Re: [REGRESSION] suspend to ram fails in 6.2-rc1 due to tpm errors

[Jarkko Sakkinen]

On Tue, Jan 10, 2023 at 06:19:48PM +0100, Vlastimil Babka wrote:
> On 1/9/23 17:08, Jason A. Donenfeld wrote:
> > Hi Thorsten,
> > 
> > On Thu, Jan 05, 2023 at 02:59:15PM +0100, Thorsten Leemhuis wrote:
> >> On 29.12.22 05:03, Jason A. Donenfeld wrote:
> >>> On Wed, Dec 28, 2022 at 06:07:25PM -0500, James Bottomley wrote:
> >>>> On Wed, 2022-12-28 at 21:22 +0100, Vlastimil Babka wrote:
> >>>>> Ugh, while the problem [1] was fixed in 6.1, it's now happening again
> >>>>> on the T460 with 6.2-rc1. Except I didn't see any oops message or
> >>>>> "tpm_try_transmit" error this time. The first indication of a problem
> >>>>> is this during a resume from suspend to ram:
> >>>>>
> >>>>> tpm tpm0: A TPM error (28) occurred continue selftest
> >>>>>
> >>>>> and then periodically 
> >>>>>
> >>>>> tpm tpm0: A TPM error (28) occurred attempting get random
> >>>>
> >>>> That's a TPM 1.2 error which means the TPM failed the selftest.  The
> >>>> original problem was reported against TPM 2.0  because of a missing
> >>>> try_get_ops().
> >>>
> >>> No, I'm pretty sure the original bug, which was fixed by "char: tpm:
> >>> Protect tpm_pm_suspend with locks" regards 1.2 as well, especially
> >>> considering it's the same hardware from Vlastimil causing this. I also
> >>> recall seeing this in 1.2 when I ran this with the TPM emulator. So
> >>> that's not correct.
> >>
> >> James, are you or some other TPM developer looking into this? Or is this
> >> deadlocked now? And if so: how can we get this unstuck to get this
> >> regression solved?
> >>
> >> Side note: I wonder if the problem that Johannes reported yesterday in
> >> this thread (
> >> https://lore.kernel.org/all/Y7VCcgHUC6JtnO2b@xxxxxxxxx/
> >> ) is related or something else, as it seems his issue happens with 6.1,
> >> while Vlastimil's problems should be fixed there. Or am I missing something?
> > 
> > So, this is now in rc3:
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=1382999aa0548a171a272ca817f6c38e797c458c
> > 
> > That should help avoid the worst of the issue -- laptop not sleeping.
> > But the race or whatever it is still does exist. So you might want to
> > keep this in your tracker to periodically nudge the TPM folks about it.
> 
> Heh, booted rc3 and managed to hit it on very first suspend to ram attempt:
> 
> tpm tpm0: A TPM error (28) occurred continue selftest
> 
> But thanks to the patch, the next suspend worked:
> 
> [  236.598900] tpm tpm0: Error (28) sending savestate before suspend
> [  236.598915] tpm_tis 00:08: Ignoring error 28 while suspending
> 
> and on resume again:
> 
> [  238.196645] tpm tpm0: A TPM error (28) occurred continue selftest
> 
> and indeed now I keep getting (as expected)
> 
> [  399.671077] tpm tpm0: A TPM error (28) occurred attempting get random
> 
> So hopefully somebody will look into the root cause at some point.

I would start by using tpm1_auto_startup() also here, and see
the effect:

https://lore.kernel.org/linux-integrity/Y8U1QxA4GYvPWDky@xxxxxxxxxx/

BR, Jarkko