Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
  21234e3a84c7 ("net/mlx5e: Fix use after free in mlx5e_fs_init()")
  c7eafc5ed068 ("net/mlx5e: Convert ethtool_steering member of flow_steering struct to pointer")
https://lore.kernel.org/all/20220825104410.67d4709c@canb.auug.org.au/
https://lore.kernel.org/all/20220823055533.334471-1-saeed@kernel.org/

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

6 files changed, 289 insertions, 9 deletions

diff --git a/tools/lib/perf/cpumap.c b/tools/lib/perf/cpumap.c
index 384d5e076ee4..6cd0be7c1bb4 100644
--- a/tools/lib/perf/cpumap.c
+++ b/tools/lib/perf/cpumap.c
@@ -309,7 +309,7 @@ bool perf_cpu_map__has(const struct perf_cpu_map *cpus, struct perf_cpu cpu)
 	return perf_cpu_map__idx(cpus, cpu) != -1;
 }
 
-struct perf_cpu perf_cpu_map__max(struct perf_cpu_map *map)
+struct perf_cpu perf_cpu_map__max(const struct perf_cpu_map *map)
 {
 	struct perf_cpu result = {
 		.cpu = -1
diff --git a/tools/lib/perf/evsel.c b/tools/lib/perf/evsel.c
index 952f3520d5c2..8ce5bbd09666 100644
--- a/tools/lib/perf/evsel.c
+++ b/tools/lib/perf/evsel.c
@@ -305,6 +305,9 @@ int perf_evsel__read_size(struct perf_evsel *evsel)
 	if (read_format & PERF_FORMAT_ID)
 		entry += sizeof(u64);
 
+	if (read_format & PERF_FORMAT_LOST)
+		entry += sizeof(u64);
+
 	if (read_format & PERF_FORMAT_GROUP) {
 		nr = evsel->nr_members;
 		size += sizeof(u64);
@@ -314,24 +317,98 @@ int perf_evsel__read_size(struct perf_evsel *evsel)
 	return size;
 }
 
+/* This only reads values for the leader */
+static int perf_evsel__read_group(struct perf_evsel *evsel, int cpu_map_idx,
+				  int thread, struct perf_counts_values *count)
+{
+	size_t size = perf_evsel__read_size(evsel);
+	int *fd = FD(evsel, cpu_map_idx, thread);
+	u64 read_format = evsel->attr.read_format;
+	u64 *data;
+	int idx = 1;
+
+	if (fd == NULL || *fd < 0)
+		return -EINVAL;
+
+	data = calloc(1, size);
+	if (data == NULL)
+		return -ENOMEM;
+
+	if (readn(*fd, data, size) <= 0) {
+		free(data);
+		return -errno;
+	}
+
+	/*
+	 * This reads only the leader event intentionally since we don't have
+	 * perf counts values for sibling events.
+	 */
+	if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+		count->ena = data[idx++];
+	if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+		count->run = data[idx++];
+
+	/* value is always available */
+	count->val = data[idx++];
+	if (read_format & PERF_FORMAT_ID)
+		count->id = data[idx++];
+	if (read_format & PERF_FORMAT_LOST)
+		count->lost = data[idx++];
+
+	free(data);
+	return 0;
+}
+
+/*
+ * The perf read format is very flexible.  It needs to set the proper
+ * values according to the read format.
+ */
+static void perf_evsel__adjust_values(struct perf_evsel *evsel, u64 *buf,
+				      struct perf_counts_values *count)
+{
+	u64 read_format = evsel->attr.read_format;
+	int n = 0;
+
+	count->val = buf[n++];
+
+	if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+		count->ena = buf[n++];
+
+	if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+		count->run = buf[n++];
+
+	if (read_format & PERF_FORMAT_ID)
+		count->id = buf[n++];
+
+	if (read_format & PERF_FORMAT_LOST)
+		count->lost = buf[n++];
+}
+
 int perf_evsel__read(struct perf_evsel *evsel, int cpu_map_idx, int thread,
 		     struct perf_counts_values *count)
 {
 	size_t size = perf_evsel__read_size(evsel);
 	int *fd = FD(evsel, cpu_map_idx, thread);
+	u64 read_format = evsel->attr.read_format;
+	struct perf_counts_values buf;
 
 	memset(count, 0, sizeof(*count));
 
 	if (fd == NULL || *fd < 0)
 		return -EINVAL;
 
+	if (read_format & PERF_FORMAT_GROUP)
+		return perf_evsel__read_group(evsel, cpu_map_idx, thread, count);
+
 	if (MMAP(evsel, cpu_map_idx, thread) &&
+	    !(read_format & (PERF_FORMAT_ID | PERF_FORMAT_LOST)) &&
 	    !perf_mmap__read_self(MMAP(evsel, cpu_map_idx, thread), count))
 		return 0;
 
-	if (readn(*fd, count->values, size) <= 0)
+	if (readn(*fd, buf.values, size) <= 0)
 		return -errno;
 
+	perf_evsel__adjust_values(evsel, buf.values, count);
 	return 0;
 }
 
diff --git a/tools/lib/perf/include/perf/cpumap.h b/tools/lib/perf/include/perf/cpumap.h
index 24de795b09bb..03aceb72a783 100644
--- a/tools/lib/perf/include/perf/cpumap.h
+++ b/tools/lib/perf/include/perf/cpumap.h
@@ -23,7 +23,7 @@ LIBPERF_API void perf_cpu_map__put(struct perf_cpu_map *map);
 LIBPERF_API struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx);
 LIBPERF_API int perf_cpu_map__nr(const struct perf_cpu_map *cpus);
 LIBPERF_API bool perf_cpu_map__empty(const struct perf_cpu_map *map);
-LIBPERF_API struct perf_cpu perf_cpu_map__max(struct perf_cpu_map *map);
+LIBPERF_API struct perf_cpu perf_cpu_map__max(const struct perf_cpu_map *map);
 LIBPERF_API bool perf_cpu_map__has(const struct perf_cpu_map *map, struct perf_cpu cpu);
 
 #define perf_cpu_map__for_each_cpu(cpu, idx, cpus)		\
diff --git a/tools/lib/perf/include/perf/event.h b/tools/lib/perf/include/perf/event.h
index 556bb06798f2..93bf93a59c99 100644
--- a/tools/lib/perf/include/perf/event.h
+++ b/tools/lib/perf/include/perf/event.h
@@ -6,6 +6,7 @@
 #include <linux/types.h>
 #include <linux/limits.h>
 #include <linux/bpf.h>
+#include <linux/compiler.h>
 #include <sys/types.h> /* pid_t */
 
 #define event_contains(obj, mem) ((obj).header.size > offsetof(typeof(obj), mem))
@@ -76,7 +77,7 @@ struct perf_record_lost_samples {
 };
 
 /*
- * PERF_FORMAT_ENABLED | PERF_FORMAT_RUNNING | PERF_FORMAT_ID
+ * PERF_FORMAT_ENABLED | PERF_FORMAT_RUNNING | PERF_FORMAT_ID | PERF_FORMAT_LOST
  */
 struct perf_record_read {
 	struct perf_event_header header;
@@ -85,6 +86,7 @@ struct perf_record_read {
 	__u64			 time_enabled;
 	__u64			 time_running;
 	__u64			 id;
+	__u64			 lost;
 };
 
 struct perf_record_throttle {
@@ -153,22 +155,60 @@ enum {
 	PERF_CPU_MAP__MASK = 1,
 };
 
+/*
+ * Array encoding of a perf_cpu_map where nr is the number of entries in cpu[]
+ * and each entry is a value for a CPU in the map.
+ */
 struct cpu_map_entries {
 	__u16			 nr;
 	__u16			 cpu[];
 };
 
-struct perf_record_record_cpu_map {
+/* Bitmap encoding of a perf_cpu_map where bitmap entries are 32-bit. */
+struct perf_record_mask_cpu_map32 {
+	/* Number of mask values. */
+	__u16			 nr;
+	/* Constant 4. */
+	__u16			 long_size;
+	/* Bitmap data. */
+	__u32			 mask[];
+};
+
+/* Bitmap encoding of a perf_cpu_map where bitmap entries are 64-bit. */
+struct perf_record_mask_cpu_map64 {
+	/* Number of mask values. */
 	__u16			 nr;
+	/* Constant 8. */
 	__u16			 long_size;
-	unsigned long		 mask[];
+	/* Legacy padding. */
+	char                     __pad[4];
+	/* Bitmap data. */
+	__u64			 mask[];
 };
 
-struct perf_record_cpu_map_data {
+/*
+ * 'struct perf_record_cpu_map_data' is packed as unfortunately an earlier
+ * version had unaligned data and we wish to retain file format compatibility.
+ * -irogers
+ */
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wpacked"
+#pragma GCC diagnostic ignored "-Wattributes"
+
+struct __packed perf_record_cpu_map_data {
 	__u16			 type;
-	char			 data[];
+	union {
+		/* Used when type == PERF_CPU_MAP__CPUS. */
+		struct cpu_map_entries cpus_data;
+		/* Used when type == PERF_CPU_MAP__MASK and long_size == 4. */
+		struct perf_record_mask_cpu_map32 mask32_data;
+		/* Used when type == PERF_CPU_MAP__MASK and long_size == 8. */
+		struct perf_record_mask_cpu_map64 mask64_data;
+	};
 };
 
+#pragma GCC diagnostic pop
+
 struct perf_record_cpu_map {
 	struct perf_event_header	 header;
 	struct perf_record_cpu_map_data	 data;
diff --git a/tools/lib/perf/include/perf/evsel.h b/tools/lib/perf/include/perf/evsel.h
index 699c0ed97d34..6f92204075c2 100644
--- a/tools/lib/perf/include/perf/evsel.h
+++ b/tools/lib/perf/include/perf/evsel.h
@@ -18,8 +18,10 @@ struct perf_counts_values {
 			uint64_t val;
 			uint64_t ena;
 			uint64_t run;
+			uint64_t id;
+			uint64_t lost;
 		};
-		uint64_t values[3];
+		uint64_t values[5];
 	};
 };
 
diff --git a/tools/lib/perf/tests/test-evsel.c b/tools/lib/perf/tests/test-evsel.c
index 89be89afb24d..a11fc51bfb68 100644
--- a/tools/lib/perf/tests/test-evsel.c
+++ b/tools/lib/perf/tests/test-evsel.c
@@ -1,10 +1,13 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <stdarg.h>
 #include <stdio.h>
+#include <string.h>
 #include <linux/perf_event.h>
+#include <linux/kernel.h>
 #include <perf/cpumap.h>
 #include <perf/threadmap.h>
 #include <perf/evsel.h>
+#include <internal/evsel.h>
 #include <internal/tests.h>
 #include "tests.h"
 
@@ -189,6 +192,163 @@ static int test_stat_user_read(int event)
 	return 0;
 }
 
+static int test_stat_read_format_single(struct perf_event_attr *attr, struct perf_thread_map *threads)
+{
+	struct perf_evsel *evsel;
+	struct perf_counts_values counts;
+	volatile int count = 0x100000;
+	int err;
+
+	evsel = perf_evsel__new(attr);
+	__T("failed to create evsel", evsel);
+
+	/* skip old kernels that don't support the format */
+	err = perf_evsel__open(evsel, NULL, threads);
+	if (err < 0)
+		return 0;
+
+	while (count--) ;
+
+	memset(&counts, -1, sizeof(counts));
+	perf_evsel__read(evsel, 0, 0, &counts);
+
+	__T("failed to read value", counts.val);
+	if (attr->read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+		__T("failed to read TOTAL_TIME_ENABLED", counts.ena);
+	if (attr->read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+		__T("failed to read TOTAL_TIME_RUNNING", counts.run);
+	if (attr->read_format & PERF_FORMAT_ID)
+		__T("failed to read ID", counts.id);
+	if (attr->read_format & PERF_FORMAT_LOST)
+		__T("failed to read LOST", counts.lost == 0);
+
+	perf_evsel__close(evsel);
+	perf_evsel__delete(evsel);
+	return 0;
+}
+
+static int test_stat_read_format_group(struct perf_event_attr *attr, struct perf_thread_map *threads)
+{
+	struct perf_evsel *leader, *member;
+	struct perf_counts_values counts;
+	volatile int count = 0x100000;
+	int err;
+
+	attr->read_format |= PERF_FORMAT_GROUP;
+	leader = perf_evsel__new(attr);
+	__T("failed to create leader", leader);
+
+	attr->read_format &= ~PERF_FORMAT_GROUP;
+	member = perf_evsel__new(attr);
+	__T("failed to create member", member);
+
+	member->leader = leader;
+	leader->nr_members = 2;
+
+	/* skip old kernels that don't support the format */
+	err = perf_evsel__open(leader, NULL, threads);
+	if (err < 0)
+		return 0;
+	err = perf_evsel__open(member, NULL, threads);
+	if (err < 0)
+		return 0;
+
+	while (count--) ;
+
+	memset(&counts, -1, sizeof(counts));
+	perf_evsel__read(leader, 0, 0, &counts);
+
+	__T("failed to read leader value", counts.val);
+	if (attr->read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+		__T("failed to read leader TOTAL_TIME_ENABLED", counts.ena);
+	if (attr->read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+		__T("failed to read leader TOTAL_TIME_RUNNING", counts.run);
+	if (attr->read_format & PERF_FORMAT_ID)
+		__T("failed to read leader ID", counts.id);
+	if (attr->read_format & PERF_FORMAT_LOST)
+		__T("failed to read leader LOST", counts.lost == 0);
+
+	memset(&counts, -1, sizeof(counts));
+	perf_evsel__read(member, 0, 0, &counts);
+
+	__T("failed to read member value", counts.val);
+	if (attr->read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+		__T("failed to read member TOTAL_TIME_ENABLED", counts.ena);
+	if (attr->read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+		__T("failed to read member TOTAL_TIME_RUNNING", counts.run);
+	if (attr->read_format & PERF_FORMAT_ID)
+		__T("failed to read member ID", counts.id);
+	if (attr->read_format & PERF_FORMAT_LOST)
+		__T("failed to read member LOST", counts.lost == 0);
+
+	perf_evsel__close(member);
+	perf_evsel__close(leader);
+	perf_evsel__delete(member);
+	perf_evsel__delete(leader);
+	return 0;
+}
+
+static int test_stat_read_format(void)
+{
+	struct perf_thread_map *threads;
+	struct perf_event_attr attr = {
+		.type	= PERF_TYPE_SOFTWARE,
+		.config	= PERF_COUNT_SW_TASK_CLOCK,
+	};
+	int err, i;
+
+#define FMT(_fmt)  PERF_FORMAT_ ## _fmt
+#define FMT_TIME  (FMT(TOTAL_TIME_ENABLED) | FMT(TOTAL_TIME_RUNNING))
+
+	uint64_t test_formats [] = {
+		0,
+		FMT_TIME,
+		FMT(ID),
+		FMT(LOST),
+		FMT_TIME | FMT(ID),
+		FMT_TIME | FMT(LOST),
+		FMT_TIME | FMT(ID) | FMT(LOST),
+		FMT(ID) | FMT(LOST),
+	};
+
+#undef FMT
+#undef FMT_TIME
+
+	threads = perf_thread_map__new_dummy();
+	__T("failed to create threads", threads);
+
+	perf_thread_map__set_pid(threads, 0, 0);
+
+	for (i = 0; i < (int)ARRAY_SIZE(test_formats); i++) {
+		attr.read_format = test_formats[i];
+		__T_VERBOSE("testing single read with read_format: %lx\n",
+			    (unsigned long)test_formats[i]);
+
+		err = test_stat_read_format_single(&attr, threads);
+		__T("failed to read single format", err == 0);
+	}
+
+	perf_thread_map__put(threads);
+
+	threads = perf_thread_map__new_array(2, NULL);
+	__T("failed to create threads", threads);
+
+	perf_thread_map__set_pid(threads, 0, 0);
+	perf_thread_map__set_pid(threads, 1, 0);
+
+	for (i = 0; i < (int)ARRAY_SIZE(test_formats); i++) {
+		attr.read_format = test_formats[i];
+		__T_VERBOSE("testing group read with read_format: %lx\n",
+			    (unsigned long)test_formats[i]);
+
+		err = test_stat_read_format_group(&attr, threads);
+		__T("failed to read group format", err == 0);
+	}
+
+	perf_thread_map__put(threads);
+	return 0;
+}
+
 int test_evsel(int argc, char **argv)
 {
 	__T_START;
@@ -200,6 +360,7 @@ int test_evsel(int argc, char **argv)
 	test_stat_thread_enable();
 	test_stat_user_read(PERF_COUNT_HW_INSTRUCTIONS);
 	test_stat_user_read(PERF_COUNT_HW_CPU_CYCLES);
+	test_stat_read_format();
 
 	__T_END;
 	return tests_failed == 0 ? 0 : -1;