:939 shmem_evict_inode+0x35f/0xca0 mm/shmem.c:1030 evict+0x46e/0x980 fs/inode.c:553 iput_final fs/inode.c:1515 [inline] iput+0x589/0xb20 fs/inode.c:1542 shmem_unused_huge_shrink+0xbad/0x1490 mm/shmem.c:446 shmem_unused_huge_scan+0x10c/0x170 mm/shmem.c:512 super_cache_scan+0x376/0x450 fs/super.c:106 do_shrink_slab mm/vmscan.c:378 [inline] shrink_slab.part.59+0x543/0xd30 mm/vmscan.c:481 shrink_slab mm/vmscan.c:2592 [inline] shrink_node+0x2c7/0x870 mm/vmscan.c:2592 shrink_zones mm/vmscan.c:2734 [inline] do_try_to_free_pages+0x369/0xc80 mm/vmscan.c:2776 try_to_free_pages+0x3c6/0x900 mm/vmscan.c:2982 __perform_reclaim mm/page_alloc.c:3301 [inline] __alloc_pages_direct_reclaim mm/page_alloc.c:3322 [inline] __alloc_pages_slowpath+0xa24/0x1c30 mm/page_alloc.c:3683 __alloc_pages_nodemask+0x544/0xae0 mm/page_alloc.c:3848 __alloc_pages include/linux/gfp.h:426 [inline] __alloc_pages_node include/linux/gfp.h:439 [inline] khugepaged_alloc_page+0xc2/0x1b0 mm/khugepaged.c:750 collapse_huge_page+0x182/0x1fe0 mm/khugepaged.c:955 khugepaged_scan_pmd+0xfdf/0x12a0 mm/khugepaged.c:1208 khugepaged_scan_mm_slot mm/khugepaged.c:1727 [inline] khugepaged_do_scan mm/khugepaged.c:1808 [inline] khugepaged+0xe9b/0x1590 mm/khugepaged.c:1853 kthread+0x326/0x3f0 kernel/kthread.c:227 ret_from_fork+0x31/0x40 arch/x86/entry/entry_64.S:430 The iput() from atomic context was a bad idea: if after igrab() somebody else calls iput() and we left with the last inode reference, our iput() would lead to inode eviction and therefore sleeping. This patch should fix the situation. Link: http://lkml.kernel.org/r/20170131093141.GA15899@node.shutemov.name Signed-off-by: Kirill A. Shutemov Reported-by: Dmitry Vyukov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/shmem.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index bb53285a1d99..3a7587a0314d 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -415,6 +415,7 @@ static unsigned long shmem_unused_huge_shrink(struct shmem_sb_info *sbinfo, struct shrink_control *sc, unsigned long nr_to_split) { LIST_HEAD(list), *pos, *next; + LIST_HEAD(to_remove); struct inode *inode; struct shmem_inode_info *info; struct page *page; @@ -441,9 +442,8 @@ static unsigned long shmem_unused_huge_shrink(struct shmem_sb_info *sbinfo, /* Check if there's anything to gain */ if (round_up(inode->i_size, PAGE_SIZE) == round_up(inode->i_size, HPAGE_PMD_SIZE)) { - list_del_init(&info->shrinklist); + list_move(&info->shrinklist, &to_remove); removed++; - iput(inode); goto next; } @@ -454,6 +454,13 @@ next: } spin_unlock(&sbinfo->shrinklist_lock); + list_for_each_safe(pos, next, &to_remove) { + info = list_entry(pos, struct shmem_inode_info, shrinklist); + inode = &info->vfs_inode; + list_del_init(&info->shrinklist); + iput(inode); + } + list_for_each_safe(pos, next, &list) { int ret; -- cgit v1.2.3 From 35f860f9ba6aac56cc38e8b18916d833a83f1157 Mon Sep 17 00:00:00 2001 From: David Lin Date: Fri, 3 Feb 2017 13:13:18 -0800 Subject: jump label: pass kbuild_cflags when checking for asm goto support Some versions of ARM GCC compiler such as Android toolchain throws in a '-fpic' flag by default. This causes the gcc-goto check script to fail although some config would have '-fno-pic' flag in the KBUILD_CFLAGS. This patch passes the KBUILD_CFLAGS to the check script so that the script does not rely on the default config from different compilers. Link: http://lkml.kernel.org/r/20170120234329.78868-1-dtwlin@google.com Signed-off-by: David Lin Acked-by: Steven Rostedt Cc: Michal Marek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 96b27a888285..d7b758a1ce14 100644 --- a/Makefile +++ b/Makefile @@ -797,7 +797,7 @@ KBUILD_CFLAGS += $(call cc-option,-Werror=incompatible-pointer-types) KBUILD_ARFLAGS := $(call ar-option,D) # check for 'asm goto' -ifeq ($(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-goto.sh $(CC)), y) +ifeq ($(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-goto.sh $(CC) $(KBUILD_CFLAGS)), y) KBUILD_CFLAGS += -DCC_HAVE_ASM_GOTO KBUILD_AFLAGS += -DCC_HAVE_ASM_GOTO endif -- cgit v1.2.3 From deb88a2a19e85842d79ba96b05031739ec327ff4 Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Fri, 3 Feb 2017 13:13:20 -0800 Subject: mm/memory_hotplug.c: check start_pfn in test_pages_in_a_zone() Patch series "fix a kernel oops when reading sysfs valid_zones", v2. A sysfs memory file is created for each 2GiB memory block on x86-64 when the system has 64GiB or more memory. [1] When the start address of a memory block is not backed by struct page, i.e. a memory range is not aligned by 2GiB, reading its 'valid_zones' attribute file leads to a kernel oops. This issue was observed on multiple x86-64 systems with more than 64GiB of memory. This patch-set fixes this issue. Patch 1 first fixes an issue in test_pages_in_a_zone(), which does not test the start section. Patch 2 then fixes the kernel oops by extending test_pages_in_a_zone() to return valid [start, end). Note for stable kernels: The memory block size change was made by commit bdee237c0343 ("x86: mm: Use 2GB memory block size on large-memory x86-64 systems"), which was accepted to 3.9. However, this patch-set depends on (and fixes) the change to test_pages_in_a_zone() made by commit 5f0f2887f4de ("mm/memory_hotplug.c: check for missing sections in test_pages_in_a_zone()"), which was accepted to 4.4. So, I recommend that we backport it up to 4.4. [1] 'Commit bdee237c0343 ("x86: mm: Use 2GB memory block size on large-memory x86-64 systems")' This patch (of 2): test_pages_in_a_zone() does not check 'start_pfn' when it is aligned by section since 'sec_end_pfn' is set equal to 'pfn'. Since this function is called for testing the range of a sysfs memory file, 'start_pfn' is always aligned by section. Fix it by properly setting 'sec_end_pfn' to the next section pfn. Also make sure that this function returns 1 only when the range belongs to a zone. Link: http://lkml.kernel.org/r/20170127222149.30893-2-toshi.kani@hpe.com Signed-off-by: Toshi Kani Cc: Andrew Banman Cc: Reza Arbab Cc: Greg KH Cc: [4.4+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory_hotplug.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index ca2723d47338..1218f73890b6 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -1483,7 +1483,7 @@ bool is_mem_section_removable(unsigned long start_pfn, unsigned long nr_pages) } /* - * Confirm all pages in a range [start, end) is belongs to the same zone. + * Confirm all pages in a range [start, end) belong to the same zone. */ int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn) { @@ -1491,9 +1491,9 @@ int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn) struct zone *zone = NULL; struct page *page; int i; - for (pfn = start_pfn, sec_end_pfn = SECTION_ALIGN_UP(start_pfn); + for (pfn = start_pfn, sec_end_pfn = SECTION_ALIGN_UP(start_pfn + 1); pfn < end_pfn; - pfn = sec_end_pfn + 1, sec_end_pfn += PAGES_PER_SECTION) { + pfn = sec_end_pfn, sec_end_pfn += PAGES_PER_SECTION) { /* Make sure the memory section is present first */ if (!present_section_nr(pfn_to_section_nr(pfn))) continue; @@ -1512,7 +1512,11 @@ int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn) zone = page_zone(page); } } - return 1; + + if (zone) + return 1; + else + return 0; } /* -- cgit v1.2.3 From a96dfddbcc04336bbed50dc2b24823e45e09e80c Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Fri, 3 Feb 2017 13:13:23 -0800 Subject: base/memory, hotplug: fix a kernel oops in show_valid_zones() Reading a sysfs "memoryN/valid_zones" file leads to the following oops when the first page of a range is not backed by struct page. show_valid_zones() assumes that 'start_pfn' is always valid for page_zone(). BUG: unable to handle kernel paging request at ffffea017a000000 IP: show_valid_zones+0x6f/0x160 This issue may happen on x86-64 systems with 64GiB or more memory since their memory block size is bumped up to 2GiB. [1] An example of such systems is desribed below. 0x3240000000 is only aligned by 1GiB and this memory block starts from 0x3200000000, which is not backed by struct page. BIOS-e820: [mem 0x0000003240000000-0x000000603fffffff] usable Since test_pages_in_a_zone() already checks holes, fix this issue by extending this function to return 'valid_start' and 'valid_end' for a given range. show_valid_zones() then proceeds with the valid range. [1] 'Commit bdee237c0343 ("x86: mm: Use 2GB memory block size on large-memory x86-64 systems")' Link: http://lkml.kernel.org/r/20170127222149.30893-3-toshi.kani@hpe.com Signed-off-by: Toshi Kani Cc: Greg Kroah-Hartman Cc: Zhang Zhen Cc: Reza Arbab Cc: David Rientjes Cc: Dan Williams Cc: [4.4+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/memory.c | 12 ++++++------ include/linux/memory_hotplug.h | 3 ++- mm/memory_hotplug.c | 20 +++++++++++++++----- 3 files changed, 23 insertions(+), 12 deletions(-) diff --git a/drivers/base/memory.c b/drivers/base/memory.c index dacb6a8418aa..fa26ffd25fa6 100644 --- a/drivers/base/memory.c +++ b/drivers/base/memory.c @@ -389,33 +389,33 @@ static ssize_t show_valid_zones(struct device *dev, { struct memory_block *mem = to_memory_block(dev); unsigned long start_pfn, end_pfn; + unsigned long valid_start, valid_end, valid_pages; unsigned long nr_pages = PAGES_PER_SECTION * sections_per_block; - struct page *first_page; struct zone *zone; int zone_shift = 0; start_pfn = section_nr_to_pfn(mem->start_section_nr); end_pfn = start_pfn + nr_pages; - first_page = pfn_to_page(start_pfn); /* The block contains more than one zone can not be offlined. */ - if (!test_pages_in_a_zone(start_pfn, end_pfn)) + if (!test_pages_in_a_zone(start_pfn, end_pfn, &valid_start, &valid_end)) return sprintf(buf, "none\n"); - zone = page_zone(first_page); + zone = page_zone(pfn_to_page(valid_start)); + valid_pages = valid_end - valid_start; /* MMOP_ONLINE_KEEP */ sprintf(buf, "%s", zone->name); /* MMOP_ONLINE_KERNEL */ - zone_can_shift(start_pfn, nr_pages, ZONE_NORMAL, &zone_shift); + zone_can_shift(valid_start, valid_pages, ZONE_NORMAL, &zone_shift); if (zone_shift) { strcat(buf, " "); strcat(buf, (zone + zone_shift)->name); } /* MMOP_ONLINE_MOVABLE */ - zone_can_shift(start_pfn, nr_pages, ZONE_MOVABLE, &zone_shift); + zone_can_shift(valid_start, valid_pages, ZONE_MOVABLE, &zone_shift); if (zone_shift) { strcat(buf, " "); strcat(buf, (zone + zone_shift)->name); diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index c1784c0b4f35..134a2f69c21a 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -85,7 +85,8 @@ extern int zone_grow_waitqueues(struct zone *zone, unsigned long nr_pages); extern int add_one_highpage(struct page *page, int pfn, int bad_ppro); /* VM interface that may be used by firmware interface */ extern int online_pages(unsigned long, unsigned long, int); -extern int test_pages_in_a_zone(unsigned long, unsigned long); +extern int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn, + unsigned long *valid_start, unsigned long *valid_end); extern void __offline_isolated_pages(unsigned long, unsigned long); typedef void (*online_page_callback_t)(struct page *page); diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 1218f73890b6..b8c11e063ff0 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -1484,10 +1484,13 @@ bool is_mem_section_removable(unsigned long start_pfn, unsigned long nr_pages) /* * Confirm all pages in a range [start, end) belong to the same zone. + * When true, return its valid [start, end). */ -int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn) +int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn, + unsigned long *valid_start, unsigned long *valid_end) { unsigned long pfn, sec_end_pfn; + unsigned long start, end; struct zone *zone = NULL; struct page *page; int i; @@ -1509,14 +1512,20 @@ int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn) page = pfn_to_page(pfn + i); if (zone && page_zone(page) != zone) return 0; + if (!zone) + start = pfn + i; zone = page_zone(page); + end = pfn + MAX_ORDER_NR_PAGES; } } - if (zone) + if (zone) { + *valid_start = start; + *valid_end = end; return 1; - else + } else { return 0; + } } /* @@ -1843,6 +1852,7 @@ static int __ref __offline_pages(unsigned long start_pfn, long offlined_pages; int ret, drain, retry_max, node; unsigned long flags; + unsigned long valid_start, valid_end; struct zone *zone; struct memory_notify arg; @@ -1853,10 +1863,10 @@ static int __ref __offline_pages(unsigned long start_pfn, return -EINVAL; /* This makes hotplug much easier...and readable. we assume this for now. .*/ - if (!test_pages_in_a_zone(start_pfn, end_pfn)) + if (!test_pages_in_a_zone(start_pfn, end_pfn, &valid_start, &valid_end)) return -EINVAL; - zone = page_zone(pfn_to_page(start_pfn)); + zone = page_zone(pfn_to_page(valid_start)); node = zone_to_nid(zone); nr_pages = end_pfn - start_pfn; -- cgit v1.2.3 From d1908f52557b3230fbd63c0429f3b4b748bf2b6d Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Fri, 3 Feb 2017 13:13:26 -0800 Subject: fs: break out of iomap_file_buffered_write on fatal signals Tetsuo has noticed that an OOM stress test which performs large write requests can cause the full memory reserves depletion. He has tracked this down to the following path __alloc_pages_nodemask+0x436/0x4d0 alloc_pages_current+0x97/0x1b0 __page_cache_alloc+0x15d/0x1a0 mm/filemap.c:728 pagecache_get_page+0x5a/0x2b0 mm/filemap.c:1331 grab_cache_page_write_begin+0x23/0x40 mm/filemap.c:2773 iomap_write_begin+0x50/0xd0 fs/iomap.c:118 iomap_write_actor+0xb5/0x1a0 fs/iomap.c:190 ? iomap_write_end+0x80/0x80 fs/iomap.c:150 iomap_apply+0xb3/0x130 fs/iomap.c:79 iomap_file_buffered_write+0x68/0xa0 fs/iomap.c:243 ? iomap_write_end+0x80/0x80 xfs_file_buffered_aio_write+0x132/0x390 [xfs] ? remove_wait_queue+0x59/0x60 xfs_file_write_iter+0x90/0x130 [xfs] __vfs_write+0xe5/0x140 vfs_write+0xc7/0x1f0 ? syscall_trace_enter+0x1d0/0x380 SyS_write+0x58/0xc0 do_syscall_64+0x6c/0x200 entry_SYSCALL64_slow_path+0x25/0x25 the oom victim has access to all memory reserves to make a forward progress to exit easier. But iomap_file_buffered_write and other callers of iomap_apply loop to complete the full request. We need to check for fatal signals and back off with a short write instead. As the iomap_apply delegates all the work down to the actor we have to hook into those. All callers that work with the page cache are calling iomap_write_begin so we will check for signals there. dax_iomap_actor has to handle the situation explicitly because it copies data to the userspace directly. Other callers like iomap_page_mkwrite work on a single page or iomap_fiemap_actor do not allocate memory based on the given len. Fixes: 68a9f5e7007c ("xfs: implement iomap based buffered write path") Link: http://lkml.kernel.org/r/20170201092706.9966-2-mhocko@kernel.org Signed-off-by: Michal Hocko Reported-by: Tetsuo Handa Reviewed-by: Christoph Hellwig Cc: Al Viro Cc: [4.8+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dax.c | 5 +++++ fs/iomap.c | 3 +++ 2 files changed, 8 insertions(+) diff --git a/fs/dax.c b/fs/dax.c index 3af2da5e64ce..c45598b912e1 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -1031,6 +1031,11 @@ dax_iomap_actor(struct inode *inode, loff_t pos, loff_t length, void *data, struct blk_dax_ctl dax = { 0 }; ssize_t map_len; + if (fatal_signal_pending(current)) { + ret = -EINTR; + break; + } + dax.sector = dax_iomap_sector(iomap, pos); dax.size = (length + offset + PAGE_SIZE - 1) & PAGE_MASK; map_len = dax_map_atomic(iomap->bdev, &dax); diff --git a/fs/iomap.c b/fs/iomap.c index 354a123f170e..a51cb4c07d4d 100644 --- a/fs/iomap.c +++ b/fs/iomap.c @@ -114,6 +114,9 @@ iomap_write_begin(struct inode *inode, loff_t pos, unsigned len, unsigned flags, BUG_ON(pos + len > iomap->offset + iomap->length); + if (fatal_signal_pending(current)) + return -EINTR; + page = grab_cache_page_write_begin(inode->i_mapping, index, flags); if (!page) return -ENOMEM; -- cgit v1.2.3 From 5abf186a30a89d5b9c18a6bf93a2c192c9fd52f6 Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Fri, 3 Feb 2017 13:13:29 -0800 Subject: mm, fs: check for fatal signals in do_generic_file_read() do_generic_file_read() can be told to perform a large request from userspace. If the system is under OOM and the reading task is the OOM victim then it has an access to memory reserves and finishing the full request can lead to the full memory depletion which is dangerous. Make sure we rather go with a short read and allow the killed task to terminate. Link: http://lkml.kernel.org/r/20170201092706.9966-3-mhocko@kernel.org Signed-off-by: Michal Hocko Reviewed-by: Christoph Hellwig Cc: Tetsuo Handa Cc: Al Viro Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/filemap.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/mm/filemap.c b/mm/filemap.c index b772a33ef640..3f9afded581b 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1791,6 +1791,11 @@ static ssize_t do_generic_file_read(struct file *filp, loff_t *ppos, cond_resched(); find_page: + if (fatal_signal_pending(current)) { + error = -EINTR; + goto out; + } + page = find_get_page(mapping, index); if (!page) { page_cache_sync_readahead(mapping, -- cgit v1.2.3 From d5adbfcd5f7bcc6fa58a41c5c5ada0e5c826ce2c Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 5 Feb 2017 15:10:58 -0800 Subject: Linux 4.10-rc7 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index d7b758a1ce14..8e223e081c9d 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 4 PATCHLEVEL = 10 SUBLEVEL = 0 -EXTRAVERSION = -rc6 +EXTRAVERSION = -rc7 NAME = Fearless Coyote # *DOCUMENTATION* -- cgit v1.2.3