linux-integrity.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [RFC PATCH] tpm: only set efi_tpm_final_log_size after successful event log parsing
@ 2019-09-18 19:16 Jerry Snitselaar
  2019-09-18 19:23 ` Jerry Snitselaar
                   ` (2 more replies)
  0 siblings, 3 replies; 6+ messages in thread
From: Jerry Snitselaar @ 2019-09-18 19:16 UTC (permalink / raw)
  To: linux-efi
  Cc: linux-kernel, linux-integrity, Matthew Garrett, Ard Biesheuvel,
	Jarkko Sakkinen

If __calc_tpm2_event_size fails to parse an event it will return 0,
resulting tpm2_calc_event_log_size returning -1. Currently
there is no check of this return value, and efi_tpm_final_log_size
can end up being set to this negative value resulting
in a panic like the following:

[    0.774340] BUG: unable to handle page fault for address: ffffbc8fc00866ad
[    0.774788] #PF: supervisor read access in kernel mode
[    0.774788] #PF: error_code(0x0000) - not-present page
[    0.774788] PGD 107d36067 P4D 107d36067 PUD 107d37067 PMD 107d38067 PTE 0
[    0.774788] Oops: 0000 [#1] SMP PTI
[    0.774788] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.3.0-0.rc2.1.elrdy.x86_64 #1
[    0.774788] Hardware name: LENOVO 20HGS22D0W/20HGS22D0W, BIOS N1WET51W (1.30 ) 09/14/2018
[    0.774788] RIP: 0010:memcpy_erms+0x6/0x10
[    0.774788] Code: 90 90 90 90 eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe
[    0.774788] RSP: 0000:ffffbc8fc0073b30 EFLAGS: 00010286
[    0.774788] RAX: ffff9b1fc7c5b367 RBX: ffff9b1fc8390000 RCX: ffffffffffffe962
[    0.774788] RDX: ffffffffffffe962 RSI: ffffbc8fc00866ad RDI: ffff9b1fc7c5b367
[    0.774788] RBP: ffff9b1c10ca7018 R08: ffffbc8fc0085fff R09: 8000000000000063
[    0.774788] R10: 0000000000001000 R11: 000fffffffe00000 R12: 0000000000003367
[    0.774788] R13: ffff9b1fcc47c010 R14: ffffbc8fc0085000 R15: 0000000000000002
[    0.774788] FS:  0000000000000000(0000) GS:ffff9b1fce200000(0000) knlGS:0000000000000000
[    0.774788] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    0.774788] CR2: ffffbc8fc00866ad CR3: 000000029f60a001 CR4: 00000000003606f0
[    0.774788] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[    0.774788] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[    0.774788] Call Trace:
[    0.774788]  tpm_read_log_efi+0x156/0x1a0
[    0.774788]  tpm_bios_log_setup+0xc8/0x190
[    0.774788]  tpm_chip_register+0x50/0x1c0
[    0.774788]  tpm_tis_core_init.cold.9+0x28c/0x466
[    0.774788]  tpm_tis_plat_probe+0xcc/0xea
[    0.774788]  platform_drv_probe+0x35/0x80
[    0.774788]  really_probe+0xef/0x390
[    0.774788]  driver_probe_device+0xb4/0x100
[    0.774788]  device_driver_attach+0x4f/0x60
[    0.774788]  __driver_attach+0x86/0x140
[    0.774788]  ? device_driver_attach+0x60/0x60
[    0.774788]  bus_for_each_dev+0x76/0xc0
[    0.774788]  ? klist_add_tail+0x3b/0x70
[    0.774788]  bus_add_driver+0x14a/0x1e0
[    0.774788]  ? tpm_init+0xea/0xea
[    0.774788]  ? do_early_param+0x8e/0x8e
[    0.774788]  driver_register+0x6b/0xb0
[    0.774788]  ? tpm_init+0xea/0xea
[    0.774788]  init_tis+0x86/0xd8
[    0.774788]  ? do_early_param+0x8e/0x8e
[    0.774788]  ? driver_register+0x94/0xb0
[    0.774788]  do_one_initcall+0x46/0x1e4
[    0.774788]  ? do_early_param+0x8e/0x8e
[    0.774788]  kernel_init_freeable+0x199/0x242
[    0.774788]  ? rest_init+0xaa/0xaa
[    0.774788]  kernel_init+0xa/0x106
[    0.774788]  ret_from_fork+0x35/0x40
[    0.774788] Modules linked in:
[    0.774788] CR2: ffffbc8fc00866ad
[    0.774788] ---[ end trace 42930799f8d6eaea ]---
[    0.774788] RIP: 0010:memcpy_erms+0x6/0x10
[    0.774788] Code: 90 90 90 90 eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe
[    0.774788] RSP: 0000:ffffbc8fc0073b30 EFLAGS: 00010286
[    0.774788] RAX: ffff9b1fc7c5b367 RBX: ffff9b1fc8390000 RCX: ffffffffffffe962
[    0.774788] RDX: ffffffffffffe962 RSI: ffffbc8fc00866ad RDI: ffff9b1fc7c5b367
[    0.774788] RBP: ffff9b1c10ca7018 R08: ffffbc8fc0085fff R09: 8000000000000063
[    0.774788] R10: 0000000000001000 R11: 000fffffffe00000 R12: 0000000000003367
[    0.774788] R13: ffff9b1fcc47c010 R14: ffffbc8fc0085000 R15: 0000000000000002
[    0.774788] FS:  0000000000000000(0000) GS:ffff9b1fce200000(0000) knlGS:0000000000000000
[    0.774788] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    0.774788] CR2: ffffbc8fc00866ad CR3: 000000029f60a001 CR4: 00000000003606f0
[    0.774788] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[    0.774788] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[    0.774788] Kernel panic - not syncing: Fatal exception
[    0.774788] Kernel Offset: 0x1d000000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
[    0.774788] ---[ end Kernel panic - not syncing: Fatal exception ]---

Fixes: c46f3405692de ("tpm: Reserve the TPM final events table")
Cc: Matthew Garrett <mjg59@google.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
---
 drivers/firmware/efi/tpm.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/firmware/efi/tpm.c b/drivers/firmware/efi/tpm.c
index 1d3f5ca3eaaf..5cd00a7833c2 100644
--- a/drivers/firmware/efi/tpm.c
+++ b/drivers/firmware/efi/tpm.c
@@ -40,8 +40,7 @@ int __init efi_tpm_eventlog_init(void)
 {
 	struct linux_efi_tpm_eventlog *log_tbl;
 	struct efi_tcg2_final_events_table *final_tbl;
-	unsigned int tbl_size;
-	int ret = 0;
+	int tbl_size, ret = 0;
 
 	if (efi.tpm_log == EFI_INVALID_TABLE_ADDR) {
 		/*
@@ -80,11 +79,17 @@ int __init efi_tpm_eventlog_init(void)
 					    + sizeof(final_tbl->nr_events),
 					    final_tbl->nr_events,
 					    log_tbl->log);
+	if (tbl_size < 0) {
+		pr_err("Failed to parse event in TPM Final Event log\n");
+		goto calc_out;
+	}
+
 	memblock_reserve((unsigned long)final_tbl,
 			 tbl_size + sizeof(*final_tbl));
-	early_memunmap(final_tbl, sizeof(*final_tbl));
 	efi_tpm_final_log_size = tbl_size;
 
+calc_out:
+	early_memunmap(final_tbl, sizeof(*final_tbl));
 out:
 	early_memunmap(log_tbl, sizeof(*log_tbl));
 	return ret;
-- 
2.23.0


^ permalink raw reply related	[flat|nested] 6+ messages in thread
* Re: [RFC PATCH] tpm: only set efi_tpm_final_log_size after successful event log parsing
  2019-09-18 19:16 [RFC PATCH] tpm: only set efi_tpm_final_log_size after successful event log parsing Jerry Snitselaar
@ 2019-09-18 19:23 ` Jerry Snitselaar
  2019-09-23 14:52   ` Laura Abbott
  2019-09-23 17:10 ` Jerry Snitselaar
  2019-09-25 10:38 ` Jarkko Sakkinen
  2 siblings, 1 reply; 6+ messages in thread
From: Jerry Snitselaar @ 2019-09-18 19:23 UTC (permalink / raw)
  To: linux-efi
  Cc: linux-kernel, linux-integrity, Matthew Garrett, Ard Biesheuvel,
	Jarkko Sakkinen

On Wed Sep 18 19, Jerry Snitselaar wrote:
>If __calc_tpm2_event_size fails to parse an event it will return 0,
>resulting tpm2_calc_event_log_size returning -1. Currently
>there is no check of this return value, and efi_tpm_final_log_size
>can end up being set to this negative value resulting
>in a panic like the following:
>
>[    0.774340] BUG: unable to handle page fault for address: ffffbc8fc00866ad
>[    0.774788] #PF: supervisor read access in kernel mode
>[    0.774788] #PF: error_code(0x0000) - not-present page
>[    0.774788] PGD 107d36067 P4D 107d36067 PUD 107d37067 PMD 107d38067 PTE 0
>[    0.774788] Oops: 0000 [#1] SMP PTI
>[    0.774788] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.3.0-0.rc2.1.elrdy.x86_64 #1
>[    0.774788] Hardware name: LENOVO 20HGS22D0W/20HGS22D0W, BIOS N1WET51W (1.30 ) 09/14/2018
>[    0.774788] RIP: 0010:memcpy_erms+0x6/0x10
>[    0.774788] Code: 90 90 90 90 eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe
>[    0.774788] RSP: 0000:ffffbc8fc0073b30 EFLAGS: 00010286
>[    0.774788] RAX: ffff9b1fc7c5b367 RBX: ffff9b1fc8390000 RCX: ffffffffffffe962
>[    0.774788] RDX: ffffffffffffe962 RSI: ffffbc8fc00866ad RDI: ffff9b1fc7c5b367
>[    0.774788] RBP: ffff9b1c10ca7018 R08: ffffbc8fc0085fff R09: 8000000000000063
>[    0.774788] R10: 0000000000001000 R11: 000fffffffe00000 R12: 0000000000003367
>[    0.774788] R13: ffff9b1fcc47c010 R14: ffffbc8fc0085000 R15: 0000000000000002
>[    0.774788] FS:  0000000000000000(0000) GS:ffff9b1fce200000(0000) knlGS:0000000000000000
>[    0.774788] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>[    0.774788] CR2: ffffbc8fc00866ad CR3: 000000029f60a001 CR4: 00000000003606f0
>[    0.774788] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>[    0.774788] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>[    0.774788] Call Trace:
>[    0.774788]  tpm_read_log_efi+0x156/0x1a0
>[    0.774788]  tpm_bios_log_setup+0xc8/0x190
>[    0.774788]  tpm_chip_register+0x50/0x1c0
>[    0.774788]  tpm_tis_core_init.cold.9+0x28c/0x466
>[    0.774788]  tpm_tis_plat_probe+0xcc/0xea
>[    0.774788]  platform_drv_probe+0x35/0x80
>[    0.774788]  really_probe+0xef/0x390
>[    0.774788]  driver_probe_device+0xb4/0x100
>[    0.774788]  device_driver_attach+0x4f/0x60
>[    0.774788]  __driver_attach+0x86/0x140
>[    0.774788]  ? device_driver_attach+0x60/0x60
>[    0.774788]  bus_for_each_dev+0x76/0xc0
>[    0.774788]  ? klist_add_tail+0x3b/0x70
>[    0.774788]  bus_add_driver+0x14a/0x1e0
>[    0.774788]  ? tpm_init+0xea/0xea
>[    0.774788]  ? do_early_param+0x8e/0x8e
>[    0.774788]  driver_register+0x6b/0xb0
>[    0.774788]  ? tpm_init+0xea/0xea
>[    0.774788]  init_tis+0x86/0xd8
>[    0.774788]  ? do_early_param+0x8e/0x8e
>[    0.774788]  ? driver_register+0x94/0xb0
>[    0.774788]  do_one_initcall+0x46/0x1e4
>[    0.774788]  ? do_early_param+0x8e/0x8e
>[    0.774788]  kernel_init_freeable+0x199/0x242
>[    0.774788]  ? rest_init+0xaa/0xaa
>[    0.774788]  kernel_init+0xa/0x106
>[    0.774788]  ret_from_fork+0x35/0x40
>[    0.774788] Modules linked in:
>[    0.774788] CR2: ffffbc8fc00866ad
>[    0.774788] ---[ end trace 42930799f8d6eaea ]---
>[    0.774788] RIP: 0010:memcpy_erms+0x6/0x10
>[    0.774788] Code: 90 90 90 90 eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe
>[    0.774788] RSP: 0000:ffffbc8fc0073b30 EFLAGS: 00010286
>[    0.774788] RAX: ffff9b1fc7c5b367 RBX: ffff9b1fc8390000 RCX: ffffffffffffe962
>[    0.774788] RDX: ffffffffffffe962 RSI: ffffbc8fc00866ad RDI: ffff9b1fc7c5b367
>[    0.774788] RBP: ffff9b1c10ca7018 R08: ffffbc8fc0085fff R09: 8000000000000063
>[    0.774788] R10: 0000000000001000 R11: 000fffffffe00000 R12: 0000000000003367
>[    0.774788] R13: ffff9b1fcc47c010 R14: ffffbc8fc0085000 R15: 0000000000000002
>[    0.774788] FS:  0000000000000000(0000) GS:ffff9b1fce200000(0000) knlGS:0000000000000000
>[    0.774788] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>[    0.774788] CR2: ffffbc8fc00866ad CR3: 000000029f60a001 CR4: 00000000003606f0
>[    0.774788] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>[    0.774788] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>[    0.774788] Kernel panic - not syncing: Fatal exception
>[    0.774788] Kernel Offset: 0x1d000000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
>[    0.774788] ---[ end Kernel panic - not syncing: Fatal exception ]---
>
>Fixes: c46f3405692de ("tpm: Reserve the TPM final events table")
>Cc: Matthew Garrett <mjg59@google.com>
>Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
>Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>---
> drivers/firmware/efi/tpm.c | 11 ++++++++---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
>diff --git a/drivers/firmware/efi/tpm.c b/drivers/firmware/efi/tpm.c
>index 1d3f5ca3eaaf..5cd00a7833c2 100644
>--- a/drivers/firmware/efi/tpm.c
>+++ b/drivers/firmware/efi/tpm.c
>@@ -40,8 +40,7 @@ int __init efi_tpm_eventlog_init(void)
> {
> 	struct linux_efi_tpm_eventlog *log_tbl;
> 	struct efi_tcg2_final_events_table *final_tbl;
>-	unsigned int tbl_size;
>-	int ret = 0;
>+	int tbl_size, ret = 0;
>
> 	if (efi.tpm_log == EFI_INVALID_TABLE_ADDR) {
> 		/*
>@@ -80,11 +79,17 @@ int __init efi_tpm_eventlog_init(void)
> 					    + sizeof(final_tbl->nr_events),
> 					    final_tbl->nr_events,
> 					    log_tbl->log);
>+	if (tbl_size < 0) {
>+		pr_err("Failed to parse event in TPM Final Event log\n");
>+		goto calc_out;
>+	}
>+
> 	memblock_reserve((unsigned long)final_tbl,
> 			 tbl_size + sizeof(*final_tbl));
>-	early_memunmap(final_tbl, sizeof(*final_tbl));
> 	efi_tpm_final_log_size = tbl_size;
>
>+calc_out:
>+	early_memunmap(final_tbl, sizeof(*final_tbl));
> out:
> 	early_memunmap(log_tbl, sizeof(*log_tbl));
> 	return ret;
>-- 
>2.23.0
>

I'm not sure what to set ret to, which currently isn't checked, or if anything
should be done with efi.tpm_final_log.


^ permalink raw reply	[flat|nested] 6+ messages in thread
* Re: [RFC PATCH] tpm: only set efi_tpm_final_log_size after successful event log parsing
  2019-09-18 19:23 ` Jerry Snitselaar
@ 2019-09-23 14:52   ` Laura Abbott
  0 siblings, 0 replies; 6+ messages in thread
From: Laura Abbott @ 2019-09-23 14:52 UTC (permalink / raw)
  To: linux-efi, linux-kernel, linux-integrity, Matthew Garrett,
	Ard Biesheuvel, Jarkko Sakkinen

On 9/18/19 8:23 PM, Jerry Snitselaar wrote:
> On Wed Sep 18 19, Jerry Snitselaar wrote:
>> If __calc_tpm2_event_size fails to parse an event it will return 0,
>> resulting tpm2_calc_event_log_size returning -1. Currently
>> there is no check of this return value, and efi_tpm_final_log_size
>> can end up being set to this negative value resulting
>> in a panic like the following:
>>
>> [    0.774340] BUG: unable to handle page fault for address: ffffbc8fc00866ad
>> [    0.774788] #PF: supervisor read access in kernel mode
>> [    0.774788] #PF: error_code(0x0000) - not-present page
>> [    0.774788] PGD 107d36067 P4D 107d36067 PUD 107d37067 PMD 107d38067 PTE 0
>> [    0.774788] Oops: 0000 [#1] SMP PTI
>> [    0.774788] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.3.0-0.rc2.1.elrdy.x86_64 #1
>> [    0.774788] Hardware name: LENOVO 20HGS22D0W/20HGS22D0W, BIOS N1WET51W (1.30 ) 09/14/2018
>> [    0.774788] RIP: 0010:memcpy_erms+0x6/0x10
>> [    0.774788] Code: 90 90 90 90 eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe
>> [    0.774788] RSP: 0000:ffffbc8fc0073b30 EFLAGS: 00010286
>> [    0.774788] RAX: ffff9b1fc7c5b367 RBX: ffff9b1fc8390000 RCX: ffffffffffffe962
>> [    0.774788] RDX: ffffffffffffe962 RSI: ffffbc8fc00866ad RDI: ffff9b1fc7c5b367
>> [    0.774788] RBP: ffff9b1c10ca7018 R08: ffffbc8fc0085fff R09: 8000000000000063
>> [    0.774788] R10: 0000000000001000 R11: 000fffffffe00000 R12: 0000000000003367
>> [    0.774788] R13: ffff9b1fcc47c010 R14: ffffbc8fc0085000 R15: 0000000000000002
>> [    0.774788] FS:  0000000000000000(0000) GS:ffff9b1fce200000(0000) knlGS:0000000000000000
>> [    0.774788] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [    0.774788] CR2: ffffbc8fc00866ad CR3: 000000029f60a001 CR4: 00000000003606f0
>> [    0.774788] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>> [    0.774788] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>> [    0.774788] Call Trace:
>> [    0.774788]  tpm_read_log_efi+0x156/0x1a0
>> [    0.774788]  tpm_bios_log_setup+0xc8/0x190
>> [    0.774788]  tpm_chip_register+0x50/0x1c0
>> [    0.774788]  tpm_tis_core_init.cold.9+0x28c/0x466
>> [    0.774788]  tpm_tis_plat_probe+0xcc/0xea
>> [    0.774788]  platform_drv_probe+0x35/0x80
>> [    0.774788]  really_probe+0xef/0x390
>> [    0.774788]  driver_probe_device+0xb4/0x100
>> [    0.774788]  device_driver_attach+0x4f/0x60
>> [    0.774788]  __driver_attach+0x86/0x140
>> [    0.774788]  ? device_driver_attach+0x60/0x60
>> [    0.774788]  bus_for_each_dev+0x76/0xc0
>> [    0.774788]  ? klist_add_tail+0x3b/0x70
>> [    0.774788]  bus_add_driver+0x14a/0x1e0
>> [    0.774788]  ? tpm_init+0xea/0xea
>> [    0.774788]  ? do_early_param+0x8e/0x8e
>> [    0.774788]  driver_register+0x6b/0xb0
>> [    0.774788]  ? tpm_init+0xea/0xea
>> [    0.774788]  init_tis+0x86/0xd8
>> [    0.774788]  ? do_early_param+0x8e/0x8e
>> [    0.774788]  ? driver_register+0x94/0xb0
>> [    0.774788]  do_one_initcall+0x46/0x1e4
>> [    0.774788]  ? do_early_param+0x8e/0x8e
>> [    0.774788]  kernel_init_freeable+0x199/0x242
>> [    0.774788]  ? rest_init+0xaa/0xaa
>> [    0.774788]  kernel_init+0xa/0x106
>> [    0.774788]  ret_from_fork+0x35/0x40
>> [    0.774788] Modules linked in:
>> [    0.774788] CR2: ffffbc8fc00866ad
>> [    0.774788] ---[ end trace 42930799f8d6eaea ]---
>> [    0.774788] RIP: 0010:memcpy_erms+0x6/0x10
>> [    0.774788] Code: 90 90 90 90 eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe
>> [    0.774788] RSP: 0000:ffffbc8fc0073b30 EFLAGS: 00010286
>> [    0.774788] RAX: ffff9b1fc7c5b367 RBX: ffff9b1fc8390000 RCX: ffffffffffffe962
>> [    0.774788] RDX: ffffffffffffe962 RSI: ffffbc8fc00866ad RDI: ffff9b1fc7c5b367
>> [    0.774788] RBP: ffff9b1c10ca7018 R08: ffffbc8fc0085fff R09: 8000000000000063
>> [    0.774788] R10: 0000000000001000 R11: 000fffffffe00000 R12: 0000000000003367
>> [    0.774788] R13: ffff9b1fcc47c010 R14: ffffbc8fc0085000 R15: 0000000000000002
>> [    0.774788] FS:  0000000000000000(0000) GS:ffff9b1fce200000(0000) knlGS:0000000000000000
>> [    0.774788] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [    0.774788] CR2: ffffbc8fc00866ad CR3: 000000029f60a001 CR4: 00000000003606f0
>> [    0.774788] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>> [    0.774788] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>> [    0.774788] Kernel panic - not syncing: Fatal exception
>> [    0.774788] Kernel Offset: 0x1d000000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
>> [    0.774788] ---[ end Kernel panic - not syncing: Fatal exception ]---
>>
>> Fixes: c46f3405692de ("tpm: Reserve the TPM final events table")
>> Cc: Matthew Garrett <mjg59@google.com>
>> Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
>> Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>> Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>> ---
>> drivers/firmware/efi/tpm.c | 11 ++++++++---
>> 1 file changed, 8 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/firmware/efi/tpm.c b/drivers/firmware/efi/tpm.c
>> index 1d3f5ca3eaaf..5cd00a7833c2 100644
>> --- a/drivers/firmware/efi/tpm.c
>> +++ b/drivers/firmware/efi/tpm.c
>> @@ -40,8 +40,7 @@ int __init efi_tpm_eventlog_init(void)
>> {
>>     struct linux_efi_tpm_eventlog *log_tbl;
>>     struct efi_tcg2_final_events_table *final_tbl;
>> -    unsigned int tbl_size;
>> -    int ret = 0;
>> +    int tbl_size, ret = 0;
>>
>>     if (efi.tpm_log == EFI_INVALID_TABLE_ADDR) {
>>         /*
>> @@ -80,11 +79,17 @@ int __init efi_tpm_eventlog_init(void)
>>                         + sizeof(final_tbl->nr_events),
>>                         final_tbl->nr_events,
>>                         log_tbl->log);
>> +    if (tbl_size < 0) {
>> +        pr_err("Failed to parse event in TPM Final Event log\n");
>> +        goto calc_out;
>> +    }
>> +
>>     memblock_reserve((unsigned long)final_tbl,
>>              tbl_size + sizeof(*final_tbl));
>> -    early_memunmap(final_tbl, sizeof(*final_tbl));
>>     efi_tpm_final_log_size = tbl_size;
>>
>> +calc_out:
>> +    early_memunmap(final_tbl, sizeof(*final_tbl));
>> out:
>>     early_memunmap(log_tbl, sizeof(*log_tbl));
>>     return ret;
>> -- 
>> 2.23.0
>>
> 
> I'm not sure what to set ret to, which currently isn't checked, or if anything
> should be done with efi.tpm_final_log.
> 

fwiw, this is being hit by a number of Fedora users
https://bugzilla.redhat.com/show_bug.cgi?id=1752961
I'd like to bring this patch to Fedora if there's general agreement
this is the correct approach.

Thanks,
Laura


^ permalink raw reply	[flat|nested] 6+ messages in thread
* Re: [RFC PATCH] tpm: only set efi_tpm_final_log_size after successful event log parsing
  2019-09-18 19:16 [RFC PATCH] tpm: only set efi_tpm_final_log_size after successful event log parsing Jerry Snitselaar
  2019-09-18 19:23 ` Jerry Snitselaar
@ 2019-09-23 17:10 ` Jerry Snitselaar
  2019-09-25 10:40   ` Jarkko Sakkinen
  2019-09-25 10:38 ` Jarkko Sakkinen
  2 siblings, 1 reply; 6+ messages in thread
From: Jerry Snitselaar @ 2019-09-23 17:10 UTC (permalink / raw)
  To: linux-efi
  Cc: linux-kernel, linux-integrity, Matthew Garrett, Ard Biesheuvel,
	Jarkko Sakkinen

Any thoughts on this? I know of at least 2 Lenovo models that are
running into this problem.

In the case of the one I have currently have access to the problem is
that the hash algorithm id for an event isn't one that is currently in
the TCG registry, and it fails to find a match when walking the
digest_sizes array. That seems like an issue for the vendor to fix in the bios,
but we should look at the return value of tpm2_calc_event_log_size and not
stick a negative value in efi_tpm_final_log_size.


^ permalink raw reply	[flat|nested] 6+ messages in thread
* Re: [RFC PATCH] tpm: only set efi_tpm_final_log_size after successful event log parsing
  2019-09-18 19:16 [RFC PATCH] tpm: only set efi_tpm_final_log_size after successful event log parsing Jerry Snitselaar
  2019-09-18 19:23 ` Jerry Snitselaar
  2019-09-23 17:10 ` Jerry Snitselaar
@ 2019-09-25 10:38 ` Jarkko Sakkinen
  2 siblings, 0 replies; 6+ messages in thread
From: Jarkko Sakkinen @ 2019-09-25 10:38 UTC (permalink / raw)
  To: Jerry Snitselaar
  Cc: linux-efi, linux-kernel, linux-integrity, Matthew Garrett,
	Ard Biesheuvel

On Wed, Sep 18, 2019 at 12:16:26PM -0700, Jerry Snitselaar wrote:
> +	if (tbl_size < 0) {
> +		pr_err("Failed to parse event in TPM Final Event log\n");

FW_BUG?

> +		goto calc_out;
> +	}
> +
>  	memblock_reserve((unsigned long)final_tbl,
>  			 tbl_size + sizeof(*final_tbl));
> -	early_memunmap(final_tbl, sizeof(*final_tbl));
>  	efi_tpm_final_log_size = tbl_size;
>  
> +calc_out:
> +	early_memunmap(final_tbl, sizeof(*final_tbl));

out_calc

>  out:
>  	early_memunmap(log_tbl, sizeof(*log_tbl));
>  	return ret;
> -- 
> 2.23.0
> 

/Jarkko

^ permalink raw reply	[flat|nested] 6+ messages in thread
* Re: [RFC PATCH] tpm: only set efi_tpm_final_log_size after successful event log parsing
  2019-09-23 17:10 ` Jerry Snitselaar
@ 2019-09-25 10:40   ` Jarkko Sakkinen
  0 siblings, 0 replies; 6+ messages in thread
From: Jarkko Sakkinen @ 2019-09-25 10:40 UTC (permalink / raw)
  To: linux-efi, linux-kernel, linux-integrity, Matthew Garrett,
	Ard Biesheuvel

On Mon, Sep 23, 2019 at 10:10:10AM -0700, Jerry Snitselaar wrote:
> Any thoughts on this? I know of at least 2 Lenovo models that are
> running into this problem.
> 
> In the case of the one I have currently have access to the problem is
> that the hash algorithm id for an event isn't one that is currently in
> the TCG registry, and it fails to find a match when walking the
> digest_sizes array. That seems like an issue for the vendor to fix in the bios,
> but we should look at the return value of tpm2_calc_event_log_size and not
> stick a negative value in efi_tpm_final_log_size.

Please use then "pr_err(FW_BUG".

Also, since you know the context you could add a comment along the
lines what you wrote.

/Jarkko

^ permalink raw reply	[flat|nested] 6+ messages in thread
end of thread, other threads:[~2019-09-25 10:40 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-09-18 19:16 [RFC PATCH] tpm: only set efi_tpm_final_log_size after successful event log parsing Jerry Snitselaar
2019-09-18 19:23 ` Jerry Snitselaar
2019-09-23 14:52   ` Laura Abbott
2019-09-23 17:10 ` Jerry Snitselaar
2019-09-25 10:40   ` Jarkko Sakkinen
2019-09-25 10:38 ` Jarkko Sakkinen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).