aboutsummaryrefslogtreecommitdiff
path: root/lib/tpm-v2.c
diff options
context:
space:
mode:
authorEddie James2023-10-24 10:43:49 -0500
committerIlias Apalodimas2023-10-27 13:14:47 +0300
commit97707f12fdabf5fab5942504dab711a665854942 (patch)
tree3e4ecb65ecf6ddd9ea7bcf9e20353dad305d737c /lib/tpm-v2.c
parent54b96e8b2bae9b61e9e178f9f10cf1b81bcda9ad (diff)
tpm: Support boot measurements
Add TPM2 functions to support boot measurement. This includes starting up the TPM, initializing/appending the event log, and measuring the U-Boot version. Much of the code was used in the EFI subsystem, so remove it there and use the common functions. Signed-off-by: Eddie James <eajames@linux.ibm.com> For the API moving around from EFI -> u-boot core Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org> For EFI testing Tested-by: Ilias Apalodimas <ilias.apalodimas@linaro.org> Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Diffstat (limited to 'lib/tpm-v2.c')
-rw-r--r--lib/tpm-v2.c814
1 files changed, 814 insertions, 0 deletions
diff --git a/lib/tpm-v2.c b/lib/tpm-v2.c
index 9ab5b46df17..bd0fb078dc3 100644
--- a/lib/tpm-v2.c
+++ b/lib/tpm-v2.c
@@ -1,16 +1,705 @@
// SPDX-License-Identifier: GPL-2.0+
/*
+ * Copyright (c) 2023 Linaro Limited
* Copyright (c) 2018 Bootlin
* Author: Miquel Raynal <miquel.raynal@bootlin.com>
*/
#include <common.h>
#include <dm.h>
+#include <dm/of_access.h>
+#include <tpm_api.h>
#include <tpm-common.h>
#include <tpm-v2.h>
+#include <u-boot/sha1.h>
+#include <u-boot/sha256.h>
+#include <u-boot/sha512.h>
+#include <version_string.h>
+#include <asm/io.h>
#include <linux/bitops.h>
+#include <linux/unaligned/be_byteshift.h>
+#include <linux/unaligned/generic.h>
+#include <linux/unaligned/le_byteshift.h>
+
#include "tpm-utils.h"
+const enum tpm2_algorithms tpm2_supported_algorithms[4] = {
+ TPM2_ALG_SHA1,
+ TPM2_ALG_SHA256,
+ TPM2_ALG_SHA384,
+ TPM2_ALG_SHA512,
+};
+
+int tcg2_get_active_pcr_banks(struct udevice *dev, u32 *active_pcr_banks)
+{
+ u32 supported = 0;
+ u32 pcr_banks = 0;
+ u32 active = 0;
+ int rc;
+
+ rc = tpm2_get_pcr_info(dev, &supported, &active, &pcr_banks);
+ if (rc)
+ return rc;
+
+ *active_pcr_banks = active;
+
+ return 0;
+}
+
+u32 tcg2_event_get_size(struct tpml_digest_values *digest_list)
+{
+ u32 len;
+ size_t i;
+
+ len = offsetof(struct tcg_pcr_event2, digests);
+ len += offsetof(struct tpml_digest_values, digests);
+ for (i = 0; i < digest_list->count; ++i) {
+ u16 l = tpm2_algorithm_to_len(digest_list->digests[i].hash_alg);
+
+ if (!l)
+ continue;
+
+ len += l + offsetof(struct tpmt_ha, digest);
+ }
+ len += sizeof(u32);
+
+ return len;
+}
+
+int tcg2_create_digest(struct udevice *dev, const u8 *input, u32 length,
+ struct tpml_digest_values *digest_list)
+{
+ u8 final[sizeof(union tpmu_ha)];
+ sha256_context ctx_256;
+ sha512_context ctx_512;
+ sha1_context ctx;
+ u32 active;
+ size_t i;
+ u32 len;
+ int rc;
+
+ rc = tcg2_get_active_pcr_banks(dev, &active);
+ if (rc)
+ return rc;
+
+ digest_list->count = 0;
+ for (i = 0; i < ARRAY_SIZE(tpm2_supported_algorithms); ++i) {
+ u32 mask =
+ tpm2_algorithm_to_mask(tpm2_supported_algorithms[i]);
+
+ if (!(active & mask))
+ continue;
+
+ switch (tpm2_supported_algorithms[i]) {
+ case TPM2_ALG_SHA1:
+ sha1_starts(&ctx);
+ sha1_update(&ctx, input, length);
+ sha1_finish(&ctx, final);
+ len = TPM2_SHA1_DIGEST_SIZE;
+ break;
+ case TPM2_ALG_SHA256:
+ sha256_starts(&ctx_256);
+ sha256_update(&ctx_256, input, length);
+ sha256_finish(&ctx_256, final);
+ len = TPM2_SHA256_DIGEST_SIZE;
+ break;
+ case TPM2_ALG_SHA384:
+ sha384_starts(&ctx_512);
+ sha384_update(&ctx_512, input, length);
+ sha384_finish(&ctx_512, final);
+ len = TPM2_SHA384_DIGEST_SIZE;
+ break;
+ case TPM2_ALG_SHA512:
+ sha512_starts(&ctx_512);
+ sha512_update(&ctx_512, input, length);
+ sha512_finish(&ctx_512, final);
+ len = TPM2_SHA512_DIGEST_SIZE;
+ break;
+ default:
+ printf("%s: unsupported algorithm %x\n", __func__,
+ tpm2_supported_algorithms[i]);
+ continue;
+ }
+
+ digest_list->digests[digest_list->count].hash_alg =
+ tpm2_supported_algorithms[i];
+ memcpy(&digest_list->digests[digest_list->count].digest, final,
+ len);
+ digest_list->count++;
+ }
+
+ return 0;
+}
+
+void tcg2_log_append(u32 pcr_index, u32 event_type,
+ struct tpml_digest_values *digest_list, u32 size,
+ const u8 *event, u8 *log)
+{
+ size_t len;
+ size_t pos;
+ u32 i;
+
+ pos = offsetof(struct tcg_pcr_event2, pcr_index);
+ put_unaligned_le32(pcr_index, log);
+ pos = offsetof(struct tcg_pcr_event2, event_type);
+ put_unaligned_le32(event_type, log + pos);
+ pos = offsetof(struct tcg_pcr_event2, digests) +
+ offsetof(struct tpml_digest_values, count);
+ put_unaligned_le32(digest_list->count, log + pos);
+
+ pos = offsetof(struct tcg_pcr_event2, digests) +
+ offsetof(struct tpml_digest_values, digests);
+ for (i = 0; i < digest_list->count; ++i) {
+ u16 hash_alg = digest_list->digests[i].hash_alg;
+
+ len = tpm2_algorithm_to_len(hash_alg);
+ if (!len)
+ continue;
+
+ pos += offsetof(struct tpmt_ha, hash_alg);
+ put_unaligned_le16(hash_alg, log + pos);
+ pos += offsetof(struct tpmt_ha, digest);
+ memcpy(log + pos, (u8 *)&digest_list->digests[i].digest, len);
+ pos += len;
+ }
+
+ put_unaligned_le32(size, log + pos);
+ pos += sizeof(u32);
+ memcpy(log + pos, event, size);
+}
+
+static int tcg2_log_append_check(struct tcg2_event_log *elog, u32 pcr_index,
+ u32 event_type,
+ struct tpml_digest_values *digest_list,
+ u32 size, const u8 *event)
+{
+ u32 event_size;
+ u8 *log;
+
+ event_size = size + tcg2_event_get_size(digest_list);
+ if (elog->log_position + event_size > elog->log_size) {
+ printf("%s: log too large: %u + %u > %u\n", __func__,
+ elog->log_position, event_size, elog->log_size);
+ return -ENOBUFS;
+ }
+
+ log = elog->log + elog->log_position;
+ elog->log_position += event_size;
+
+ tcg2_log_append(pcr_index, event_type, digest_list, size, event, log);
+
+ return 0;
+}
+
+static int tcg2_log_init(struct udevice *dev, struct tcg2_event_log *elog)
+{
+ struct tcg_efi_spec_id_event *ev;
+ struct tcg_pcr_event *log;
+ u32 event_size;
+ u32 count = 0;
+ u32 log_size;
+ u32 active;
+ u32 mask;
+ size_t i;
+ u16 len;
+ int rc;
+
+ rc = tcg2_get_active_pcr_banks(dev, &active);
+ if (rc)
+ return rc;
+
+ event_size = offsetof(struct tcg_efi_spec_id_event, digest_sizes);
+ for (i = 0; i < ARRAY_SIZE(tpm2_supported_algorithms); ++i) {
+ mask = tpm2_algorithm_to_mask(tpm2_supported_algorithms[i]);
+
+ if (!(active & mask))
+ continue;
+
+ switch (tpm2_supported_algorithms[i]) {
+ case TPM2_ALG_SHA1:
+ case TPM2_ALG_SHA256:
+ case TPM2_ALG_SHA384:
+ case TPM2_ALG_SHA512:
+ count++;
+ break;
+ default:
+ continue;
+ }
+ }
+
+ event_size += 1 +
+ (sizeof(struct tcg_efi_spec_id_event_algorithm_size) * count);
+ log_size = offsetof(struct tcg_pcr_event, event) + event_size;
+
+ if (log_size > elog->log_size) {
+ printf("%s: log too large: %u > %u\n", __func__, log_size,
+ elog->log_size);
+ return -ENOBUFS;
+ }
+
+ log = (struct tcg_pcr_event *)elog->log;
+ put_unaligned_le32(0, &log->pcr_index);
+ put_unaligned_le32(EV_NO_ACTION, &log->event_type);
+ memset(&log->digest, 0, sizeof(log->digest));
+ put_unaligned_le32(event_size, &log->event_size);
+
+ ev = (struct tcg_efi_spec_id_event *)log->event;
+ strlcpy((char *)ev->signature, TCG_EFI_SPEC_ID_EVENT_SIGNATURE_03,
+ sizeof(ev->signature));
+ put_unaligned_le32(0, &ev->platform_class);
+ ev->spec_version_minor = TCG_EFI_SPEC_ID_EVENT_SPEC_VERSION_MINOR_TPM2;
+ ev->spec_version_major = TCG_EFI_SPEC_ID_EVENT_SPEC_VERSION_MAJOR_TPM2;
+ ev->spec_errata = TCG_EFI_SPEC_ID_EVENT_SPEC_VERSION_ERRATA_TPM2;
+ ev->uintn_size = sizeof(size_t) / sizeof(u32);
+ put_unaligned_le32(count, &ev->number_of_algorithms);
+
+ count = 0;
+ for (i = 0; i < ARRAY_SIZE(tpm2_supported_algorithms); ++i) {
+ mask = tpm2_algorithm_to_mask(tpm2_supported_algorithms[i]);
+
+ if (!(active & mask))
+ continue;
+
+ len = tpm2_algorithm_to_len(tpm2_supported_algorithms[i]);
+ if (!len)
+ continue;
+
+ put_unaligned_le16(tpm2_supported_algorithms[i],
+ &ev->digest_sizes[count].algorithm_id);
+ put_unaligned_le16(len, &ev->digest_sizes[count].digest_size);
+ count++;
+ }
+
+ *((u8 *)ev + (event_size - 1)) = 0;
+ elog->log_position = log_size;
+
+ return 0;
+}
+
+static int tcg2_replay_eventlog(struct tcg2_event_log *elog,
+ struct udevice *dev,
+ struct tpml_digest_values *digest_list,
+ u32 log_position)
+{
+ const u32 offset = offsetof(struct tcg_pcr_event2, digests) +
+ offsetof(struct tpml_digest_values, digests);
+ u32 event_size;
+ u32 count;
+ u16 algo;
+ u32 pcr;
+ u32 pos;
+ u16 len;
+ u8 *log;
+ int rc;
+ u32 i;
+
+ while (log_position + offset < elog->log_size) {
+ log = elog->log + log_position;
+
+ pos = offsetof(struct tcg_pcr_event2, pcr_index);
+ pcr = get_unaligned_le32(log + pos);
+ pos = offsetof(struct tcg_pcr_event2, event_type);
+ if (!get_unaligned_le32(log + pos))
+ return 0;
+
+ pos = offsetof(struct tcg_pcr_event2, digests) +
+ offsetof(struct tpml_digest_values, count);
+ count = get_unaligned_le32(log + pos);
+ if (count > ARRAY_SIZE(tpm2_supported_algorithms) ||
+ (digest_list->count && digest_list->count != count))
+ return 0;
+
+ pos = offsetof(struct tcg_pcr_event2, digests) +
+ offsetof(struct tpml_digest_values, digests);
+ for (i = 0; i < count; ++i) {
+ pos += offsetof(struct tpmt_ha, hash_alg);
+ if (log_position + pos + sizeof(u16) >= elog->log_size)
+ return 0;
+
+ algo = get_unaligned_le16(log + pos);
+ pos += offsetof(struct tpmt_ha, digest);
+ switch (algo) {
+ case TPM2_ALG_SHA1:
+ case TPM2_ALG_SHA256:
+ case TPM2_ALG_SHA384:
+ case TPM2_ALG_SHA512:
+ len = tpm2_algorithm_to_len(algo);
+ break;
+ default:
+ return 0;
+ }
+
+ if (digest_list->count) {
+ if (algo != digest_list->digests[i].hash_alg ||
+ log_position + pos + len >= elog->log_size)
+ return 0;
+
+ memcpy(digest_list->digests[i].digest.sha512,
+ log + pos, len);
+ }
+
+ pos += len;
+ }
+
+ if (log_position + pos + sizeof(u32) >= elog->log_size)
+ return 0;
+
+ event_size = get_unaligned_le32(log + pos);
+ pos += event_size + sizeof(u32);
+ if (log_position + pos > elog->log_size)
+ return 0;
+
+ if (digest_list->count) {
+ rc = tcg2_pcr_extend(dev, pcr, digest_list);
+ if (rc)
+ return rc;
+ }
+
+ log_position += pos;
+ }
+
+ elog->log_position = log_position;
+ elog->found = true;
+ return 0;
+}
+
+static int tcg2_log_parse(struct udevice *dev, struct tcg2_event_log *elog)
+{
+ struct tpml_digest_values digest_list;
+ struct tcg_efi_spec_id_event *event;
+ struct tcg_pcr_event *log;
+ u32 log_active;
+ u32 calc_size;
+ u32 active;
+ u32 count;
+ u32 evsz;
+ u32 mask;
+ u16 algo;
+ u16 len;
+ int rc;
+ u32 i;
+ u16 j;
+
+ if (elog->log_size <= offsetof(struct tcg_pcr_event, event))
+ return 0;
+
+ log = (struct tcg_pcr_event *)elog->log;
+ if (get_unaligned_le32(&log->pcr_index) != 0 ||
+ get_unaligned_le32(&log->event_type) != EV_NO_ACTION)
+ return 0;
+
+ for (i = 0; i < sizeof(log->digest); i++) {
+ if (log->digest[i])
+ return 0;
+ }
+
+ evsz = get_unaligned_le32(&log->event_size);
+ if (evsz < offsetof(struct tcg_efi_spec_id_event, digest_sizes) ||
+ evsz + offsetof(struct tcg_pcr_event, event) > elog->log_size)
+ return 0;
+
+ event = (struct tcg_efi_spec_id_event *)log->event;
+ if (memcmp(event->signature, TCG_EFI_SPEC_ID_EVENT_SIGNATURE_03,
+ sizeof(TCG_EFI_SPEC_ID_EVENT_SIGNATURE_03)))
+ return 0;
+
+ if (event->spec_version_minor != TCG_EFI_SPEC_ID_EVENT_SPEC_VERSION_MINOR_TPM2 ||
+ event->spec_version_major != TCG_EFI_SPEC_ID_EVENT_SPEC_VERSION_MAJOR_TPM2)
+ return 0;
+
+ count = get_unaligned_le32(&event->number_of_algorithms);
+ if (count > ARRAY_SIZE(tpm2_supported_algorithms))
+ return 0;
+
+ calc_size = offsetof(struct tcg_efi_spec_id_event, digest_sizes) +
+ (sizeof(struct tcg_efi_spec_id_event_algorithm_size) * count) +
+ 1;
+ if (evsz != calc_size)
+ return 0;
+
+ rc = tcg2_get_active_pcr_banks(dev, &active);
+ if (rc)
+ return rc;
+
+ digest_list.count = 0;
+ log_active = 0;
+
+ for (i = 0; i < count; ++i) {
+ algo = get_unaligned_le16(&event->digest_sizes[i].algorithm_id);
+ mask = tpm2_algorithm_to_mask(algo);
+
+ if (!(active & mask))
+ return 0;
+
+ switch (algo) {
+ case TPM2_ALG_SHA1:
+ case TPM2_ALG_SHA256:
+ case TPM2_ALG_SHA384:
+ case TPM2_ALG_SHA512:
+ len = get_unaligned_le16(&event->digest_sizes[i].digest_size);
+ if (tpm2_algorithm_to_len(algo) != len)
+ return 0;
+ digest_list.digests[digest_list.count++].hash_alg = algo;
+ break;
+ default:
+ return 0;
+ }
+
+ log_active |= mask;
+ }
+
+ /* Ensure the previous firmware extended all the PCRs. */
+ if (log_active != active)
+ return 0;
+
+ /* Read PCR0 to check if previous firmware extended the PCRs or not. */
+ rc = tcg2_pcr_read(dev, 0, &digest_list);
+ if (rc)
+ return rc;
+
+ for (i = 0; i < digest_list.count; ++i) {
+ len = tpm2_algorithm_to_len(digest_list.digests[i].hash_alg);
+ for (j = 0; j < len; ++j) {
+ if (digest_list.digests[i].digest.sha512[j])
+ break;
+ }
+
+ /* PCR is non-zero; it has been extended, so skip extending. */
+ if (j != len) {
+ digest_list.count = 0;
+ break;
+ }
+ }
+
+ return tcg2_replay_eventlog(elog, dev, &digest_list,
+ offsetof(struct tcg_pcr_event, event) +
+ evsz);
+}
+
+int tcg2_pcr_extend(struct udevice *dev, u32 pcr_index,
+ struct tpml_digest_values *digest_list)
+{
+ u32 rc;
+ u32 i;
+
+ for (i = 0; i < digest_list->count; i++) {
+ u32 alg = digest_list->digests[i].hash_alg;
+
+ rc = tpm2_pcr_extend(dev, pcr_index, alg,
+ (u8 *)&digest_list->digests[i].digest,
+ tpm2_algorithm_to_len(alg));
+ if (rc) {
+ printf("%s: error pcr:%u alg:%08x\n", __func__,
+ pcr_index, alg);
+ return rc;
+ }
+ }
+
+ return 0;
+}
+
+int tcg2_pcr_read(struct udevice *dev, u32 pcr_index,
+ struct tpml_digest_values *digest_list)
+{
+ struct tpm_chip_priv *priv;
+ u32 rc;
+ u32 i;
+
+ priv = dev_get_uclass_priv(dev);
+ if (!priv)
+ return -ENODEV;
+
+ for (i = 0; i < digest_list->count; i++) {
+ u32 alg = digest_list->digests[i].hash_alg;
+ u8 *digest = (u8 *)&digest_list->digests[i].digest;
+
+ rc = tpm2_pcr_read(dev, pcr_index, priv->pcr_select_min, alg,
+ digest, tpm2_algorithm_to_len(alg), NULL);
+ if (rc) {
+ printf("%s: error pcr:%u alg:%08x\n", __func__,
+ pcr_index, alg);
+ return rc;
+ }
+ }
+
+ return 0;
+}
+
+int tcg2_measure_data(struct udevice *dev, struct tcg2_event_log *elog,
+ u32 pcr_index, u32 size, const u8 *data, u32 event_type,
+ u32 event_size, const u8 *event)
+{
+ struct tpml_digest_values digest_list;
+ int rc;
+
+ if (data)
+ rc = tcg2_create_digest(dev, data, size, &digest_list);
+ else
+ rc = tcg2_create_digest(dev, event, event_size, &digest_list);
+ if (rc)
+ return rc;
+
+ rc = tcg2_pcr_extend(dev, pcr_index, &digest_list);
+ if (rc)
+ return rc;
+
+ return tcg2_log_append_check(elog, pcr_index, event_type, &digest_list,
+ event_size, event);
+}
+
+int tcg2_log_prepare_buffer(struct udevice *dev, struct tcg2_event_log *elog,
+ bool ignore_existing_log)
+{
+ struct tcg2_event_log log;
+ int rc;
+
+ elog->log_position = 0;
+ elog->found = false;
+
+ rc = tcg2_platform_get_log(dev, (void **)&log.log, &log.log_size);
+ if (!rc) {
+ log.log_position = 0;
+ log.found = false;
+
+ if (!ignore_existing_log) {
+ rc = tcg2_log_parse(dev, &log);
+ if (rc)
+ return rc;
+ }
+
+ if (elog->log_size) {
+ if (log.found) {
+ if (elog->log_size < log.log_position)
+ return -ENOSPC;
+
+ /*
+ * Copy the discovered log into the user buffer
+ * if there's enough space.
+ */
+ memcpy(elog->log, log.log, log.log_position);
+ }
+
+ unmap_physmem(log.log, MAP_NOCACHE);
+ } else {
+ elog->log = log.log;
+ elog->log_size = log.log_size;
+ }
+
+ elog->log_position = log.log_position;
+ elog->found = log.found;
+ }
+
+ /*
+ * Initialize the log buffer if no log was discovered and the buffer is
+ * valid. User's can pass in their own buffer as a fallback if no
+ * memory region is found.
+ */
+ if (!elog->found && elog->log_size)
+ rc = tcg2_log_init(dev, elog);
+
+ return rc;
+}
+
+int tcg2_measurement_init(struct udevice **dev, struct tcg2_event_log *elog,
+ bool ignore_existing_log)
+{
+ int rc;
+
+ rc = tcg2_platform_get_tpm2(dev);
+ if (rc)
+ return rc;
+
+ rc = tpm_auto_start(*dev);
+ if (rc)
+ return rc;
+
+ rc = tcg2_log_prepare_buffer(*dev, elog, ignore_existing_log);
+ if (rc) {
+ tcg2_measurement_term(*dev, elog, true);
+ return rc;
+ }
+
+ rc = tcg2_measure_event(*dev, elog, 0, EV_S_CRTM_VERSION,
+ strlen(version_string) + 1,
+ (u8 *)version_string);
+ if (rc) {
+ tcg2_measurement_term(*dev, elog, true);
+ return rc;
+ }
+
+ return 0;
+}
+
+void tcg2_measurement_term(struct udevice *dev, struct tcg2_event_log *elog,
+ bool error)
+{
+ u32 event = error ? 0x1 : 0xffffffff;
+ int i;
+
+ for (i = 0; i < 8; ++i)
+ tcg2_measure_event(dev, elog, i, EV_SEPARATOR, sizeof(event),
+ (const u8 *)&event);
+
+ if (elog->log)
+ unmap_physmem(elog->log, MAP_NOCACHE);
+}
+
+__weak int tcg2_platform_get_log(struct udevice *dev, void **addr, u32 *size)
+{
+ const __be32 *addr_prop;
+ const __be32 *size_prop;
+ int asize;
+ int ssize;
+
+ *addr = NULL;
+ *size = 0;
+
+ addr_prop = dev_read_prop(dev, "tpm_event_log_addr", &asize);
+ if (!addr_prop)
+ addr_prop = dev_read_prop(dev, "linux,sml-base", &asize);
+
+ size_prop = dev_read_prop(dev, "tpm_event_log_size", &ssize);
+ if (!size_prop)
+ size_prop = dev_read_prop(dev, "linux,sml-size", &ssize);
+
+ if (addr_prop && size_prop) {
+ u64 a = of_read_number(addr_prop, asize / sizeof(__be32));
+ u64 s = of_read_number(size_prop, ssize / sizeof(__be32));
+
+ *addr = map_physmem(a, s, MAP_NOCACHE);
+ *size = (u32)s;
+ } else {
+ struct ofnode_phandle_args args;
+ phys_addr_t a;
+ fdt_size_t s;
+
+ if (dev_read_phandle_with_args(dev, "memory-region", NULL, 0,
+ 0, &args))
+ return -ENODEV;
+
+ a = ofnode_get_addr_size(args.node, "reg", &s);
+ if (a == FDT_ADDR_T_NONE)
+ return -ENOMEM;
+
+ *addr = map_physmem(a, s, MAP_NOCACHE);
+ *size = (u32)s;
+ }
+
+ return 0;
+}
+
+__weak int tcg2_platform_get_tpm2(struct udevice **dev)
+{
+ for_each_tpm_device(*dev) {
+ if (tpm_get_version(*dev) == TPM_V2)
+ return 0;
+ }
+
+ return -ENODEV;
+}
+
+__weak void tcg2_platform_startup_error(struct udevice *dev, int rc) {}
+
u32 tpm2_startup(struct udevice *dev, enum tpm2_startup_types mode)
{
const u8 command_v2[12] = {
@@ -359,6 +1048,131 @@ u32 tpm2_get_capability(struct udevice *dev, u32 capability, u32 property,
return 0;
}
+static int tpm2_get_num_pcr(struct udevice *dev, u32 *num_pcr)
+{
+ u8 response[(sizeof(struct tpms_capability_data) -
+ offsetof(struct tpms_capability_data, data))];
+ u32 properties_offset =
+ offsetof(struct tpml_tagged_tpm_property, tpm_property) +
+ offsetof(struct tpms_tagged_property, value);
+ u32 ret;
+
+ memset(response, 0, sizeof(response));
+ ret = tpm2_get_capability(dev, TPM2_CAP_TPM_PROPERTIES,
+ TPM2_PT_PCR_COUNT, response, 1);
+ if (ret)
+ return ret;
+
+ *num_pcr = get_unaligned_be32(response + properties_offset);
+ if (*num_pcr > TPM2_MAX_PCRS) {
+ printf("%s: too many pcrs: %u\n", __func__, *num_pcr);
+ return -E2BIG;
+ }
+
+ return 0;
+}
+
+static bool tpm2_is_active_pcr(struct tpms_pcr_selection *selection)
+{
+ int i;
+
+ /*
+ * check the pcr_select. If at least one of the PCRs supports the
+ * algorithm add it on the active ones
+ */
+ for (i = 0; i < selection->size_of_select; i++) {
+ if (selection->pcr_select[i])
+ return true;
+ }
+
+ return false;
+}
+
+int tpm2_get_pcr_info(struct udevice *dev, u32 *supported_pcr, u32 *active_pcr,
+ u32 *pcr_banks)
+{
+ u8 response[(sizeof(struct tpms_capability_data) -
+ offsetof(struct tpms_capability_data, data))];
+ struct tpml_pcr_selection pcrs;
+ u32 num_pcr;
+ size_t i;
+ u32 ret;
+
+ *supported_pcr = 0;
+ *active_pcr = 0;
+ *pcr_banks = 0;
+ memset(response, 0, sizeof(response));
+ ret = tpm2_get_capability(dev, TPM2_CAP_PCRS, 0, response, 1);
+ if (ret)
+ return ret;
+
+ pcrs.count = get_unaligned_be32(response);
+ /*
+ * We only support 5 algorithms for now so check against that
+ * instead of TPM2_NUM_PCR_BANKS
+ */
+ if (pcrs.count > ARRAY_SIZE(tpm2_supported_algorithms) ||
+ pcrs.count < 1) {
+ printf("%s: too many pcrs: %u\n", __func__, pcrs.count);
+ return -EMSGSIZE;
+ }
+
+ ret = tpm2_get_num_pcr(dev, &num_pcr);
+ if (ret)
+ return ret;
+
+ for (i = 0; i < pcrs.count; i++) {
+ /*
+ * Definition of TPMS_PCR_SELECTION Structure
+ * hash: u16
+ * size_of_select: u8
+ * pcr_select: u8 array
+ *
+ * The offsets depend on the number of the device PCRs
+ * so we have to calculate them based on that
+ */
+ u32 hash_offset = offsetof(struct tpml_pcr_selection, selection) +
+ i * offsetof(struct tpms_pcr_selection, pcr_select) +
+ i * ((num_pcr + 7) / 8);
+ u32 size_select_offset =
+ hash_offset + offsetof(struct tpms_pcr_selection,
+ size_of_select);
+ u32 pcr_select_offset =
+ hash_offset + offsetof(struct tpms_pcr_selection,
+ pcr_select);
+
+ pcrs.selection[i].hash =
+ get_unaligned_be16(response + hash_offset);
+ pcrs.selection[i].size_of_select =
+ __get_unaligned_be(response + size_select_offset);
+ if (pcrs.selection[i].size_of_select > TPM2_PCR_SELECT_MAX) {
+ printf("%s: pcrs selection too large: %u\n", __func__,
+ pcrs.selection[i].size_of_select);
+ return -ENOBUFS;
+ }
+ /* copy the array of pcr_select */
+ memcpy(pcrs.selection[i].pcr_select, response + pcr_select_offset,
+ pcrs.selection[i].size_of_select);
+ }
+
+ for (i = 0; i < pcrs.count; i++) {
+ u32 hash_mask = tpm2_algorithm_to_mask(pcrs.selection[i].hash);
+
+ if (hash_mask) {
+ *supported_pcr |= hash_mask;
+ if (tpm2_is_active_pcr(&pcrs.selection[i]))
+ *active_pcr |= hash_mask;
+ } else {
+ printf("%s: unknown algorithm %x\n", __func__,
+ pcrs.selection[i].hash);
+ }
+ }
+
+ *pcr_banks = pcrs.count;
+
+ return 0;
+}
+
u32 tpm2_dam_reset(struct udevice *dev, const char *pw, const ssize_t pw_sz)
{
u8 command_v2[COMMAND_BUFFER_SIZE] = {