[PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

From: Lukasz Majczak <lma@semihalf.com>

This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
(SLB 9670):

[    4.324298] TPM returned invalid status
[    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f

Background
==========

TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
Interface Locality Usage per Register, Table 39 Register Behavior Based on
Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
0xFF in case of lack of locality. The described situation manifests itself
with the following warning trace:

The fix
=======

Add the proper decorations to tpm_tis_gen_interrupt() and
tpm_get_timeouts().

Cc: Laurent Bigonville <bigon@debian.org>
Fixes: a3fbfae82b4c ("tpm: take TPM chip power gating out of tpm_transmit()")
Signed-off-by: Lukasz Majczak <lma@semihalf.com>
Co-developed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
---
It was easier to write a new patch based on v5. Since v5 had a crippled
change log I just restarted the version count.
 drivers/char/tpm/tpm-interface.c | 16 ++++++++++++++--
 drivers/char/tpm/tpm_tis_core.c  | 16 +++++++++++++---
 2 files changed, 27 insertions(+), 5 deletions(-)

diff --git a/drivers/char/tpm/tpm-interface.c b/drivers/char/tpm/tpm-interface.c
index 1621ce818705..9ab155ae5726 100644
--- a/drivers/char/tpm/tpm-interface.c
+++ b/drivers/char/tpm/tpm-interface.c
@@ -238,13 +238,25 @@ EXPORT_SYMBOL_GPL(tpm_transmit_cmd);
 
 int tpm_get_timeouts(struct tpm_chip *chip)
 {
+	int ret;
+
 	if (chip->flags & TPM_CHIP_FLAG_HAVE_TIMEOUTS)
 		return 0;
 
+	/* TPM 2.0 */
 	if (chip->flags & TPM_CHIP_FLAG_TPM2)
 		return tpm2_get_timeouts(chip);
-	else
-		return tpm1_get_timeouts(chip);
+
+
+	ret = tpm_chip_start(chip);
+	if (ret)
+		return ret;
+
+	ret = tpm1_get_timeouts(chip);
+
+	tpm_chip_stop(chip);
+
+	return ret;
 }
 EXPORT_SYMBOL_GPL(tpm_get_timeouts);
 
diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
index 431919d5f48a..ce80e7a2d420 100644
--- a/drivers/char/tpm/tpm_tis_core.c
+++ b/drivers/char/tpm/tpm_tis_core.c
@@ -707,12 +707,22 @@ static int tpm_tis_gen_interrupt(struct tpm_chip *chip)
 	const char *desc = "attempting to generate an interrupt";
 	u32 cap2;
 	cap_t cap;
+	int ret;
 
+	/* TPM 2.0 */
 	if (chip->flags & TPM_CHIP_FLAG_TPM2)
 		return tpm2_get_tpm_pt(chip, 0x100, &cap2, desc);
-	else
-		return tpm1_getcap(chip, TPM_CAP_PROP_TIS_TIMEOUT, &cap, desc,
-				  0);
+
+	/* TPM 1.2 */
+	ret = request_locality(chip, 0);
+	if (ret < 0)
+		return ret;
+
+	ret = tpm1_getcap(chip, TPM_CAP_PROP_TIS_TIMEOUT, &cap, desc, 0);
+
+	release_locality(chip, 0);
+
+	return ret;
 }
 
 /* Register the IRQ and issue a command that will cause an interrupt. If an
-- 
2.30.1

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Paul Menzel]

Dear Jarkko,


Am 16.02.21 um 09:17 schrieb Jarkko Sakkinen:
> From: Lukasz Majczak <lma@semihalf.com>
> 
> This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> (SLB 9670):
> 
> [    4.324298] TPM returned invalid status
> [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> 
> Background
> ==========
> 
> TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> Interface Locality Usage per Register, Table 39 Register Behavior Based on
> Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> 0xFF in case of lack of locality. The described situation manifests itself
> with the following warning trace:

Did you miss to paste the trace?

[…]


Kind regards,

Paul

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Tue, Feb 16, 2021 at 09:26:49AM +0100, Paul Menzel wrote:
> Dear Jarkko,
> 
> 
> Am 16.02.21 um 09:17 schrieb Jarkko Sakkinen:
> > From: Lukasz Majczak <lma@semihalf.com>
> > 
> > This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> > (SLB 9670):
> > 
> > [    4.324298] TPM returned invalid status
> > [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> > 
> > Background
> > ==========
> > 
> > TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> > Interface Locality Usage per Register, Table 39 Register Behavior Based on
> > Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> > 0xFF in case of lack of locality. The described situation manifests itself
> > with the following warning trace:
> 
> Did you miss to paste the trace?
> 
> […]
> 
> 
> Kind regards,
> 
> Paul

There is partial trace, which I took from the Lukasz patch. IMHO that is
sufficient.

And I have no means to generate the full trace anyway.

/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Paul Menzel]

Dear Jarko,


Am 16.02.21 um 09:39 schrieb Jarkko Sakkinen:
> On Tue, Feb 16, 2021 at 09:26:49AM +0100, Paul Menzel wrote:

>> Am 16.02.21 um 09:17 schrieb Jarkko Sakkinen:
>>> From: Lukasz Majczak <lma@semihalf.com>
>>>
>>> This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
>>> (SLB 9670):
>>>
>>> [    4.324298] TPM returned invalid status
>>> [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
>>>
>>> Background
>>> ==========
>>>
>>> TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
>>> Interface Locality Usage per Register, Table 39 Register Behavior Based on
>>> Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
>>> 0xFF in case of lack of locality. The described situation manifests itself
>>> with the following warning trace:
>>
>> Did you miss to paste the trace?
>>
>> […]

> There is partial trace, which I took from the Lukasz patch. IMHO that is
> sufficient.

Reading the paragraph with “ with the following warning trace:” (and 
colon), I would expect to see some trace, and not the next section 
header *The fix*.

> And I have no means to generate the full trace anyway.

No problem.


Kind regards,

Paul

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Tue, Feb 16, 2021 at 09:50:40AM +0100, Paul Menzel wrote:
> Dear Jarko,
> 
> 
> Am 16.02.21 um 09:39 schrieb Jarkko Sakkinen:
> > On Tue, Feb 16, 2021 at 09:26:49AM +0100, Paul Menzel wrote:
> 
> > > Am 16.02.21 um 09:17 schrieb Jarkko Sakkinen:
> > > > From: Lukasz Majczak <lma@semihalf.com>
> > > > 
> > > > This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> > > > (SLB 9670):
> > > > 
> > > > [    4.324298] TPM returned invalid status
> > > > [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> > > > 
> > > > Background
> > > > ==========
> > > > 
> > > > TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> > > > Interface Locality Usage per Register, Table 39 Register Behavior Based on
> > > > Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> > > > 0xFF in case of lack of locality. The described situation manifests itself
> > > > with the following warning trace:
> > > 
> > > Did you miss to paste the trace?
> > > 
> > > […]
> 
> > There is partial trace, which I took from the Lukasz patch. IMHO that is
> > sufficient.
> 
> Reading the paragraph with “ with the following warning trace:” (and colon),
> I would expect to see some trace, and not the next section header *The fix*.


Aah! I took the paragraph from the original commit. So, what if I just
remove the last sentence? I see what you mean.

> 
> > And I have no means to generate the full trace anyway.
> 
> No problem.
> 
> 
> Kind regards,
> 
> Paul
> 

/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Laurent Bigonville]

Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
> From: Lukasz Majczak <lma@semihalf.com>
>
> This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> (SLB 9670):
>
> [    4.324298] TPM returned invalid status
> [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
>
> Background
> ==========
>
> TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> Interface Locality Usage per Register, Table 39 Register Behavior Based on
> Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> 0xFF in case of lack of locality. The described situation manifests itself
> with the following warning trace:
>
> The fix
> =======
>
> Add the proper decorations to tpm_tis_gen_interrupt() and
> tpm_get_timeouts().

I tried that patch (alone on the top of the HEAD of Linus master) and I 
still get the same trace in dmesg

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Laurent Bigonville]

Le 16/02/21 à 12:17, Lukasz Majczak a écrit :
> Hi Laurent,
Hello Lukasz,
>
> I think your case is different, as the trace shows the issue in the 
> place already guarded by tpm_chip_start()/stop(). Can you past more 
> dmesg log? (With tpm prefix?). I believe tpm_tis_status() might return 
> something different than 0xff here.

I see "WARNING: CPU: 5 PID: 449 at drivers/char/tpm/tpm_tis_core.c:249 
tpm_tis_status+0x86/0xa0 [tpm_tis_core]" in dmesg

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Lukasz Majczak]

Hi Laurent,

I think your case is different, as the trace shows the issue occurs in
the place already guarded by tpm_chip_start()/stop() - tpm2_probe().
Can you paste more dmesg log? (With tpm prefix?). I believe
tpm_tis_status() might return something different than 0xff here.
Please ensure you have applied Jarkko's patch that logs status
(https://patchwork.kernel.org/project/linux-integrity/patch/20210202222150.120664-1-jarkko@kernel.org/)

Best regards
Lukasz

wt., 16 lut 2021 o 12:02 Laurent Bigonville <bigon@debian.org> napisał(a):
>
> Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
> > From: Lukasz Majczak <lma@semihalf.com>
> >
> > This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> > (SLB 9670):
> >
> > [    4.324298] TPM returned invalid status
> > [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> >
> > Background
> > ==========
> >
> > TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> > Interface Locality Usage per Register, Table 39 Register Behavior Based on
> > Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> > 0xFF in case of lack of locality. The described situation manifests itself
> > with the following warning trace:
> >
> > The fix
> > =======
> >
> > Add the proper decorations to tpm_tis_gen_interrupt() and
> > tpm_get_timeouts().
>
> I tried that patch (alone on the top of the HEAD of Linus master) and I
> still get the same trace in dmesg
>

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Tue, Feb 16, 2021 at 03:09:49PM +0100, Lukasz Majczak wrote:
> Hi Laurent,
> 
> I think your case is different, as the trace shows the issue occurs in
> the place already guarded by tpm_chip_start()/stop() - tpm2_probe().
> Can you paste more dmesg log? (With tpm prefix?). I believe
> tpm_tis_status() might return something different than 0xff here.
> Please ensure you have applied Jarkko's patch that logs status
> (https://patchwork.kernel.org/project/linux-integrity/patch/20210202222150.120664-1-jarkko@kernel.org/)


I would politely ask you:

- Not to top-post. Quote and reponse properly.
- No HTML emails.

> Best regards
> Lukasz


/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
> Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
> > From: Lukasz Majczak <lma@semihalf.com>
> > 
> > This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> > (SLB 9670):
> > 
> > [    4.324298] TPM returned invalid status
> > [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> > 
> > Background
> > ==========
> > 
> > TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> > Interface Locality Usage per Register, Table 39 Register Behavior Based on
> > Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> > 0xFF in case of lack of locality. The described situation manifests itself
> > with the following warning trace:
> > 
> > The fix
> > =======
> > 
> > Add the proper decorations to tpm_tis_gen_interrupt() and
> > tpm_get_timeouts().
> 
> I tried that patch (alone on the top of the HEAD of Linus master) and I
> still get the same trace in dmesg

Can you give a shot to

git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git

It has couple of more fixes, and paste the log.

/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Tue, Feb 16, 2021 at 06:16:58PM +0200, Jarkko Sakkinen wrote:
> On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
> > Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
> > > From: Lukasz Majczak <lma@semihalf.com>
> > > 
> > > This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> > > (SLB 9670):
> > > 
> > > [    4.324298] TPM returned invalid status
> > > [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> > > 
> > > Background
> > > ==========
> > > 
> > > TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> > > Interface Locality Usage per Register, Table 39 Register Behavior Based on
> > > Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> > > 0xFF in case of lack of locality. The described situation manifests itself
> > > with the following warning trace:
> > > 
> > > The fix
> > > =======
> > > 
> > > Add the proper decorations to tpm_tis_gen_interrupt() and
> > > tpm_get_timeouts().
> > 
> > I tried that patch (alone on the top of the HEAD of Linus master) and I
> > still get the same trace in dmesg
> 
> Can you give a shot to
> 
> git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
> 
> It has couple of more fixes, and paste the log.

And if possible a full stack trace :-)

/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Tue, Feb 16, 2021 at 06:26:54PM +0200, Jarkko Sakkinen wrote:
> On Tue, Feb 16, 2021 at 06:16:58PM +0200, Jarkko Sakkinen wrote:
> > On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
> > > Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
> > > > From: Lukasz Majczak <lma@semihalf.com>
> > > > 
> > > > This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> > > > (SLB 9670):
> > > > 
> > > > [    4.324298] TPM returned invalid status
> > > > [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> > > > 
> > > > Background
> > > > ==========
> > > > 
> > > > TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> > > > Interface Locality Usage per Register, Table 39 Register Behavior Based on
> > > > Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> > > > 0xFF in case of lack of locality. The described situation manifests itself
> > > > with the following warning trace:
> > > > 
> > > > The fix
> > > > =======
> > > > 
> > > > Add the proper decorations to tpm_tis_gen_interrupt() and
> > > > tpm_get_timeouts().
> > > 
> > > I tried that patch (alone on the top of the HEAD of Linus master) and I
> > > still get the same trace in dmesg
> > 
> > Can you give a shot to
> > 
> > git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
> > 
> > It has couple of more fixes, and paste the log.
> 
> And if possible a full stack trace :-)

And apply this patch on top. Cannot apply it there before it's reviewed.

/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Laurent Bigonville]

Le 16/02/21 à 17:34, Jarkko Sakkinen a écrit :
> On Tue, Feb 16, 2021 at 06:26:54PM +0200, Jarkko Sakkinen wrote:
>> On Tue, Feb 16, 2021 at 06:16:58PM +0200, Jarkko Sakkinen wrote:
>>> On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
>>>> Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
>>>>> From: Lukasz Majczak <lma@semihalf.com>
>>>>>
>>>>> This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
>>>>> (SLB 9670):
>>>>>
>>>>> [    4.324298] TPM returned invalid status
>>>>> [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
>>>>>
>>>>> Background
>>>>> ==========
>>>>>
>>>>> TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
>>>>> Interface Locality Usage per Register, Table 39 Register Behavior Based on
>>>>> Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
>>>>> 0xFF in case of lack of locality. The described situation manifests itself
>>>>> with the following warning trace:
>>>>>
>>>>> The fix
>>>>> =======
>>>>>
>>>>> Add the proper decorations to tpm_tis_gen_interrupt() and
>>>>> tpm_get_timeouts().
>>>> I tried that patch (alone on the top of the HEAD of Linus master) and I
>>>> still get the same trace in dmesg
>>> Can you give a shot to
>>>
>>> git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
>>>
>>> It has couple of more fixes, and paste the log.
>> And if possible a full stack trace :-)
> And apply this patch on top. Cannot apply it there before it's reviewed.
>
I get the following stacktrace with your branch and the patch here:

[   13.498925] ------------[ cut here ]------------
[   13.498930] TPM returned invalid status
[   13.498953] WARNING: CPU: 1 PID: 459 at 
drivers/char/tpm/tpm_tis_core.c:205 tpm_tis_status+0x86/0xa0 [tpm_tis_core]
[   13.498963] Modules linked in: libiscsi_tcp(E) tpm_tis(E+) 
tpm_tis_core(E) libiscsi(E) snd_timer(E) tpm(E) joydev(E) snd(E) 
scsi_transport_iscsi(E) pcc_cpufreq(E-) fjes(E-) soundcore(E) 
i7core_edac(E) i5500_temp(E) asus_atk0110(E) rng_core(E) 
acpi_cpufreq(E-) evdev(E) loop(E) firewire_sbp2(E) msr(E) parport_pc(E) 
ppdev(E) lp(E) parport(E) fuse(E) configfs(E) sunrpc(E) ip_tables(E) 
x_tables(E) autofs4(E) ext4(E) crc16(E) mbcache(E) jbd2(E) btrfs(E) 
blake2b_generic(E) zstd_compress(E) efivars(E) raid10(E) raid456(E) 
async_raid6_recov(E) async_memcpy(E) async_pq(E) async_xor(E) 
async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) crc32c_generic(E) raid1(E) 
raid0(E) multipath(E) linear(E) md_mod(E) dm_mod(E) sr_mod(E) sd_mod(E) 
cdrom(E) t10_pi(E) hid_generic(E) usbhid(E) hid(E) amdgpu(E) ahci(E) 
libahci(E) libata(E) gpu_sched(E) i2c_algo_bit(E) drm_ttm_helper(E) 
ttm(E) uhci_hcd(E) ehci_pci(E) firewire_ohci(E) crc32c_intel(E) 
ehci_hcd(E) drm_kms_helper(E) firewire_core(E) i2c_i801(E) psmouse(E)
[   13.499044]  scsi_mod(E) cec(E) i2c_smbus(E) mxm_wmi(E) lpc_ich(E) 
crc_itu_t(E) sky2(E) usbcore(E) drm(E) mfd_core(E) wmi(E) button(E)
[   13.499058] CPU: 1 PID: 459 Comm: systemd-udevd Tainted: G          I 
E     5.11.0+ #4
[   13.499062] Hardware name: System manufacturer System Product 
Name/P6T DELUXE V2, BIOS 0406    04/24/2009
[   13.499064] RIP: 0010:tpm_tis_status+0x86/0xa0 [tpm_tis_core]
[   13.499069] Code: 00 75 30 48 83 c4 18 c3 31 c0 80 3d 83 48 00 00 00 
75 e0 48 c7 c7 4c 53 49 c1 88 44 24 07 c6 05 6f 48 00 00 01 e8 b2 24 ba 
cd <0f> 0b 0f b6 44 24 07 eb c0 e8 fc 99 bd cd 66 66 2e 0f 1f 84 00 00
[   13.499072] RSP: 0018:ffffbaeb80fb3aa0 EFLAGS: 00010286
[   13.499075] RAX: 0000000000000000 RBX: ffff9704034e5000 RCX: 
ffff9707ada58bc8
[   13.499078] RDX: 00000000ffffffd8 RSI: 0000000000000027 RDI: 
ffff9707ada58bc0
[   13.499080] RBP: 00000000ffff5d40 R08: 0000000000000000 R09: 
ffffbaeb80fb38c0
[   13.499082] R10: ffffbaeb80fb38b8 R11: ffffffff8fac2588 R12: 
0000000000000016
[   13.499085] R13: ffff97040c38d000 R14: 0000000000001000 R15: 
ffffbaeb80fb3ada
[   13.499087] FS:  00007fe1bcfeb8c0(0000) GS:ffff9707ada40000(0000) 
knlGS:0000000000000000
[   13.499090] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   13.499092] CR2: 000055e5f7048590 CR3: 00000001084fe000 CR4: 
00000000000006e0
[   13.499095] Call Trace:
[   13.499099]  tpm_transmit+0x15f/0x3d0 [tpm]
[   13.499113]  tpm_transmit_cmd+0x25/0x90 [tpm]
[   13.499121]  tpm2_probe+0xe2/0x140 [tpm]
[   13.499130]  tpm_tis_core_init+0x1d5/0x2b0 [tpm_tis_core]
[   13.499135]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
[   13.499141]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
[   13.499148]  pnp_device_probe+0xaf/0x140
[   13.499154]  really_probe+0xf2/0x440
[   13.499160]  driver_probe_device+0xe1/0x150
[   13.499163]  device_driver_attach+0xa1/0xb0
[   13.499167]  __driver_attach+0x8a/0x150
[   13.499171]  ? device_driver_attach+0xb0/0xb0
[   13.499174]  ? device_driver_attach+0xb0/0xb0
[   13.499177]  bus_for_each_dev+0x78/0xc0
[   13.499181]  bus_add_driver+0x12b/0x1e0
[   13.499184]  driver_register+0x8b/0xe0
[   13.499188]  ? 0xffffffffc1413000
[   13.499191]  init_tis+0xa0/0x1000 [tpm_tis]
[   13.499197]  do_one_initcall+0x44/0x1d0
[   13.499202]  ? do_init_module+0x23/0x260
[   13.499206]  ? kmem_cache_alloc_trace+0xf5/0x200
[   13.499212]  do_init_module+0x5c/0x260
[   13.499215]  __do_sys_finit_module+0xb1/0x110
[   13.499225]  do_syscall_64+0x33/0x80
[   13.499229]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
[   13.499234] RIP: 0033:0x7fe1bd4a49b9
[   13.499237] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 
48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 
05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
[   13.499240] RSP: 002b:00007ffca5bfa108 EFLAGS: 00000246 ORIG_RAX: 
0000000000000139
[   13.499243] RAX: ffffffffffffffda RBX: 000055e5f7036920 RCX: 
00007fe1bd4a49b9
[   13.499245] RDX: 0000000000000000 RSI: 00007fe1bd62fe2d RDI: 
0000000000000012
[   13.499247] RBP: 0000000000020000 R08: 0000000000000000 R09: 
000055e5f6e7a0b0
[   13.499249] R10: 0000000000000012 R11: 0000000000000246 R12: 
00007fe1bd62fe2d
[   13.499251] R13: 0000000000000000 R14: 000055e5f7033420 R15: 
000055e5f7036920
[   13.499255] ---[ end trace 7e963e5c3647102e ]---

If I also apply the patch to change the WARN_ONCE in dev_warn_once, and 
grep on the string "tpm", I get the following:

Feb 16 19:51:22 fornost kernel: tpm tpm0: TPM returned invalid status: 0xa0
Feb 16 19:51:22 fornost kernel: tpm_tis 00:06: 1.2 TPM (device-id 
0x6871, rev-id 1)
Feb 16 19:51:24 fornost kernel: tpm tpm0: tpm_try_transmit: send(): 
error -62
Feb 16 19:51:24 fornost kernel: tpm tpm0: A TPM error (-62) occurred 
attempting to determine the timeouts
Feb 16 19:51:24 fornost kernel: tpm_tis: probe of 00:06 failed with 
error -62
Feb 16 19:51:24 fornost kernel: tpm_inf_pnp 00:06: Found TPM with ID IFX0102

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Lukasz Majczak]

wt., 16 lut 2021 o 20:06 Laurent Bigonville <bigon@debian.org> napisał(a):
>
> Le 16/02/21 à 17:34, Jarkko Sakkinen a écrit :
> > On Tue, Feb 16, 2021 at 06:26:54PM +0200, Jarkko Sakkinen wrote:
> >> On Tue, Feb 16, 2021 at 06:16:58PM +0200, Jarkko Sakkinen wrote:
> >>> On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
> >>>> Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
> >>>>> From: Lukasz Majczak <lma@semihalf.com>
> >>>>>
> >>>>> This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> >>>>> (SLB 9670):
> >>>>>
> >>>>> [    4.324298] TPM returned invalid status
> >>>>> [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> >>>>>
> >>>>> Background
> >>>>> ==========
> >>>>>
> >>>>> TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> >>>>> Interface Locality Usage per Register, Table 39 Register Behavior Based on
> >>>>> Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> >>>>> 0xFF in case of lack of locality. The described situation manifests itself
> >>>>> with the following warning trace:
> >>>>>
> >>>>> The fix
> >>>>> =======
> >>>>>
> >>>>> Add the proper decorations to tpm_tis_gen_interrupt() and
> >>>>> tpm_get_timeouts().
> >>>> I tried that patch (alone on the top of the HEAD of Linus master) and I
> >>>> still get the same trace in dmesg
> >>> Can you give a shot to
> >>>
> >>> git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
> >>>
> >>> It has couple of more fixes, and paste the log.
> >> And if possible a full stack trace :-)
> > And apply this patch on top. Cannot apply it there before it's reviewed.
> >
> I get the following stacktrace with your branch and the patch here:
>
> [   13.498925] ------------[ cut here ]------------
> [   13.498930] TPM returned invalid status
> [   13.498953] WARNING: CPU: 1 PID: 459 at
> drivers/char/tpm/tpm_tis_core.c:205 tpm_tis_status+0x86/0xa0 [tpm_tis_core]
> [   13.498963] Modules linked in: libiscsi_tcp(E) tpm_tis(E+)
> tpm_tis_core(E) libiscsi(E) snd_timer(E) tpm(E) joydev(E) snd(E)
> scsi_transport_iscsi(E) pcc_cpufreq(E-) fjes(E-) soundcore(E)
> i7core_edac(E) i5500_temp(E) asus_atk0110(E) rng_core(E)
> acpi_cpufreq(E-) evdev(E) loop(E) firewire_sbp2(E) msr(E) parport_pc(E)
> ppdev(E) lp(E) parport(E) fuse(E) configfs(E) sunrpc(E) ip_tables(E)
> x_tables(E) autofs4(E) ext4(E) crc16(E) mbcache(E) jbd2(E) btrfs(E)
> blake2b_generic(E) zstd_compress(E) efivars(E) raid10(E) raid456(E)
> async_raid6_recov(E) async_memcpy(E) async_pq(E) async_xor(E)
> async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) crc32c_generic(E) raid1(E)
> raid0(E) multipath(E) linear(E) md_mod(E) dm_mod(E) sr_mod(E) sd_mod(E)
> cdrom(E) t10_pi(E) hid_generic(E) usbhid(E) hid(E) amdgpu(E) ahci(E)
> libahci(E) libata(E) gpu_sched(E) i2c_algo_bit(E) drm_ttm_helper(E)
> ttm(E) uhci_hcd(E) ehci_pci(E) firewire_ohci(E) crc32c_intel(E)
> ehci_hcd(E) drm_kms_helper(E) firewire_core(E) i2c_i801(E) psmouse(E)
> [   13.499044]  scsi_mod(E) cec(E) i2c_smbus(E) mxm_wmi(E) lpc_ich(E)
> crc_itu_t(E) sky2(E) usbcore(E) drm(E) mfd_core(E) wmi(E) button(E)
> [   13.499058] CPU: 1 PID: 459 Comm: systemd-udevd Tainted: G          I
> E     5.11.0+ #4
> [   13.499062] Hardware name: System manufacturer System Product
> Name/P6T DELUXE V2, BIOS 0406    04/24/2009
> [   13.499064] RIP: 0010:tpm_tis_status+0x86/0xa0 [tpm_tis_core]
> [   13.499069] Code: 00 75 30 48 83 c4 18 c3 31 c0 80 3d 83 48 00 00 00
> 75 e0 48 c7 c7 4c 53 49 c1 88 44 24 07 c6 05 6f 48 00 00 01 e8 b2 24 ba
> cd <0f> 0b 0f b6 44 24 07 eb c0 e8 fc 99 bd cd 66 66 2e 0f 1f 84 00 00
> [   13.499072] RSP: 0018:ffffbaeb80fb3aa0 EFLAGS: 00010286
> [   13.499075] RAX: 0000000000000000 RBX: ffff9704034e5000 RCX:
> ffff9707ada58bc8
> [   13.499078] RDX: 00000000ffffffd8 RSI: 0000000000000027 RDI:
> ffff9707ada58bc0
> [   13.499080] RBP: 00000000ffff5d40 R08: 0000000000000000 R09:
> ffffbaeb80fb38c0
> [   13.499082] R10: ffffbaeb80fb38b8 R11: ffffffff8fac2588 R12:
> 0000000000000016
> [   13.499085] R13: ffff97040c38d000 R14: 0000000000001000 R15:
> ffffbaeb80fb3ada
> [   13.499087] FS:  00007fe1bcfeb8c0(0000) GS:ffff9707ada40000(0000)
> knlGS:0000000000000000
> [   13.499090] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   13.499092] CR2: 000055e5f7048590 CR3: 00000001084fe000 CR4:
> 00000000000006e0
> [   13.499095] Call Trace:
> [   13.499099]  tpm_transmit+0x15f/0x3d0 [tpm]
> [   13.499113]  tpm_transmit_cmd+0x25/0x90 [tpm]
> [   13.499121]  tpm2_probe+0xe2/0x140 [tpm]
> [   13.499130]  tpm_tis_core_init+0x1d5/0x2b0 [tpm_tis_core]
> [   13.499135]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
> [   13.499141]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
> [   13.499148]  pnp_device_probe+0xaf/0x140
> [   13.499154]  really_probe+0xf2/0x440
> [   13.499160]  driver_probe_device+0xe1/0x150
> [   13.499163]  device_driver_attach+0xa1/0xb0
> [   13.499167]  __driver_attach+0x8a/0x150
> [   13.499171]  ? device_driver_attach+0xb0/0xb0
> [   13.499174]  ? device_driver_attach+0xb0/0xb0
> [   13.499177]  bus_for_each_dev+0x78/0xc0
> [   13.499181]  bus_add_driver+0x12b/0x1e0
> [   13.499184]  driver_register+0x8b/0xe0
> [   13.499188]  ? 0xffffffffc1413000
> [   13.499191]  init_tis+0xa0/0x1000 [tpm_tis]
> [   13.499197]  do_one_initcall+0x44/0x1d0
> [   13.499202]  ? do_init_module+0x23/0x260
> [   13.499206]  ? kmem_cache_alloc_trace+0xf5/0x200
> [   13.499212]  do_init_module+0x5c/0x260
> [   13.499215]  __do_sys_finit_module+0xb1/0x110
> [   13.499225]  do_syscall_64+0x33/0x80
> [   13.499229]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
> [   13.499234] RIP: 0033:0x7fe1bd4a49b9
> [   13.499237] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00
> 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f
> 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
> [   13.499240] RSP: 002b:00007ffca5bfa108 EFLAGS: 00000246 ORIG_RAX:
> 0000000000000139
> [   13.499243] RAX: ffffffffffffffda RBX: 000055e5f7036920 RCX:
> 00007fe1bd4a49b9
> [   13.499245] RDX: 0000000000000000 RSI: 00007fe1bd62fe2d RDI:
> 0000000000000012
> [   13.499247] RBP: 0000000000020000 R08: 0000000000000000 R09:
> 000055e5f6e7a0b0
> [   13.499249] R10: 0000000000000012 R11: 0000000000000246 R12:
> 00007fe1bd62fe2d
> [   13.499251] R13: 0000000000000000 R14: 000055e5f7033420 R15:
> 000055e5f7036920
> [   13.499255] ---[ end trace 7e963e5c3647102e ]---
>
> If I also apply the patch to change the WARN_ONCE in dev_warn_once, and
> grep on the string "tpm", I get the following:
>
> Feb 16 19:51:22 fornost kernel: tpm tpm0: TPM returned invalid status: 0xa0
> Feb 16 19:51:22 fornost kernel: tpm_tis 00:06: 1.2 TPM (device-id
> 0x6871, rev-id 1)
> Feb 16 19:51:24 fornost kernel: tpm tpm0: tpm_try_transmit: send():
> error -62
> Feb 16 19:51:24 fornost kernel: tpm tpm0: A TPM error (-62) occurred
> attempting to determine the timeouts
> Feb 16 19:51:24 fornost kernel: tpm_tis: probe of 00:06 failed with
> error -62
> Feb 16 19:51:24 fornost kernel: tpm_inf_pnp 00:06: Found TPM with ID IFX0102
>

Hi,
According to TCG PC Client Platform TPM Profile Specification for TPM 2.0
(https://trustedcomputinggroup.org/wp-content/uploads/PC-Client-Specific-Platform-TPM-Profile-for-TPM-2p0-v1p05p_r14_pub.pdf)
page 82-82, TPS_STS register bits description, it states, that in your
case (0xa0) bits 7 (stsValid, read-only) and 5 (tpmGo, write-only) are
set...
Both bits, in theirs descriptions have "Valid indicator: N/A".
The "if" statement (that guards this warning) masks TPM_STS with
TPM_STS_READ_ZERO (0x23), so it checks bit 0,1 and 5 (tpmGo) which,
in this case, is set causing the warning message.
I think the mask (TPM_STS_READ_ZERO) should have value 0x3 (skipping
bit 5) as tpmGo is write-only.
What do you think ?

Best regards,
Lukasz

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Tue, Feb 16, 2021 at 08:06:04PM +0100, Laurent Bigonville wrote:
> Le 16/02/21 à 17:34, Jarkko Sakkinen a écrit :
> > On Tue, Feb 16, 2021 at 06:26:54PM +0200, Jarkko Sakkinen wrote:
> > > On Tue, Feb 16, 2021 at 06:16:58PM +0200, Jarkko Sakkinen wrote:
> > > > On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
> > > > > Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
> > > > > > From: Lukasz Majczak <lma@semihalf.com>
> > > > > > 
> > > > > > This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> > > > > > (SLB 9670):
> > > > > > 
> > > > > > [    4.324298] TPM returned invalid status
> > > > > > [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> > > > > > 
> > > > > > Background
> > > > > > ==========
> > > > > > 
> > > > > > TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> > > > > > Interface Locality Usage per Register, Table 39 Register Behavior Based on
> > > > > > Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> > > > > > 0xFF in case of lack of locality. The described situation manifests itself
> > > > > > with the following warning trace:
> > > > > > 
> > > > > > The fix
> > > > > > =======
> > > > > > 
> > > > > > Add the proper decorations to tpm_tis_gen_interrupt() and
> > > > > > tpm_get_timeouts().
> > > > > I tried that patch (alone on the top of the HEAD of Linus master) and I
> > > > > still get the same trace in dmesg
> > > > Can you give a shot to
> > > > 
> > > > git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
> > > > 
> > > > It has couple of more fixes, and paste the log.
> > > And if possible a full stack trace :-)
> > And apply this patch on top. Cannot apply it there before it's reviewed.
> > 
> I get the following stacktrace with your branch and the patch here:
> 
> [   13.498925] ------------[ cut here ]------------
> [   13.498930] TPM returned invalid status
> [   13.498953] WARNING: CPU: 1 PID: 459 at
> drivers/char/tpm/tpm_tis_core.c:205 tpm_tis_status+0x86/0xa0 [tpm_tis_core]
> [   13.498963] Modules linked in: libiscsi_tcp(E) tpm_tis(E+)
> tpm_tis_core(E) libiscsi(E) snd_timer(E) tpm(E) joydev(E) snd(E)
> scsi_transport_iscsi(E) pcc_cpufreq(E-) fjes(E-) soundcore(E) i7core_edac(E)
> i5500_temp(E) asus_atk0110(E) rng_core(E) acpi_cpufreq(E-) evdev(E) loop(E)
> firewire_sbp2(E) msr(E) parport_pc(E) ppdev(E) lp(E) parport(E) fuse(E)
> configfs(E) sunrpc(E) ip_tables(E) x_tables(E) autofs4(E) ext4(E) crc16(E)
> mbcache(E) jbd2(E) btrfs(E) blake2b_generic(E) zstd_compress(E) efivars(E)
> raid10(E) raid456(E) async_raid6_recov(E) async_memcpy(E) async_pq(E)
> async_xor(E) async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) crc32c_generic(E)
> raid1(E) raid0(E) multipath(E) linear(E) md_mod(E) dm_mod(E) sr_mod(E)
> sd_mod(E) cdrom(E) t10_pi(E) hid_generic(E) usbhid(E) hid(E) amdgpu(E)
> ahci(E) libahci(E) libata(E) gpu_sched(E) i2c_algo_bit(E) drm_ttm_helper(E)
> ttm(E) uhci_hcd(E) ehci_pci(E) firewire_ohci(E) crc32c_intel(E) ehci_hcd(E)
> drm_kms_helper(E) firewire_core(E) i2c_i801(E) psmouse(E)
> [   13.499044]  scsi_mod(E) cec(E) i2c_smbus(E) mxm_wmi(E) lpc_ich(E)
> crc_itu_t(E) sky2(E) usbcore(E) drm(E) mfd_core(E) wmi(E) button(E)
> [   13.499058] CPU: 1 PID: 459 Comm: systemd-udevd Tainted: G          I
> E     5.11.0+ #4
> [   13.499062] Hardware name: System manufacturer System Product Name/P6T
> DELUXE V2, BIOS 0406    04/24/2009
> [   13.499064] RIP: 0010:tpm_tis_status+0x86/0xa0 [tpm_tis_core]
> [   13.499069] Code: 00 75 30 48 83 c4 18 c3 31 c0 80 3d 83 48 00 00 00 75
> e0 48 c7 c7 4c 53 49 c1 88 44 24 07 c6 05 6f 48 00 00 01 e8 b2 24 ba cd <0f>
> 0b 0f b6 44 24 07 eb c0 e8 fc 99 bd cd 66 66 2e 0f 1f 84 00 00
> [   13.499072] RSP: 0018:ffffbaeb80fb3aa0 EFLAGS: 00010286
> [   13.499075] RAX: 0000000000000000 RBX: ffff9704034e5000 RCX:
> ffff9707ada58bc8
> [   13.499078] RDX: 00000000ffffffd8 RSI: 0000000000000027 RDI:
> ffff9707ada58bc0
> [   13.499080] RBP: 00000000ffff5d40 R08: 0000000000000000 R09:
> ffffbaeb80fb38c0
> [   13.499082] R10: ffffbaeb80fb38b8 R11: ffffffff8fac2588 R12:
> 0000000000000016
> [   13.499085] R13: ffff97040c38d000 R14: 0000000000001000 R15:
> ffffbaeb80fb3ada
> [   13.499087] FS:  00007fe1bcfeb8c0(0000) GS:ffff9707ada40000(0000)
> knlGS:0000000000000000
> [   13.499090] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   13.499092] CR2: 000055e5f7048590 CR3: 00000001084fe000 CR4:
> 00000000000006e0
> [   13.499095] Call Trace:
> [   13.499099]  tpm_transmit+0x15f/0x3d0 [tpm]
> [   13.499113]  tpm_transmit_cmd+0x25/0x90 [tpm]
> [   13.499121]  tpm2_probe+0xe2/0x140 [tpm]
> [   13.499130]  tpm_tis_core_init+0x1d5/0x2b0 [tpm_tis_core]
> [   13.499135]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
> [   13.499141]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
> [   13.499148]  pnp_device_probe+0xaf/0x140
> [   13.499154]  really_probe+0xf2/0x440
> [   13.499160]  driver_probe_device+0xe1/0x150
> [   13.499163]  device_driver_attach+0xa1/0xb0
> [   13.499167]  __driver_attach+0x8a/0x150
> [   13.499171]  ? device_driver_attach+0xb0/0xb0
> [   13.499174]  ? device_driver_attach+0xb0/0xb0
> [   13.499177]  bus_for_each_dev+0x78/0xc0
> [   13.499181]  bus_add_driver+0x12b/0x1e0
> [   13.499184]  driver_register+0x8b/0xe0
> [   13.499188]  ? 0xffffffffc1413000
> [   13.499191]  init_tis+0xa0/0x1000 [tpm_tis]
> [   13.499197]  do_one_initcall+0x44/0x1d0
> [   13.499202]  ? do_init_module+0x23/0x260
> [   13.499206]  ? kmem_cache_alloc_trace+0xf5/0x200
> [   13.499212]  do_init_module+0x5c/0x260
> [   13.499215]  __do_sys_finit_module+0xb1/0x110
> [   13.499225]  do_syscall_64+0x33/0x80
> [   13.499229]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
> [   13.499234] RIP: 0033:0x7fe1bd4a49b9
> [   13.499237] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48
> 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48>
> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
> [   13.499240] RSP: 002b:00007ffca5bfa108 EFLAGS: 00000246 ORIG_RAX:
> 0000000000000139
> [   13.499243] RAX: ffffffffffffffda RBX: 000055e5f7036920 RCX:
> 00007fe1bd4a49b9
> [   13.499245] RDX: 0000000000000000 RSI: 00007fe1bd62fe2d RDI:
> 0000000000000012
> [   13.499247] RBP: 0000000000020000 R08: 0000000000000000 R09:
> 000055e5f6e7a0b0
> [   13.499249] R10: 0000000000000012 R11: 0000000000000246 R12:
> 00007fe1bd62fe2d
> [   13.499251] R13: 0000000000000000 R14: 000055e5f7033420 R15:
> 000055e5f7036920
> [   13.499255] ---[ end trace 7e963e5c3647102e ]---
> 
> If I also apply the patch to change the WARN_ONCE in dev_warn_once, and grep
> on the string "tpm", I get the following:
> 
> Feb 16 19:51:22 fornost kernel: tpm tpm0: TPM returned invalid status: 0xa0
> Feb 16 19:51:22 fornost kernel: tpm_tis 00:06: 1.2 TPM (device-id 0x6871,
> rev-id 1)
> Feb 16 19:51:24 fornost kernel: tpm tpm0: tpm_try_transmit: send(): error
> -62
> Feb 16 19:51:24 fornost kernel: tpm tpm0: A TPM error (-62) occurred
> attempting to determine the timeouts
> Feb 16 19:51:24 fornost kernel: tpm_tis: probe of 00:06 failed with error
> -62
> Feb 16 19:51:24 fornost kernel: tpm_inf_pnp 00:06: Found TPM with ID IFX0102
> 
> 

No immediate idea why this happens as the code should take locality.

I noticed that the original patch bundles incorrectly two fixes, so I 
send mu PT without the current fix and turn into a patch set. Let's
try to debug this for rc2/rc3.


/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Laurent Bigonville]

Le 17/02/21 à 23:29, Jarkko Sakkinen a écrit :
> On Tue, Feb 16, 2021 at 08:06:04PM +0100, Laurent Bigonville wrote:
>> Le 16/02/21 à 17:34, Jarkko Sakkinen a écrit :
>>> On Tue, Feb 16, 2021 at 06:26:54PM +0200, Jarkko Sakkinen wrote:
>>>> On Tue, Feb 16, 2021 at 06:16:58PM +0200, Jarkko Sakkinen wrote:
>>>>> On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
>>>>>> Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
>>>>>>> From: Lukasz Majczak <lma@semihalf.com>
>>>>>>>
>>>>>>> This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
>>>>>>> (SLB 9670):
>>>>>>>
>>>>>>> [    4.324298] TPM returned invalid status
>>>>>>> [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
>>>>>>>
>>>>>>> Background
>>>>>>> ==========
>>>>>>>
>>>>>>> TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
>>>>>>> Interface Locality Usage per Register, Table 39 Register Behavior Based on
>>>>>>> Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
>>>>>>> 0xFF in case of lack of locality. The described situation manifests itself
>>>>>>> with the following warning trace:
>>>>>>>
>>>>>>> The fix
>>>>>>> =======
>>>>>>>
>>>>>>> Add the proper decorations to tpm_tis_gen_interrupt() and
>>>>>>> tpm_get_timeouts().
>>>>>> I tried that patch (alone on the top of the HEAD of Linus master) and I
>>>>>> still get the same trace in dmesg
>>>>> Can you give a shot to
>>>>>
>>>>> git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
>>>>>
>>>>> It has couple of more fixes, and paste the log.
>>>> And if possible a full stack trace :-)
>>> And apply this patch on top. Cannot apply it there before it's reviewed.
>>>
>> I get the following stacktrace with your branch and the patch here:
>>
>> [   13.498925] ------------[ cut here ]------------
>> [   13.498930] TPM returned invalid status
>> [   13.498953] WARNING: CPU: 1 PID: 459 at
>> drivers/char/tpm/tpm_tis_core.c:205 tpm_tis_status+0x86/0xa0 [tpm_tis_core]
>> [   13.498963] Modules linked in: libiscsi_tcp(E) tpm_tis(E+)
>> tpm_tis_core(E) libiscsi(E) snd_timer(E) tpm(E) joydev(E) snd(E)
>> scsi_transport_iscsi(E) pcc_cpufreq(E-) fjes(E-) soundcore(E) i7core_edac(E)
>> i5500_temp(E) asus_atk0110(E) rng_core(E) acpi_cpufreq(E-) evdev(E) loop(E)
>> firewire_sbp2(E) msr(E) parport_pc(E) ppdev(E) lp(E) parport(E) fuse(E)
>> configfs(E) sunrpc(E) ip_tables(E) x_tables(E) autofs4(E) ext4(E) crc16(E)
>> mbcache(E) jbd2(E) btrfs(E) blake2b_generic(E) zstd_compress(E) efivars(E)
>> raid10(E) raid456(E) async_raid6_recov(E) async_memcpy(E) async_pq(E)
>> async_xor(E) async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) crc32c_generic(E)
>> raid1(E) raid0(E) multipath(E) linear(E) md_mod(E) dm_mod(E) sr_mod(E)
>> sd_mod(E) cdrom(E) t10_pi(E) hid_generic(E) usbhid(E) hid(E) amdgpu(E)
>> ahci(E) libahci(E) libata(E) gpu_sched(E) i2c_algo_bit(E) drm_ttm_helper(E)
>> ttm(E) uhci_hcd(E) ehci_pci(E) firewire_ohci(E) crc32c_intel(E) ehci_hcd(E)
>> drm_kms_helper(E) firewire_core(E) i2c_i801(E) psmouse(E)
>> [   13.499044]  scsi_mod(E) cec(E) i2c_smbus(E) mxm_wmi(E) lpc_ich(E)
>> crc_itu_t(E) sky2(E) usbcore(E) drm(E) mfd_core(E) wmi(E) button(E)
>> [   13.499058] CPU: 1 PID: 459 Comm: systemd-udevd Tainted: G          I
>> E     5.11.0+ #4
>> [   13.499062] Hardware name: System manufacturer System Product Name/P6T
>> DELUXE V2, BIOS 0406    04/24/2009
>> [   13.499064] RIP: 0010:tpm_tis_status+0x86/0xa0 [tpm_tis_core]
>> [   13.499069] Code: 00 75 30 48 83 c4 18 c3 31 c0 80 3d 83 48 00 00 00 75
>> e0 48 c7 c7 4c 53 49 c1 88 44 24 07 c6 05 6f 48 00 00 01 e8 b2 24 ba cd <0f>
>> 0b 0f b6 44 24 07 eb c0 e8 fc 99 bd cd 66 66 2e 0f 1f 84 00 00
>> [   13.499072] RSP: 0018:ffffbaeb80fb3aa0 EFLAGS: 00010286
>> [   13.499075] RAX: 0000000000000000 RBX: ffff9704034e5000 RCX:
>> ffff9707ada58bc8
>> [   13.499078] RDX: 00000000ffffffd8 RSI: 0000000000000027 RDI:
>> ffff9707ada58bc0
>> [   13.499080] RBP: 00000000ffff5d40 R08: 0000000000000000 R09:
>> ffffbaeb80fb38c0
>> [   13.499082] R10: ffffbaeb80fb38b8 R11: ffffffff8fac2588 R12:
>> 0000000000000016
>> [   13.499085] R13: ffff97040c38d000 R14: 0000000000001000 R15:
>> ffffbaeb80fb3ada
>> [   13.499087] FS:  00007fe1bcfeb8c0(0000) GS:ffff9707ada40000(0000)
>> knlGS:0000000000000000
>> [   13.499090] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [   13.499092] CR2: 000055e5f7048590 CR3: 00000001084fe000 CR4:
>> 00000000000006e0
>> [   13.499095] Call Trace:
>> [   13.499099]  tpm_transmit+0x15f/0x3d0 [tpm]
>> [   13.499113]  tpm_transmit_cmd+0x25/0x90 [tpm]
>> [   13.499121]  tpm2_probe+0xe2/0x140 [tpm]
>> [   13.499130]  tpm_tis_core_init+0x1d5/0x2b0 [tpm_tis_core]
>> [   13.499135]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
>> [   13.499141]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
>> [   13.499148]  pnp_device_probe+0xaf/0x140
>> [   13.499154]  really_probe+0xf2/0x440
>> [   13.499160]  driver_probe_device+0xe1/0x150
>> [   13.499163]  device_driver_attach+0xa1/0xb0
>> [   13.499167]  __driver_attach+0x8a/0x150
>> [   13.499171]  ? device_driver_attach+0xb0/0xb0
>> [   13.499174]  ? device_driver_attach+0xb0/0xb0
>> [   13.499177]  bus_for_each_dev+0x78/0xc0
>> [   13.499181]  bus_add_driver+0x12b/0x1e0
>> [   13.499184]  driver_register+0x8b/0xe0
>> [   13.499188]  ? 0xffffffffc1413000
>> [   13.499191]  init_tis+0xa0/0x1000 [tpm_tis]
>> [   13.499197]  do_one_initcall+0x44/0x1d0
>> [   13.499202]  ? do_init_module+0x23/0x260
>> [   13.499206]  ? kmem_cache_alloc_trace+0xf5/0x200
>> [   13.499212]  do_init_module+0x5c/0x260
>> [   13.499215]  __do_sys_finit_module+0xb1/0x110
>> [   13.499225]  do_syscall_64+0x33/0x80
>> [   13.499229]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
>> [   13.499234] RIP: 0033:0x7fe1bd4a49b9
>> [   13.499237] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48
>> 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48>
>> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
>> [   13.499240] RSP: 002b:00007ffca5bfa108 EFLAGS: 00000246 ORIG_RAX:
>> 0000000000000139
>> [   13.499243] RAX: ffffffffffffffda RBX: 000055e5f7036920 RCX:
>> 00007fe1bd4a49b9
>> [   13.499245] RDX: 0000000000000000 RSI: 00007fe1bd62fe2d RDI:
>> 0000000000000012
>> [   13.499247] RBP: 0000000000020000 R08: 0000000000000000 R09:
>> 000055e5f6e7a0b0
>> [   13.499249] R10: 0000000000000012 R11: 0000000000000246 R12:
>> 00007fe1bd62fe2d
>> [   13.499251] R13: 0000000000000000 R14: 000055e5f7033420 R15:
>> 000055e5f7036920
>> [   13.499255] ---[ end trace 7e963e5c3647102e ]---
>>
>> If I also apply the patch to change the WARN_ONCE in dev_warn_once, and grep
>> on the string "tpm", I get the following:
>>
>> Feb 16 19:51:22 fornost kernel: tpm tpm0: TPM returned invalid status: 0xa0
>> Feb 16 19:51:22 fornost kernel: tpm_tis 00:06: 1.2 TPM (device-id 0x6871,
>> rev-id 1)
>> Feb 16 19:51:24 fornost kernel: tpm tpm0: tpm_try_transmit: send(): error
>> -62
>> Feb 16 19:51:24 fornost kernel: tpm tpm0: A TPM error (-62) occurred
>> attempting to determine the timeouts
>> Feb 16 19:51:24 fornost kernel: tpm_tis: probe of 00:06 failed with error
>> -62
>> Feb 16 19:51:24 fornost kernel: tpm_inf_pnp 00:06: Found TPM with ID IFX0102
>>
>>
> No immediate idea why this happens as the code should take locality.
>
> I noticed that the original patch bundles incorrectly two fixes, so I
> send mu PT without the current fix and turn into a patch set. Let's
> try to debug this for rc2/rc3.
Hey, I'm coming back regarding this issue, I'm running 5.10 (from debian 
unstable) and still getting this issue

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Tue, Jun 01, 2021 at 11:17:14PM +0200, Laurent Bigonville wrote:
> Le 17/02/21 à 23:29, Jarkko Sakkinen a écrit :
> > On Tue, Feb 16, 2021 at 08:06:04PM +0100, Laurent Bigonville wrote:
> > > Le 16/02/21 à 17:34, Jarkko Sakkinen a écrit :
> > > > On Tue, Feb 16, 2021 at 06:26:54PM +0200, Jarkko Sakkinen wrote:
> > > > > On Tue, Feb 16, 2021 at 06:16:58PM +0200, Jarkko Sakkinen wrote:
> > > > > > On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
> > > > > > > Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
> > > > > > > > From: Lukasz Majczak <lma@semihalf.com>
> > > > > > > > 
> > > > > > > > This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> > > > > > > > (SLB 9670):
> > > > > > > > 
> > > > > > > > [    4.324298] TPM returned invalid status
> > > > > > > > [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> > > > > > > > 
> > > > > > > > Background
> > > > > > > > ==========
> > > > > > > > 
> > > > > > > > TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> > > > > > > > Interface Locality Usage per Register, Table 39 Register Behavior Based on
> > > > > > > > Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> > > > > > > > 0xFF in case of lack of locality. The described situation manifests itself
> > > > > > > > with the following warning trace:
> > > > > > > > 
> > > > > > > > The fix
> > > > > > > > =======
> > > > > > > > 
> > > > > > > > Add the proper decorations to tpm_tis_gen_interrupt() and
> > > > > > > > tpm_get_timeouts().
> > > > > > > I tried that patch (alone on the top of the HEAD of Linus master) and I
> > > > > > > still get the same trace in dmesg
> > > > > > Can you give a shot to
> > > > > > 
> > > > > > git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
> > > > > > 
> > > > > > It has couple of more fixes, and paste the log.
> > > > > And if possible a full stack trace :-)
> > > > And apply this patch on top. Cannot apply it there before it's reviewed.
> > > > 
> > > I get the following stacktrace with your branch and the patch here:
> > > 
> > > [   13.498925] ------------[ cut here ]------------
> > > [   13.498930] TPM returned invalid status
> > > [   13.498953] WARNING: CPU: 1 PID: 459 at
> > > drivers/char/tpm/tpm_tis_core.c:205 tpm_tis_status+0x86/0xa0 [tpm_tis_core]
> > > [   13.498963] Modules linked in: libiscsi_tcp(E) tpm_tis(E+)
> > > tpm_tis_core(E) libiscsi(E) snd_timer(E) tpm(E) joydev(E) snd(E)
> > > scsi_transport_iscsi(E) pcc_cpufreq(E-) fjes(E-) soundcore(E) i7core_edac(E)
> > > i5500_temp(E) asus_atk0110(E) rng_core(E) acpi_cpufreq(E-) evdev(E) loop(E)
> > > firewire_sbp2(E) msr(E) parport_pc(E) ppdev(E) lp(E) parport(E) fuse(E)
> > > configfs(E) sunrpc(E) ip_tables(E) x_tables(E) autofs4(E) ext4(E) crc16(E)
> > > mbcache(E) jbd2(E) btrfs(E) blake2b_generic(E) zstd_compress(E) efivars(E)
> > > raid10(E) raid456(E) async_raid6_recov(E) async_memcpy(E) async_pq(E)
> > > async_xor(E) async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) crc32c_generic(E)
> > > raid1(E) raid0(E) multipath(E) linear(E) md_mod(E) dm_mod(E) sr_mod(E)
> > > sd_mod(E) cdrom(E) t10_pi(E) hid_generic(E) usbhid(E) hid(E) amdgpu(E)
> > > ahci(E) libahci(E) libata(E) gpu_sched(E) i2c_algo_bit(E) drm_ttm_helper(E)
> > > ttm(E) uhci_hcd(E) ehci_pci(E) firewire_ohci(E) crc32c_intel(E) ehci_hcd(E)
> > > drm_kms_helper(E) firewire_core(E) i2c_i801(E) psmouse(E)
> > > [   13.499044]  scsi_mod(E) cec(E) i2c_smbus(E) mxm_wmi(E) lpc_ich(E)
> > > crc_itu_t(E) sky2(E) usbcore(E) drm(E) mfd_core(E) wmi(E) button(E)
> > > [   13.499058] CPU: 1 PID: 459 Comm: systemd-udevd Tainted: G          I
> > > E     5.11.0+ #4
> > > [   13.499062] Hardware name: System manufacturer System Product Name/P6T
> > > DELUXE V2, BIOS 0406    04/24/2009
> > > [   13.499064] RIP: 0010:tpm_tis_status+0x86/0xa0 [tpm_tis_core]
> > > [   13.499069] Code: 00 75 30 48 83 c4 18 c3 31 c0 80 3d 83 48 00 00 00 75
> > > e0 48 c7 c7 4c 53 49 c1 88 44 24 07 c6 05 6f 48 00 00 01 e8 b2 24 ba cd <0f>
> > > 0b 0f b6 44 24 07 eb c0 e8 fc 99 bd cd 66 66 2e 0f 1f 84 00 00
> > > [   13.499072] RSP: 0018:ffffbaeb80fb3aa0 EFLAGS: 00010286
> > > [   13.499075] RAX: 0000000000000000 RBX: ffff9704034e5000 RCX:
> > > ffff9707ada58bc8
> > > [   13.499078] RDX: 00000000ffffffd8 RSI: 0000000000000027 RDI:
> > > ffff9707ada58bc0
> > > [   13.499080] RBP: 00000000ffff5d40 R08: 0000000000000000 R09:
> > > ffffbaeb80fb38c0
> > > [   13.499082] R10: ffffbaeb80fb38b8 R11: ffffffff8fac2588 R12:
> > > 0000000000000016
> > > [   13.499085] R13: ffff97040c38d000 R14: 0000000000001000 R15:
> > > ffffbaeb80fb3ada
> > > [   13.499087] FS:  00007fe1bcfeb8c0(0000) GS:ffff9707ada40000(0000)
> > > knlGS:0000000000000000
> > > [   13.499090] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > > [   13.499092] CR2: 000055e5f7048590 CR3: 00000001084fe000 CR4:
> > > 00000000000006e0
> > > [   13.499095] Call Trace:
> > > [   13.499099]  tpm_transmit+0x15f/0x3d0 [tpm]
> > > [   13.499113]  tpm_transmit_cmd+0x25/0x90 [tpm]
> > > [   13.499121]  tpm2_probe+0xe2/0x140 [tpm]
> > > [   13.499130]  tpm_tis_core_init+0x1d5/0x2b0 [tpm_tis_core]
> > > [   13.499135]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
> > > [   13.499141]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
> > > [   13.499148]  pnp_device_probe+0xaf/0x140
> > > [   13.499154]  really_probe+0xf2/0x440
> > > [   13.499160]  driver_probe_device+0xe1/0x150
> > > [   13.499163]  device_driver_attach+0xa1/0xb0
> > > [   13.499167]  __driver_attach+0x8a/0x150
> > > [   13.499171]  ? device_driver_attach+0xb0/0xb0
> > > [   13.499174]  ? device_driver_attach+0xb0/0xb0
> > > [   13.499177]  bus_for_each_dev+0x78/0xc0
> > > [   13.499181]  bus_add_driver+0x12b/0x1e0
> > > [   13.499184]  driver_register+0x8b/0xe0
> > > [   13.499188]  ? 0xffffffffc1413000
> > > [   13.499191]  init_tis+0xa0/0x1000 [tpm_tis]
> > > [   13.499197]  do_one_initcall+0x44/0x1d0
> > > [   13.499202]  ? do_init_module+0x23/0x260
> > > [   13.499206]  ? kmem_cache_alloc_trace+0xf5/0x200
> > > [   13.499212]  do_init_module+0x5c/0x260
> > > [   13.499215]  __do_sys_finit_module+0xb1/0x110
> > > [   13.499225]  do_syscall_64+0x33/0x80
> > > [   13.499229]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
> > > [   13.499234] RIP: 0033:0x7fe1bd4a49b9
> > > [   13.499237] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48
> > > 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48>
> > > 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
> > > [   13.499240] RSP: 002b:00007ffca5bfa108 EFLAGS: 00000246 ORIG_RAX:
> > > 0000000000000139
> > > [   13.499243] RAX: ffffffffffffffda RBX: 000055e5f7036920 RCX:
> > > 00007fe1bd4a49b9
> > > [   13.499245] RDX: 0000000000000000 RSI: 00007fe1bd62fe2d RDI:
> > > 0000000000000012
> > > [   13.499247] RBP: 0000000000020000 R08: 0000000000000000 R09:
> > > 000055e5f6e7a0b0
> > > [   13.499249] R10: 0000000000000012 R11: 0000000000000246 R12:
> > > 00007fe1bd62fe2d
> > > [   13.499251] R13: 0000000000000000 R14: 000055e5f7033420 R15:
> > > 000055e5f7036920
> > > [   13.499255] ---[ end trace 7e963e5c3647102e ]---
> > > 
> > > If I also apply the patch to change the WARN_ONCE in dev_warn_once, and grep
> > > on the string "tpm", I get the following:
> > > 
> > > Feb 16 19:51:22 fornost kernel: tpm tpm0: TPM returned invalid status: 0xa0
> > > Feb 16 19:51:22 fornost kernel: tpm_tis 00:06: 1.2 TPM (device-id 0x6871,
> > > rev-id 1)
> > > Feb 16 19:51:24 fornost kernel: tpm tpm0: tpm_try_transmit: send(): error
> > > -62
> > > Feb 16 19:51:24 fornost kernel: tpm tpm0: A TPM error (-62) occurred
> > > attempting to determine the timeouts
> > > Feb 16 19:51:24 fornost kernel: tpm_tis: probe of 00:06 failed with error
> > > -62
> > > Feb 16 19:51:24 fornost kernel: tpm_inf_pnp 00:06: Found TPM with ID IFX0102
> > > 
> > > 
> > No immediate idea why this happens as the code should take locality.
> > 
> > I noticed that the original patch bundles incorrectly two fixes, so I
> > send mu PT without the current fix and turn into a patch set. Let's
> > try to debug this for rc2/rc3.
> Hey, I'm coming back regarding this issue, I'm running 5.10 (from debian
> unstable) and still getting this issue

Many of the fixes are landing to 5.13, stable kernels do not acquire
out-of-tree fixes unless downstream maintainers decide to apply them
by hand.

/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Laurent Bigonville]

Le 3/06/21 à 07:28, Jarkko Sakkinen a écrit :
> On Tue, Jun 01, 2021 at 11:17:14PM +0200, Laurent Bigonville wrote:
>> Le 17/02/21 à 23:29, Jarkko Sakkinen a écrit :
>>> On Tue, Feb 16, 2021 at 08:06:04PM +0100, Laurent Bigonville wrote:
>>>> Le 16/02/21 à 17:34, Jarkko Sakkinen a écrit :
>>>>> On Tue, Feb 16, 2021 at 06:26:54PM +0200, Jarkko Sakkinen wrote:
>>>>>> On Tue, Feb 16, 2021 at 06:16:58PM +0200, Jarkko Sakkinen wrote:
>>>>>>> On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
>>>>>>>> Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
>>>>>>>>> From: Lukasz Majczak <lma@semihalf.com>
>>>>>>>>>
>>>>>>>>> This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
>>>>>>>>> (SLB 9670):
>>>>>>>>>
>>>>>>>>> [    4.324298] TPM returned invalid status
>>>>>>>>> [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
>>>>>>>>>
>>>>>>>>> Background
>>>>>>>>> ==========
>>>>>>>>>
>>>>>>>>> TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
>>>>>>>>> Interface Locality Usage per Register, Table 39 Register Behavior Based on
>>>>>>>>> Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
>>>>>>>>> 0xFF in case of lack of locality. The described situation manifests itself
>>>>>>>>> with the following warning trace:
>>>>>>>>>
>>>>>>>>> The fix
>>>>>>>>> =======
>>>>>>>>>
>>>>>>>>> Add the proper decorations to tpm_tis_gen_interrupt() and
>>>>>>>>> tpm_get_timeouts().
>>>>>>>> I tried that patch (alone on the top of the HEAD of Linus master) and I
>>>>>>>> still get the same trace in dmesg
>>>>>>> Can you give a shot to
>>>>>>>
>>>>>>> git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
>>>>>>>
>>>>>>> It has couple of more fixes, and paste the log.
>>>>>> And if possible a full stack trace :-)
>>>>> And apply this patch on top. Cannot apply it there before it's reviewed.
>>>>>
>>>> I get the following stacktrace with your branch and the patch here:
>>>>
>>>> [   13.498925] ------------[ cut here ]------------
>>>> [   13.498930] TPM returned invalid status
>>>> [   13.498953] WARNING: CPU: 1 PID: 459 at
>>>> drivers/char/tpm/tpm_tis_core.c:205 tpm_tis_status+0x86/0xa0 [tpm_tis_core]
>>>> [   13.498963] Modules linked in: libiscsi_tcp(E) tpm_tis(E+)
>>>> tpm_tis_core(E) libiscsi(E) snd_timer(E) tpm(E) joydev(E) snd(E)
>>>> scsi_transport_iscsi(E) pcc_cpufreq(E-) fjes(E-) soundcore(E) i7core_edac(E)
>>>> i5500_temp(E) asus_atk0110(E) rng_core(E) acpi_cpufreq(E-) evdev(E) loop(E)
>>>> firewire_sbp2(E) msr(E) parport_pc(E) ppdev(E) lp(E) parport(E) fuse(E)
>>>> configfs(E) sunrpc(E) ip_tables(E) x_tables(E) autofs4(E) ext4(E) crc16(E)
>>>> mbcache(E) jbd2(E) btrfs(E) blake2b_generic(E) zstd_compress(E) efivars(E)
>>>> raid10(E) raid456(E) async_raid6_recov(E) async_memcpy(E) async_pq(E)
>>>> async_xor(E) async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) crc32c_generic(E)
>>>> raid1(E) raid0(E) multipath(E) linear(E) md_mod(E) dm_mod(E) sr_mod(E)
>>>> sd_mod(E) cdrom(E) t10_pi(E) hid_generic(E) usbhid(E) hid(E) amdgpu(E)
>>>> ahci(E) libahci(E) libata(E) gpu_sched(E) i2c_algo_bit(E) drm_ttm_helper(E)
>>>> ttm(E) uhci_hcd(E) ehci_pci(E) firewire_ohci(E) crc32c_intel(E) ehci_hcd(E)
>>>> drm_kms_helper(E) firewire_core(E) i2c_i801(E) psmouse(E)
>>>> [   13.499044]  scsi_mod(E) cec(E) i2c_smbus(E) mxm_wmi(E) lpc_ich(E)
>>>> crc_itu_t(E) sky2(E) usbcore(E) drm(E) mfd_core(E) wmi(E) button(E)
>>>> [   13.499058] CPU: 1 PID: 459 Comm: systemd-udevd Tainted: G          I
>>>> E     5.11.0+ #4
>>>> [   13.499062] Hardware name: System manufacturer System Product Name/P6T
>>>> DELUXE V2, BIOS 0406    04/24/2009
>>>> [   13.499064] RIP: 0010:tpm_tis_status+0x86/0xa0 [tpm_tis_core]
>>>> [   13.499069] Code: 00 75 30 48 83 c4 18 c3 31 c0 80 3d 83 48 00 00 00 75
>>>> e0 48 c7 c7 4c 53 49 c1 88 44 24 07 c6 05 6f 48 00 00 01 e8 b2 24 ba cd <0f>
>>>> 0b 0f b6 44 24 07 eb c0 e8 fc 99 bd cd 66 66 2e 0f 1f 84 00 00
>>>> [   13.499072] RSP: 0018:ffffbaeb80fb3aa0 EFLAGS: 00010286
>>>> [   13.499075] RAX: 0000000000000000 RBX: ffff9704034e5000 RCX:
>>>> ffff9707ada58bc8
>>>> [   13.499078] RDX: 00000000ffffffd8 RSI: 0000000000000027 RDI:
>>>> ffff9707ada58bc0
>>>> [   13.499080] RBP: 00000000ffff5d40 R08: 0000000000000000 R09:
>>>> ffffbaeb80fb38c0
>>>> [   13.499082] R10: ffffbaeb80fb38b8 R11: ffffffff8fac2588 R12:
>>>> 0000000000000016
>>>> [   13.499085] R13: ffff97040c38d000 R14: 0000000000001000 R15:
>>>> ffffbaeb80fb3ada
>>>> [   13.499087] FS:  00007fe1bcfeb8c0(0000) GS:ffff9707ada40000(0000)
>>>> knlGS:0000000000000000
>>>> [   13.499090] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>>> [   13.499092] CR2: 000055e5f7048590 CR3: 00000001084fe000 CR4:
>>>> 00000000000006e0
>>>> [   13.499095] Call Trace:
>>>> [   13.499099]  tpm_transmit+0x15f/0x3d0 [tpm]
>>>> [   13.499113]  tpm_transmit_cmd+0x25/0x90 [tpm]
>>>> [   13.499121]  tpm2_probe+0xe2/0x140 [tpm]
>>>> [   13.499130]  tpm_tis_core_init+0x1d5/0x2b0 [tpm_tis_core]
>>>> [   13.499135]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
>>>> [   13.499141]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
>>>> [   13.499148]  pnp_device_probe+0xaf/0x140
>>>> [   13.499154]  really_probe+0xf2/0x440
>>>> [   13.499160]  driver_probe_device+0xe1/0x150
>>>> [   13.499163]  device_driver_attach+0xa1/0xb0
>>>> [   13.499167]  __driver_attach+0x8a/0x150
>>>> [   13.499171]  ? device_driver_attach+0xb0/0xb0
>>>> [   13.499174]  ? device_driver_attach+0xb0/0xb0
>>>> [   13.499177]  bus_for_each_dev+0x78/0xc0
>>>> [   13.499181]  bus_add_driver+0x12b/0x1e0
>>>> [   13.499184]  driver_register+0x8b/0xe0
>>>> [   13.499188]  ? 0xffffffffc1413000
>>>> [   13.499191]  init_tis+0xa0/0x1000 [tpm_tis]
>>>> [   13.499197]  do_one_initcall+0x44/0x1d0
>>>> [   13.499202]  ? do_init_module+0x23/0x260
>>>> [   13.499206]  ? kmem_cache_alloc_trace+0xf5/0x200
>>>> [   13.499212]  do_init_module+0x5c/0x260
>>>> [   13.499215]  __do_sys_finit_module+0xb1/0x110
>>>> [   13.499225]  do_syscall_64+0x33/0x80
>>>> [   13.499229]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
>>>> [   13.499234] RIP: 0033:0x7fe1bd4a49b9
>>>> [   13.499237] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48
>>>> 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48>
>>>> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
>>>> [   13.499240] RSP: 002b:00007ffca5bfa108 EFLAGS: 00000246 ORIG_RAX:
>>>> 0000000000000139
>>>> [   13.499243] RAX: ffffffffffffffda RBX: 000055e5f7036920 RCX:
>>>> 00007fe1bd4a49b9
>>>> [   13.499245] RDX: 0000000000000000 RSI: 00007fe1bd62fe2d RDI:
>>>> 0000000000000012
>>>> [   13.499247] RBP: 0000000000020000 R08: 0000000000000000 R09:
>>>> 000055e5f6e7a0b0
>>>> [   13.499249] R10: 0000000000000012 R11: 0000000000000246 R12:
>>>> 00007fe1bd62fe2d
>>>> [   13.499251] R13: 0000000000000000 R14: 000055e5f7033420 R15:
>>>> 000055e5f7036920
>>>> [   13.499255] ---[ end trace 7e963e5c3647102e ]---
>>>>
>>>> If I also apply the patch to change the WARN_ONCE in dev_warn_once, and grep
>>>> on the string "tpm", I get the following:
>>>>
>>>> Feb 16 19:51:22 fornost kernel: tpm tpm0: TPM returned invalid status: 0xa0
>>>> Feb 16 19:51:22 fornost kernel: tpm_tis 00:06: 1.2 TPM (device-id 0x6871,
>>>> rev-id 1)
>>>> Feb 16 19:51:24 fornost kernel: tpm tpm0: tpm_try_transmit: send(): error
>>>> -62
>>>> Feb 16 19:51:24 fornost kernel: tpm tpm0: A TPM error (-62) occurred
>>>> attempting to determine the timeouts
>>>> Feb 16 19:51:24 fornost kernel: tpm_tis: probe of 00:06 failed with error
>>>> -62
>>>> Feb 16 19:51:24 fornost kernel: tpm_inf_pnp 00:06: Found TPM with ID IFX0102
>>>>
>>>>
>>> No immediate idea why this happens as the code should take locality.
>>>
>>> I noticed that the original patch bundles incorrectly two fixes, so I
>>> send mu PT without the current fix and turn into a patch set. Let's
>>> try to debug this for rc2/rc3.
>> Hey, I'm coming back regarding this issue, I'm running 5.10 (from debian
>> unstable) and still getting this issue
> Many of the fixes are landing to 5.13, stable kernels do not acquire
> out-of-tree fixes unless downstream maintainers decide to apply them
> by hand.
OK I tried again with linus master branch (5.13.0-rc4+) and your 
linux-tpmdd branch and the problem persists

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Thu, Jun 03, 2021 at 11:42:15AM +0200, Laurent Bigonville wrote:
> 
> Le 3/06/21 à 07:28, Jarkko Sakkinen a écrit :
> > On Tue, Jun 01, 2021 at 11:17:14PM +0200, Laurent Bigonville wrote:
> > > Le 17/02/21 à 23:29, Jarkko Sakkinen a écrit :
> > > > On Tue, Feb 16, 2021 at 08:06:04PM +0100, Laurent Bigonville wrote:
> > > > > Le 16/02/21 à 17:34, Jarkko Sakkinen a écrit :
> > > > > > On Tue, Feb 16, 2021 at 06:26:54PM +0200, Jarkko Sakkinen wrote:
> > > > > > > On Tue, Feb 16, 2021 at 06:16:58PM +0200, Jarkko Sakkinen wrote:
> > > > > > > > On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
> > > > > > > > > Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
> > > > > > > > > > From: Lukasz Majczak <lma@semihalf.com>
> > > > > > > > > > 
> > > > > > > > > > This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
> > > > > > > > > > (SLB 9670):
> > > > > > > > > > 
> > > > > > > > > > [    4.324298] TPM returned invalid status
> > > > > > > > > > [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
> > > > > > > > > > 
> > > > > > > > > > Background
> > > > > > > > > > ==========
> > > > > > > > > > 
> > > > > > > > > > TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
> > > > > > > > > > Interface Locality Usage per Register, Table 39 Register Behavior Based on
> > > > > > > > > > Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
> > > > > > > > > > 0xFF in case of lack of locality. The described situation manifests itself
> > > > > > > > > > with the following warning trace:
> > > > > > > > > > 
> > > > > > > > > > The fix
> > > > > > > > > > =======
> > > > > > > > > > 
> > > > > > > > > > Add the proper decorations to tpm_tis_gen_interrupt() and
> > > > > > > > > > tpm_get_timeouts().
> > > > > > > > > I tried that patch (alone on the top of the HEAD of Linus master) and I
> > > > > > > > > still get the same trace in dmesg
> > > > > > > > Can you give a shot to
> > > > > > > > 
> > > > > > > > git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
> > > > > > > > 
> > > > > > > > It has couple of more fixes, and paste the log.
> > > > > > > And if possible a full stack trace :-)
> > > > > > And apply this patch on top. Cannot apply it there before it's reviewed.
> > > > > > 
> > > > > I get the following stacktrace with your branch and the patch here:
> > > > > 
> > > > > [   13.498925] ------------[ cut here ]------------
> > > > > [   13.498930] TPM returned invalid status
> > > > > [   13.498953] WARNING: CPU: 1 PID: 459 at
> > > > > drivers/char/tpm/tpm_tis_core.c:205 tpm_tis_status+0x86/0xa0 [tpm_tis_core]
> > > > > [   13.498963] Modules linked in: libiscsi_tcp(E) tpm_tis(E+)
> > > > > tpm_tis_core(E) libiscsi(E) snd_timer(E) tpm(E) joydev(E) snd(E)
> > > > > scsi_transport_iscsi(E) pcc_cpufreq(E-) fjes(E-) soundcore(E) i7core_edac(E)
> > > > > i5500_temp(E) asus_atk0110(E) rng_core(E) acpi_cpufreq(E-) evdev(E) loop(E)
> > > > > firewire_sbp2(E) msr(E) parport_pc(E) ppdev(E) lp(E) parport(E) fuse(E)
> > > > > configfs(E) sunrpc(E) ip_tables(E) x_tables(E) autofs4(E) ext4(E) crc16(E)
> > > > > mbcache(E) jbd2(E) btrfs(E) blake2b_generic(E) zstd_compress(E) efivars(E)
> > > > > raid10(E) raid456(E) async_raid6_recov(E) async_memcpy(E) async_pq(E)
> > > > > async_xor(E) async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) crc32c_generic(E)
> > > > > raid1(E) raid0(E) multipath(E) linear(E) md_mod(E) dm_mod(E) sr_mod(E)
> > > > > sd_mod(E) cdrom(E) t10_pi(E) hid_generic(E) usbhid(E) hid(E) amdgpu(E)
> > > > > ahci(E) libahci(E) libata(E) gpu_sched(E) i2c_algo_bit(E) drm_ttm_helper(E)
> > > > > ttm(E) uhci_hcd(E) ehci_pci(E) firewire_ohci(E) crc32c_intel(E) ehci_hcd(E)
> > > > > drm_kms_helper(E) firewire_core(E) i2c_i801(E) psmouse(E)
> > > > > [   13.499044]  scsi_mod(E) cec(E) i2c_smbus(E) mxm_wmi(E) lpc_ich(E)
> > > > > crc_itu_t(E) sky2(E) usbcore(E) drm(E) mfd_core(E) wmi(E) button(E)
> > > > > [   13.499058] CPU: 1 PID: 459 Comm: systemd-udevd Tainted: G          I
> > > > > E     5.11.0+ #4
> > > > > [   13.499062] Hardware name: System manufacturer System Product Name/P6T
> > > > > DELUXE V2, BIOS 0406    04/24/2009
> > > > > [   13.499064] RIP: 0010:tpm_tis_status+0x86/0xa0 [tpm_tis_core]
> > > > > [   13.499069] Code: 00 75 30 48 83 c4 18 c3 31 c0 80 3d 83 48 00 00 00 75
> > > > > e0 48 c7 c7 4c 53 49 c1 88 44 24 07 c6 05 6f 48 00 00 01 e8 b2 24 ba cd <0f>
> > > > > 0b 0f b6 44 24 07 eb c0 e8 fc 99 bd cd 66 66 2e 0f 1f 84 00 00
> > > > > [   13.499072] RSP: 0018:ffffbaeb80fb3aa0 EFLAGS: 00010286
> > > > > [   13.499075] RAX: 0000000000000000 RBX: ffff9704034e5000 RCX:
> > > > > ffff9707ada58bc8
> > > > > [   13.499078] RDX: 00000000ffffffd8 RSI: 0000000000000027 RDI:
> > > > > ffff9707ada58bc0
> > > > > [   13.499080] RBP: 00000000ffff5d40 R08: 0000000000000000 R09:
> > > > > ffffbaeb80fb38c0
> > > > > [   13.499082] R10: ffffbaeb80fb38b8 R11: ffffffff8fac2588 R12:
> > > > > 0000000000000016
> > > > > [   13.499085] R13: ffff97040c38d000 R14: 0000000000001000 R15:
> > > > > ffffbaeb80fb3ada
> > > > > [   13.499087] FS:  00007fe1bcfeb8c0(0000) GS:ffff9707ada40000(0000)
> > > > > knlGS:0000000000000000
> > > > > [   13.499090] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > > > > [   13.499092] CR2: 000055e5f7048590 CR3: 00000001084fe000 CR4:
> > > > > 00000000000006e0
> > > > > [   13.499095] Call Trace:
> > > > > [   13.499099]  tpm_transmit+0x15f/0x3d0 [tpm]
> > > > > [   13.499113]  tpm_transmit_cmd+0x25/0x90 [tpm]
> > > > > [   13.499121]  tpm2_probe+0xe2/0x140 [tpm]
> > > > > [   13.499130]  tpm_tis_core_init+0x1d5/0x2b0 [tpm_tis_core]
> > > > > [   13.499135]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
> > > > > [   13.499141]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
> > > > > [   13.499148]  pnp_device_probe+0xaf/0x140
> > > > > [   13.499154]  really_probe+0xf2/0x440
> > > > > [   13.499160]  driver_probe_device+0xe1/0x150
> > > > > [   13.499163]  device_driver_attach+0xa1/0xb0
> > > > > [   13.499167]  __driver_attach+0x8a/0x150
> > > > > [   13.499171]  ? device_driver_attach+0xb0/0xb0
> > > > > [   13.499174]  ? device_driver_attach+0xb0/0xb0
> > > > > [   13.499177]  bus_for_each_dev+0x78/0xc0
> > > > > [   13.499181]  bus_add_driver+0x12b/0x1e0
> > > > > [   13.499184]  driver_register+0x8b/0xe0
> > > > > [   13.499188]  ? 0xffffffffc1413000
> > > > > [   13.499191]  init_tis+0xa0/0x1000 [tpm_tis]
> > > > > [   13.499197]  do_one_initcall+0x44/0x1d0
> > > > > [   13.499202]  ? do_init_module+0x23/0x260
> > > > > [   13.499206]  ? kmem_cache_alloc_trace+0xf5/0x200
> > > > > [   13.499212]  do_init_module+0x5c/0x260
> > > > > [   13.499215]  __do_sys_finit_module+0xb1/0x110
> > > > > [   13.499225]  do_syscall_64+0x33/0x80
> > > > > [   13.499229]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
> > > > > [   13.499234] RIP: 0033:0x7fe1bd4a49b9
> > > > > [   13.499237] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48
> > > > > 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48>
> > > > > 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
> > > > > [   13.499240] RSP: 002b:00007ffca5bfa108 EFLAGS: 00000246 ORIG_RAX:
> > > > > 0000000000000139
> > > > > [   13.499243] RAX: ffffffffffffffda RBX: 000055e5f7036920 RCX:
> > > > > 00007fe1bd4a49b9
> > > > > [   13.499245] RDX: 0000000000000000 RSI: 00007fe1bd62fe2d RDI:
> > > > > 0000000000000012
> > > > > [   13.499247] RBP: 0000000000020000 R08: 0000000000000000 R09:
> > > > > 000055e5f6e7a0b0
> > > > > [   13.499249] R10: 0000000000000012 R11: 0000000000000246 R12:
> > > > > 00007fe1bd62fe2d
> > > > > [   13.499251] R13: 0000000000000000 R14: 000055e5f7033420 R15:
> > > > > 000055e5f7036920
> > > > > [   13.499255] ---[ end trace 7e963e5c3647102e ]---
> > > > > 
> > > > > If I also apply the patch to change the WARN_ONCE in dev_warn_once, and grep
> > > > > on the string "tpm", I get the following:
> > > > > 
> > > > > Feb 16 19:51:22 fornost kernel: tpm tpm0: TPM returned invalid status: 0xa0
> > > > > Feb 16 19:51:22 fornost kernel: tpm_tis 00:06: 1.2 TPM (device-id 0x6871,
> > > > > rev-id 1)
> > > > > Feb 16 19:51:24 fornost kernel: tpm tpm0: tpm_try_transmit: send(): error
> > > > > -62
> > > > > Feb 16 19:51:24 fornost kernel: tpm tpm0: A TPM error (-62) occurred
> > > > > attempting to determine the timeouts
> > > > > Feb 16 19:51:24 fornost kernel: tpm_tis: probe of 00:06 failed with error
> > > > > -62
> > > > > Feb 16 19:51:24 fornost kernel: tpm_inf_pnp 00:06: Found TPM with ID IFX0102
> > > > > 
> > > > > 
> > > > No immediate idea why this happens as the code should take locality.
> > > > 
> > > > I noticed that the original patch bundles incorrectly two fixes, so I
> > > > send mu PT without the current fix and turn into a patch set. Let's
> > > > try to debug this for rc2/rc3.
> > > Hey, I'm coming back regarding this issue, I'm running 5.10 (from debian
> > > unstable) and still getting this issue
> > Many of the fixes are landing to 5.13, stable kernels do not acquire
> > out-of-tree fixes unless downstream maintainers decide to apply them
> > by hand.
> OK I tried again with linus master branch (5.13.0-rc4+) and your linux-tpmdd
> branch and the problem persists

Do have a klog for this?

/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Laurent Bigonville]

Le 9/06/21 à 14:43, Jarkko Sakkinen a écrit :
> On Thu, Jun 03, 2021 at 11:42:15AM +0200, Laurent Bigonville wrote:
>> Le 3/06/21 à 07:28, Jarkko Sakkinen a écrit :
>>> On Tue, Jun 01, 2021 at 11:17:14PM +0200, Laurent Bigonville wrote:
>>>> Le 17/02/21 à 23:29, Jarkko Sakkinen a écrit :
>>>>> On Tue, Feb 16, 2021 at 08:06:04PM +0100, Laurent Bigonville wrote:
>>>>>> Le 16/02/21 à 17:34, Jarkko Sakkinen a écrit :
>>>>>>> On Tue, Feb 16, 2021 at 06:26:54PM +0200, Jarkko Sakkinen wrote:
>>>>>>>> On Tue, Feb 16, 2021 at 06:16:58PM +0200, Jarkko Sakkinen wrote:
>>>>>>>>> On Tue, Feb 16, 2021 at 12:02:24PM +0100, Laurent Bigonville wrote:
>>>>>>>>>> Le 16/02/21 à 09:17, Jarkko Sakkinen a écrit :
>>>>>>>>>>> From: Lukasz Majczak <lma@semihalf.com>
>>>>>>>>>>>
>>>>>>>>>>> This is shown with Samsung Chromebook Pro (Caroline) with TPM 1.2
>>>>>>>>>>> (SLB 9670):
>>>>>>>>>>>
>>>>>>>>>>> [    4.324298] TPM returned invalid status
>>>>>>>>>>> [    4.324806] WARNING: CPU: 2 PID: 1 at drivers/char/tpm/tpm_tis_core.c:275 tpm_tis_status+0x86/0x8f
>>>>>>>>>>>
>>>>>>>>>>> Background
>>>>>>>>>>> ==========
>>>>>>>>>>>
>>>>>>>>>>> TCG PC Client Platform TPM Profile (PTP) Specification, paragraph 6.1 FIFO
>>>>>>>>>>> Interface Locality Usage per Register, Table 39 Register Behavior Based on
>>>>>>>>>>> Locality Setting for FIFO - a read attempt to TPM_STS_x Registers returns
>>>>>>>>>>> 0xFF in case of lack of locality. The described situation manifests itself
>>>>>>>>>>> with the following warning trace:
>>>>>>>>>>>
>>>>>>>>>>> The fix
>>>>>>>>>>> =======
>>>>>>>>>>>
>>>>>>>>>>> Add the proper decorations to tpm_tis_gen_interrupt() and
>>>>>>>>>>> tpm_get_timeouts().
>>>>>>>>>> I tried that patch (alone on the top of the HEAD of Linus master) and I
>>>>>>>>>> still get the same trace in dmesg
>>>>>>>>> Can you give a shot to
>>>>>>>>>
>>>>>>>>> git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
>>>>>>>>>
>>>>>>>>> It has couple of more fixes, and paste the log.
>>>>>>>> And if possible a full stack trace :-)
>>>>>>> And apply this patch on top. Cannot apply it there before it's reviewed.
>>>>>>>
>>>>>> I get the following stacktrace with your branch and the patch here:
>>>>>>
>>>>>> [   13.498925] ------------[ cut here ]------------
>>>>>> [   13.498930] TPM returned invalid status
>>>>>> [   13.498953] WARNING: CPU: 1 PID: 459 at
>>>>>> drivers/char/tpm/tpm_tis_core.c:205 tpm_tis_status+0x86/0xa0 [tpm_tis_core]
>>>>>> [   13.498963] Modules linked in: libiscsi_tcp(E) tpm_tis(E+)
>>>>>> tpm_tis_core(E) libiscsi(E) snd_timer(E) tpm(E) joydev(E) snd(E)
>>>>>> scsi_transport_iscsi(E) pcc_cpufreq(E-) fjes(E-) soundcore(E) i7core_edac(E)
>>>>>> i5500_temp(E) asus_atk0110(E) rng_core(E) acpi_cpufreq(E-) evdev(E) loop(E)
>>>>>> firewire_sbp2(E) msr(E) parport_pc(E) ppdev(E) lp(E) parport(E) fuse(E)
>>>>>> configfs(E) sunrpc(E) ip_tables(E) x_tables(E) autofs4(E) ext4(E) crc16(E)
>>>>>> mbcache(E) jbd2(E) btrfs(E) blake2b_generic(E) zstd_compress(E) efivars(E)
>>>>>> raid10(E) raid456(E) async_raid6_recov(E) async_memcpy(E) async_pq(E)
>>>>>> async_xor(E) async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) crc32c_generic(E)
>>>>>> raid1(E) raid0(E) multipath(E) linear(E) md_mod(E) dm_mod(E) sr_mod(E)
>>>>>> sd_mod(E) cdrom(E) t10_pi(E) hid_generic(E) usbhid(E) hid(E) amdgpu(E)
>>>>>> ahci(E) libahci(E) libata(E) gpu_sched(E) i2c_algo_bit(E) drm_ttm_helper(E)
>>>>>> ttm(E) uhci_hcd(E) ehci_pci(E) firewire_ohci(E) crc32c_intel(E) ehci_hcd(E)
>>>>>> drm_kms_helper(E) firewire_core(E) i2c_i801(E) psmouse(E)
>>>>>> [   13.499044]  scsi_mod(E) cec(E) i2c_smbus(E) mxm_wmi(E) lpc_ich(E)
>>>>>> crc_itu_t(E) sky2(E) usbcore(E) drm(E) mfd_core(E) wmi(E) button(E)
>>>>>> [   13.499058] CPU: 1 PID: 459 Comm: systemd-udevd Tainted: G          I
>>>>>> E     5.11.0+ #4
>>>>>> [   13.499062] Hardware name: System manufacturer System Product Name/P6T
>>>>>> DELUXE V2, BIOS 0406    04/24/2009
>>>>>> [   13.499064] RIP: 0010:tpm_tis_status+0x86/0xa0 [tpm_tis_core]
>>>>>> [   13.499069] Code: 00 75 30 48 83 c4 18 c3 31 c0 80 3d 83 48 00 00 00 75
>>>>>> e0 48 c7 c7 4c 53 49 c1 88 44 24 07 c6 05 6f 48 00 00 01 e8 b2 24 ba cd <0f>
>>>>>> 0b 0f b6 44 24 07 eb c0 e8 fc 99 bd cd 66 66 2e 0f 1f 84 00 00
>>>>>> [   13.499072] RSP: 0018:ffffbaeb80fb3aa0 EFLAGS: 00010286
>>>>>> [   13.499075] RAX: 0000000000000000 RBX: ffff9704034e5000 RCX:
>>>>>> ffff9707ada58bc8
>>>>>> [   13.499078] RDX: 00000000ffffffd8 RSI: 0000000000000027 RDI:
>>>>>> ffff9707ada58bc0
>>>>>> [   13.499080] RBP: 00000000ffff5d40 R08: 0000000000000000 R09:
>>>>>> ffffbaeb80fb38c0
>>>>>> [   13.499082] R10: ffffbaeb80fb38b8 R11: ffffffff8fac2588 R12:
>>>>>> 0000000000000016
>>>>>> [   13.499085] R13: ffff97040c38d000 R14: 0000000000001000 R15:
>>>>>> ffffbaeb80fb3ada
>>>>>> [   13.499087] FS:  00007fe1bcfeb8c0(0000) GS:ffff9707ada40000(0000)
>>>>>> knlGS:0000000000000000
>>>>>> [   13.499090] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>>>>> [   13.499092] CR2: 000055e5f7048590 CR3: 00000001084fe000 CR4:
>>>>>> 00000000000006e0
>>>>>> [   13.499095] Call Trace:
>>>>>> [   13.499099]  tpm_transmit+0x15f/0x3d0 [tpm]
>>>>>> [   13.499113]  tpm_transmit_cmd+0x25/0x90 [tpm]
>>>>>> [   13.499121]  tpm2_probe+0xe2/0x140 [tpm]
>>>>>> [   13.499130]  tpm_tis_core_init+0x1d5/0x2b0 [tpm_tis_core]
>>>>>> [   13.499135]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
>>>>>> [   13.499141]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
>>>>>> [   13.499148]  pnp_device_probe+0xaf/0x140
>>>>>> [   13.499154]  really_probe+0xf2/0x440
>>>>>> [   13.499160]  driver_probe_device+0xe1/0x150
>>>>>> [   13.499163]  device_driver_attach+0xa1/0xb0
>>>>>> [   13.499167]  __driver_attach+0x8a/0x150
>>>>>> [   13.499171]  ? device_driver_attach+0xb0/0xb0
>>>>>> [   13.499174]  ? device_driver_attach+0xb0/0xb0
>>>>>> [   13.499177]  bus_for_each_dev+0x78/0xc0
>>>>>> [   13.499181]  bus_add_driver+0x12b/0x1e0
>>>>>> [   13.499184]  driver_register+0x8b/0xe0
>>>>>> [   13.499188]  ? 0xffffffffc1413000
>>>>>> [   13.499191]  init_tis+0xa0/0x1000 [tpm_tis]
>>>>>> [   13.499197]  do_one_initcall+0x44/0x1d0
>>>>>> [   13.499202]  ? do_init_module+0x23/0x260
>>>>>> [   13.499206]  ? kmem_cache_alloc_trace+0xf5/0x200
>>>>>> [   13.499212]  do_init_module+0x5c/0x260
>>>>>> [   13.499215]  __do_sys_finit_module+0xb1/0x110
>>>>>> [   13.499225]  do_syscall_64+0x33/0x80
>>>>>> [   13.499229]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
>>>>>> [   13.499234] RIP: 0033:0x7fe1bd4a49b9
>>>>>> [   13.499237] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48
>>>>>> 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48>
>>>>>> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
>>>>>> [   13.499240] RSP: 002b:00007ffca5bfa108 EFLAGS: 00000246 ORIG_RAX:
>>>>>> 0000000000000139
>>>>>> [   13.499243] RAX: ffffffffffffffda RBX: 000055e5f7036920 RCX:
>>>>>> 00007fe1bd4a49b9
>>>>>> [   13.499245] RDX: 0000000000000000 RSI: 00007fe1bd62fe2d RDI:
>>>>>> 0000000000000012
>>>>>> [   13.499247] RBP: 0000000000020000 R08: 0000000000000000 R09:
>>>>>> 000055e5f6e7a0b0
>>>>>> [   13.499249] R10: 0000000000000012 R11: 0000000000000246 R12:
>>>>>> 00007fe1bd62fe2d
>>>>>> [   13.499251] R13: 0000000000000000 R14: 000055e5f7033420 R15:
>>>>>> 000055e5f7036920
>>>>>> [   13.499255] ---[ end trace 7e963e5c3647102e ]---
>>>>>>
>>>>>> If I also apply the patch to change the WARN_ONCE in dev_warn_once, and grep
>>>>>> on the string "tpm", I get the following:
>>>>>>
>>>>>> Feb 16 19:51:22 fornost kernel: tpm tpm0: TPM returned invalid status: 0xa0
>>>>>> Feb 16 19:51:22 fornost kernel: tpm_tis 00:06: 1.2 TPM (device-id 0x6871,
>>>>>> rev-id 1)
>>>>>> Feb 16 19:51:24 fornost kernel: tpm tpm0: tpm_try_transmit: send(): error
>>>>>> -62
>>>>>> Feb 16 19:51:24 fornost kernel: tpm tpm0: A TPM error (-62) occurred
>>>>>> attempting to determine the timeouts
>>>>>> Feb 16 19:51:24 fornost kernel: tpm_tis: probe of 00:06 failed with error
>>>>>> -62
>>>>>> Feb 16 19:51:24 fornost kernel: tpm_inf_pnp 00:06: Found TPM with ID IFX0102
>>>>>>
>>>>>>
>>>>> No immediate idea why this happens as the code should take locality.
>>>>>
>>>>> I noticed that the original patch bundles incorrectly two fixes, so I
>>>>> send mu PT without the current fix and turn into a patch set. Let's
>>>>> try to debug this for rc2/rc3.
>>>> Hey, I'm coming back regarding this issue, I'm running 5.10 (from debian
>>>> unstable) and still getting this issue
>>> Many of the fixes are landing to 5.13, stable kernels do not acquire
>>> out-of-tree fixes unless downstream maintainers decide to apply them
>>> by hand.
>> OK I tried again with linus master branch (5.13.0-rc4+) and your linux-tpmdd
>> branch and the problem persists
> Do have a klog for this?

In dmesg I see the following:

[   13.018702] ------------[ cut here ]------------
[   13.018706] TPM returned invalid status
[   13.018720] WARNING: CPU: 1 PID: 450 at 
drivers/char/tpm/tpm_tis_core.c:205 tpm_tis_status+0x86/0xa0 [tpm_tis_core]
[   13.018728] Modules linked in: tpm_tis(E+) tpm_tis_core(E) tpm(E) 
asus_atk0110(E) rng_core(E) fjes(E-) evdev(E) acpi_cpufreq(E-) loop(E+) 
firewire_sbp2(E) msr(E) parport_pc(E) ppdev(E) lp(E) parport(E) fuse(E) 
configfs(E) sunrpc(E) ip_tables(E) x_tables(E) autofs4(E) ext4(E) 
crc16(E) mbcache(E) jbd2(E) btrfs(E) blake2b_generic(E) zstd_compress(E) 
efivars(E) raid10(E) raid456(E) async_raid6_recov(E) async_memcpy(E) 
async_pq(E) async_xor(E) async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) 
crc32c_generic(E) raid1(E) raid0(E) multipath(E) linear(E) md_mod(E) 
dm_mod(E) sr_mod(E) sd_mod(E) cdrom(E) t10_pi(E) hid_generic(E) 
usbhid(E) hid(E) amdgpu(E) gpu_sched(E) i2c_algo_bit(E) 
drm_ttm_helper(E) ahci(E) ttm(E) libahci(E) firewire_ohci(E) i2c_i801(E) 
libata(E) drm_kms_helper(E) uhci_hcd(E) ehci_pci(E) cec(E) 
crc32c_intel(E) firewire_core(E) mxm_wmi(E) psmouse(E) ehci_hcd(E) 
crc_itu_t(E) i2c_smbus(E) scsi_mod(E) lpc_ich(E) sky2(E) usbcore(E) 
drm(E) mfd_core(E) wmi(E) button(E)
[   13.018782] CPU: 1 PID: 450 Comm: systemd-udevd Tainted: G          I 
E     5.13.0-rc4+ #6
[   13.018784] Hardware name: System manufacturer System Product 
Name/P6T DELUXE V2, BIOS 0406    04/24/2009
[   13.018786] RIP: 0010:tpm_tis_status+0x86/0xa0 [tpm_tis_core]
[   13.018790] Code: 00 75 30 48 83 c4 18 c3 31 c0 80 3d c3 49 00 00 00 
75 e0 48 c7 c7 64 03 4d c1 88 44 24 07 c6 05 af 49 00 00 01 e8 42 ab b8 
f0 <0f> 0b 0f b6 44 24 07 eb c0 e8 9c 36 bc f0 66 66 2e 0f 1f 84 00 00
[   13.018792] RSP: 0000:ffffae84807f3a98 EFLAGS: 00010282
[   13.018794] RAX: 0000000000000000 RBX: ffff96e5491b9000 RCX: 
ffff96e8eda58788
[   13.018795] RDX: 00000000ffffffd8 RSI: 0000000000000027 RDI: 
ffff96e8eda58780
[   13.018797] RBP: 00000000ffff5cf4 R08: 0000000000000000 R09: 
ffffae84807f38c0
[   13.018798] R10: ffffae84807f38b8 R11: ffffffffb2ac3888 R12: 
0000000000000016
[   13.018800] R13: ffff96e550d1e000 R14: 0000000000001000 R15: 
ffffae84807f3ad2
[   13.018801] FS:  00007f75db70e8c0(0000) GS:ffff96e8eda40000(0000) 
knlGS:0000000000000000
[   13.018803] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   13.018805] CR2: 00007f75da19d077 CR3: 000000010675e000 CR4: 
00000000000006e0
[   13.018807] Call Trace:
[   13.018810]  tpm_transmit+0x15f/0x3d0 [tpm]
[   13.018818]  tpm_transmit_cmd+0x25/0x90 [tpm]
[   13.018824]  tpm2_probe+0xe2/0x140 [tpm]
[   13.018830]  tpm_tis_core_init+0x1d5/0x2aa [tpm_tis_core]
[   13.018834]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
[   13.018838]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
[   13.018843]  pnp_device_probe+0xaf/0x140
[   13.018847]  really_probe+0xf2/0x460
[   13.018852]  driver_probe_device+0xe8/0x160
[   13.018854]  device_driver_attach+0xa1/0xb0
[   13.018866]  __driver_attach+0x8f/0x150
[   13.018869]  ? device_driver_attach+0xb0/0xb0
[   13.018871]  ? device_driver_attach+0xb0/0xb0
[   13.018873]  bus_for_each_dev+0x78/0xc0
[   13.018876]  bus_add_driver+0x12b/0x1e0
[   13.018879]  driver_register+0x8b/0xe0
[   13.018881]  ? 0xffffffffc14de000
[   13.018883]  init_tis+0xa0/0x1000 [tpm_tis]
[   13.018888]  do_one_initcall+0x44/0x1d0
[   13.018892]  ? kmem_cache_alloc_trace+0x119/0x240
[   13.018896]  do_init_module+0x5c/0x260
[   13.018900]  __do_sys_finit_module+0xb1/0x110
[   13.018904]  do_syscall_64+0x40/0xb0
[   13.018908]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[   13.018911] RIP: 0033:0x7f75dbbc79b9
[   13.018913] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 
48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 
05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
[   13.018926] RSP: 002b:00007ffe79a15068 EFLAGS: 00000246 ORIG_RAX: 
0000000000000139
[   13.018929] RAX: ffffffffffffffda RBX: 0000555b250d19a0 RCX: 
00007f75dbbc79b9
[   13.018930] RDX: 0000000000000000 RSI: 00007f75dbd52e2d RDI: 
0000000000000012
[   13.018936] RBP: 0000000000020000 R08: 0000000000000000 R09: 
0000555b250d0060
[   13.018938] R10: 0000000000000012 R11: 0000000000000246 R12: 
00007f75dbd52e2d
[   13.018939] R13: 0000000000000000 R14: 0000555b250d1420 R15: 
0000555b250d19a0
[   13.018944] ---[ end trace 6abb891c8492a76e ]---

And

# dmesg |grep -i tpm

[   13.019986] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)
[   15.016198] tpm tpm0: tpm_try_transmit: send(): error -62
[   15.016208] tpm tpm0: A TPM error (-62) occurred attempting to 
determine the timeouts
[   15.016239] tpm_tis: probe of 00:06 failed with error -62
[   15.053255] tpm_inf_pnp 00:06: Found TPM with ID IFX0102

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

[-- Attachment #1: Type: text/plain, Size: 639 bytes --]

On Thu, Jun 10, 2021 at 01:35:47PM +0200, Laurent Bigonville wrote:
> # dmesg |grep -i tpm
> 
> [   13.019986] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)
> [   15.016198] tpm tpm0: tpm_try_transmit: send(): error -62
> [   15.016208] tpm tpm0: A TPM error (-62) occurred attempting to determine
> the timeouts
> [   15.016239] tpm_tis: probe of 00:06 failed with error -62
> [   15.053255] tpm_inf_pnp 00:06: Found TPM with ID IFX0102

If possible, can you check what happens when you apply the attached patch.
I'm not proposing it as a bug fix but it just simplifies the flow a lot,
and might help to observe something.

/Jarkko

[-- Attachment #2: 0001-tpm_tis-Use-request_locality-before-calling-tpm2_pro.patch --]
[-- Type: text/plain, Size: 1297 bytes --]

From 0b0dfc48b47e44d547282b20f6d8c97103fadbe6 Mon Sep 17 00:00:00 2001
From: Jarkko Sakkinen <jarkko@kernel.org>
Date: Thu, 10 Jun 2021 15:24:49 +0300
Subject: [PATCH] tpm_tis: Use request_locality() before calling tpm2_probe()

Avoid unnecessary round-tripping in and out of the TIS driver by calling
request_locality() directly, instead of going through tpm_chip_start(), and
for the consistency sake (other sites in initialization code work this
way).

Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
---
 drivers/char/tpm/tpm_tis_core.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
index 69579efb247b..3b5a03f9efce 100644
--- a/drivers/char/tpm/tpm_tis_core.c
+++ b/drivers/char/tpm/tpm_tis_core.c
@@ -980,13 +980,11 @@ int tpm_tis_core_init(struct device *dev, struct tpm_tis_data *priv, int irq,
 	intmask &= ~TPM_GLOBAL_INT_ENABLE;
 	tpm_tis_write32(priv, TPM_INT_ENABLE(priv->locality), intmask);
 
-	rc = tpm_chip_start(chip);
-	if (rc)
+	rc = request_locality(chip, 0);
+	if (rc < 0)
 		goto out_err;
 	rc = tpm2_probe(chip);
-	tpm_chip_stop(chip);
-	if (rc)
-		goto out_err;
+	release_locality(chip, 0);
 
 	rc = tpm_tis_read32(priv, TPM_DID_VID(0), &vendor);
 	if (rc < 0)
-- 
2.31.1

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Laurent Bigonville]

Le 10/06/21 à 14:29, Jarkko Sakkinen a écrit :
> On Thu, Jun 10, 2021 at 01:35:47PM +0200, Laurent Bigonville wrote:
>> # dmesg |grep -i tpm
>>
>> [   13.019986] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)
>> [   15.016198] tpm tpm0: tpm_try_transmit: send(): error -62
>> [   15.016208] tpm tpm0: A TPM error (-62) occurred attempting to determine
>> the timeouts
>> [   15.016239] tpm_tis: probe of 00:06 failed with error -62
>> [   15.053255] tpm_inf_pnp 00:06: Found TPM with ID IFX0102
> If possible, can you check what happens when you apply the attached patch.
> I'm not proposing it as a bug fix but it just simplifies the flow a lot,
> and might help to observe something.
>
With the patch on the top of HEAD of master and the patch from your 
branch, I'm getting this:

[   13.140548] tpm tpm0: invalid TPM_STS.x 0xa0, dumping stack for forensics
[   13.140617] CPU: 4 PID: 462 Comm: systemd-udevd Tainted: G          I 
E     5.13.0-rc5+ #7
[   13.140621] Hardware name: System manufacturer System Product 
Name/P6T DELUXE V2, BIOS 0406    04/24/2009
[   13.140623] Call Trace:
[   13.140628]  dump_stack+0x76/0x94
[   13.140637]  tpm_tis_status.cold+0x19/0x20 [tpm_tis_core]
[   13.140643]  tpm_transmit+0x15f/0x3d0 [tpm]
[   13.140657]  tpm_transmit_cmd+0x25/0x90 [tpm]
[   13.140666]  tpm2_probe+0xe2/0x140 [tpm]
[   13.140676]  tpm_tis_core_init+0x1d7/0x2ba [tpm_tis_core]
[   13.140681]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
[   13.140688]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
[   13.140695]  pnp_device_probe+0xaf/0x140
[   13.140702]  really_probe+0xf2/0x460
[   13.140708]  driver_probe_device+0xe8/0x160
[   13.140712]  device_driver_attach+0xa1/0xb0
[   13.140716]  __driver_attach+0x8f/0x150
[   13.140720]  ? device_driver_attach+0xb0/0xb0
[   13.140723]  ? device_driver_attach+0xb0/0xb0
[   13.140726]  bus_for_each_dev+0x78/0xc0
[   13.140730]  bus_add_driver+0x12b/0x1e0
[   13.140734]  driver_register+0x8b/0xe0
[   13.140738]  ? 0xffffffffc11c4000
[   13.140741]  init_tis+0xa0/0x1000 [tpm_tis]
[   13.140748]  do_one_initcall+0x44/0x1d0
[   13.140754]  ? kmem_cache_alloc_trace+0x119/0x240
[   13.140759]  do_init_module+0x5c/0x260
[   13.140764]  __do_sys_finit_module+0xb1/0x110
[   13.140769]  do_syscall_64+0x40/0xb0
[   13.140775]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[   13.140780] RIP: 0033:0x7f082957b9b9
[   13.140783] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 
48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 
05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
[   13.140786] RSP: 002b:00007fffd1b5a048 EFLAGS: 00000246 ORIG_RAX: 
0000000000000139
[   13.140791] RAX: ffffffffffffffda RBX: 000055bd4f3f6960 RCX: 
00007f082957b9b9
[   13.140793] RDX: 0000000000000000 RSI: 00007f0829706e2d RDI: 
0000000000000012
[   13.140796] RBP: 0000000000020000 R08: 0000000000000000 R09: 
000055bd4f238350
[   13.140798] R10: 0000000000000012 R11: 0000000000000246 R12: 
00007f0829706e2d
[   13.140800] R13: 0000000000000000 R14: 000055bd4f2381c0 R15: 
000055bd4f3f6960
[   13.141844] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)

# dmesg |grep -i tpm

[...]
[   15.140121] tpm tpm0: tpm_try_transmit: send(): error -62
[   15.141852] tpm tpm0: A TPM error (-62) occurred attempting to 
determine the timeouts
[   15.143596] tpm_tis: probe of 00:06 failed with error -62
[   15.185451] tpm_inf_pnp 00:06: Found TPM with ID IFX0102

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Thu, Jun 10, 2021 at 04:00:27PM +0200, Laurent Bigonville wrote:
> Le 10/06/21 à 14:29, Jarkko Sakkinen a écrit :
> > On Thu, Jun 10, 2021 at 01:35:47PM +0200, Laurent Bigonville wrote:
> > > # dmesg |grep -i tpm
> > > 
> > > [   13.019986] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)
> > > [   15.016198] tpm tpm0: tpm_try_transmit: send(): error -62
> > > [   15.016208] tpm tpm0: A TPM error (-62) occurred attempting to determine
> > > the timeouts
> > > [   15.016239] tpm_tis: probe of 00:06 failed with error -62
> > > [   15.053255] tpm_inf_pnp 00:06: Found TPM with ID IFX0102
> > If possible, can you check what happens when you apply the attached patch.
> > I'm not proposing it as a bug fix but it just simplifies the flow a lot,
> > and might help to observe something.
> > 
> With the patch on the top of HEAD of master and the patch from your branch,
> I'm getting this:
> 
> [   13.140548] tpm tpm0: invalid TPM_STS.x 0xa0, dumping stack for forensics
> [   13.140617] CPU: 4 PID: 462 Comm: systemd-udevd Tainted: G          I
> E     5.13.0-rc5+ #7
> [   13.140621] Hardware name: System manufacturer System Product Name/P6T
> DELUXE V2, BIOS 0406    04/24/2009
> [   13.140623] Call Trace:
> [   13.140628]  dump_stack+0x76/0x94
> [   13.140637]  tpm_tis_status.cold+0x19/0x20 [tpm_tis_core]
> [   13.140643]  tpm_transmit+0x15f/0x3d0 [tpm]
> [   13.140657]  tpm_transmit_cmd+0x25/0x90 [tpm]
> [   13.140666]  tpm2_probe+0xe2/0x140 [tpm]

It's weird because it's the first transaction with the TPM, not likely
to have any sort of mismatch in power gating.

I'm suspecting this:

	if (wait_startup(chip, 0) != 0) {
		rc = -ENODEV;
		goto out_err;
	}

I did not find anything from [*] stating that reading Access Register,
could cause any change in the burstCount (0xa0) but still might be worth
of testing to couple that with

	release_locality(chip, 0);

I.e. add that just after the first snippet.

[*] https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/

/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Laurent Bigonville]

Le 14/06/21 à 22:40, Jarkko Sakkinen a écrit :
> On Thu, Jun 10, 2021 at 04:00:27PM +0200, Laurent Bigonville wrote:
>> Le 10/06/21 à 14:29, Jarkko Sakkinen a écrit :
>>> On Thu, Jun 10, 2021 at 01:35:47PM +0200, Laurent Bigonville wrote:
>>>> # dmesg |grep -i tpm
>>>>
>>>> [   13.019986] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)
>>>> [   15.016198] tpm tpm0: tpm_try_transmit: send(): error -62
>>>> [   15.016208] tpm tpm0: A TPM error (-62) occurred attempting to determine
>>>> the timeouts
>>>> [   15.016239] tpm_tis: probe of 00:06 failed with error -62
>>>> [   15.053255] tpm_inf_pnp 00:06: Found TPM with ID IFX0102
>>> If possible, can you check what happens when you apply the attached patch.
>>> I'm not proposing it as a bug fix but it just simplifies the flow a lot,
>>> and might help to observe something.
>>>
>> With the patch on the top of HEAD of master and the patch from your branch,
>> I'm getting this:
>>
>> [   13.140548] tpm tpm0: invalid TPM_STS.x 0xa0, dumping stack for forensics
>> [   13.140617] CPU: 4 PID: 462 Comm: systemd-udevd Tainted: G          I
>> E     5.13.0-rc5+ #7
>> [   13.140621] Hardware name: System manufacturer System Product Name/P6T
>> DELUXE V2, BIOS 0406    04/24/2009
>> [   13.140623] Call Trace:
>> [   13.140628]  dump_stack+0x76/0x94
>> [   13.140637]  tpm_tis_status.cold+0x19/0x20 [tpm_tis_core]
>> [   13.140643]  tpm_transmit+0x15f/0x3d0 [tpm]
>> [   13.140657]  tpm_transmit_cmd+0x25/0x90 [tpm]
>> [   13.140666]  tpm2_probe+0xe2/0x140 [tpm]
> It's weird because it's the first transaction with the TPM, not likely
> to have any sort of mismatch in power gating.
>
> I'm suspecting this:
>
> 	if (wait_startup(chip, 0) != 0) {
> 		rc = -ENODEV;
> 		goto out_err;
> 	}
>
> I did not find anything from [*] stating that reading Access Register,
> could cause any change in the burstCount (0xa0) but still might be worth
> of testing to couple that with
>
> 	release_locality(chip, 0);
>
> I.e. add that just after the first snippet.
>
You mean this, right?

diff --git a/drivers/char/tpm/tpm_tis_core.c 
b/drivers/char/tpm/tpm_tis_core.c
index 3b5a03f9efce..857ea790e9d3 100644
--- a/drivers/char/tpm/tpm_tis_core.c
+++ b/drivers/char/tpm/tpm_tis_core.c
@@ -969,6 +969,7 @@ int tpm_tis_core_init(struct device *dev, struct 
tpm_tis_data *priv, int irq,
                 rc = -ENODEV;
                 goto out_err;
         }
+       release_locality(chip, 0);

         /* Take control of the TPM's interrupt hardware and shut it off */
         rc = tpm_tis_read32(priv, TPM_INT_ENABLE(priv->locality), 
&intmask);


With this change (and with or without "tpm_tis: Use request_locality() 
before calling tpm2_probe()"), I get the following:

[   13.420485] tpm tpm0: invalid TPM_STS.x 0xff, dumping stack for forensics
[   13.420494] CPU: 5 PID: 461 Comm: systemd-udevd Tainted: G          I 
E     5.13.0-rc6+ #8
[   13.420498] Hardware name: System manufacturer System Product 
Name/P6T DELUXE V2, BIOS 0406    04/24/2009
[   13.420500] Call Trace:
[   13.420505]  dump_stack+0x76/0x94
[   13.420512]  tpm_tis_status.cold+0x19/0x20 [tpm_tis_core]
[   13.420519]  wait_for_tpm_stat+0x49/0x200 [tpm_tis_core]
[   13.420524]  ? get_page_from_freelist+0x228/0xed0
[   13.420530]  ? tpm_tcg_read_bytes+0x30/0x50 [tpm_tis]
[   13.420536]  tpm_tis_send_data+0x90/0x250 [tpm_tis_core]
[   13.420542]  tpm_tis_send_main+0x2e/0xf0 [tpm_tis_core]
[   13.420547]  tpm_transmit+0xd8/0x3d0 [tpm]
[   13.420559]  tpm_transmit_cmd+0x25/0x90 [tpm]
[   13.420567]  tpm2_probe+0xe2/0x140 [tpm]
[   13.420577]  tpm_tis_core_init+0x1fd/0x2e0 [tpm_tis_core]
[   13.420582]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
[   13.420587]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
[   13.420594]  pnp_device_probe+0xaf/0x140
[   13.420600]  really_probe+0xf2/0x460
[   13.420606]  driver_probe_device+0xe8/0x160
[   13.420610]  device_driver_attach+0xa1/0xb0
[   13.420614]  __driver_attach+0x8f/0x150
[   13.420618]  ? device_driver_attach+0xb0/0xb0
[   13.420621]  ? device_driver_attach+0xb0/0xb0
[   13.420624]  bus_for_each_dev+0x78/0xc0
[   13.420628]  bus_add_driver+0x12b/0x1e0
[   13.420632]  driver_register+0x8b/0xe0
[   13.420636]  ? 0xffffffffc1413000
[   13.420639]  init_tis+0xa0/0x1000 [tpm_tis]
[   13.420646]  do_one_initcall+0x44/0x1d0
[   13.420652]  ? kmem_cache_alloc_trace+0x119/0x240
[   13.420657]  do_init_module+0x5c/0x260
[   13.420662]  __do_sys_finit_module+0xb1/0x110
[   13.420668]  do_syscall_64+0x40/0xb0
[   13.420673]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[   13.420678] RIP: 0033:0x7f15fe84e9b9
[   13.420681] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 
48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 
05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
[   13.420685] RSP: 002b:00007fffc3aca668 EFLAGS: 00000246 ORIG_RAX: 
0000000000000139
[   13.420689] RAX: ffffffffffffffda RBX: 00005561627275b0 RCX: 
00007f15fe84e9b9
[   13.420692] RDX: 0000000000000000 RSI: 00007f15fe9d9e2d RDI: 
0000000000000012
[   13.420694] RBP: 0000000000020000 R08: 0000000000000000 R09: 
0000556161c2f330
[   13.420696] R10: 0000000000000012 R11: 0000000000000246 R12: 
00007f15fe9d9e2d
[   13.420698] R13: 0000000000000000 R14: 000055616271dab0 R15: 
00005561627275b0
[...]
[   14.185406] tpm tpm0: tpm_try_transmit: send(): error -62
[   14.185422] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)
[   15.021233] tpm tpm0: tpm_try_transmit: send(): error -62
[   24.381575] tpm tpm0: tpm_try_transmit: send(): error -62
[...]

But now the /dev/tpm0 device appears (but doesn't seem to work)

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Jarkko Sakkinen]

On Tue, Jun 15, 2021 at 01:19:56PM +0200, Laurent Bigonville wrote:
> Le 14/06/21 à 22:40, Jarkko Sakkinen a écrit :
> > On Thu, Jun 10, 2021 at 04:00:27PM +0200, Laurent Bigonville wrote:
> > > Le 10/06/21 à 14:29, Jarkko Sakkinen a écrit :
> > > > On Thu, Jun 10, 2021 at 01:35:47PM +0200, Laurent Bigonville wrote:
> > > > > # dmesg |grep -i tpm
> > > > > 
> > > > > [   13.019986] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)
> > > > > [   15.016198] tpm tpm0: tpm_try_transmit: send(): error -62
> > > > > [   15.016208] tpm tpm0: A TPM error (-62) occurred attempting to determine
> > > > > the timeouts
> > > > > [   15.016239] tpm_tis: probe of 00:06 failed with error -62
> > > > > [   15.053255] tpm_inf_pnp 00:06: Found TPM with ID IFX0102
> > > > If possible, can you check what happens when you apply the attached patch.
> > > > I'm not proposing it as a bug fix but it just simplifies the flow a lot,
> > > > and might help to observe something.
> > > > 
> > > With the patch on the top of HEAD of master and the patch from your branch,
> > > I'm getting this:
> > > 
> > > [   13.140548] tpm tpm0: invalid TPM_STS.x 0xa0, dumping stack for forensics
> > > [   13.140617] CPU: 4 PID: 462 Comm: systemd-udevd Tainted: G          I
> > > E     5.13.0-rc5+ #7
> > > [   13.140621] Hardware name: System manufacturer System Product Name/P6T
> > > DELUXE V2, BIOS 0406    04/24/2009
> > > [   13.140623] Call Trace:
> > > [   13.140628]  dump_stack+0x76/0x94
> > > [   13.140637]  tpm_tis_status.cold+0x19/0x20 [tpm_tis_core]
> > > [   13.140643]  tpm_transmit+0x15f/0x3d0 [tpm]
> > > [   13.140657]  tpm_transmit_cmd+0x25/0x90 [tpm]
> > > [   13.140666]  tpm2_probe+0xe2/0x140 [tpm]
> > It's weird because it's the first transaction with the TPM, not likely
> > to have any sort of mismatch in power gating.
> > 
> > I'm suspecting this:
> > 
> > 	if (wait_startup(chip, 0) != 0) {
> > 		rc = -ENODEV;
> > 		goto out_err;
> > 	}
> > 
> > I did not find anything from [*] stating that reading Access Register,
> > could cause any change in the burstCount (0xa0) but still might be worth
> > of testing to couple that with
> > 
> > 	release_locality(chip, 0);
> > 
> > I.e. add that just after the first snippet.
> > 
> You mean this, right?
> 
> diff --git a/drivers/char/tpm/tpm_tis_core.c
> b/drivers/char/tpm/tpm_tis_core.c
> index 3b5a03f9efce..857ea790e9d3 100644
> --- a/drivers/char/tpm/tpm_tis_core.c
> +++ b/drivers/char/tpm/tpm_tis_core.c
> @@ -969,6 +969,7 @@ int tpm_tis_core_init(struct device *dev, struct
> tpm_tis_data *priv, int irq,
>                 rc = -ENODEV;
>                 goto out_err;
>         }
> +       release_locality(chip, 0);
> 
>         /* Take control of the TPM's interrupt hardware and shut it off */
>         rc = tpm_tis_read32(priv, TPM_INT_ENABLE(priv->locality), &intmask);
> 
> 
> With this change (and with or without "tpm_tis: Use request_locality()
> before calling tpm2_probe()"), I get the following:
> 
> [   13.420485] tpm tpm0: invalid TPM_STS.x 0xff, dumping stack for forensics
> [   13.420494] CPU: 5 PID: 461 Comm: systemd-udevd Tainted: G          I
> E     5.13.0-rc6+ #8
> [   13.420498] Hardware name: System manufacturer System Product Name/P6T
> DELUXE V2, BIOS 0406    04/24/2009
> [   13.420500] Call Trace:
> [   13.420505]  dump_stack+0x76/0x94
> [   13.420512]  tpm_tis_status.cold+0x19/0x20 [tpm_tis_core]
> [   13.420519]  wait_for_tpm_stat+0x49/0x200 [tpm_tis_core]
> [   13.420524]  ? get_page_from_freelist+0x228/0xed0
> [   13.420530]  ? tpm_tcg_read_bytes+0x30/0x50 [tpm_tis]
> [   13.420536]  tpm_tis_send_data+0x90/0x250 [tpm_tis_core]
> [   13.420542]  tpm_tis_send_main+0x2e/0xf0 [tpm_tis_core]
> [   13.420547]  tpm_transmit+0xd8/0x3d0 [tpm]
> [   13.420559]  tpm_transmit_cmd+0x25/0x90 [tpm]
> [   13.420567]  tpm2_probe+0xe2/0x140 [tpm]
> [   13.420577]  tpm_tis_core_init+0x1fd/0x2e0 [tpm_tis_core]
> [   13.420582]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
> [   13.420587]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
> [   13.420594]  pnp_device_probe+0xaf/0x140
> [   13.420600]  really_probe+0xf2/0x460
> [   13.420606]  driver_probe_device+0xe8/0x160
> [   13.420610]  device_driver_attach+0xa1/0xb0
> [   13.420614]  __driver_attach+0x8f/0x150
> [   13.420618]  ? device_driver_attach+0xb0/0xb0
> [   13.420621]  ? device_driver_attach+0xb0/0xb0
> [   13.420624]  bus_for_each_dev+0x78/0xc0
> [   13.420628]  bus_add_driver+0x12b/0x1e0
> [   13.420632]  driver_register+0x8b/0xe0
> [   13.420636]  ? 0xffffffffc1413000
> [   13.420639]  init_tis+0xa0/0x1000 [tpm_tis]
> [   13.420646]  do_one_initcall+0x44/0x1d0
> [   13.420652]  ? kmem_cache_alloc_trace+0x119/0x240
> [   13.420657]  do_init_module+0x5c/0x260
> [   13.420662]  __do_sys_finit_module+0xb1/0x110
> [   13.420668]  do_syscall_64+0x40/0xb0
> [   13.420673]  entry_SYSCALL_64_after_hwframe+0x44/0xae
> [   13.420678] RIP: 0033:0x7f15fe84e9b9
> [   13.420681] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48
> 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48>
> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
> [   13.420685] RSP: 002b:00007fffc3aca668 EFLAGS: 00000246 ORIG_RAX:
> 0000000000000139
> [   13.420689] RAX: ffffffffffffffda RBX: 00005561627275b0 RCX:
> 00007f15fe84e9b9
> [   13.420692] RDX: 0000000000000000 RSI: 00007f15fe9d9e2d RDI:
> 0000000000000012
> [   13.420694] RBP: 0000000000020000 R08: 0000000000000000 R09:
> 0000556161c2f330
> [   13.420696] R10: 0000000000000012 R11: 0000000000000246 R12:
> 00007f15fe9d9e2d
> [   13.420698] R13: 0000000000000000 R14: 000055616271dab0 R15:
> 00005561627275b0
> [...]
> [   14.185406] tpm tpm0: tpm_try_transmit: send(): error -62
> [   14.185422] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)
> [   15.021233] tpm tpm0: tpm_try_transmit: send(): error -62
> [   24.381575] tpm tpm0: tpm_try_transmit: send(): error -62
> [...]
> 
> But now the /dev/tpm0 device appears (but doesn't seem to work)

I'm sorry, I made a mistake: release_locality() call should be exactly
right before tpm_chip_start(). The reason is that we access TPM_INT_ENABLE,
i.e. locality is released too early.

In summary, I think I would try this:

	tpm_chip_stop(chip);

	rc = tpm_chip_start(chip);
	if (rc)
		goto out_err;
	rc = tpm2_probe(chip);
	tpm_chip_stop(chip);
	if (rc)
		goto out_err;

tpm_chip_stop() is makes more sense it takes care also of disabling
clk_enable.

I.e. TPM should be shut down right after intmask mangling...

/Jarkko

Re: [PATCH] tpm, tpm_tis: Acquire locality in tpm_tis_gen_interrupt() and tpm_get_timeouts()

[Laurent Bigonville]

Le 15/06/21 à 14:56, Jarkko Sakkinen a écrit :
> On Tue, Jun 15, 2021 at 01:19:56PM +0200, Laurent Bigonville wrote:
>> Le 14/06/21 à 22:40, Jarkko Sakkinen a écrit :
>>> On Thu, Jun 10, 2021 at 04:00:27PM +0200, Laurent Bigonville wrote:
>>>> Le 10/06/21 à 14:29, Jarkko Sakkinen a écrit :
>>>>> On Thu, Jun 10, 2021 at 01:35:47PM +0200, Laurent Bigonville wrote:
>>>>>> # dmesg |grep -i tpm
>>>>>>
>>>>>> [   13.019986] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)
>>>>>> [   15.016198] tpm tpm0: tpm_try_transmit: send(): error -62
>>>>>> [   15.016208] tpm tpm0: A TPM error (-62) occurred attempting to determine
>>>>>> the timeouts
>>>>>> [   15.016239] tpm_tis: probe of 00:06 failed with error -62
>>>>>> [   15.053255] tpm_inf_pnp 00:06: Found TPM with ID IFX0102
>>>>> If possible, can you check what happens when you apply the attached patch.
>>>>> I'm not proposing it as a bug fix but it just simplifies the flow a lot,
>>>>> and might help to observe something.
>>>>>
>>>> With the patch on the top of HEAD of master and the patch from your branch,
>>>> I'm getting this:
>>>>
>>>> [   13.140548] tpm tpm0: invalid TPM_STS.x 0xa0, dumping stack for forensics
>>>> [   13.140617] CPU: 4 PID: 462 Comm: systemd-udevd Tainted: G          I
>>>> E     5.13.0-rc5+ #7
>>>> [   13.140621] Hardware name: System manufacturer System Product Name/P6T
>>>> DELUXE V2, BIOS 0406    04/24/2009
>>>> [   13.140623] Call Trace:
>>>> [   13.140628]  dump_stack+0x76/0x94
>>>> [   13.140637]  tpm_tis_status.cold+0x19/0x20 [tpm_tis_core]
>>>> [   13.140643]  tpm_transmit+0x15f/0x3d0 [tpm]
>>>> [   13.140657]  tpm_transmit_cmd+0x25/0x90 [tpm]
>>>> [   13.140666]  tpm2_probe+0xe2/0x140 [tpm]
>>> It's weird because it's the first transaction with the TPM, not likely
>>> to have any sort of mismatch in power gating.
>>>
>>> I'm suspecting this:
>>>
>>> 	if (wait_startup(chip, 0) != 0) {
>>> 		rc = -ENODEV;
>>> 		goto out_err;
>>> 	}
>>>
>>> I did not find anything from [*] stating that reading Access Register,
>>> could cause any change in the burstCount (0xa0) but still might be worth
>>> of testing to couple that with
>>>
>>> 	release_locality(chip, 0);
>>>
>>> I.e. add that just after the first snippet.
>>>
>> You mean this, right?
>>
>> diff --git a/drivers/char/tpm/tpm_tis_core.c
>> b/drivers/char/tpm/tpm_tis_core.c
>> index 3b5a03f9efce..857ea790e9d3 100644
>> --- a/drivers/char/tpm/tpm_tis_core.c
>> +++ b/drivers/char/tpm/tpm_tis_core.c
>> @@ -969,6 +969,7 @@ int tpm_tis_core_init(struct device *dev, struct
>> tpm_tis_data *priv, int irq,
>>                  rc = -ENODEV;
>>                  goto out_err;
>>          }
>> +       release_locality(chip, 0);
>>
>>          /* Take control of the TPM's interrupt hardware and shut it off */
>>          rc = tpm_tis_read32(priv, TPM_INT_ENABLE(priv->locality), &intmask);
>>
>>
>> With this change (and with or without "tpm_tis: Use request_locality()
>> before calling tpm2_probe()"), I get the following:
>>
>> [   13.420485] tpm tpm0: invalid TPM_STS.x 0xff, dumping stack for forensics
>> [   13.420494] CPU: 5 PID: 461 Comm: systemd-udevd Tainted: G          I
>> E     5.13.0-rc6+ #8
>> [   13.420498] Hardware name: System manufacturer System Product Name/P6T
>> DELUXE V2, BIOS 0406    04/24/2009
>> [   13.420500] Call Trace:
>> [   13.420505]  dump_stack+0x76/0x94
>> [   13.420512]  tpm_tis_status.cold+0x19/0x20 [tpm_tis_core]
>> [   13.420519]  wait_for_tpm_stat+0x49/0x200 [tpm_tis_core]
>> [   13.420524]  ? get_page_from_freelist+0x228/0xed0
>> [   13.420530]  ? tpm_tcg_read_bytes+0x30/0x50 [tpm_tis]
>> [   13.420536]  tpm_tis_send_data+0x90/0x250 [tpm_tis_core]
>> [   13.420542]  tpm_tis_send_main+0x2e/0xf0 [tpm_tis_core]
>> [   13.420547]  tpm_transmit+0xd8/0x3d0 [tpm]
>> [   13.420559]  tpm_transmit_cmd+0x25/0x90 [tpm]
>> [   13.420567]  tpm2_probe+0xe2/0x140 [tpm]
>> [   13.420577]  tpm_tis_core_init+0x1fd/0x2e0 [tpm_tis_core]
>> [   13.420582]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
>> [   13.420587]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
>> [   13.420594]  pnp_device_probe+0xaf/0x140
>> [   13.420600]  really_probe+0xf2/0x460
>> [   13.420606]  driver_probe_device+0xe8/0x160
>> [   13.420610]  device_driver_attach+0xa1/0xb0
>> [   13.420614]  __driver_attach+0x8f/0x150
>> [   13.420618]  ? device_driver_attach+0xb0/0xb0
>> [   13.420621]  ? device_driver_attach+0xb0/0xb0
>> [   13.420624]  bus_for_each_dev+0x78/0xc0
>> [   13.420628]  bus_add_driver+0x12b/0x1e0
>> [   13.420632]  driver_register+0x8b/0xe0
>> [   13.420636]  ? 0xffffffffc1413000
>> [   13.420639]  init_tis+0xa0/0x1000 [tpm_tis]
>> [   13.420646]  do_one_initcall+0x44/0x1d0
>> [   13.420652]  ? kmem_cache_alloc_trace+0x119/0x240
>> [   13.420657]  do_init_module+0x5c/0x260
>> [   13.420662]  __do_sys_finit_module+0xb1/0x110
>> [   13.420668]  do_syscall_64+0x40/0xb0
>> [   13.420673]  entry_SYSCALL_64_after_hwframe+0x44/0xae
>> [   13.420678] RIP: 0033:0x7f15fe84e9b9
>> [   13.420681] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48
>> 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48>
>> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
>> [   13.420685] RSP: 002b:00007fffc3aca668 EFLAGS: 00000246 ORIG_RAX:
>> 0000000000000139
>> [   13.420689] RAX: ffffffffffffffda RBX: 00005561627275b0 RCX:
>> 00007f15fe84e9b9
>> [   13.420692] RDX: 0000000000000000 RSI: 00007f15fe9d9e2d RDI:
>> 0000000000000012
>> [   13.420694] RBP: 0000000000020000 R08: 0000000000000000 R09:
>> 0000556161c2f330
>> [   13.420696] R10: 0000000000000012 R11: 0000000000000246 R12:
>> 00007f15fe9d9e2d
>> [   13.420698] R13: 0000000000000000 R14: 000055616271dab0 R15:
>> 00005561627275b0
>> [...]
>> [   14.185406] tpm tpm0: tpm_try_transmit: send(): error -62
>> [   14.185422] tpm_tis 00:06: 1.2 TPM (device-id 0x6871, rev-id 1)
>> [   15.021233] tpm tpm0: tpm_try_transmit: send(): error -62
>> [   24.381575] tpm tpm0: tpm_try_transmit: send(): error -62
>> [...]
>>
>> But now the /dev/tpm0 device appears (but doesn't seem to work)
> I'm sorry, I made a mistake: release_locality() call should be exactly
> right before tpm_chip_start(). The reason is that we access TPM_INT_ENABLE,
> i.e. locality is released too early.
>
> In summary, I think I would try this:
>
> 	tpm_chip_stop(chip);
>
> 	rc = tpm_chip_start(chip);
> 	if (rc)
> 		goto out_err;
> 	rc = tpm2_probe(chip);
> 	tpm_chip_stop(chip);
> 	if (rc)
> 		goto out_err;
>
> tpm_chip_stop() is makes more sense it takes care also of disabling
> clk_enable.
>
> I.e. TPM should be shut down right after intmask mangling...

Sorry to bring bad news but it's still not working :(

With the following change:

diff --git a/drivers/char/tpm/tpm_tis_core.c 
b/drivers/char/tpm/tpm_tis_core.c
index 69579efb247b..45ad813ce840 100644
--- a/drivers/char/tpm/tpm_tis_core.c
+++ b/drivers/char/tpm/tpm_tis_core.c
@@ -980,6 +980,8 @@ int tpm_tis_core_init(struct device *dev, struct 
tpm_tis_data *priv, int irq,
         intmask &= ~TPM_GLOBAL_INT_ENABLE;
         tpm_tis_write32(priv, TPM_INT_ENABLE(priv->locality), intmask);

+       tpm_chip_stop(chip);
+
         rc = tpm_chip_start(chip);
         if (rc)
                 goto out_err;

I get now a Oops

[   13.438251] BUG: unable to handle page fault for address: 
ffff9e8141157000
[   13.438257] #PF: supervisor write access in kernel mode
[   13.438259] #PF: error_code(0x0002) - not-present page
[   13.438261] PGD 100000067 P4D 100000067 PUD 0
[   13.438265] Oops: 0002 [#1] SMP PTI
[   13.438268] CPU: 7 PID: 469 Comm: systemd-udevd Tainted: G          I 
E     5.13.0-rc6+ #8
[   13.438271] Hardware name: System manufacturer System Product 
Name/P6T DELUXE V2, BIOS 0406    04/24/2009
[   13.438273] RIP: 0010:iowrite8+0x9/0x50
[   13.438282] Code: 48 c7 c2 e8 6a 10 98 48 89 04 24 e8 ac 5e 3f 00 0f 
0b 48 8b 04 24 48 83 c4 08 c3 66 0f 1f 44 00 00 48 81 fe ff ff 03 00 76 
04 <40> 88 3e c3 48 81 fe 00 00 01 00 76 07 0f b7 d6 89 f8 ee c3 8b 05
[   13.438286] RSP: 0018:ffff9e8040f23b10 EFLAGS: 00010292
[   13.438289] RAX: ffffffffc15440b0 RBX: ffff9e8040f23b48 RCX: 
ffff9e8040f23b47
[   13.438292] RDX: 0000000000000001 RSI: ffff9e8141157000 RDI: 
0000000000000020
[   13.438295] RBP: ffff9e8040f23b48 R08: ffff88f2108bf018 R09: 
ffff88f20104b290
[   13.438297] R10: 0000000000000002 R11: 0000000000000004 R12: 
ffff88f2108bf018
[   13.438300] R13: 00000000fffff000 R14: ffff88f20104b000 R15: 
00000000ffffffff
[   13.438303] FS:  00007fafaab758c0(0000) GS:ffff88f5adbc0000(0000) 
knlGS:0000000000000000
[   13.438307] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   13.438310] CR2: ffff9e8141157000 CR3: 00000001065ca000 CR4: 
00000000000006e0
[   13.438313] Call Trace:
[   13.438318]  tpm_tcg_write_bytes+0x33/0x50 [tpm_tis]
[   13.438326]  release_locality+0x3c/0x60 [tpm_tis_core]
[   13.438331]  tpm_relinquish_locality+0x24/0x40 [tpm]
[   13.438342]  tpm_chip_stop+0x26/0x50 [tpm]
[   13.438351]  tpm_tis_core_init+0x1c2/0x2b2 [tpm_tis_core]
[   13.438357]  ? tpm_tis_init.part.0+0x130/0x130 [tpm_tis]
[   13.438363]  tpm_tis_pnp_init+0xe1/0x110 [tpm_tis]
[   13.438370]  pnp_device_probe+0xaf/0x140
[   13.438377]  really_probe+0xf2/0x460
[   13.438384]  driver_probe_device+0xe8/0x160
[   13.438388]  device_driver_attach+0xa1/0xb0
[   13.438393]  __driver_attach+0x8f/0x150
[   13.438397]  ? device_driver_attach+0xb0/0xb0
[   13.438402]  ? device_driver_attach+0xb0/0xb0
[   13.438405]  bus_for_each_dev+0x78/0xc0
[   13.438410]  bus_add_driver+0x12b/0x1e0
[   13.438415]  driver_register+0x8b/0xe0
[   13.438419]  ? 0xffffffffc1549000
[   13.438422]  init_tis+0xa0/0x1000 [tpm_tis]
[   13.438430]  do_one_initcall+0x44/0x1d0
[   13.438437]  ? kmem_cache_alloc_trace+0x119/0x240
[   13.438444]  do_init_module+0x5c/0x260
[   13.438450]  __do_sys_finit_module+0xb1/0x110
[   13.438456]  do_syscall_64+0x40/0xb0
[   13.438462]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[   13.438467] RIP: 0033:0x7fafab02e9b9
[   13.438471] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 
48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 
05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a7 54 0c 00 f7 d8 64 89 01 48
[   13.438475] RSP: 002b:00007ffe03c50528 EFLAGS: 00000246 ORIG_RAX: 
0000000000000139
[   13.438480] RAX: ffffffffffffffda RBX: 00005596198bfc00 RCX: 
00007fafab02e9b9
[   13.438483] RDX: 0000000000000000 RSI: 00007fafab1b9e2d RDI: 
0000000000000012
[   13.438486] RBP: 0000000000020000 R08: 0000000000000000 R09: 
00005596198baaa0
[   13.438489] R10: 0000000000000012 R11: 0000000000000246 R12: 
00007fafab1b9e2d
[   13.438492] R13: 0000000000000000 R14: 00005596198bab20 R15: 
00005596198bfc00
[   13.438497] Modules linked in: tpm_tis(E+) tpm_tis_core(E) tpm(E) 
asus_atk0110(E) rng_core(E) fjes(E-) evdev(E) acpi_cpufreq(E-) loop(E+) 
firewire_sbp2(E) msr(E) parport_pc(E) ppdev(E) lp(E) parport(E) fuse(E) 
configfs(E) sunrpc(E) ip_tables(E) x_tables(E) autofs4(E) ext4(E) 
crc16(E) mbcache(E) jbd2(E) btrfs(E) blake2b_generic(E) zstd_compress(E) 
efivars(E) raid10(E) raid456(E) async_raid6_recov(E) async_memcpy(E) 
async_pq(E) async_xor(E) async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) 
crc32c_generic(E) raid1(E) raid0(E) multipath(E) linear(E) md_mod(E) 
dm_mod(E) sr_mod(E) sd_mod(E) cdrom(E) t10_pi(E) hid_generic(E) 
usbhid(E) hid(E) amdgpu(E) gpu_sched(E) i2c_algo_bit(E) 
drm_ttm_helper(E) ttm(E) ahci(E) libahci(E) drm_kms_helper(E) 
i2c_i801(E) firewire_ohci(E) ehci_pci(E) uhci_hcd(E) libata(E) 
firewire_core(E) cec(E) crc32c_intel(E) mxm_wmi(E) psmouse(E) 
ehci_hcd(E) crc_itu_t(E) scsi_mod(E) i2c_smbus(E) lpc_ich(E) sky2(E) 
usbcore(E) drm(E) mfd_core(E) wmi(E) button(E)
[   13.438584] CR2: ffff9e8141157000
[   13.438587] ---[ end trace 25e12430fd1718f9 ]---
[   13.438589] RIP: 0010:iowrite8+0x9/0x50
[   13.438594] Code: 48 c7 c2 e8 6a 10 98 48 89 04 24 e8 ac 5e 3f 00 0f 
0b 48 8b 04 24 48 83 c4 08 c3 66 0f 1f 44 00 00 48 81 fe ff ff 03 00 76 
04 <40> 88 3e c3 48 81 fe 00 00 01 00 76 07 0f b7 d6 89 f8 ee c3 8b 05
[   13.438597] RSP: 0018:ffff9e8040f23b10 EFLAGS: 00010292
[   13.438601] RAX: ffffffffc15440b0 RBX: ffff9e8040f23b48 RCX: 
ffff9e8040f23b47
[   13.438604] RDX: 0000000000000001 RSI: ffff9e8141157000 RDI: 
0000000000000020
[   13.438606] RBP: ffff9e8040f23b48 R08: ffff88f2108bf018 R09: 
ffff88f20104b290
[   13.438609] R10: 0000000000000002 R11: 0000000000000004 R12: 
ffff88f2108bf018
[   13.438612] R13: 00000000fffff000 R14: ffff88f20104b000 R15: 
00000000ffffffff
[   13.438615] FS:  00007fafaab758c0(0000) GS:ffff88f5adbc0000(0000) 
knlGS:0000000000000000
[   13.438619] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   13.438622] CR2: ffff9e8141157000 CR3: 00000001065ca000 CR4: 
00000000000006e0