Skip to contentSkip to conversation

tpm: return false from tpm_amd_is_rng_defective on non-x86 platforms

PATCH In review Jerry Snitselaar 4 participants 7 messages Jun 29, 2023 – Jul 10, 2023

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 tpm: return false from tpm_amd_is_rng_defective on non-x86 platforms

tpm_amd_is_rng_defective is for dealing with an issue related to the AMD firmware TPM, so on non-x86 architectures just have it inline and return false.

Cc Jarkko Sakkinen <[email protected]>
Cc "Jason A. Donenfeld" <[email protected]>
Cc Jason Gunthorpe <[email protected]>
Cc Peter Huewe <[email protected]>
Cc Linux regressions mailing list <[email protected]>
Cc Mario Limonciello <[email protected]>
Fixes f1324bbc4011 ("tpm: disable hwrng for fTPM on some AMD designs")
---
 drivers/char/tpm/tpm-chip.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/char/tpm/tpm-chip.c +7 -0
index cd48033b804a..cf5499e51999 100644
@@ -518,6 +518,7 @@ static int tpm_add_legacy_sysfs(struct tpm_chip *chip)
  * 6.x.y.z series: 6.0.18.6 +
  * 3.x.y.z series: 3.57.y.5 +
  */
+#ifdef CONFIG_X86
 static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 {
 	u32 val1, val2;
@@ -566,6 +567,12 @@ static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 
 	return true;
 }
+#else
+static inline bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
+{
+	return false;
+}
+#endif /* CONFIG_X86 */
 
 static int tpm_hwrng_read(struct hwrng *rng, void *data, size_t max, bool wait)
 {
Signature · 1 line
2.38.1
maintainer
#1 lore raw
Quoted 47 lines · Jerry Snitselaar

tpm_amd_is_rng_defective is for dealing with an issue related to the AMD firmware TPM, so on non-x86 architectures just have it inline and return false.

Cc Jarkko Sakkinen <[email protected]>
Cc "Jason A. Donenfeld" <[email protected]>
Cc Jason Gunthorpe <[email protected]>
Cc Peter Huewe <[email protected]>
Cc Linux regressions mailing list <[email protected]>
Cc Mario Limonciello <[email protected]>
Fixes f1324bbc4011 ("tpm: disable hwrng for fTPM on some AMD designs")
---
 drivers/char/tpm/tpm-chip.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/char/tpm/tpm-chip.c +7 -0
index cd48033b804a..cf5499e51999 100644
@@ -518,6 +518,7 @@ static int tpm_add_legacy_sysfs(struct tpm_chip *chip)
  * 6.x.y.z series: 6.0.18.6 +
  * 3.x.y.z series: 3.57.y.5 +
  */
+#ifdef CONFIG_X86
 static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 {
 	u32 val1, val2;
@@ -566,6 +567,12 @@ static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 
 	return true;
 }
+#else
+static inline bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
+{
+	return false;
+}
+#endif /* CONFIG_X86 */
 
 static int tpm_hwrng_read(struct hwrng *rng, void *data, size_t max, bool wait)
 {
Signature · 1 line
2.38.1

Sanity check, this was the right patch, right?

I'll apply it.

BR, Jarkko

#2 lore raw

[Public]

Quoted 62 lines

-----Original Message-----
From: Jarkko Sakkinen <[email protected]>
Sent: Friday, June 30, 2023 5:07 AM
To: Jerry Snitselaar <[email protected]>; [email protected]; [email protected]
Cc: Jason A. Donenfeld <[email protected]>; Jason Gunthorpe <[email protected]>; Peter Huewe <[email protected]>; [email protected]; Linux regressions mailing list <[email protected]>; Limonciello, Mario <[email protected]>; Aneesh Kumar K . V <[email protected]>; Sachin Sant <[email protected]>
Subject: Re: [PATCH] tpm: return false from tpm_amd_is_rng_defective on non-x86 platforms

Quoted 4 lines · Jerry Snitselaar

tpm_amd_is_rng_defective is for dealing with an issue related to the AMD firmware TPM, so on non-x86 architectures just have it inline and return false.

Quoted 10 lines
Cc Jarkko Sakkinen <[email protected]>
Cc "Jason A. Donenfeld" <[email protected]>
Cc Jason Gunthorpe <[email protected]>
Cc Peter Huewe <[email protected]>
Cc Linux regressions mailing list <[email protected]>
Cc Mario Limonciello <[email protected]>

[email protected]/

Quoted 1 line
Fixes f1324bbc4011 ("tpm: disable hwrng for fTPM on some AMD

designs")

drivers/char/tpm/tpm-chip.c static int tpm_add_legacy_sysfs(struct tpm_chip

Quoted 10 lines
---
 drivers/char/tpm/tpm-chip.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/char/tpm/tpm-chip.c +0 -0
index cd48033b804a..cf5499e51999 100644
@@ -518,6 +518,7 @@ static int tpm_add_legacy_sysfs(struct tpm_chip

*chip)

drivers/char/tpm/tpm-chip.c static bool tpm_amd_is_rng_defective(struct

Quoted 8 lines
drivers/char/tpm/tpm-chip.c +1 -0
  * 6.x.y.z series: 6.0.18.6 +
  * 3.x.y.z series: 3.57.y.5 +
  */
+#ifdef CONFIG_X86
 static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 {
    u32 val1, val2;
@@ -566,6 +567,12 @@ static bool tpm_amd_is_rng_defective(struct

tpm_chip *chip)

Quoted 10 lines
drivers/char/tpm/tpm-chip.c +6 -0
    return true;
 }
+#else
+static inline bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
+{
+   return false;
+}
+#endif /* CONFIG_X86 */
 
Quoted 8 lines

drivers/char/tpm/tpm-chip.c static bool tpm_amd_is_rng_defective(struct · line 576

Quoted 1 line
drivers/char/tpm/tpm-chip.c +0 -0
 static int tpm_hwrng_read(struct hwrng *rng, void *data, size_t max, bool

wait)

Quoted 3 lines
drivers/char/tpm/tpm-chip.c +0 -1
 {
--

2.38.1

Sanity check, this was the right patch, right?

Yup.

Quoted 3 lines

I'll apply it.

BR, Jarkko

Thanks!

#3 lore raw
In reply to Jarkko Sakkinen
Quoted 51 lines · Jarkko Sakkinen
Quoted 47 lines · Jerry Snitselaar

tpm_amd_is_rng_defective is for dealing with an issue related to the AMD firmware TPM, so on non-x86 architectures just have it inline and return false.

Cc Jarkko Sakkinen <[email protected]>
Cc "Jason A. Donenfeld" <[email protected]>
Cc Jason Gunthorpe <[email protected]>
Cc Peter Huewe <[email protected]>
Cc Linux regressions mailing list <[email protected]>
Cc Mario Limonciello <[email protected]>
Fixes f1324bbc4011 ("tpm: disable hwrng for fTPM on some AMD designs")
---
 drivers/char/tpm/tpm-chip.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/char/tpm/tpm-chip.c +7 -0
index cd48033b804a..cf5499e51999 100644
@@ -518,6 +518,7 @@ static int tpm_add_legacy_sysfs(struct tpm_chip *chip)
  * 6.x.y.z series: 6.0.18.6 +
  * 3.x.y.z series: 3.57.y.5 +
  */
+#ifdef CONFIG_X86
 static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 {
 	u32 val1, val2;
@@ -566,6 +567,12 @@ static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 
 	return true;
 }
+#else
+static inline bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
+{
+	return false;
+}
+#endif /* CONFIG_X86 */
 
 static int tpm_hwrng_read(struct hwrng *rng, void *data, size_t max, bool wait)
 {
Signature · 1 line
2.38.1

Sanity check, this was the right patch, right?

Quoted 3 lines

I'll apply it.

BR, Jarkko

Sorry, I've been dealing with a family health issue the past week. It wasn't clear to me why chip->ops was null when I first took a look, but I think I understand now looking at it again this morning. The stack trace shows it in the device_shutdown() path:

    [ 34.381674] NIP [c0000000009db1e4] tpm_amd_is_rng_defective+0x74/0x240
    [ 34.381681] LR [c0000000009db928] tpm_chip_unregister+0x138/0x160
    [ 34.381685] Call Trace:
    [ 34.381686] [c00000009742faa0] [c0000000009db928] tpm_chip_unregister+0x138/0x160
    [ 34.381690] [c00000009742fae0] [c0000000009eab94] tpm_ibmvtpm_remove+0x34/0x130
    [ 34.381695] [c00000009742fb50] [c000000000115738] vio_bus_remove+0x58/0xd0
    [ 34.381701] [c00000009742fb90] [c000000000a01ecc] device_shutdown+0x21c/0x39c
    [ 34.381705] [c00000009742fc20] [c0000000001a2684] kernel_restart_prepare+0x54/0x70
    [ 34.381710] [c00000009742fc40] [c000000000292c48] kernel_kexec+0xa8/0x100
    [ 34.381714] [c00000009742fcb0] [c0000000001a2cd4] __do_sys_reboot+0x214/0x2c0
    [ 34.381718] [c00000009742fe10] [c000000000034adc] system_call_exception+0x13c/0x340
    [ 34.381723] [c00000009742fe50] [c00000000000d05c] system_call_vectored_common+0x15c/0x2ec

So I think what happened is:

device_shutdown -> dev->class->shutdown_pre (tpm_class_shutdown) // clears chip->ops

                -> dev->bus->shutdown (vio_bus_shutdown) -> vio_bus_remove -> viodrv->remove (tpm_ibmvtpm_remove) -> tpm_chip_unregister -> tpm_amd_is_rng_defective -> oops!

I guess anything that gets called in the tpm_chip_unregister path should be doing a check of chip->ops prior to using it. So I think Mario's patch would still be a good thing to have.

Regards,
Jerry

#4 lore raw

Hi, Thorsten here, the Linux kernel's regression tracker. Top-posting for once, to make this easily accessible to everyone.

So what's the way forward now? It sounded like Jarkko wanted to apply the patch from this thread days ago, but that didn't happen afaics. Then below message showed up, but Marios patch also wasn't applied.

Is this intentional, or did something somewhere fall through the cracks?

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat) --
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
If I did something stupid, please tell me, as explained on that page.

#regzbot poke

Quoted 88 lines · Jerry Snitselaar
Quoted 51 lines · Jarkko Sakkinen
Quoted 47 lines · Jerry Snitselaar

tpm_amd_is_rng_defective is for dealing with an issue related to the AMD firmware TPM, so on non-x86 architectures just have it inline and return false.

Cc Jarkko Sakkinen <[email protected]>
Cc "Jason A. Donenfeld" <[email protected]>
Cc Jason Gunthorpe <[email protected]>
Cc Peter Huewe <[email protected]>
Cc Linux regressions mailing list <[email protected]>
Cc Mario Limonciello <[email protected]>
Fixes f1324bbc4011 ("tpm: disable hwrng for fTPM on some AMD designs")
---
 drivers/char/tpm/tpm-chip.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/char/tpm/tpm-chip.c +7 -0
index cd48033b804a..cf5499e51999 100644
@@ -518,6 +518,7 @@ static int tpm_add_legacy_sysfs(struct tpm_chip *chip)
  * 6.x.y.z series: 6.0.18.6 +
  * 3.x.y.z series: 3.57.y.5 +
  */
+#ifdef CONFIG_X86
 static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 {
 	u32 val1, val2;
@@ -566,6 +567,12 @@ static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 
 	return true;
 }
+#else
+static inline bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
+{
+	return false;
+}
+#endif /* CONFIG_X86 */
 
 static int tpm_hwrng_read(struct hwrng *rng, void *data, size_t max, bool wait)
 {
Signature · 1 line
2.38.1

Sanity check, this was the right patch, right?

Quoted 3 lines

I'll apply it.

BR, Jarkko

Sorry, I've been dealing with a family health issue the past week. It wasn't clear to me why chip->ops was null when I first took a look, but I think I understand now looking at it again this morning. The stack trace shows it in the device_shutdown() path:

    [ 34.381674] NIP [c0000000009db1e4] tpm_amd_is_rng_defective+0x74/0x240
    [ 34.381681] LR [c0000000009db928] tpm_chip_unregister+0x138/0x160
    [ 34.381685] Call Trace:
    [ 34.381686] [c00000009742faa0] [c0000000009db928] tpm_chip_unregister+0x138/0x160
    [ 34.381690] [c00000009742fae0] [c0000000009eab94] tpm_ibmvtpm_remove+0x34/0x130
    [ 34.381695] [c00000009742fb50] [c000000000115738] vio_bus_remove+0x58/0xd0
    [ 34.381701] [c00000009742fb90] [c000000000a01ecc] device_shutdown+0x21c/0x39c
    [ 34.381705] [c00000009742fc20] [c0000000001a2684] kernel_restart_prepare+0x54/0x70
    [ 34.381710] [c00000009742fc40] [c000000000292c48] kernel_kexec+0xa8/0x100
    [ 34.381714] [c00000009742fcb0] [c0000000001a2cd4] __do_sys_reboot+0x214/0x2c0
    [ 34.381718] [c00000009742fe10] [c000000000034adc] system_call_exception+0x13c/0x340
    [ 34.381723] [c00000009742fe50] [c00000000000d05c] system_call_vectored_common+0x15c/0x2ec

So I think what happened is:

device_shutdown -> dev->class->shutdown_pre (tpm_class_shutdown) // clears chip->ops

                -> dev->bus->shutdown (vio_bus_shutdown) -> vio_bus_remove -> viodrv->remove (tpm_ibmvtpm_remove) -> tpm_chip_unregister -> tpm_amd_is_rng_defective -> oops!

I guess anything that gets called in the tpm_chip_unregister path should be doing a check of chip->ops prior to using it. So I think Mario's patch would still be a good thing to have.

Regards,
Jerry

#5 lore raw
Quoted 10 lines · Linux regression tracking (Thorsten Leemhuis)

Hi, Thorsten here, the Linux kernel's regression tracker. Top-posting for once, to make this easily accessible to everyone.

So what's the way forward now? It sounded like Jarkko wanted to apply the patch from this thread days ago, but that didn't happen afaics. Then below message showed up, but Marios patch also wasn't applied.

Is this intentional, or did something somewhere fall through the cracks?

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)

I haven't seen any update to Jarkko's repo.

My patch resolves the immediate issue being seen on the ppc system, and was mostly just me asking why even go through this amd specific code on non-x86 systems.

The vio bus shutdown code only does the remove call when kexec is in progress. The pnp and platform bus type shutdown calls do not do something similar so maybe the check in Mario's patch isn't needed, but I don't think it would hurt to have it in there.

Regards,
Jerry

Quoted 96 lines

--
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
If I did something stupid, please tell me, as explained on that page.

#regzbot poke

Quoted 88 lines · Jerry Snitselaar
Quoted 51 lines · Jarkko Sakkinen
Quoted 47 lines · Jerry Snitselaar

tpm_amd_is_rng_defective is for dealing with an issue related to the AMD firmware TPM, so on non-x86 architectures just have it inline and return false.

Cc Jarkko Sakkinen <[email protected]>
Cc "Jason A. Donenfeld" <[email protected]>
Cc Jason Gunthorpe <[email protected]>
Cc Peter Huewe <[email protected]>
Cc Linux regressions mailing list <[email protected]>
Cc Mario Limonciello <[email protected]>
Fixes f1324bbc4011 ("tpm: disable hwrng for fTPM on some AMD designs")
---
 drivers/char/tpm/tpm-chip.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/char/tpm/tpm-chip.c +7 -0
index cd48033b804a..cf5499e51999 100644
@@ -518,6 +518,7 @@ static int tpm_add_legacy_sysfs(struct tpm_chip *chip)
  * 6.x.y.z series: 6.0.18.6 +
  * 3.x.y.z series: 3.57.y.5 +
  */
+#ifdef CONFIG_X86
 static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 {
 	u32 val1, val2;
@@ -566,6 +567,12 @@ static bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
 
 	return true;
 }
+#else
+static inline bool tpm_amd_is_rng_defective(struct tpm_chip *chip)
+{
+	return false;
+}
+#endif /* CONFIG_X86 */
 
 static int tpm_hwrng_read(struct hwrng *rng, void *data, size_t max, bool wait)
 {
Signature · 1 line
2.38.1

Sanity check, this was the right patch, right?

Quoted 3 lines

I'll apply it.

BR, Jarkko

Sorry, I've been dealing with a family health issue the past week. It wasn't clear to me why chip->ops was null when I first took a look, but I think I understand now looking at it again this morning. The stack trace shows it in the device_shutdown() path:

    [ 34.381674] NIP [c0000000009db1e4] tpm_amd_is_rng_defective+0x74/0x240
    [ 34.381681] LR [c0000000009db928] tpm_chip_unregister+0x138/0x160
    [ 34.381685] Call Trace:
    [ 34.381686] [c00000009742faa0] [c0000000009db928] tpm_chip_unregister+0x138/0x160
    [ 34.381690] [c00000009742fae0] [c0000000009eab94] tpm_ibmvtpm_remove+0x34/0x130
    [ 34.381695] [c00000009742fb50] [c000000000115738] vio_bus_remove+0x58/0xd0
    [ 34.381701] [c00000009742fb90] [c000000000a01ecc] device_shutdown+0x21c/0x39c
    [ 34.381705] [c00000009742fc20] [c0000000001a2684] kernel_restart_prepare+0x54/0x70
    [ 34.381710] [c00000009742fc40] [c000000000292c48] kernel_kexec+0xa8/0x100
    [ 34.381714] [c00000009742fcb0] [c0000000001a2cd4] __do_sys_reboot+0x214/0x2c0
    [ 34.381718] [c00000009742fe10] [c000000000034adc] system_call_exception+0x13c/0x340
    [ 34.381723] [c00000009742fe50] [c00000000000d05c] system_call_vectored_common+0x15c/0x2ec

So I think what happened is:

device_shutdown -> dev->class->shutdown_pre (tpm_class_shutdown) // clears chip->ops

                -> dev->bus->shutdown (vio_bus_shutdown) -> vio_bus_remove -> viodrv->remove (tpm_ibmvtpm_remove) -> tpm_chip_unregister -> tpm_amd_is_rng_defective -> oops!

I guess anything that gets called in the tpm_chip_unregister path should be doing a check of chip->ops prior to using it. So I think Mario's patch would still be a good thing to have.

Regards,
Jerry

maintainer
#6 lore raw
Quoted 14 lines · Jerry Snitselaar
Quoted 10 lines · Linux regression tracking (Thorsten Leemhuis)

Hi, Thorsten here, the Linux kernel's regression tracker. Top-posting for once, to make this easily accessible to everyone.

So what's the way forward now? It sounded like Jarkko wanted to apply the patch from this thread days ago, but that didn't happen afaics. Then below message showed up, but Marios patch also wasn't applied.

Is this intentional, or did something somewhere fall through the cracks?

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)

I haven't seen any update to Jarkko's repo.

Quoted 8 lines

My patch resolves the immediate issue being seen on the ppc system, and was mostly just me asking why even go through this amd specific code on non-x86 systems.

The vio bus shutdown code only does the remove call when kexec is in progress. The pnp and platform bus type shutdown calls do not do something similar so maybe the check in Mario's patch isn't needed, but I don't think it would hurt to have it in there.

The patch has been applied. I'm sending pull request for rc2 today.

The merge window was a bit conflicting with the summer season, and also relocating.

BR, Jarkko