Re: [tpmdd-devel] tpm device not showing up in /dev anymore

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Wed Sep 06 17, Jarkko Sakkinen wrote:
>On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville wrote:
>> Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>> > On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>> > > > Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>> > > > > Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a ecrit :
>> > > > >>> An idea how to troubleshoot this?
>> > > > >> Can you run git bisect on the changes between 4.11 and 4.12, so that
>> > > > >> we find the offending commit? It is probably sufficient to limit the
>> > > > >> search to commits that touch something in drivers/char/tpm.
>> > > > >
>> > > > > I'll try and keep you posted.
>> > > >
>> > > > OK I've been able to bisect the problem and the bad commit is:
>> > > >
>> > > > e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first bad commit
>> > > > commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>> > > > Author: Jerry Snitselaar <jsnitsel@redhat.com>
>> > > > Date:   Mon Mar 27 08:46:04 2017 -0700
>> > > >
>> > > >      tpm_tis: convert to using locality callbacks
>> > > >
>> > > >      This patch converts tpm_tis to use of the new tpm class ops
>> > > >      request_locality, and relinquish_locality.
>> > > >
>> > > >      With the move to using the callbacks, release_locality is
>> > > > changed so
>> > > >      that we now release the locality even if there is no
>> > > > request pending.
>> > > >
>> > > >      This required some changes to the tpm_tis_core_init code path to
>> > > >      make sure locality is requested when needed:
>> > > >
>> > > >        - tpm2_probe code path will end up calling
>> > > > request/release through
>> > > >          callbacks, so request_locality prior to tpm2_probe not needed.
>> > > >
>> > > >        - probe_itpm makes calls to tpm_tis_send_data which no
>> > > > longer calls
>> > > >          request_locality, so add request_locality prior to
>> > > > tpm_tis_send_data
>> > > >          calls. Also drop release_locality call in middleof
>> > > > probe_itpm, and
>> > > >          keep locality until release_locality called at end of
>> > > > probe_itpm.
>> > > >
>> > > >      Cc: Peter Huewe <peterhuewe@gmx.de>
>> > > >      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>> > > >      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>> > > >      Cc: Marcel Selhorst <tpmdd@selhorst.net>
>> > > >      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>> > > >      Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>> > > >      Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>> > > >      Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>> > > >
>> > > > :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>> > > > 72f21b446e45ea1003de75902b0553deb99157fd M    drivers
>> > > >
>> > >
>> > > I've looked again at the code in question, but could not find
>> > > anything that is obviously wrong there. Locality is now
>> > > requested/released at slightly different points in the process than
>> > > before, but that's it. It does not seem to cause problems with the
>> > > majority of TPMs, since you are the first to report any, so maybe it
>> > > is a quirk that only affects this device.
>> > >
>> > > Perhaps Jerry can help, since this is his change?
>> > >
>> > > Alexander
>> >
>> > Getting some caffeine in me, and starting to take a look. Adding
>> > Jarkko as well since this might involve the general locality changes.
>> >
>> > Laurent, if I send you a patch with some debugging code added, would
>> > you be able to run it on that system? I wasn't running into issues
>> > on the system I had with a 1.2 device, but I no longer have access
>> > to it. I'll see if I can find one in our labs and reproduce it there.
>>
>> Yes I should be able to do that
>
>Any findings?
>
>/Jarkko

Okay, finally getting back to this. Looking at the code it isn't clear to me
why the change is causing this. So while I stare at this some more Laurent
could you reproduce it with this patch so I can see what the status and
access registers look like? Does anyone else on here happen to have a Sinosun
tpm device? The systems I have access to with TPM1.2 devices don't have this
issue.

--8<--

diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
index fdde971bc810..7d60a7e4b50a 100644
--- a/drivers/char/tpm/tpm_tis_core.c
+++ b/drivers/char/tpm/tpm_tis_core.c
@@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct tpm_chip *chip, const u8 *buf, size_t len)
 	int rc, status, burstcnt;
 	size_t count = 0;
 	bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
+	u8 access;
 
 	status = tpm_tis_status(chip);
 	if ((status & TPM_STS_COMMAND_READY) == 0) {
@@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip, const u8 *buf, size_t len)
 		}
 		status = tpm_tis_status(chip);
 		if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
+			rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), &access);
+			if (rc < 0)
+				dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: read failure TPM_ACCESS(%d)\n", priv->locality);
+			else
+				dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: locality: %d status: %x access: %x\n", priv->locality, status, access);
 			rc = -EIO;
 			goto out_err;
 		}
@@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip, const u8 *buf, size_t len)
 	}
 	status = tpm_tis_status(chip);
 	if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
+		rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), &access);
+		if (rc < 0)
+			dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read failure TPM_ACCESS(%d)\n", priv->locality);
+		else
+			dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: locality: %d status: %x access: %x\n", priv->locality, status, access);
 		rc = -EIO;
 		goto out_err;
 	}
-- 
2.15.0.rc0

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: multipart/mixed; boundary="------------A9FE3B2D3EA7C477C0DA4586", Size: 6419 bytes --]

Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
> On Wed Sep 06 17, Jarkko Sakkinen wrote:
>> On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville wrote:
>>> Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>> > On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>> > > > Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>> > > > > Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a ecrit :
>>> > > > >>> An idea how to troubleshoot this?
>>> > > > >> Can you run git bisect on the changes between 4.11 and 
>>> 4.12, so that
>>> > > > >> we find the offending commit? It is probably sufficient to 
>>> limit the
>>> > > > >> search to commits that touch something in drivers/char/tpm.
>>> > > > >
>>> > > > > I'll try and keep you posted.
>>> > > >
>>> > > > OK I've been able to bisect the problem and the bad commit is:
>>> > > >
>>> > > > e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first bad commit
>>> > > > commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>> > > > Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>> > > > Date:   Mon Mar 27 08:46:04 2017 -0700
>>> > > >
>>> > > >      tpm_tis: convert to using locality callbacks
>>> > > >
>>> > > >      This patch converts tpm_tis to use of the new tpm class ops
>>> > > >      request_locality, and relinquish_locality.
>>> > > >
>>> > > >      With the move to using the callbacks, release_locality is
>>> > > > changed so
>>> > > >      that we now release the locality even if there is no
>>> > > > request pending.
>>> > > >
>>> > > >      This required some changes to the tpm_tis_core_init code 
>>> path to
>>> > > >      make sure locality is requested when needed:
>>> > > >
>>> > > >        - tpm2_probe code path will end up calling
>>> > > > request/release through
>>> > > >          callbacks, so request_locality prior to tpm2_probe 
>>> not needed.
>>> > > >
>>> > > >        - probe_itpm makes calls to tpm_tis_send_data which no
>>> > > > longer calls
>>> > > >          request_locality, so add request_locality prior to
>>> > > > tpm_tis_send_data
>>> > > >          calls. Also drop release_locality call in middleof
>>> > > > probe_itpm, and
>>> > > >          keep locality until release_locality called at end of
>>> > > > probe_itpm.
>>> > > >
>>> > > >      Cc: Peter Huewe <peterhuewe@gmx.de>
>>> > > >      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>> > > >      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>>> > > >      Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>> > > >      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>>> > > >      Reviewed-by: Jarkko Sakkinen 
>>> <jarkko.sakkinen@linux.intel.com>
>>> > > >      Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>> > > >      Signed-off-by: Jarkko Sakkinen 
>>> <jarkko.sakkinen@linux.intel.com>
>>> > > >
>>> > > > :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>> > > > 72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>> > > >
>>> > >
>>> > > I've looked again at the code in question, but could not find
>>> > > anything that is obviously wrong there. Locality is now
>>> > > requested/released at slightly different points in the process than
>>> > > before, but that's it. It does not seem to cause problems with the
>>> > > majority of TPMs, since you are the first to report any, so 
>>> maybe it
>>> > > is a quirk that only affects this device.
>>> > >
>>> > > Perhaps Jerry can help, since this is his change?
>>> > >
>>> > > Alexander
>>> >
>>> > Getting some caffeine in me, and starting to take a look. Adding
>>> > Jarkko as well since this might involve the general locality changes.
>>> >
>>> > Laurent, if I send you a patch with some debugging code added, would
>>> > you be able to run it on that system? I wasn't running into issues
>>> > on the system I had with a 1.2 device, but I no longer have access
>>> > to it. I'll see if I can find one in our labs and reproduce it there.
>>>
>>> Yes I should be able to do that
>>
>> Any findings?
>>
>> /Jarkko
>
> Okay, finally getting back to this. Looking at the code it isn't clear 
> to me
> why the change is causing this. So while I stare at this some more 
> Laurent
> could you reproduce it with this patch so I can see what the status and
> access registers look like? Does anyone else on here happen to have a 
> Sinosun
> tpm device? The systems I have access to with TPM1.2 devices don't 
> have this
> issue.
>
> --8<--
>
> diff --git a/drivers/char/tpm/tpm_tis_core.c 
> b/drivers/char/tpm/tpm_tis_core.c
> index fdde971bc810..7d60a7e4b50a 100644
> --- a/drivers/char/tpm/tpm_tis_core.c
> +++ b/drivers/char/tpm/tpm_tis_core.c
> @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct tpm_chip 
> *chip, const u8 *buf, size_t len)
>     int rc, status, burstcnt;
>     size_t count = 0;
>     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
> +    u8 access;
>
>     status = tpm_tis_status(chip);
>     if ((status & TPM_STS_COMMAND_READY) == 0) {
> @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct tpm_chip 
> *chip, const u8 *buf, size_t len)
>         }
>         status = tpm_tis_status(chip);
>         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
> +            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), 
> &access);
> +            if (rc < 0)
> +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: read 
> failure TPM_ACCESS(%d)\n", priv->locality);
> +            else
> +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: 
> locality: %d status: %x access: %x\n", priv->locality, status, access);
>             rc = -EIO;
>             goto out_err;
>         }
> @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct tpm_chip 
> *chip, const u8 *buf, size_t len)
>     }
>     status = tpm_tis_status(chip);
>     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
> +        rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), &access);
> +        if (rc < 0)
> +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read 
> failure TPM_ACCESS(%d)\n", priv->locality);
> +        else
> +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: locality: 
> %d status: %x access: %x\n", priv->locality, status, access);
>         rc = -EIO;
>         goto out_err;
>     }

Please find here the dmesg output of the patched kernel


    [ Part 2, Application/GZIP (Name: "dmesg.txt.gz") 20 KB. ]
    [ Unable to print this part. ]

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
> Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
> > On Wed Sep 06 17, Jarkko Sakkinen wrote:
> > > On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville wrote:
> > > > Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
> > > > > On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
> > > > > > > Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
> > > > > > > > Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a ecrit :
> > > > > > > >>> An idea how to troubleshoot this?
> > > > > > > >> Can you run git bisect on the changes between 4.11 and
> > > > 4.12, so that
> > > > > > > >> we find the offending commit? It is probably sufficient
> > > > to limit the
> > > > > > > >> search to commits that touch something in drivers/char/tpm.
> > > > > > > >
> > > > > > > > I'll try and keep you posted.
> > > > > > >
> > > > > > > OK I've been able to bisect the problem and the bad commit is:
> > > > > > >
> > > > > > > e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first bad commit
> > > > > > > commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
> > > > > > > Author: Jerry Snitselaar <jsnitsel@redhat.com>
> > > > > > > Date:   Mon Mar 27 08:46:04 2017 -0700
> > > > > > >
> > > > > > >      tpm_tis: convert to using locality callbacks
> > > > > > >
> > > > > > >      This patch converts tpm_tis to use of the new tpm class ops
> > > > > > >      request_locality, and relinquish_locality.
> > > > > > >
> > > > > > >      With the move to using the callbacks, release_locality is
> > > > > > > changed so
> > > > > > >      that we now release the locality even if there is no
> > > > > > > request pending.
> > > > > > >
> > > > > > >      This required some changes to the tpm_tis_core_init
> > > > code path to
> > > > > > >      make sure locality is requested when needed:
> > > > > > >
> > > > > > >        - tpm2_probe code path will end up calling
> > > > > > > request/release through
> > > > > > >          callbacks, so request_locality prior to
> > > > tpm2_probe not needed.
> > > > > > >
> > > > > > >        - probe_itpm makes calls to tpm_tis_send_data which no
> > > > > > > longer calls
> > > > > > >          request_locality, so add request_locality prior to
> > > > > > > tpm_tis_send_data
> > > > > > >          calls. Also drop release_locality call in middleof
> > > > > > > probe_itpm, and
> > > > > > >          keep locality until release_locality called at end of
> > > > > > > probe_itpm.
> > > > > > >
> > > > > > >      Cc: Peter Huewe <peterhuewe@gmx.de>
> > > > > > >      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
> > > > > > >      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
> > > > > > >      Cc: Marcel Selhorst <tpmdd@selhorst.net>
> > > > > > >      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
> > > > > > >      Reviewed-by: Jarkko Sakkinen
> > > > <jarkko.sakkinen@linux.intel.com>
> > > > > > >      Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
> > > > > > >      Signed-off-by: Jarkko Sakkinen
> > > > <jarkko.sakkinen@linux.intel.com>
> > > > > > >
> > > > > > > :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
> > > > > > > 72f21b446e45ea1003de75902b0553deb99157fd M drivers
> > > > > > >
> > > > > >
> > > > > > I've looked again at the code in question, but could not find
> > > > > > anything that is obviously wrong there. Locality is now
> > > > > > requested/released at slightly different points in the process than
> > > > > > before, but that's it. It does not seem to cause problems with the
> > > > > > majority of TPMs, since you are the first to report any, so
> > > > maybe it
> > > > > > is a quirk that only affects this device.
> > > > > >
> > > > > > Perhaps Jerry can help, since this is his change?
> > > > > >
> > > > > > Alexander
> > > > >
> > > > > Getting some caffeine in me, and starting to take a look. Adding
> > > > > Jarkko as well since this might involve the general locality changes.
> > > > >
> > > > > Laurent, if I send you a patch with some debugging code added, would
> > > > > you be able to run it on that system? I wasn't running into issues
> > > > > on the system I had with a 1.2 device, but I no longer have access
> > > > > to it. I'll see if I can find one in our labs and reproduce it there.
> > > > 
> > > > Yes I should be able to do that
> > > 
> > > Any findings?
> > > 
> > > /Jarkko
> > 
> > Okay, finally getting back to this. Looking at the code it isn't clear
> > to me
> > why the change is causing this. So while I stare at this some more
> > Laurent
> > could you reproduce it with this patch so I can see what the status and
> > access registers look like? Does anyone else on here happen to have a
> > Sinosun
> > tpm device? The systems I have access to with TPM1.2 devices don't have
> > this
> > issue.
> > 
> > --8<--
> > 
> > diff --git a/drivers/char/tpm/tpm_tis_core.c
> > b/drivers/char/tpm/tpm_tis_core.c
> > index fdde971bc810..7d60a7e4b50a 100644
> > --- a/drivers/char/tpm/tpm_tis_core.c
> > +++ b/drivers/char/tpm/tpm_tis_core.c
> > @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
> > const u8 *buf, size_t len)
> >     int rc, status, burstcnt;
> >     size_t count = 0;
> >     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
> > +    u8 access;
> > 
> >     status = tpm_tis_status(chip);
> >     if ((status & TPM_STS_COMMAND_READY) == 0) {
> > @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
> > const u8 *buf, size_t len)
> >         }
> >         status = tpm_tis_status(chip);
> >         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
> > +            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
> > &access);
> > +            if (rc < 0)
> > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: read
> > failure TPM_ACCESS(%d)\n", priv->locality);
> > +            else
> > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
> > locality: %d status: %x access: %x\n", priv->locality, status, access);
> >             rc = -EIO;
> >             goto out_err;
> >         }
> > @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
> > const u8 *buf, size_t len)
> >     }
> >     status = tpm_tis_status(chip);
> >     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
> > +        rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), &access);
> > +        if (rc < 0)
> > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
> > failure TPM_ACCESS(%d)\n", priv->locality);
> > +        else
> > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: locality:
> > %d status: %x access: %x\n", priv->locality, status, access);
> >         rc = -EIO;
> >         goto out_err;
> >     }
> 
> Please find here the dmesg output of the patched kernel

At least 0xff is corrupted value in senseful way. CPU fills the read
with ones for example for unaligned bus read. See table 19 in PC client
spec. This can happen when you do unaligned read for example.

Maybe TPM is unreachable i.e. powered off. Bit busy with stuff ATM but
would probably make sense to compare that 0x81 to table 18 in the same
spec.

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Mon Oct 23 17, Jarkko Sakkinen wrote:
>On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
>> Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>> > On Wed Sep 06 17, Jarkko Sakkinen wrote:
>> > > On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville wrote:
>> > > > Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>> > > > > On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>> > > > > > > Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>> > > > > > > > Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a ecrit :
>> > > > > > > >>> An idea how to troubleshoot this?
>> > > > > > > >> Can you run git bisect on the changes between 4.11 and
>> > > > 4.12, so that
>> > > > > > > >> we find the offending commit? It is probably sufficient
>> > > > to limit the
>> > > > > > > >> search to commits that touch something in drivers/char/tpm.
>> > > > > > > >
>> > > > > > > > I'll try and keep you posted.
>> > > > > > >
>> > > > > > > OK I've been able to bisect the problem and the bad commit is:
>> > > > > > >
>> > > > > > > e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first bad commit
>> > > > > > > commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>> > > > > > > Author: Jerry Snitselaar <jsnitsel@redhat.com>
>> > > > > > > Date:   Mon Mar 27 08:46:04 2017 -0700
>> > > > > > >
>> > > > > > >      tpm_tis: convert to using locality callbacks
>> > > > > > >
>> > > > > > >      This patch converts tpm_tis to use of the new tpm class ops
>> > > > > > >      request_locality, and relinquish_locality.
>> > > > > > >
>> > > > > > >      With the move to using the callbacks, release_locality is
>> > > > > > > changed so
>> > > > > > >      that we now release the locality even if there is no
>> > > > > > > request pending.
>> > > > > > >
>> > > > > > >      This required some changes to the tpm_tis_core_init
>> > > > code path to
>> > > > > > >      make sure locality is requested when needed:
>> > > > > > >
>> > > > > > >        - tpm2_probe code path will end up calling
>> > > > > > > request/release through
>> > > > > > >          callbacks, so request_locality prior to
>> > > > tpm2_probe not needed.
>> > > > > > >
>> > > > > > >        - probe_itpm makes calls to tpm_tis_send_data which no
>> > > > > > > longer calls
>> > > > > > >          request_locality, so add request_locality prior to
>> > > > > > > tpm_tis_send_data
>> > > > > > >          calls. Also drop release_locality call in middleof
>> > > > > > > probe_itpm, and
>> > > > > > >          keep locality until release_locality called at end of
>> > > > > > > probe_itpm.
>> > > > > > >
>> > > > > > >      Cc: Peter Huewe <peterhuewe@gmx.de>
>> > > > > > >      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>> > > > > > >      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>> > > > > > >      Cc: Marcel Selhorst <tpmdd@selhorst.net>
>> > > > > > >      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>> > > > > > >      Reviewed-by: Jarkko Sakkinen
>> > > > <jarkko.sakkinen@linux.intel.com>
>> > > > > > >      Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>> > > > > > >      Signed-off-by: Jarkko Sakkinen
>> > > > <jarkko.sakkinen@linux.intel.com>
>> > > > > > >
>> > > > > > > :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>> > > > > > > 72f21b446e45ea1003de75902b0553deb99157fd M drivers
>> > > > > > >
>> > > > > >
>> > > > > > I've looked again at the code in question, but could not find
>> > > > > > anything that is obviously wrong there. Locality is now
>> > > > > > requested/released at slightly different points in the process than
>> > > > > > before, but that's it. It does not seem to cause problems with the
>> > > > > > majority of TPMs, since you are the first to report any, so
>> > > > maybe it
>> > > > > > is a quirk that only affects this device.
>> > > > > >
>> > > > > > Perhaps Jerry can help, since this is his change?
>> > > > > >
>> > > > > > Alexander
>> > > > >
>> > > > > Getting some caffeine in me, and starting to take a look. Adding
>> > > > > Jarkko as well since this might involve the general locality changes.
>> > > > >
>> > > > > Laurent, if I send you a patch with some debugging code added, would
>> > > > > you be able to run it on that system? I wasn't running into issues
>> > > > > on the system I had with a 1.2 device, but I no longer have access
>> > > > > to it. I'll see if I can find one in our labs and reproduce it there.
>> > > >
>> > > > Yes I should be able to do that
>> > >
>> > > Any findings?
>> > >
>> > > /Jarkko
>> >
>> > Okay, finally getting back to this. Looking at the code it isn't clear
>> > to me
>> > why the change is causing this. So while I stare at this some more
>> > Laurent
>> > could you reproduce it with this patch so I can see what the status and
>> > access registers look like? Does anyone else on here happen to have a
>> > Sinosun
>> > tpm device? The systems I have access to with TPM1.2 devices don't have
>> > this
>> > issue.
>> >
>> > --8<--
>> >
>> > diff --git a/drivers/char/tpm/tpm_tis_core.c
>> > b/drivers/char/tpm/tpm_tis_core.c
>> > index fdde971bc810..7d60a7e4b50a 100644
>> > --- a/drivers/char/tpm/tpm_tis_core.c
>> > +++ b/drivers/char/tpm/tpm_tis_core.c
>> > @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>> > const u8 *buf, size_t len)
>> >     int rc, status, burstcnt;
>> >     size_t count = 0;
>> >     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>> > +    u8 access;
>> >
>> >     status = tpm_tis_status(chip);
>> >     if ((status & TPM_STS_COMMAND_READY) == 0) {
>> > @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>> > const u8 *buf, size_t len)
>> >         }
>> >         status = tpm_tis_status(chip);
>> >         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>> > +            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
>> > &access);
>> > +            if (rc < 0)
>> > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: read
>> > failure TPM_ACCESS(%d)\n", priv->locality);
>> > +            else
>> > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
>> > locality: %d status: %x access: %x\n", priv->locality, status, access);
>> >             rc = -EIO;
>> >             goto out_err;
>> >         }
>> > @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>> > const u8 *buf, size_t len)
>> >     }
>> >     status = tpm_tis_status(chip);
>> >     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>> > +        rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), &access);
>> > +        if (rc < 0)
>> > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
>> > failure TPM_ACCESS(%d)\n", priv->locality);
>> > +        else
>> > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: locality:
>> > %d status: %x access: %x\n", priv->locality, status, access);
>> >         rc = -EIO;
>> >         goto out_err;
>> >     }
>>
>> Please find here the dmesg output of the patched kernel
>
>At least 0xff is corrupted value in senseful way. CPU fills the read
>with ones for example for unaligned bus read. See table 19 in PC client
>spec. This can happen when you do unaligned read for example.
>
>Maybe TPM is unreachable i.e. powered off. Bit busy with stuff ATM but
>would probably make sense to compare that 0x81 to table 18 in the same
>spec.
>
>/Jarkko

0x81 is saying the access register status is valid, and the locality
is not active. That first bit means A Dynamic OS has not been previously
established on the platform. Normally we would see 0xa1, which would
mean valid register status, and the locality is active.

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

Le 23/10/17 a 15:45, Jerry Snitselaar a ecrit :
> On Mon Oct 23 17, Jarkko Sakkinen wrote:
>> On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
>>> Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>>> > On Wed Sep 06 17, Jarkko Sakkinen wrote:
>>> > > On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville wrote:
>>> > > > Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>> > > > > On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>> > > > > > > Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>> > > > > > > > Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a 
>>> ecrit :
>>> > > > > > > >>> An idea how to troubleshoot this?
>>> > > > > > > >> Can you run git bisect on the changes between 4.11 and
>>> > > > 4.12, so that
>>> > > > > > > >> we find the offending commit? It is probably sufficient
>>> > > > to limit the
>>> > > > > > > >> search to commits that touch something in 
>>> drivers/char/tpm.
>>> > > > > > > >
>>> > > > > > > > I'll try and keep you posted.
>>> > > > > > >
>>> > > > > > > OK I've been able to bisect the problem and the bad 
>>> commit is:
>>> > > > > > >
>>> > > > > > > e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first 
>>> bad commit
>>> > > > > > > commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>> > > > > > > Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>> > > > > > > Date:   Mon Mar 27 08:46:04 2017 -0700
>>> > > > > > >
>>> > > > > > >      tpm_tis: convert to using locality callbacks
>>> > > > > > >
>>> > > > > > >      This patch converts tpm_tis to use of the new tpm 
>>> class ops
>>> > > > > > >      request_locality, and relinquish_locality.
>>> > > > > > >
>>> > > > > > >      With the move to using the callbacks, 
>>> release_locality is
>>> > > > > > > changed so
>>> > > > > > >      that we now release the locality even if there is no
>>> > > > > > > request pending.
>>> > > > > > >
>>> > > > > > >      This required some changes to the tpm_tis_core_init
>>> > > > code path to
>>> > > > > > >      make sure locality is requested when needed:
>>> > > > > > >
>>> > > > > > >        - tpm2_probe code path will end up calling
>>> > > > > > > request/release through
>>> > > > > > >          callbacks, so request_locality prior to
>>> > > > tpm2_probe not needed.
>>> > > > > > >
>>> > > > > > >        - probe_itpm makes calls to tpm_tis_send_data 
>>> which no
>>> > > > > > > longer calls
>>> > > > > > >          request_locality, so add request_locality prior to
>>> > > > > > > tpm_tis_send_data
>>> > > > > > >          calls. Also drop release_locality call in middleof
>>> > > > > > > probe_itpm, and
>>> > > > > > >          keep locality until release_locality called at 
>>> end of
>>> > > > > > > probe_itpm.
>>> > > > > > >
>>> > > > > > >      Cc: Peter Huewe <peterhuewe@gmx.de>
>>> > > > > > >      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>> > > > > > >      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>>> > > > > > >      Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>> > > > > > >      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>>> > > > > > >      Reviewed-by: Jarkko Sakkinen
>>> > > > <jarkko.sakkinen@linux.intel.com>
>>> > > > > > >      Tested-by: Jarkko Sakkinen 
>>> <jarkko.sakkinen@linux.intel.com>
>>> > > > > > >      Signed-off-by: Jarkko Sakkinen
>>> > > > <jarkko.sakkinen@linux.intel.com>
>>> > > > > > >
>>> > > > > > > :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>> > > > > > > 72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>> > > > > > >
>>> > > > > >
>>> > > > > > I've looked again at the code in question, but could not find
>>> > > > > > anything that is obviously wrong there. Locality is now
>>> > > > > > requested/released at slightly different points in the 
>>> process than
>>> > > > > > before, but that's it. It does not seem to cause problems 
>>> with the
>>> > > > > > majority of TPMs, since you are the first to report any, so
>>> > > > maybe it
>>> > > > > > is a quirk that only affects this device.
>>> > > > > >
>>> > > > > > Perhaps Jerry can help, since this is his change?
>>> > > > > >
>>> > > > > > Alexander
>>> > > > >
>>> > > > > Getting some caffeine in me, and starting to take a look. 
>>> Adding
>>> > > > > Jarkko as well since this might involve the general locality 
>>> changes.
>>> > > > >
>>> > > > > Laurent, if I send you a patch with some debugging code 
>>> added, would
>>> > > > > you be able to run it on that system? I wasn't running into 
>>> issues
>>> > > > > on the system I had with a 1.2 device, but I no longer have 
>>> access
>>> > > > > to it. I'll see if I can find one in our labs and reproduce 
>>> it there.
>>> > > >
>>> > > > Yes I should be able to do that
>>> > >
>>> > > Any findings?
>>> > >
>>> > > /Jarkko
>>> >
>>> > Okay, finally getting back to this. Looking at the code it isn't 
>>> clear
>>> > to me
>>> > why the change is causing this. So while I stare at this some more
>>> > Laurent
>>> > could you reproduce it with this patch so I can see what the 
>>> status and
>>> > access registers look like? Does anyone else on here happen to have a
>>> > Sinosun
>>> > tpm device? The systems I have access to with TPM1.2 devices don't 
>>> have
>>> > this
>>> > issue.
>>> >
>>> > --8<--
>>> >
>>> > diff --git a/drivers/char/tpm/tpm_tis_core.c
>>> > b/drivers/char/tpm/tpm_tis_core.c
>>> > index fdde971bc810..7d60a7e4b50a 100644
>>> > --- a/drivers/char/tpm/tpm_tis_core.c
>>> > +++ b/drivers/char/tpm/tpm_tis_core.c
>>> > @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct tpm_chip 
>>> *chip,
>>> > const u8 *buf, size_t len)
>>> >     int rc, status, burstcnt;
>>> >     size_t count = 0;
>>> >     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>>> > +    u8 access;
>>> >
>>> >     status = tpm_tis_status(chip);
>>> >     if ((status & TPM_STS_COMMAND_READY) == 0) {
>>> > @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct tpm_chip 
>>> *chip,
>>> > const u8 *buf, size_t len)
>>> >         }
>>> >         status = tpm_tis_status(chip);
>>> >         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>>> > +            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
>>> > &access);
>>> > +            if (rc < 0)
>>> > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: read
>>> > failure TPM_ACCESS(%d)\n", priv->locality);
>>> > +            else
>>> > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
>>> > locality: %d status: %x access: %x\n", priv->locality, status, 
>>> access);
>>> >             rc = -EIO;
>>> >             goto out_err;
>>> >         }
>>> > @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct tpm_chip 
>>> *chip,
>>> > const u8 *buf, size_t len)
>>> >     }
>>> >     status = tpm_tis_status(chip);
>>> >     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>>> > +        rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), 
>>> &access);
>>> > +        if (rc < 0)
>>> > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
>>> > failure TPM_ACCESS(%d)\n", priv->locality);
>>> > +        else
>>> > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: 
>>> locality:
>>> > %d status: %x access: %x\n", priv->locality, status, access);
>>> >         rc = -EIO;
>>> >         goto out_err;
>>> >     }
>>>
>>> Please find here the dmesg output of the patched kernel
>>
>> At least 0xff is corrupted value in senseful way. CPU fills the read
>> with ones for example for unaligned bus read. See table 19 in PC client
>> spec. This can happen when you do unaligned read for example.
>>
>> Maybe TPM is unreachable i.e. powered off. Bit busy with stuff ATM but
>> would probably make sense to compare that 0x81 to table 18 in the same
>> spec.
>>
>> /Jarkko
>
> 0x81 is saying the access register status is valid, and the locality
> is not active. That first bit means A Dynamic OS has not been previously
> established on the platform. Normally we would see 0xa1, which would
> mean valid register status, and the locality is active.

FTR, the ownership has been claimed using tpm-tools on linux, but I 
configured windows 10 (dual-boot) on the machine to do its magic to 
recognize the tpm chip by entering the owner password manually there, 
not sure if that matters.

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
> On Mon Oct 23 17, Jarkko Sakkinen wrote:
> > On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
> > > Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
> > > > On Wed Sep 06 17, Jarkko Sakkinen wrote:
> > > > > On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville wrote:
> > > > > > Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
> > > > > > > On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
> > > > > > > > > Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
> > > > > > > > > > Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a ecrit :
> > > > > > > > > >>> An idea how to troubleshoot this?
> > > > > > > > > >> Can you run git bisect on the changes between 4.11 and
> > > > > > 4.12, so that
> > > > > > > > > >> we find the offending commit? It is probably sufficient
> > > > > > to limit the
> > > > > > > > > >> search to commits that touch something in drivers/char/tpm.
> > > > > > > > > >
> > > > > > > > > > I'll try and keep you posted.
> > > > > > > > >
> > > > > > > > > OK I've been able to bisect the problem and the bad commit is:
> > > > > > > > >
> > > > > > > > > e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first bad commit
> > > > > > > > > commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
> > > > > > > > > Author: Jerry Snitselaar <jsnitsel@redhat.com>
> > > > > > > > > Date:   Mon Mar 27 08:46:04 2017 -0700
> > > > > > > > >
> > > > > > > > >      tpm_tis: convert to using locality callbacks
> > > > > > > > >
> > > > > > > > >      This patch converts tpm_tis to use of the new tpm class ops
> > > > > > > > >      request_locality, and relinquish_locality.
> > > > > > > > >
> > > > > > > > >      With the move to using the callbacks, release_locality is
> > > > > > > > > changed so
> > > > > > > > >      that we now release the locality even if there is no
> > > > > > > > > request pending.
> > > > > > > > >
> > > > > > > > >      This required some changes to the tpm_tis_core_init
> > > > > > code path to
> > > > > > > > >      make sure locality is requested when needed:
> > > > > > > > >
> > > > > > > > >        - tpm2_probe code path will end up calling
> > > > > > > > > request/release through
> > > > > > > > >          callbacks, so request_locality prior to
> > > > > > tpm2_probe not needed.
> > > > > > > > >
> > > > > > > > >        - probe_itpm makes calls to tpm_tis_send_data which no
> > > > > > > > > longer calls
> > > > > > > > >          request_locality, so add request_locality prior to
> > > > > > > > > tpm_tis_send_data
> > > > > > > > >          calls. Also drop release_locality call in middleof
> > > > > > > > > probe_itpm, and
> > > > > > > > >          keep locality until release_locality called at end of
> > > > > > > > > probe_itpm.
> > > > > > > > >
> > > > > > > > >      Cc: Peter Huewe <peterhuewe@gmx.de>
> > > > > > > > >      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
> > > > > > > > >      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
> > > > > > > > >      Cc: Marcel Selhorst <tpmdd@selhorst.net>
> > > > > > > > >      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
> > > > > > > > >      Reviewed-by: Jarkko Sakkinen
> > > > > > <jarkko.sakkinen@linux.intel.com>
> > > > > > > > >      Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
> > > > > > > > >      Signed-off-by: Jarkko Sakkinen
> > > > > > <jarkko.sakkinen@linux.intel.com>
> > > > > > > > >
> > > > > > > > > :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
> > > > > > > > > 72f21b446e45ea1003de75902b0553deb99157fd M drivers
> > > > > > > > >
> > > > > > > >
> > > > > > > > I've looked again at the code in question, but could not find
> > > > > > > > anything that is obviously wrong there. Locality is now
> > > > > > > > requested/released at slightly different points in the process than
> > > > > > > > before, but that's it. It does not seem to cause problems with the
> > > > > > > > majority of TPMs, since you are the first to report any, so
> > > > > > maybe it
> > > > > > > > is a quirk that only affects this device.
> > > > > > > >
> > > > > > > > Perhaps Jerry can help, since this is his change?
> > > > > > > >
> > > > > > > > Alexander
> > > > > > >
> > > > > > > Getting some caffeine in me, and starting to take a look. Adding
> > > > > > > Jarkko as well since this might involve the general locality changes.
> > > > > > >
> > > > > > > Laurent, if I send you a patch with some debugging code added, would
> > > > > > > you be able to run it on that system? I wasn't running into issues
> > > > > > > on the system I had with a 1.2 device, but I no longer have access
> > > > > > > to it. I'll see if I can find one in our labs and reproduce it there.
> > > > > >
> > > > > > Yes I should be able to do that
> > > > >
> > > > > Any findings?
> > > > >
> > > > > /Jarkko
> > > >
> > > > Okay, finally getting back to this. Looking at the code it isn't clear
> > > > to me
> > > > why the change is causing this. So while I stare at this some more
> > > > Laurent
> > > > could you reproduce it with this patch so I can see what the status and
> > > > access registers look like? Does anyone else on here happen to have a
> > > > Sinosun
> > > > tpm device? The systems I have access to with TPM1.2 devices don't have
> > > > this
> > > > issue.
> > > >
> > > > --8<--
> > > >
> > > > diff --git a/drivers/char/tpm/tpm_tis_core.c
> > > > b/drivers/char/tpm/tpm_tis_core.c
> > > > index fdde971bc810..7d60a7e4b50a 100644
> > > > --- a/drivers/char/tpm/tpm_tis_core.c
> > > > +++ b/drivers/char/tpm/tpm_tis_core.c
> > > > @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
> > > > const u8 *buf, size_t len)
> > > >     int rc, status, burstcnt;
> > > >     size_t count = 0;
> > > >     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
> > > > +    u8 access;
> > > >
> > > >     status = tpm_tis_status(chip);
> > > >     if ((status & TPM_STS_COMMAND_READY) == 0) {
> > > > @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
> > > > const u8 *buf, size_t len)
> > > >         }
> > > >         status = tpm_tis_status(chip);
> > > >         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
> > > > +            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
> > > > &access);
> > > > +            if (rc < 0)
> > > > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: read
> > > > failure TPM_ACCESS(%d)\n", priv->locality);
> > > > +            else
> > > > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
> > > > locality: %d status: %x access: %x\n", priv->locality, status, access);
> > > >             rc = -EIO;
> > > >             goto out_err;
> > > >         }
> > > > @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
> > > > const u8 *buf, size_t len)
> > > >     }
> > > >     status = tpm_tis_status(chip);
> > > >     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
> > > > +        rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), &access);
> > > > +        if (rc < 0)
> > > > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
> > > > failure TPM_ACCESS(%d)\n", priv->locality);
> > > > +        else
> > > > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: locality:
> > > > %d status: %x access: %x\n", priv->locality, status, access);
> > > >         rc = -EIO;
> > > >         goto out_err;
> > > >     }
> > > 
> > > Please find here the dmesg output of the patched kernel
> > 
> > At least 0xff is corrupted value in senseful way. CPU fills the read
> > with ones for example for unaligned bus read. See table 19 in PC client
> > spec. This can happen when you do unaligned read for example.
> > 
> > Maybe TPM is unreachable i.e. powered off. Bit busy with stuff ATM but
> > would probably make sense to compare that 0x81 to table 18 in the same
> > spec.
> > 
> > /Jarkko
> 
> 0x81 is saying the access register status is valid, and the locality
> is not active. That first bit means A Dynamic OS has not been previously
> established on the platform. Normally we would see 0xa1, which would
> mean valid register status, and the locality is active.

I think the important thing to note here is that STS has bits set that
should never be set. So we can conclude that TPM might be either

1. Powered off
2. In some transition state?

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Tue Oct 24 17, Jarkko Sakkinen wrote:
>On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
>> On Mon Oct 23 17, Jarkko Sakkinen wrote:
>> > On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
>> > > Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>> > > > On Wed Sep 06 17, Jarkko Sakkinen wrote:
>> > > > > On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville wrote:
>> > > > > > Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>> > > > > > > On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>> > > > > > > > > Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>> > > > > > > > > > Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a ecrit :
>> > > > > > > > > >>> An idea how to troubleshoot this?
>> > > > > > > > > >> Can you run git bisect on the changes between 4.11 and
>> > > > > > 4.12, so that
>> > > > > > > > > >> we find the offending commit? It is probably sufficient
>> > > > > > to limit the
>> > > > > > > > > >> search to commits that touch something in drivers/char/tpm.
>> > > > > > > > > >
>> > > > > > > > > > I'll try and keep you posted.
>> > > > > > > > >
>> > > > > > > > > OK I've been able to bisect the problem and the bad commit is:
>> > > > > > > > >
>> > > > > > > > > e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first bad commit
>> > > > > > > > > commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>> > > > > > > > > Author: Jerry Snitselaar <jsnitsel@redhat.com>
>> > > > > > > > > Date:   Mon Mar 27 08:46:04 2017 -0700
>> > > > > > > > >
>> > > > > > > > >      tpm_tis: convert to using locality callbacks
>> > > > > > > > >
>> > > > > > > > >      This patch converts tpm_tis to use of the new tpm class ops
>> > > > > > > > >      request_locality, and relinquish_locality.
>> > > > > > > > >
>> > > > > > > > >      With the move to using the callbacks, release_locality is
>> > > > > > > > > changed so
>> > > > > > > > >      that we now release the locality even if there is no
>> > > > > > > > > request pending.
>> > > > > > > > >
>> > > > > > > > >      This required some changes to the tpm_tis_core_init
>> > > > > > code path to
>> > > > > > > > >      make sure locality is requested when needed:
>> > > > > > > > >
>> > > > > > > > >        - tpm2_probe code path will end up calling
>> > > > > > > > > request/release through
>> > > > > > > > >          callbacks, so request_locality prior to
>> > > > > > tpm2_probe not needed.
>> > > > > > > > >
>> > > > > > > > >        - probe_itpm makes calls to tpm_tis_send_data which no
>> > > > > > > > > longer calls
>> > > > > > > > >          request_locality, so add request_locality prior to
>> > > > > > > > > tpm_tis_send_data
>> > > > > > > > >          calls. Also drop release_locality call in middleof
>> > > > > > > > > probe_itpm, and
>> > > > > > > > >          keep locality until release_locality called at end of
>> > > > > > > > > probe_itpm.
>> > > > > > > > >
>> > > > > > > > >      Cc: Peter Huewe <peterhuewe@gmx.de>
>> > > > > > > > >      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>> > > > > > > > >      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>> > > > > > > > >      Cc: Marcel Selhorst <tpmdd@selhorst.net>
>> > > > > > > > >      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>> > > > > > > > >      Reviewed-by: Jarkko Sakkinen
>> > > > > > <jarkko.sakkinen@linux.intel.com>
>> > > > > > > > >      Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>> > > > > > > > >      Signed-off-by: Jarkko Sakkinen
>> > > > > > <jarkko.sakkinen@linux.intel.com>
>> > > > > > > > >
>> > > > > > > > > :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>> > > > > > > > > 72f21b446e45ea1003de75902b0553deb99157fd M drivers
>> > > > > > > > >
>> > > > > > > >
>> > > > > > > > I've looked again at the code in question, but could not find
>> > > > > > > > anything that is obviously wrong there. Locality is now
>> > > > > > > > requested/released at slightly different points in the process than
>> > > > > > > > before, but that's it. It does not seem to cause problems with the
>> > > > > > > > majority of TPMs, since you are the first to report any, so
>> > > > > > maybe it
>> > > > > > > > is a quirk that only affects this device.
>> > > > > > > >
>> > > > > > > > Perhaps Jerry can help, since this is his change?
>> > > > > > > >
>> > > > > > > > Alexander
>> > > > > > >
>> > > > > > > Getting some caffeine in me, and starting to take a look. Adding
>> > > > > > > Jarkko as well since this might involve the general locality changes.
>> > > > > > >
>> > > > > > > Laurent, if I send you a patch with some debugging code added, would
>> > > > > > > you be able to run it on that system? I wasn't running into issues
>> > > > > > > on the system I had with a 1.2 device, but I no longer have access
>> > > > > > > to it. I'll see if I can find one in our labs and reproduce it there.
>> > > > > >
>> > > > > > Yes I should be able to do that
>> > > > >
>> > > > > Any findings?
>> > > > >
>> > > > > /Jarkko
>> > > >
>> > > > Okay, finally getting back to this. Looking at the code it isn't clear
>> > > > to me
>> > > > why the change is causing this. So while I stare at this some more
>> > > > Laurent
>> > > > could you reproduce it with this patch so I can see what the status and
>> > > > access registers look like? Does anyone else on here happen to have a
>> > > > Sinosun
>> > > > tpm device? The systems I have access to with TPM1.2 devices don't have
>> > > > this
>> > > > issue.
>> > > >
>> > > > --8<--
>> > > >
>> > > > diff --git a/drivers/char/tpm/tpm_tis_core.c
>> > > > b/drivers/char/tpm/tpm_tis_core.c
>> > > > index fdde971bc810..7d60a7e4b50a 100644
>> > > > --- a/drivers/char/tpm/tpm_tis_core.c
>> > > > +++ b/drivers/char/tpm/tpm_tis_core.c
>> > > > @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>> > > > const u8 *buf, size_t len)
>> > > >     int rc, status, burstcnt;
>> > > >     size_t count = 0;
>> > > >     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>> > > > +    u8 access;
>> > > >
>> > > >     status = tpm_tis_status(chip);
>> > > >     if ((status & TPM_STS_COMMAND_READY) == 0) {
>> > > > @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>> > > > const u8 *buf, size_t len)
>> > > >         }
>> > > >         status = tpm_tis_status(chip);
>> > > >         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>> > > > +            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
>> > > > &access);
>> > > > +            if (rc < 0)
>> > > > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: read
>> > > > failure TPM_ACCESS(%d)\n", priv->locality);
>> > > > +            else
>> > > > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
>> > > > locality: %d status: %x access: %x\n", priv->locality, status, access);
>> > > >             rc = -EIO;
>> > > >             goto out_err;
>> > > >         }
>> > > > @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>> > > > const u8 *buf, size_t len)
>> > > >     }
>> > > >     status = tpm_tis_status(chip);
>> > > >     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>> > > > +        rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), &access);
>> > > > +        if (rc < 0)
>> > > > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
>> > > > failure TPM_ACCESS(%d)\n", priv->locality);
>> > > > +        else
>> > > > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: locality:
>> > > > %d status: %x access: %x\n", priv->locality, status, access);
>> > > >         rc = -EIO;
>> > > >         goto out_err;
>> > > >     }
>> > >
>> > > Please find here the dmesg output of the patched kernel
>> >
>> > At least 0xff is corrupted value in senseful way. CPU fills the read
>> > with ones for example for unaligned bus read. See table 19 in PC client
>> > spec. This can happen when you do unaligned read for example.
>> >
>> > Maybe TPM is unreachable i.e. powered off. Bit busy with stuff ATM but
>> > would probably make sense to compare that 0x81 to table 18 in the same
>> > spec.
>> >
>> > /Jarkko
>>
>> 0x81 is saying the access register status is valid, and the locality
>> is not active. That first bit means A Dynamic OS has not been previously
>> established on the platform. Normally we would see 0xa1, which would
>> mean valid register status, and the locality is active.
>
>I think the important thing to note here is that STS has bits set that
>should never be set. So we can conclude that TPM might be either
>
>1. Powered off
>2. In some transition state?
>
>/Jarkko

If it was powered off would we be getting a valid read from the access register?

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Tue, Oct 24, 2017 at 07:57:06AM -0700, Jerry Snitselaar wrote:
> On Tue Oct 24 17, Jarkko Sakkinen wrote:
> > On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
> > > On Mon Oct 23 17, Jarkko Sakkinen wrote:
> > > > On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
> > > > > Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
> > > > > > On Wed Sep 06 17, Jarkko Sakkinen wrote:
> > > > > > > On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville wrote:
> > > > > > > > Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
> > > > > > > > > On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
> > > > > > > > > > > Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
> > > > > > > > > > > > Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a ecrit :
> > > > > > > > > > > >>> An idea how to troubleshoot this?
> > > > > > > > > > > >> Can you run git bisect on the changes between 4.11 and
> > > > > > > > 4.12, so that
> > > > > > > > > > > >> we find the offending commit? It is probably sufficient
> > > > > > > > to limit the
> > > > > > > > > > > >> search to commits that touch something in drivers/char/tpm.
> > > > > > > > > > > >
> > > > > > > > > > > > I'll try and keep you posted.
> > > > > > > > > > >
> > > > > > > > > > > OK I've been able to bisect the problem and the bad commit is:
> > > > > > > > > > >
> > > > > > > > > > > e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first bad commit
> > > > > > > > > > > commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
> > > > > > > > > > > Author: Jerry Snitselaar <jsnitsel@redhat.com>
> > > > > > > > > > > Date:   Mon Mar 27 08:46:04 2017 -0700
> > > > > > > > > > >
> > > > > > > > > > >      tpm_tis: convert to using locality callbacks
> > > > > > > > > > >
> > > > > > > > > > >      This patch converts tpm_tis to use of the new tpm class ops
> > > > > > > > > > >      request_locality, and relinquish_locality.
> > > > > > > > > > >
> > > > > > > > > > >      With the move to using the callbacks, release_locality is
> > > > > > > > > > > changed so
> > > > > > > > > > >      that we now release the locality even if there is no
> > > > > > > > > > > request pending.
> > > > > > > > > > >
> > > > > > > > > > >      This required some changes to the tpm_tis_core_init
> > > > > > > > code path to
> > > > > > > > > > >      make sure locality is requested when needed:
> > > > > > > > > > >
> > > > > > > > > > >        - tpm2_probe code path will end up calling
> > > > > > > > > > > request/release through
> > > > > > > > > > >          callbacks, so request_locality prior to
> > > > > > > > tpm2_probe not needed.
> > > > > > > > > > >
> > > > > > > > > > >        - probe_itpm makes calls to tpm_tis_send_data which no
> > > > > > > > > > > longer calls
> > > > > > > > > > >          request_locality, so add request_locality prior to
> > > > > > > > > > > tpm_tis_send_data
> > > > > > > > > > >          calls. Also drop release_locality call in middleof
> > > > > > > > > > > probe_itpm, and
> > > > > > > > > > >          keep locality until release_locality called at end of
> > > > > > > > > > > probe_itpm.
> > > > > > > > > > >
> > > > > > > > > > >      Cc: Peter Huewe <peterhuewe@gmx.de>
> > > > > > > > > > >      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
> > > > > > > > > > >      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
> > > > > > > > > > >      Cc: Marcel Selhorst <tpmdd@selhorst.net>
> > > > > > > > > > >      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
> > > > > > > > > > >      Reviewed-by: Jarkko Sakkinen
> > > > > > > > <jarkko.sakkinen@linux.intel.com>
> > > > > > > > > > >      Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
> > > > > > > > > > >      Signed-off-by: Jarkko Sakkinen
> > > > > > > > <jarkko.sakkinen@linux.intel.com>
> > > > > > > > > > >
> > > > > > > > > > > :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
> > > > > > > > > > > 72f21b446e45ea1003de75902b0553deb99157fd M drivers
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > I've looked again at the code in question, but could not find
> > > > > > > > > > anything that is obviously wrong there. Locality is now
> > > > > > > > > > requested/released at slightly different points in the process than
> > > > > > > > > > before, but that's it. It does not seem to cause problems with the
> > > > > > > > > > majority of TPMs, since you are the first to report any, so
> > > > > > > > maybe it
> > > > > > > > > > is a quirk that only affects this device.
> > > > > > > > > >
> > > > > > > > > > Perhaps Jerry can help, since this is his change?
> > > > > > > > > >
> > > > > > > > > > Alexander
> > > > > > > > >
> > > > > > > > > Getting some caffeine in me, and starting to take a look. Adding
> > > > > > > > > Jarkko as well since this might involve the general locality changes.
> > > > > > > > >
> > > > > > > > > Laurent, if I send you a patch with some debugging code added, would
> > > > > > > > > you be able to run it on that system? I wasn't running into issues
> > > > > > > > > on the system I had with a 1.2 device, but I no longer have access
> > > > > > > > > to it. I'll see if I can find one in our labs and reproduce it there.
> > > > > > > >
> > > > > > > > Yes I should be able to do that
> > > > > > >
> > > > > > > Any findings?
> > > > > > >
> > > > > > > /Jarkko
> > > > > >
> > > > > > Okay, finally getting back to this. Looking at the code it isn't clear
> > > > > > to me
> > > > > > why the change is causing this. So while I stare at this some more
> > > > > > Laurent
> > > > > > could you reproduce it with this patch so I can see what the status and
> > > > > > access registers look like? Does anyone else on here happen to have a
> > > > > > Sinosun
> > > > > > tpm device? The systems I have access to with TPM1.2 devices don't have
> > > > > > this
> > > > > > issue.
> > > > > >
> > > > > > --8<--
> > > > > >
> > > > > > diff --git a/drivers/char/tpm/tpm_tis_core.c
> > > > > > b/drivers/char/tpm/tpm_tis_core.c
> > > > > > index fdde971bc810..7d60a7e4b50a 100644
> > > > > > --- a/drivers/char/tpm/tpm_tis_core.c
> > > > > > +++ b/drivers/char/tpm/tpm_tis_core.c
> > > > > > @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
> > > > > > const u8 *buf, size_t len)
> > > > > >     int rc, status, burstcnt;
> > > > > >     size_t count = 0;
> > > > > >     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
> > > > > > +    u8 access;
> > > > > >
> > > > > >     status = tpm_tis_status(chip);
> > > > > >     if ((status & TPM_STS_COMMAND_READY) == 0) {
> > > > > > @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
> > > > > > const u8 *buf, size_t len)
> > > > > >         }
> > > > > >         status = tpm_tis_status(chip);
> > > > > >         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
> > > > > > +            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
> > > > > > &access);
> > > > > > +            if (rc < 0)
> > > > > > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: read
> > > > > > failure TPM_ACCESS(%d)\n", priv->locality);
> > > > > > +            else
> > > > > > +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
> > > > > > locality: %d status: %x access: %x\n", priv->locality, status, access);
> > > > > >             rc = -EIO;
> > > > > >             goto out_err;
> > > > > >         }
> > > > > > @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
> > > > > > const u8 *buf, size_t len)
> > > > > >     }
> > > > > >     status = tpm_tis_status(chip);
> > > > > >     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
> > > > > > +        rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), &access);
> > > > > > +        if (rc < 0)
> > > > > > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
> > > > > > failure TPM_ACCESS(%d)\n", priv->locality);
> > > > > > +        else
> > > > > > +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: locality:
> > > > > > %d status: %x access: %x\n", priv->locality, status, access);
> > > > > >         rc = -EIO;
> > > > > >         goto out_err;
> > > > > >     }
> > > > >
> > > > > Please find here the dmesg output of the patched kernel
> > > >
> > > > At least 0xff is corrupted value in senseful way. CPU fills the read
> > > > with ones for example for unaligned bus read. See table 19 in PC client
> > > > spec. This can happen when you do unaligned read for example.
> > > >
> > > > Maybe TPM is unreachable i.e. powered off. Bit busy with stuff ATM but
> > > > would probably make sense to compare that 0x81 to table 18 in the same
> > > > spec.
> > > >
> > > > /Jarkko
> > > 
> > > 0x81 is saying the access register status is valid, and the locality
> > > is not active. That first bit means A Dynamic OS has not been previously
> > > established on the platform. Normally we would see 0xa1, which would
> > > mean valid register status, and the locality is active.
> > 
> > I think the important thing to note here is that STS has bits set that
> > should never be set. So we can conclude that TPM might be either
> > 
> > 1. Powered off
> > 2. In some transition state?
> > 
> > /Jarkko
> 
> If it was powered off would we be getting a valid read from the access
> register?

I think there is no universal answer to that :-)

Maybe adding a extra delay would be next test to make? If for random
reason it is in-between states...

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

Le 24/10/17 a 18:07, Jarkko Sakkinen a ecrit :
> On Tue, Oct 24, 2017 at 07:57:06AM -0700, Jerry Snitselaar wrote:
>> On Tue Oct 24 17, Jarkko Sakkinen wrote:
>>> On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
>>>> On Mon Oct 23 17, Jarkko Sakkinen wrote:
>>>>> On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
>>>>>> Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>>>>>>> On Wed Sep 06 17, Jarkko Sakkinen wrote:
>>>>>>>> On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville wrote:
>>>>>>>>> Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>>>>>>>>> On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>>>>>>>>>>> Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>>>>>>>>>>>> Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a ecrit :
>>>>>>>>>>>>>>> An idea how to troubleshoot this?
>>>>>>>>>>>>>> Can you run git bisect on the changes between 4.11 and
>>>>>>>>> 4.12, so that
>>>>>>>>>>>>>> we find the offending commit? It is probably sufficient
>>>>>>>>> to limit the
>>>>>>>>>>>>>> search to commits that touch something in drivers/char/tpm.
>>>>>>>>>>>>> I'll try and keep you posted.
>>>>>>>>>>>> OK I've been able to bisect the problem and the bad commit is:
>>>>>>>>>>>>
>>>>>>>>>>>> e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first bad commit
>>>>>>>>>>>> commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>>>>>>>>>>> Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>> Date:   Mon Mar 27 08:46:04 2017 -0700
>>>>>>>>>>>>
>>>>>>>>>>>>       tpm_tis: convert to using locality callbacks
>>>>>>>>>>>>
>>>>>>>>>>>>       This patch converts tpm_tis to use of the new tpm class ops
>>>>>>>>>>>>       request_locality, and relinquish_locality.
>>>>>>>>>>>>
>>>>>>>>>>>>       With the move to using the callbacks, release_locality is
>>>>>>>>>>>> changed so
>>>>>>>>>>>>       that we now release the locality even if there is no
>>>>>>>>>>>> request pending.
>>>>>>>>>>>>
>>>>>>>>>>>>       This required some changes to the tpm_tis_core_init
>>>>>>>>> code path to
>>>>>>>>>>>>       make sure locality is requested when needed:
>>>>>>>>>>>>
>>>>>>>>>>>>         - tpm2_probe code path will end up calling
>>>>>>>>>>>> request/release through
>>>>>>>>>>>>           callbacks, so request_locality prior to
>>>>>>>>> tpm2_probe not needed.
>>>>>>>>>>>>         - probe_itpm makes calls to tpm_tis_send_data which no
>>>>>>>>>>>> longer calls
>>>>>>>>>>>>           request_locality, so add request_locality prior to
>>>>>>>>>>>> tpm_tis_send_data
>>>>>>>>>>>>           calls. Also drop release_locality call in middleof
>>>>>>>>>>>> probe_itpm, and
>>>>>>>>>>>>           keep locality until release_locality called at end of
>>>>>>>>>>>> probe_itpm.
>>>>>>>>>>>>
>>>>>>>>>>>>       Cc: Peter Huewe <peterhuewe@gmx.de>
>>>>>>>>>>>>       Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>       Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>>>>>>>>>>>>       Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>>>>>>>>>>>       Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>       Reviewed-by: Jarkko Sakkinen
>>>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>       Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>       Signed-off-by: Jarkko Sakkinen
>>>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>> :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>>>>>>>>>>> 72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>>>>>>>>>>>
>>>>>>>>>>> I've looked again at the code in question, but could not find
>>>>>>>>>>> anything that is obviously wrong there. Locality is now
>>>>>>>>>>> requested/released at slightly different points in the process than
>>>>>>>>>>> before, but that's it. It does not seem to cause problems with the
>>>>>>>>>>> majority of TPMs, since you are the first to report any, so
>>>>>>>>> maybe it
>>>>>>>>>>> is a quirk that only affects this device.
>>>>>>>>>>>
>>>>>>>>>>> Perhaps Jerry can help, since this is his change?
>>>>>>>>>>>
>>>>>>>>>>> Alexander
>>>>>>>>>> Getting some caffeine in me, and starting to take a look. Adding
>>>>>>>>>> Jarkko as well since this might involve the general locality changes.
>>>>>>>>>>
>>>>>>>>>> Laurent, if I send you a patch with some debugging code added, would
>>>>>>>>>> you be able to run it on that system? I wasn't running into issues
>>>>>>>>>> on the system I had with a 1.2 device, but I no longer have access
>>>>>>>>>> to it. I'll see if I can find one in our labs and reproduce it there.
>>>>>>>>> Yes I should be able to do that
>>>>>>>> Any findings?
>>>>>>>>
>>>>>>>> /Jarkko
>>>>>>> Okay, finally getting back to this. Looking at the code it isn't clear
>>>>>>> to me
>>>>>>> why the change is causing this. So while I stare at this some more
>>>>>>> Laurent
>>>>>>> could you reproduce it with this patch so I can see what the status and
>>>>>>> access registers look like? Does anyone else on here happen to have a
>>>>>>> Sinosun
>>>>>>> tpm device? The systems I have access to with TPM1.2 devices don't have
>>>>>>> this
>>>>>>> issue.
>>>>>>>
>>>>>>> --8<--
>>>>>>>
>>>>>>> diff --git a/drivers/char/tpm/tpm_tis_core.c
>>>>>>> b/drivers/char/tpm/tpm_tis_core.c
>>>>>>> index fdde971bc810..7d60a7e4b50a 100644
>>>>>>> --- a/drivers/char/tpm/tpm_tis_core.c
>>>>>>> +++ b/drivers/char/tpm/tpm_tis_core.c
>>>>>>> @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>> const u8 *buf, size_t len)
>>>>>>>      int rc, status, burstcnt;
>>>>>>>      size_t count = 0;
>>>>>>>      bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>>>>>>> +    u8 access;
>>>>>>>
>>>>>>>      status = tpm_tis_status(chip);
>>>>>>>      if ((status & TPM_STS_COMMAND_READY) == 0) {
>>>>>>> @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>> const u8 *buf, size_t len)
>>>>>>>          }
>>>>>>>          status = tpm_tis_status(chip);
>>>>>>>          if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>>>>>>> +            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
>>>>>>> &access);
>>>>>>> +            if (rc < 0)
>>>>>>> +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: read
>>>>>>> failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>> +            else
>>>>>>> +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
>>>>>>> locality: %d status: %x access: %x\n", priv->locality, status, access);
>>>>>>>              rc = -EIO;
>>>>>>>              goto out_err;
>>>>>>>          }
>>>>>>> @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>> const u8 *buf, size_t len)
>>>>>>>      }
>>>>>>>      status = tpm_tis_status(chip);
>>>>>>>      if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>>>>>>> +        rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), &access);
>>>>>>> +        if (rc < 0)
>>>>>>> +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
>>>>>>> failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>> +        else
>>>>>>> +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: locality:
>>>>>>> %d status: %x access: %x\n", priv->locality, status, access);
>>>>>>>          rc = -EIO;
>>>>>>>          goto out_err;
>>>>>>>      }
>>>>>> Please find here the dmesg output of the patched kernel
>>>>> At least 0xff is corrupted value in senseful way. CPU fills the read
>>>>> with ones for example for unaligned bus read. See table 19 in PC client
>>>>> spec. This can happen when you do unaligned read for example.
>>>>>
>>>>> Maybe TPM is unreachable i.e. powered off. Bit busy with stuff ATM but
>>>>> would probably make sense to compare that 0x81 to table 18 in the same
>>>>> spec.
>>>>>
>>>>> /Jarkko
>>>> 0x81 is saying the access register status is valid, and the locality
>>>> is not active. That first bit means A Dynamic OS has not been previously
>>>> established on the platform. Normally we would see 0xa1, which would
>>>> mean valid register status, and the locality is active.
>>> I think the important thing to note here is that STS has bits set that
>>> should never be set. So we can conclude that TPM might be either
>>>
>>> 1. Powered off
>>> 2. In some transition state?
>>>
>>> /Jarkko
>> If it was powered off would we be getting a valid read from the access
>> register?
> I think there is no universal answer to that :-)
>
> Maybe adding a extra delay would be next test to make? If for random
> reason it is in-between states...
Any more ideas?

The chip is definitely in a weird state :/ I tried several ways to reset 
the chip (windows, tpm-tools,...).

I've been able to reset the chip via the bios (which now shows unowned) 
but chip is still locked apparently.

But still with < 4.12 I'm able to get /some/ information Public EK, 
PCR,... out of the chip so it's not completely broken...

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

Le 09/11/17 a 01:04, Laurent Bigonville a ecrit :
>
>
> Le 24/10/17 a 18:07, Jarkko Sakkinen a ecrit :
>> On Tue, Oct 24, 2017 at 07:57:06AM -0700, Jerry Snitselaar wrote:
>>> On Tue Oct 24 17, Jarkko Sakkinen wrote:
>>>> On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
>>>>> On Mon Oct 23 17, Jarkko Sakkinen wrote:
>>>>>> On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
>>>>>>> Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>>>>>>>> On Wed Sep 06 17, Jarkko Sakkinen wrote:
>>>>>>>>> On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville 
>>>>>>>>> wrote:
>>>>>>>>>> Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>>>>>>>>>> On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>>>>>>>>>>>> Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>>>>>>>>>>>>> Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a 
>>>>>>>>>>>>>> ecrit :
>>>>>>>>>>>>>>>> An idea how to troubleshoot this?
>>>>>>>>>>>>>>> Can you run git bisect on the changes between 4.11 and
>>>>>>>>>> 4.12, so that
>>>>>>>>>>>>>>> we find the offending commit? It is probably sufficient
>>>>>>>>>> to limit the
>>>>>>>>>>>>>>> search to commits that touch something in drivers/char/tpm.
>>>>>>>>>>>>>> I'll try and keep you posted.
>>>>>>>>>>>>> OK I've been able to bisect the problem and the bad commit 
>>>>>>>>>>>>> is:
>>>>>>>>>>>>>
>>>>>>>>>>>>> e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first bad 
>>>>>>>>>>>>> commit
>>>>>>>>>>>>> commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>>>>>>>>>>>> Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>> Date:   Mon Mar 27 08:46:04 2017 -0700
>>>>>>>>>>>>>
>>>>>>>>>>>>>       tpm_tis: convert to using locality callbacks
>>>>>>>>>>>>>
>>>>>>>>>>>>>       This patch converts tpm_tis to use of the new tpm 
>>>>>>>>>>>>> class ops
>>>>>>>>>>>>>       request_locality, and relinquish_locality.
>>>>>>>>>>>>>
>>>>>>>>>>>>>       With the move to using the callbacks, 
>>>>>>>>>>>>> release_locality is
>>>>>>>>>>>>> changed so
>>>>>>>>>>>>>       that we now release the locality even if there is no
>>>>>>>>>>>>> request pending.
>>>>>>>>>>>>>
>>>>>>>>>>>>>       This required some changes to the tpm_tis_core_init
>>>>>>>>>> code path to
>>>>>>>>>>>>>       make sure locality is requested when needed:
>>>>>>>>>>>>>
>>>>>>>>>>>>>         - tpm2_probe code path will end up calling
>>>>>>>>>>>>> request/release through
>>>>>>>>>>>>>           callbacks, so request_locality prior to
>>>>>>>>>> tpm2_probe not needed.
>>>>>>>>>>>>>         - probe_itpm makes calls to tpm_tis_send_data 
>>>>>>>>>>>>> which no
>>>>>>>>>>>>> longer calls
>>>>>>>>>>>>>           request_locality, so add request_locality prior to
>>>>>>>>>>>>> tpm_tis_send_data
>>>>>>>>>>>>>           calls. Also drop release_locality call in middleof
>>>>>>>>>>>>> probe_itpm, and
>>>>>>>>>>>>>           keep locality until release_locality called at 
>>>>>>>>>>>>> end of
>>>>>>>>>>>>> probe_itpm.
>>>>>>>>>>>>>
>>>>>>>>>>>>>       Cc: Peter Huewe <peterhuewe@gmx.de>
>>>>>>>>>>>>>       Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>       Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>>>>>>>>>>>>>       Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>>>>>>>>>>>>       Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>       Reviewed-by: Jarkko Sakkinen
>>>>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>       Tested-by: Jarkko Sakkinen 
>>>>>>>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>       Signed-off-by: Jarkko Sakkinen
>>>>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>> :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>>>>>>>>>>>> 72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>>>>>>>>>>>>
>>>>>>>>>>>> I've looked again at the code in question, but could not find
>>>>>>>>>>>> anything that is obviously wrong there. Locality is now
>>>>>>>>>>>> requested/released at slightly different points in the 
>>>>>>>>>>>> process than
>>>>>>>>>>>> before, but that's it. It does not seem to cause problems 
>>>>>>>>>>>> with the
>>>>>>>>>>>> majority of TPMs, since you are the first to report any, so
>>>>>>>>>> maybe it
>>>>>>>>>>>> is a quirk that only affects this device.
>>>>>>>>>>>>
>>>>>>>>>>>> Perhaps Jerry can help, since this is his change?
>>>>>>>>>>>>
>>>>>>>>>>>> Alexander
>>>>>>>>>>> Getting some caffeine in me, and starting to take a look. 
>>>>>>>>>>> Adding
>>>>>>>>>>> Jarkko as well since this might involve the general locality 
>>>>>>>>>>> changes.
>>>>>>>>>>>
>>>>>>>>>>> Laurent, if I send you a patch with some debugging code 
>>>>>>>>>>> added, would
>>>>>>>>>>> you be able to run it on that system? I wasn't running into 
>>>>>>>>>>> issues
>>>>>>>>>>> on the system I had with a 1.2 device, but I no longer have 
>>>>>>>>>>> access
>>>>>>>>>>> to it. I'll see if I can find one in our labs and reproduce 
>>>>>>>>>>> it there.
>>>>>>>>>> Yes I should be able to do that
>>>>>>>>> Any findings?
>>>>>>>>>
>>>>>>>>> /Jarkko
>>>>>>>> Okay, finally getting back to this. Looking at the code it 
>>>>>>>> isn't clear
>>>>>>>> to me
>>>>>>>> why the change is causing this. So while I stare at this some more
>>>>>>>> Laurent
>>>>>>>> could you reproduce it with this patch so I can see what the 
>>>>>>>> status and
>>>>>>>> access registers look like? Does anyone else on here happen to 
>>>>>>>> have a
>>>>>>>> Sinosun
>>>>>>>> tpm device? The systems I have access to with TPM1.2 devices 
>>>>>>>> don't have
>>>>>>>> this
>>>>>>>> issue.
>>>>>>>>
>>>>>>>> --8<--
>>>>>>>>
>>>>>>>> diff --git a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>> b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>> index fdde971bc810..7d60a7e4b50a 100644
>>>>>>>> --- a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>> +++ b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>> @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct 
>>>>>>>> tpm_chip *chip,
>>>>>>>> const u8 *buf, size_t len)
>>>>>>>>      int rc, status, burstcnt;
>>>>>>>>      size_t count = 0;
>>>>>>>>      bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>>>>>>>> +    u8 access;
>>>>>>>>
>>>>>>>>      status = tpm_tis_status(chip);
>>>>>>>>      if ((status & TPM_STS_COMMAND_READY) == 0) {
>>>>>>>> @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct 
>>>>>>>> tpm_chip *chip,
>>>>>>>> const u8 *buf, size_t len)
>>>>>>>>          }
>>>>>>>>          status = tpm_tis_status(chip);
>>>>>>>>          if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>>>>>>>> +            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
>>>>>>>> &access);
>>>>>>>> +            if (rc < 0)
>>>>>>>> +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 
>>>>>>>> 0: read
>>>>>>>> failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>> +            else
>>>>>>>> +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
>>>>>>>> locality: %d status: %x access: %x\n", priv->locality, status, 
>>>>>>>> access);
>>>>>>>>              rc = -EIO;
>>>>>>>>              goto out_err;
>>>>>>>>          }
>>>>>>>> @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct 
>>>>>>>> tpm_chip *chip,
>>>>>>>> const u8 *buf, size_t len)
>>>>>>>>      }
>>>>>>>>      status = tpm_tis_status(chip);
>>>>>>>>      if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>>>>>>>> +        rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), 
>>>>>>>> &access);
>>>>>>>> +        if (rc < 0)
>>>>>>>> +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
>>>>>>>> failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>> +        else
>>>>>>>> +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: 
>>>>>>>> locality:
>>>>>>>> %d status: %x access: %x\n", priv->locality, status, access);
>>>>>>>>          rc = -EIO;
>>>>>>>>          goto out_err;
>>>>>>>>      }
>>>>>>> Please find here the dmesg output of the patched kernel
>>>>>> At least 0xff is corrupted value in senseful way. CPU fills the read
>>>>>> with ones for example for unaligned bus read. See table 19 in PC 
>>>>>> client
>>>>>> spec. This can happen when you do unaligned read for example.
>>>>>>
>>>>>> Maybe TPM is unreachable i.e. powered off. Bit busy with stuff 
>>>>>> ATM but
>>>>>> would probably make sense to compare that 0x81 to table 18 in the 
>>>>>> same
>>>>>> spec.
>>>>>>
>>>>>> /Jarkko
>>>>> 0x81 is saying the access register status is valid, and the locality
>>>>> is not active. That first bit means A Dynamic OS has not been 
>>>>> previously
>>>>> established on the platform. Normally we would see 0xa1, which would
>>>>> mean valid register status, and the locality is active.
>>>> I think the important thing to note here is that STS has bits set that
>>>> should never be set. So we can conclude that TPM might be either
>>>>
>>>> 1. Powered off
>>>> 2. In some transition state?
>>>>
>>>> /Jarkko
>>> If it was powered off would we be getting a valid read from the access
>>> register?
>> I think there is no universal answer to that :-)
>>
>> Maybe adding a extra delay would be next test to make? If for random
>> reason it is in-between states...
> Any more ideas?
>
> The chip is definitely in a weird state :/ I tried several ways to 
> reset the chip (windows, tpm-tools,...).
>
> I've been able to reset the chip via the bios (which now shows 
> unowned) but chip is still locked apparently.
>
> But still with < 4.12 I'm able to get /some/ information Public EK, 
> PCR,... out of the chip so it's not completely broken...

OK correction the TPM is now unlocked (I let the computer running for 
more than 24h with nothing accessing the TPM) and with 4.9 I've been 
able to take the ownership again.

Under 4.12 I still have the same errors as mentioned originally

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Thu Nov 09 17, Laurent Bigonville wrote:
>Le 09/11/17 a 01:04, Laurent Bigonville a ecrit :
>>
>>
>>Le 24/10/17 a 18:07, Jarkko Sakkinen a ecrit :
>>>On Tue, Oct 24, 2017 at 07:57:06AM -0700, Jerry Snitselaar wrote:
>>>>On Tue Oct 24 17, Jarkko Sakkinen wrote:
>>>>>On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
>>>>>>On Mon Oct 23 17, Jarkko Sakkinen wrote:
>>>>>>>On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
>>>>>>>>Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>>>>>>>>>On Wed Sep 06 17, Jarkko Sakkinen wrote:
>>>>>>>>>>On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent 
>>>>>>>>>>Bigonville wrote:
>>>>>>>>>>>Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>>>>>>>>>>>On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>>>>>>>>>>>>>Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>>>>>>>>>>>>>>Le 29/08/17 a 18:00, 
>>>>>>>>>>>>>>>Alexander.Steffen@infineon.com a ecrit :
>>>>>>>>>>>>>>>>>An idea how to troubleshoot this?
>>>>>>>>>>>>>>>>Can you run git bisect on the changes between 4.11 and
>>>>>>>>>>>4.12, so that
>>>>>>>>>>>>>>>>we find the offending commit? It is probably sufficient
>>>>>>>>>>>to limit the
>>>>>>>>>>>>>>>>search to commits that touch something in drivers/char/tpm.
>>>>>>>>>>>>>>>I'll try and keep you posted.
>>>>>>>>>>>>>>OK I've been able to bisect the problem and 
>>>>>>>>>>>>>>the bad commit is:
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is 
>>>>>>>>>>>>>>the first bad commit
>>>>>>>>>>>>>>commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>>>>>>>>>>>>>Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>>Date:   Mon Mar 27 08:46:04 2017 -0700
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>      tpm_tis: convert to using locality callbacks
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>      This patch converts tpm_tis to use of 
>>>>>>>>>>>>>>the new tpm class ops
>>>>>>>>>>>>>>      request_locality, and relinquish_locality.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>      With the move to using the callbacks, 
>>>>>>>>>>>>>>release_locality is
>>>>>>>>>>>>>>changed so
>>>>>>>>>>>>>>      that we now release the locality even if there is no
>>>>>>>>>>>>>>request pending.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>      This required some changes to the tpm_tis_core_init
>>>>>>>>>>>code path to
>>>>>>>>>>>>>>      make sure locality is requested when needed:
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>        - tpm2_probe code path will end up calling
>>>>>>>>>>>>>>request/release through
>>>>>>>>>>>>>>          callbacks, so request_locality prior to
>>>>>>>>>>>tpm2_probe not needed.
>>>>>>>>>>>>>>        - probe_itpm makes calls to 
>>>>>>>>>>>>>>tpm_tis_send_data which no
>>>>>>>>>>>>>>longer calls
>>>>>>>>>>>>>>          request_locality, so add request_locality prior to
>>>>>>>>>>>>>>tpm_tis_send_data
>>>>>>>>>>>>>>          calls. Also drop release_locality call in middleof
>>>>>>>>>>>>>>probe_itpm, and
>>>>>>>>>>>>>>          keep locality until 
>>>>>>>>>>>>>>release_locality called at end of
>>>>>>>>>>>>>>probe_itpm.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>      Cc: Peter Huewe <peterhuewe@gmx.de>
>>>>>>>>>>>>>>      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>>>>>>>>>>>>>>      Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>>>>>>>>>>>>>      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>>      Reviewed-by: Jarkko Sakkinen
>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>      Tested-by: Jarkko Sakkinen 
>>>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>      Signed-off-by: Jarkko Sakkinen
>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>:040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>>>>>>>>>>>>>72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>>>>>>>>>>>>>
>>>>>>>>>>>>>I've looked again at the code in question, but could not find
>>>>>>>>>>>>>anything that is obviously wrong there. Locality is now
>>>>>>>>>>>>>requested/released at slightly different 
>>>>>>>>>>>>>points in the process than
>>>>>>>>>>>>>before, but that's it. It does not seem to 
>>>>>>>>>>>>>cause problems with the
>>>>>>>>>>>>>majority of TPMs, since you are the first to report any, so
>>>>>>>>>>>maybe it
>>>>>>>>>>>>>is a quirk that only affects this device.
>>>>>>>>>>>>>
>>>>>>>>>>>>>Perhaps Jerry can help, since this is his change?
>>>>>>>>>>>>>
>>>>>>>>>>>>>Alexander
>>>>>>>>>>>>Getting some caffeine in me, and starting to 
>>>>>>>>>>>>take a look. Adding
>>>>>>>>>>>>Jarkko as well since this might involve the 
>>>>>>>>>>>>general locality changes.
>>>>>>>>>>>>
>>>>>>>>>>>>Laurent, if I send you a patch with some 
>>>>>>>>>>>>debugging code added, would
>>>>>>>>>>>>you be able to run it on that system? I wasn't 
>>>>>>>>>>>>running into issues
>>>>>>>>>>>>on the system I had with a 1.2 device, but I no 
>>>>>>>>>>>>longer have access
>>>>>>>>>>>>to it. I'll see if I can find one in our labs 
>>>>>>>>>>>>and reproduce it there.
>>>>>>>>>>>Yes I should be able to do that
>>>>>>>>>>Any findings?
>>>>>>>>>>
>>>>>>>>>>/Jarkko
>>>>>>>>>Okay, finally getting back to this. Looking at the 
>>>>>>>>>code it isn't clear
>>>>>>>>>to me
>>>>>>>>>why the change is causing this. So while I stare at this some more
>>>>>>>>>Laurent
>>>>>>>>>could you reproduce it with this patch so I can see 
>>>>>>>>>what the status and
>>>>>>>>>access registers look like? Does anyone else on here 
>>>>>>>>>happen to have a
>>>>>>>>>Sinosun
>>>>>>>>>tpm device? The systems I have access to with TPM1.2 
>>>>>>>>>devices don't have
>>>>>>>>>this
>>>>>>>>>issue.
>>>>>>>>>
>>>>>>>>>--8<--
>>>>>>>>>
>>>>>>>>>diff --git a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>index fdde971bc810..7d60a7e4b50a 100644
>>>>>>>>>--- a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>+++ b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>@@ -258,6 +258,7 @@ static int 
>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>     int rc, status, burstcnt;
>>>>>>>>>     size_t count = 0;
>>>>>>>>>     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>>>>>>>>>+    u8 access;
>>>>>>>>>
>>>>>>>>>     status = tpm_tis_status(chip);
>>>>>>>>>     if ((status & TPM_STS_COMMAND_READY) == 0) {
>>>>>>>>>@@ -292,6 +293,11 @@ static int 
>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>         }
>>>>>>>>>         status = tpm_tis_status(chip);
>>>>>>>>>         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>>>>>>>>>+            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
>>>>>>>>>&access);
>>>>>>>>>+            if (rc < 0)
>>>>>>>>>+                dev_info(&chip->dev, 
>>>>>>>>>"TPM_STS_DATA_EXPECT == 0: read
>>>>>>>>>failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>+            else
>>>>>>>>>+                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
>>>>>>>>>locality: %d status: %x access: %x\n", priv->locality, 
>>>>>>>>>status, access);
>>>>>>>>>             rc = -EIO;
>>>>>>>>>             goto out_err;
>>>>>>>>>         }
>>>>>>>>>@@ -309,6 +315,11 @@ static int 
>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>     }
>>>>>>>>>     status = tpm_tis_status(chip);
>>>>>>>>>     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>>>>>>>>>+        rc = tpm_tis_read8(priv, 
>>>>>>>>>TPM_ACCESS(priv->locality), &access);
>>>>>>>>>+        if (rc < 0)
>>>>>>>>>+            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
>>>>>>>>>failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>+        else
>>>>>>>>>+            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT 
>>>>>>>>>!= 0: locality:
>>>>>>>>>%d status: %x access: %x\n", priv->locality, status, access);
>>>>>>>>>         rc = -EIO;
>>>>>>>>>         goto out_err;
>>>>>>>>>     }
>>>>>>>>Please find here the dmesg output of the patched kernel
>>>>>>>At least 0xff is corrupted value in senseful way. CPU fills the read
>>>>>>>with ones for example for unaligned bus read. See table 19 
>>>>>>>in PC client
>>>>>>>spec. This can happen when you do unaligned read for example.
>>>>>>>
>>>>>>>Maybe TPM is unreachable i.e. powered off. Bit busy with 
>>>>>>>stuff ATM but
>>>>>>>would probably make sense to compare that 0x81 to table 18 
>>>>>>>in the same
>>>>>>>spec.
>>>>>>>
>>>>>>>/Jarkko
>>>>>>0x81 is saying the access register status is valid, and the locality
>>>>>>is not active. That first bit means A Dynamic OS has not 
>>>>>>been previously
>>>>>>established on the platform. Normally we would see 0xa1, which would
>>>>>>mean valid register status, and the locality is active.
>>>>>I think the important thing to note here is that STS has bits set that
>>>>>should never be set. So we can conclude that TPM might be either
>>>>>
>>>>>1. Powered off
>>>>>2. In some transition state?
>>>>>
>>>>>/Jarkko
>>>>If it was powered off would we be getting a valid read from the access
>>>>register?
>>>I think there is no universal answer to that :-)
>>>
>>>Maybe adding a extra delay would be next test to make? If for random
>>>reason it is in-between states...
>>Any more ideas?
>>
>>The chip is definitely in a weird state :/ I tried several ways to 
>>reset the chip (windows, tpm-tools,...).
>>
>>I've been able to reset the chip via the bios (which now shows 
>>unowned) but chip is still locked apparently.
>>
>>But still with < 4.12 I'm able to get /some/ information Public EK, 
>>PCR,... out of the chip so it's not completely broken...
>
>OK correction the TPM is now unlocked (I let the computer running for 
>more than 24h with nothing accessing the TPM) and with 4.9 I've been 
>able to take the ownership again.
>
>Under 4.12 I still have the same errors as mentioned originally


Still thinking about how to attack this. I don't think extending the timeout
would change anything because it is reading the register, and thinking the
status is valid because it reads 0xff. So it would still have that problem,
right?

Maybe a sanity check could be added to wait_for_tpm_stat() to
first check that it didn't read 0xff before checking for the value it
is looking for. Would that make sense Jarkko?

diff --git a/drivers/char/tpm/tpm-interface.c b/drivers/char/tpm/tpm-interface.c
index 1d6729be4cd6..8b77fa2003ce 100644
--- a/drivers/char/tpm/tpm-interface.c
+++ b/drivers/char/tpm/tpm-interface.c
@@ -1087,7 +1087,7 @@ int wait_for_tpm_stat(struct tpm_chip *chip, u8 mask, unsigned long timeout,
                do {
                        tpm_msleep(TPM_TIMEOUT);
                        status = chip->ops->status(chip);
-                       if ((status & mask) == mask)
+                       if ((status != 0xff) && (status & mask) == mask)
                                return 0;
                } while (time_before(jiffies, stop));
        }

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Thu Nov 09 17, Jerry Snitselaar wrote:
>On Thu Nov 09 17, Laurent Bigonville wrote:
>>Le 09/11/17 a 01:04, Laurent Bigonville a ecrit :
>>>
>>>
>>>Le 24/10/17 a 18:07, Jarkko Sakkinen a ecrit :
>>>>On Tue, Oct 24, 2017 at 07:57:06AM -0700, Jerry Snitselaar wrote:
>>>>>On Tue Oct 24 17, Jarkko Sakkinen wrote:
>>>>>>On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
>>>>>>>On Mon Oct 23 17, Jarkko Sakkinen wrote:
>>>>>>>>On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
>>>>>>>>>Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>>>>>>>>>>On Wed Sep 06 17, Jarkko Sakkinen wrote:
>>>>>>>>>>>On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent 
>>>>>>>>>>>Bigonville wrote:
>>>>>>>>>>>>Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>>>>>>>>>>>>On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>>>>>>>>>>>>>>Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>>>>>>>>>>>>>>>Le 29/08/17 a 18:00, 
>>>>>>>>>>>>>>>>Alexander.Steffen@infineon.com a ecrit :
>>>>>>>>>>>>>>>>>>An idea how to troubleshoot this?
>>>>>>>>>>>>>>>>>Can you run git bisect on the changes between 4.11 and
>>>>>>>>>>>>4.12, so that
>>>>>>>>>>>>>>>>>we find the offending commit? It is probably sufficient
>>>>>>>>>>>>to limit the
>>>>>>>>>>>>>>>>>search to commits that touch something in drivers/char/tpm.
>>>>>>>>>>>>>>>>I'll try and keep you posted.
>>>>>>>>>>>>>>>OK I've been able to bisect the problem 
>>>>>>>>>>>>>>>and the bad commit is:
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 
>>>>>>>>>>>>>>>is the first bad commit
>>>>>>>>>>>>>>>commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>>>>>>>>>>>>>>Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>>>Date:   Mon Mar 27 08:46:04 2017 -0700
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>      tpm_tis: convert to using locality callbacks
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>      This patch converts tpm_tis to use 
>>>>>>>>>>>>>>>of the new tpm class ops
>>>>>>>>>>>>>>>      request_locality, and relinquish_locality.
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>      With the move to using the 
>>>>>>>>>>>>>>>callbacks, release_locality is
>>>>>>>>>>>>>>>changed so
>>>>>>>>>>>>>>>      that we now release the locality even if there is no
>>>>>>>>>>>>>>>request pending.
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>      This required some changes to the tpm_tis_core_init
>>>>>>>>>>>>code path to
>>>>>>>>>>>>>>>      make sure locality is requested when needed:
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>        - tpm2_probe code path will end up calling
>>>>>>>>>>>>>>>request/release through
>>>>>>>>>>>>>>>          callbacks, so request_locality prior to
>>>>>>>>>>>>tpm2_probe not needed.
>>>>>>>>>>>>>>>        - probe_itpm makes calls to 
>>>>>>>>>>>>>>>tpm_tis_send_data which no
>>>>>>>>>>>>>>>longer calls
>>>>>>>>>>>>>>>          request_locality, so add request_locality prior to
>>>>>>>>>>>>>>>tpm_tis_send_data
>>>>>>>>>>>>>>>          calls. Also drop release_locality call in middleof
>>>>>>>>>>>>>>>probe_itpm, and
>>>>>>>>>>>>>>>          keep locality until 
>>>>>>>>>>>>>>>release_locality called at end of
>>>>>>>>>>>>>>>probe_itpm.
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>      Cc: Peter Huewe <peterhuewe@gmx.de>
>>>>>>>>>>>>>>>      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>>>>>>>>>>>>>>>      Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>>>>>>>>>>>>>>      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>>>      Reviewed-by: Jarkko Sakkinen
>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>      Tested-by: Jarkko Sakkinen 
>>>>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>      Signed-off-by: Jarkko Sakkinen
>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>:040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>>>>>>>>>>>>>>72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>I've looked again at the code in question, but could not find
>>>>>>>>>>>>>>anything that is obviously wrong there. Locality is now
>>>>>>>>>>>>>>requested/released at slightly different 
>>>>>>>>>>>>>>points in the process than
>>>>>>>>>>>>>>before, but that's it. It does not seem to 
>>>>>>>>>>>>>>cause problems with the
>>>>>>>>>>>>>>majority of TPMs, since you are the first to report any, so
>>>>>>>>>>>>maybe it
>>>>>>>>>>>>>>is a quirk that only affects this device.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>Perhaps Jerry can help, since this is his change?
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>Alexander
>>>>>>>>>>>>>Getting some caffeine in me, and starting to 
>>>>>>>>>>>>>take a look. Adding
>>>>>>>>>>>>>Jarkko as well since this might involve the 
>>>>>>>>>>>>>general locality changes.
>>>>>>>>>>>>>
>>>>>>>>>>>>>Laurent, if I send you a patch with some 
>>>>>>>>>>>>>debugging code added, would
>>>>>>>>>>>>>you be able to run it on that system? I wasn't 
>>>>>>>>>>>>>running into issues
>>>>>>>>>>>>>on the system I had with a 1.2 device, but I 
>>>>>>>>>>>>>no longer have access
>>>>>>>>>>>>>to it. I'll see if I can find one in our labs 
>>>>>>>>>>>>>and reproduce it there.
>>>>>>>>>>>>Yes I should be able to do that
>>>>>>>>>>>Any findings?
>>>>>>>>>>>
>>>>>>>>>>>/Jarkko
>>>>>>>>>>Okay, finally getting back to this. Looking at the 
>>>>>>>>>>code it isn't clear
>>>>>>>>>>to me
>>>>>>>>>>why the change is causing this. So while I stare at this some more
>>>>>>>>>>Laurent
>>>>>>>>>>could you reproduce it with this patch so I can see 
>>>>>>>>>>what the status and
>>>>>>>>>>access registers look like? Does anyone else on here 
>>>>>>>>>>happen to have a
>>>>>>>>>>Sinosun
>>>>>>>>>>tpm device? The systems I have access to with TPM1.2 
>>>>>>>>>>devices don't have
>>>>>>>>>>this
>>>>>>>>>>issue.
>>>>>>>>>>
>>>>>>>>>>--8<--
>>>>>>>>>>
>>>>>>>>>>diff --git a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>index fdde971bc810..7d60a7e4b50a 100644
>>>>>>>>>>--- a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>+++ b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>@@ -258,6 +258,7 @@ static int 
>>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>>     int rc, status, burstcnt;
>>>>>>>>>>     size_t count = 0;
>>>>>>>>>>     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>>>>>>>>>>+    u8 access;
>>>>>>>>>>
>>>>>>>>>>     status = tpm_tis_status(chip);
>>>>>>>>>>     if ((status & TPM_STS_COMMAND_READY) == 0) {
>>>>>>>>>>@@ -292,6 +293,11 @@ static int 
>>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>>         }
>>>>>>>>>>         status = tpm_tis_status(chip);
>>>>>>>>>>         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>>>>>>>>>>+            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
>>>>>>>>>>&access);
>>>>>>>>>>+            if (rc < 0)
>>>>>>>>>>+                dev_info(&chip->dev, 
>>>>>>>>>>"TPM_STS_DATA_EXPECT == 0: read
>>>>>>>>>>failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>>+            else
>>>>>>>>>>+                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
>>>>>>>>>>locality: %d status: %x access: %x\n", 
>>>>>>>>>>priv->locality, status, access);
>>>>>>>>>>             rc = -EIO;
>>>>>>>>>>             goto out_err;
>>>>>>>>>>         }
>>>>>>>>>>@@ -309,6 +315,11 @@ static int 
>>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>>     }
>>>>>>>>>>     status = tpm_tis_status(chip);
>>>>>>>>>>     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>>>>>>>>>>+        rc = tpm_tis_read8(priv, 
>>>>>>>>>>TPM_ACCESS(priv->locality), &access);
>>>>>>>>>>+        if (rc < 0)
>>>>>>>>>>+            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
>>>>>>>>>>failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>>+        else
>>>>>>>>>>+            dev_info(&chip->dev, 
>>>>>>>>>>"TPM_STS_DATA_EXPECT != 0: locality:
>>>>>>>>>>%d status: %x access: %x\n", priv->locality, status, access);
>>>>>>>>>>         rc = -EIO;
>>>>>>>>>>         goto out_err;
>>>>>>>>>>     }
>>>>>>>>>Please find here the dmesg output of the patched kernel
>>>>>>>>At least 0xff is corrupted value in senseful way. CPU fills the read
>>>>>>>>with ones for example for unaligned bus read. See table 
>>>>>>>>19 in PC client
>>>>>>>>spec. This can happen when you do unaligned read for example.
>>>>>>>>
>>>>>>>>Maybe TPM is unreachable i.e. powered off. Bit busy with 
>>>>>>>>stuff ATM but
>>>>>>>>would probably make sense to compare that 0x81 to table 
>>>>>>>>18 in the same
>>>>>>>>spec.
>>>>>>>>
>>>>>>>>/Jarkko
>>>>>>>0x81 is saying the access register status is valid, and the locality
>>>>>>>is not active. That first bit means A Dynamic OS has not 
>>>>>>>been previously
>>>>>>>established on the platform. Normally we would see 0xa1, which would
>>>>>>>mean valid register status, and the locality is active.
>>>>>>I think the important thing to note here is that STS has bits set that
>>>>>>should never be set. So we can conclude that TPM might be either
>>>>>>
>>>>>>1. Powered off
>>>>>>2. In some transition state?
>>>>>>
>>>>>>/Jarkko
>>>>>If it was powered off would we be getting a valid read from the access
>>>>>register?
>>>>I think there is no universal answer to that :-)
>>>>
>>>>Maybe adding a extra delay would be next test to make? If for random
>>>>reason it is in-between states...
>>>Any more ideas?
>>>
>>>The chip is definitely in a weird state :/ I tried several ways to 
>>>reset the chip (windows, tpm-tools,...).
>>>
>>>I've been able to reset the chip via the bios (which now shows 
>>>unowned) but chip is still locked apparently.
>>>
>>>But still with < 4.12 I'm able to get /some/ information Public 
>>>EK, PCR,... out of the chip so it's not completely broken...
>>
>>OK correction the TPM is now unlocked (I let the computer running 
>>for more than 24h with nothing accessing the TPM) and with 4.9 I've 
>>been able to take the ownership again.
>>
>>Under 4.12 I still have the same errors as mentioned originally
>
>
>Still thinking about how to attack this. I don't think extending the timeout
>would change anything because it is reading the register, and thinking the
>status is valid because it reads 0xff. So it would still have that problem,
>right?
>
>Maybe a sanity check could be added to wait_for_tpm_stat() to
>first check that it didn't read 0xff before checking for the value it
>is looking for. Would that make sense Jarkko?
>
>diff --git a/drivers/char/tpm/tpm-interface.c b/drivers/char/tpm/tpm-interface.c
>index 1d6729be4cd6..8b77fa2003ce 100644
>--- a/drivers/char/tpm/tpm-interface.c
>+++ b/drivers/char/tpm/tpm-interface.c
>@@ -1087,7 +1087,7 @@ int wait_for_tpm_stat(struct tpm_chip *chip, u8 mask, unsigned long timeout,
>               do {
>                       tpm_msleep(TPM_TIMEOUT);
>                       status = chip->ops->status(chip);
>-                       if ((status & mask) == mask)
>+                       if ((status != 0xff) && (status & mask) == mask)
>                               return 0;
>               } while (time_before(jiffies, stop));
>       }
>

Actually I'm reading through the tpm profile spec again, and 6.1 FIFO
Interface Locality Usage per Register has a table that I think
explains what is going on.

Since the access register is showing 0x81, which means access register
bits are valid and the locality is not active. Table 39 in 6.1 says in
the case where active locality is not set, or set for another
locality, TPM_STS_x register reads return FFh.

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Thu Nov 09 17, Laurent Bigonville wrote:
>Le 09/11/17 a 01:04, Laurent Bigonville a ecrit :
>>
>>
>>Le 24/10/17 a 18:07, Jarkko Sakkinen a ecrit :
>>>On Tue, Oct 24, 2017 at 07:57:06AM -0700, Jerry Snitselaar wrote:
>>>>On Tue Oct 24 17, Jarkko Sakkinen wrote:
>>>>>On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
>>>>>>On Mon Oct 23 17, Jarkko Sakkinen wrote:
>>>>>>>On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
>>>>>>>>Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>>>>>>>>>On Wed Sep 06 17, Jarkko Sakkinen wrote:
>>>>>>>>>>On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent 
>>>>>>>>>>Bigonville wrote:
>>>>>>>>>>>Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>>>>>>>>>>>On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>>>>>>>>>>>>>Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>>>>>>>>>>>>>>Le 29/08/17 a 18:00, 
>>>>>>>>>>>>>>>Alexander.Steffen@infineon.com a ecrit :
>>>>>>>>>>>>>>>>>An idea how to troubleshoot this?
>>>>>>>>>>>>>>>>Can you run git bisect on the changes between 4.11 and
>>>>>>>>>>>4.12, so that
>>>>>>>>>>>>>>>>we find the offending commit? It is probably sufficient
>>>>>>>>>>>to limit the
>>>>>>>>>>>>>>>>search to commits that touch something in drivers/char/tpm.
>>>>>>>>>>>>>>>I'll try and keep you posted.
>>>>>>>>>>>>>>OK I've been able to bisect the problem and 
>>>>>>>>>>>>>>the bad commit is:
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is 
>>>>>>>>>>>>>>the first bad commit
>>>>>>>>>>>>>>commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>>>>>>>>>>>>>Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>>Date:   Mon Mar 27 08:46:04 2017 -0700
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>      tpm_tis: convert to using locality callbacks
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>      This patch converts tpm_tis to use of 
>>>>>>>>>>>>>>the new tpm class ops
>>>>>>>>>>>>>>      request_locality, and relinquish_locality.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>      With the move to using the callbacks, 
>>>>>>>>>>>>>>release_locality is
>>>>>>>>>>>>>>changed so
>>>>>>>>>>>>>>      that we now release the locality even if there is no
>>>>>>>>>>>>>>request pending.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>      This required some changes to the tpm_tis_core_init
>>>>>>>>>>>code path to
>>>>>>>>>>>>>>      make sure locality is requested when needed:
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>        - tpm2_probe code path will end up calling
>>>>>>>>>>>>>>request/release through
>>>>>>>>>>>>>>          callbacks, so request_locality prior to
>>>>>>>>>>>tpm2_probe not needed.
>>>>>>>>>>>>>>        - probe_itpm makes calls to 
>>>>>>>>>>>>>>tpm_tis_send_data which no
>>>>>>>>>>>>>>longer calls
>>>>>>>>>>>>>>          request_locality, so add request_locality prior to
>>>>>>>>>>>>>>tpm_tis_send_data
>>>>>>>>>>>>>>          calls. Also drop release_locality call in middleof
>>>>>>>>>>>>>>probe_itpm, and
>>>>>>>>>>>>>>          keep locality until 
>>>>>>>>>>>>>>release_locality called at end of
>>>>>>>>>>>>>>probe_itpm.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>      Cc: Peter Huewe <peterhuewe@gmx.de>
>>>>>>>>>>>>>>      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>>>>>>>>>>>>>>      Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>>>>>>>>>>>>>      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>>      Reviewed-by: Jarkko Sakkinen
>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>      Tested-by: Jarkko Sakkinen 
>>>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>      Signed-off-by: Jarkko Sakkinen
>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>:040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>>>>>>>>>>>>>72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>>>>>>>>>>>>>
>>>>>>>>>>>>>I've looked again at the code in question, but could not find
>>>>>>>>>>>>>anything that is obviously wrong there. Locality is now
>>>>>>>>>>>>>requested/released at slightly different 
>>>>>>>>>>>>>points in the process than
>>>>>>>>>>>>>before, but that's it. It does not seem to 
>>>>>>>>>>>>>cause problems with the
>>>>>>>>>>>>>majority of TPMs, since you are the first to report any, so
>>>>>>>>>>>maybe it
>>>>>>>>>>>>>is a quirk that only affects this device.
>>>>>>>>>>>>>
>>>>>>>>>>>>>Perhaps Jerry can help, since this is his change?
>>>>>>>>>>>>>
>>>>>>>>>>>>>Alexander
>>>>>>>>>>>>Getting some caffeine in me, and starting to 
>>>>>>>>>>>>take a look. Adding
>>>>>>>>>>>>Jarkko as well since this might involve the 
>>>>>>>>>>>>general locality changes.
>>>>>>>>>>>>
>>>>>>>>>>>>Laurent, if I send you a patch with some 
>>>>>>>>>>>>debugging code added, would
>>>>>>>>>>>>you be able to run it on that system? I wasn't 
>>>>>>>>>>>>running into issues
>>>>>>>>>>>>on the system I had with a 1.2 device, but I no 
>>>>>>>>>>>>longer have access
>>>>>>>>>>>>to it. I'll see if I can find one in our labs 
>>>>>>>>>>>>and reproduce it there.
>>>>>>>>>>>Yes I should be able to do that
>>>>>>>>>>Any findings?
>>>>>>>>>>
>>>>>>>>>>/Jarkko
>>>>>>>>>Okay, finally getting back to this. Looking at the 
>>>>>>>>>code it isn't clear
>>>>>>>>>to me
>>>>>>>>>why the change is causing this. So while I stare at this some more
>>>>>>>>>Laurent
>>>>>>>>>could you reproduce it with this patch so I can see 
>>>>>>>>>what the status and
>>>>>>>>>access registers look like? Does anyone else on here 
>>>>>>>>>happen to have a
>>>>>>>>>Sinosun
>>>>>>>>>tpm device? The systems I have access to with TPM1.2 
>>>>>>>>>devices don't have
>>>>>>>>>this
>>>>>>>>>issue.
>>>>>>>>>
>>>>>>>>>--8<--
>>>>>>>>>
>>>>>>>>>diff --git a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>index fdde971bc810..7d60a7e4b50a 100644
>>>>>>>>>--- a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>+++ b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>@@ -258,6 +258,7 @@ static int 
>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>     int rc, status, burstcnt;
>>>>>>>>>     size_t count = 0;
>>>>>>>>>     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>>>>>>>>>+    u8 access;
>>>>>>>>>
>>>>>>>>>     status = tpm_tis_status(chip);
>>>>>>>>>     if ((status & TPM_STS_COMMAND_READY) == 0) {
>>>>>>>>>@@ -292,6 +293,11 @@ static int 
>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>         }
>>>>>>>>>         status = tpm_tis_status(chip);
>>>>>>>>>         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>>>>>>>>>+            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
>>>>>>>>>&access);
>>>>>>>>>+            if (rc < 0)
>>>>>>>>>+                dev_info(&chip->dev, 
>>>>>>>>>"TPM_STS_DATA_EXPECT == 0: read
>>>>>>>>>failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>+            else
>>>>>>>>>+                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
>>>>>>>>>locality: %d status: %x access: %x\n", priv->locality, 
>>>>>>>>>status, access);
>>>>>>>>>             rc = -EIO;
>>>>>>>>>             goto out_err;
>>>>>>>>>         }
>>>>>>>>>@@ -309,6 +315,11 @@ static int 
>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>     }
>>>>>>>>>     status = tpm_tis_status(chip);
>>>>>>>>>     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>>>>>>>>>+        rc = tpm_tis_read8(priv, 
>>>>>>>>>TPM_ACCESS(priv->locality), &access);
>>>>>>>>>+        if (rc < 0)
>>>>>>>>>+            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
>>>>>>>>>failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>+        else
>>>>>>>>>+            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT 
>>>>>>>>>!= 0: locality:
>>>>>>>>>%d status: %x access: %x\n", priv->locality, status, access);
>>>>>>>>>         rc = -EIO;
>>>>>>>>>         goto out_err;
>>>>>>>>>     }
>>>>>>>>Please find here the dmesg output of the patched kernel
>>>>>>>At least 0xff is corrupted value in senseful way. CPU fills the read
>>>>>>>with ones for example for unaligned bus read. See table 19 
>>>>>>>in PC client
>>>>>>>spec. This can happen when you do unaligned read for example.
>>>>>>>
>>>>>>>Maybe TPM is unreachable i.e. powered off. Bit busy with 
>>>>>>>stuff ATM but
>>>>>>>would probably make sense to compare that 0x81 to table 18 
>>>>>>>in the same
>>>>>>>spec.
>>>>>>>
>>>>>>>/Jarkko
>>>>>>0x81 is saying the access register status is valid, and the locality
>>>>>>is not active. That first bit means A Dynamic OS has not 
>>>>>>been previously
>>>>>>established on the platform. Normally we would see 0xa1, which would
>>>>>>mean valid register status, and the locality is active.
>>>>>I think the important thing to note here is that STS has bits set that
>>>>>should never be set. So we can conclude that TPM might be either
>>>>>
>>>>>1. Powered off
>>>>>2. In some transition state?
>>>>>
>>>>>/Jarkko
>>>>If it was powered off would we be getting a valid read from the access
>>>>register?
>>>I think there is no universal answer to that :-)
>>>
>>>Maybe adding a extra delay would be next test to make? If for random
>>>reason it is in-between states...
>>Any more ideas?
>>
>>The chip is definitely in a weird state :/ I tried several ways to 
>>reset the chip (windows, tpm-tools,...).
>>
>>I've been able to reset the chip via the bios (which now shows 
>>unowned) but chip is still locked apparently.
>>
>>But still with < 4.12 I'm able to get /some/ information Public EK, 
>>PCR,... out of the chip so it's not completely broken...
>
>OK correction the TPM is now unlocked (I let the computer running for 
>more than 24h with nothing accessing the TPM) and with 4.9 I've been 
>able to take the ownership again.
>
>Under 4.12 I still have the same errors as mentioned originally

Hi Laurent,

Would it be possible for you to run ftrace from boot with the following kernel parameters:

ftrace=function_graph ftrace_filter=tpm*

and then send me the results of 'cat /sys/kernel/debug/tracing/trace' ?

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Thu Nov 09 17, Jerry Snitselaar wrote:
>On Thu Nov 09 17, Laurent Bigonville wrote:
>>Le 09/11/17 a 01:04, Laurent Bigonville a ecrit :
>>>
>>>
>>>Le 24/10/17 a 18:07, Jarkko Sakkinen a ecrit :
>>>>On Tue, Oct 24, 2017 at 07:57:06AM -0700, Jerry Snitselaar wrote:
>>>>>On Tue Oct 24 17, Jarkko Sakkinen wrote:
>>>>>>On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
>>>>>>>On Mon Oct 23 17, Jarkko Sakkinen wrote:
>>>>>>>>On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
>>>>>>>>>Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>>>>>>>>>>On Wed Sep 06 17, Jarkko Sakkinen wrote:
>>>>>>>>>>>On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent 
>>>>>>>>>>>Bigonville wrote:
>>>>>>>>>>>>Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>>>>>>>>>>>>On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>>>>>>>>>>>>>>Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>>>>>>>>>>>>>>>Le 29/08/17 a 18:00, 
>>>>>>>>>>>>>>>>Alexander.Steffen@infineon.com a ecrit :
>>>>>>>>>>>>>>>>>>An idea how to troubleshoot this?
>>>>>>>>>>>>>>>>>Can you run git bisect on the changes between 4.11 and
>>>>>>>>>>>>4.12, so that
>>>>>>>>>>>>>>>>>we find the offending commit? It is probably sufficient
>>>>>>>>>>>>to limit the
>>>>>>>>>>>>>>>>>search to commits that touch something in drivers/char/tpm.
>>>>>>>>>>>>>>>>I'll try and keep you posted.
>>>>>>>>>>>>>>>OK I've been able to bisect the problem 
>>>>>>>>>>>>>>>and the bad commit is:
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 
>>>>>>>>>>>>>>>is the first bad commit
>>>>>>>>>>>>>>>commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>>>>>>>>>>>>>>Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>>>Date:   Mon Mar 27 08:46:04 2017 -0700
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>      tpm_tis: convert to using locality callbacks
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>      This patch converts tpm_tis to use 
>>>>>>>>>>>>>>>of the new tpm class ops
>>>>>>>>>>>>>>>      request_locality, and relinquish_locality.
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>      With the move to using the 
>>>>>>>>>>>>>>>callbacks, release_locality is
>>>>>>>>>>>>>>>changed so
>>>>>>>>>>>>>>>      that we now release the locality even if there is no
>>>>>>>>>>>>>>>request pending.
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>      This required some changes to the tpm_tis_core_init
>>>>>>>>>>>>code path to
>>>>>>>>>>>>>>>      make sure locality is requested when needed:
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>        - tpm2_probe code path will end up calling
>>>>>>>>>>>>>>>request/release through
>>>>>>>>>>>>>>>          callbacks, so request_locality prior to
>>>>>>>>>>>>tpm2_probe not needed.
>>>>>>>>>>>>>>>        - probe_itpm makes calls to 
>>>>>>>>>>>>>>>tpm_tis_send_data which no
>>>>>>>>>>>>>>>longer calls
>>>>>>>>>>>>>>>          request_locality, so add request_locality prior to
>>>>>>>>>>>>>>>tpm_tis_send_data
>>>>>>>>>>>>>>>          calls. Also drop release_locality call in middleof
>>>>>>>>>>>>>>>probe_itpm, and
>>>>>>>>>>>>>>>          keep locality until 
>>>>>>>>>>>>>>>release_locality called at end of
>>>>>>>>>>>>>>>probe_itpm.
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>      Cc: Peter Huewe <peterhuewe@gmx.de>
>>>>>>>>>>>>>>>      Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>      Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>>>>>>>>>>>>>>>      Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>>>>>>>>>>>>>>      Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>>>      Reviewed-by: Jarkko Sakkinen
>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>      Tested-by: Jarkko Sakkinen 
>>>>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>      Signed-off-by: Jarkko Sakkinen
>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>:040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>>>>>>>>>>>>>>72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>I've looked again at the code in question, but could not find
>>>>>>>>>>>>>>anything that is obviously wrong there. Locality is now
>>>>>>>>>>>>>>requested/released at slightly different 
>>>>>>>>>>>>>>points in the process than
>>>>>>>>>>>>>>before, but that's it. It does not seem to 
>>>>>>>>>>>>>>cause problems with the
>>>>>>>>>>>>>>majority of TPMs, since you are the first to report any, so
>>>>>>>>>>>>maybe it
>>>>>>>>>>>>>>is a quirk that only affects this device.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>Perhaps Jerry can help, since this is his change?
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>Alexander
>>>>>>>>>>>>>Getting some caffeine in me, and starting to 
>>>>>>>>>>>>>take a look. Adding
>>>>>>>>>>>>>Jarkko as well since this might involve the 
>>>>>>>>>>>>>general locality changes.
>>>>>>>>>>>>>
>>>>>>>>>>>>>Laurent, if I send you a patch with some 
>>>>>>>>>>>>>debugging code added, would
>>>>>>>>>>>>>you be able to run it on that system? I wasn't 
>>>>>>>>>>>>>running into issues
>>>>>>>>>>>>>on the system I had with a 1.2 device, but I 
>>>>>>>>>>>>>no longer have access
>>>>>>>>>>>>>to it. I'll see if I can find one in our labs 
>>>>>>>>>>>>>and reproduce it there.
>>>>>>>>>>>>Yes I should be able to do that
>>>>>>>>>>>Any findings?
>>>>>>>>>>>
>>>>>>>>>>>/Jarkko
>>>>>>>>>>Okay, finally getting back to this. Looking at the 
>>>>>>>>>>code it isn't clear
>>>>>>>>>>to me
>>>>>>>>>>why the change is causing this. So while I stare at this some more
>>>>>>>>>>Laurent
>>>>>>>>>>could you reproduce it with this patch so I can see 
>>>>>>>>>>what the status and
>>>>>>>>>>access registers look like? Does anyone else on here 
>>>>>>>>>>happen to have a
>>>>>>>>>>Sinosun
>>>>>>>>>>tpm device? The systems I have access to with TPM1.2 
>>>>>>>>>>devices don't have
>>>>>>>>>>this
>>>>>>>>>>issue.
>>>>>>>>>>
>>>>>>>>>>--8<--
>>>>>>>>>>
>>>>>>>>>>diff --git a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>index fdde971bc810..7d60a7e4b50a 100644
>>>>>>>>>>--- a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>+++ b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>@@ -258,6 +258,7 @@ static int 
>>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>>     int rc, status, burstcnt;
>>>>>>>>>>     size_t count = 0;
>>>>>>>>>>     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>>>>>>>>>>+    u8 access;
>>>>>>>>>>
>>>>>>>>>>     status = tpm_tis_status(chip);
>>>>>>>>>>     if ((status & TPM_STS_COMMAND_READY) == 0) {
>>>>>>>>>>@@ -292,6 +293,11 @@ static int 
>>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>>         }
>>>>>>>>>>         status = tpm_tis_status(chip);
>>>>>>>>>>         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>>>>>>>>>>+            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
>>>>>>>>>>&access);
>>>>>>>>>>+            if (rc < 0)
>>>>>>>>>>+                dev_info(&chip->dev, 
>>>>>>>>>>"TPM_STS_DATA_EXPECT == 0: read
>>>>>>>>>>failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>>+            else
>>>>>>>>>>+                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
>>>>>>>>>>locality: %d status: %x access: %x\n", 
>>>>>>>>>>priv->locality, status, access);
>>>>>>>>>>             rc = -EIO;
>>>>>>>>>>             goto out_err;
>>>>>>>>>>         }
>>>>>>>>>>@@ -309,6 +315,11 @@ static int 
>>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>>     }
>>>>>>>>>>     status = tpm_tis_status(chip);
>>>>>>>>>>     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>>>>>>>>>>+        rc = tpm_tis_read8(priv, 
>>>>>>>>>>TPM_ACCESS(priv->locality), &access);
>>>>>>>>>>+        if (rc < 0)
>>>>>>>>>>+            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
>>>>>>>>>>failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>>+        else
>>>>>>>>>>+            dev_info(&chip->dev, 
>>>>>>>>>>"TPM_STS_DATA_EXPECT != 0: locality:
>>>>>>>>>>%d status: %x access: %x\n", priv->locality, status, access);
>>>>>>>>>>         rc = -EIO;
>>>>>>>>>>         goto out_err;
>>>>>>>>>>     }
>>>>>>>>>Please find here the dmesg output of the patched kernel
>>>>>>>>At least 0xff is corrupted value in senseful way. CPU fills the read
>>>>>>>>with ones for example for unaligned bus read. See table 
>>>>>>>>19 in PC client
>>>>>>>>spec. This can happen when you do unaligned read for example.
>>>>>>>>
>>>>>>>>Maybe TPM is unreachable i.e. powered off. Bit busy with 
>>>>>>>>stuff ATM but
>>>>>>>>would probably make sense to compare that 0x81 to table 
>>>>>>>>18 in the same
>>>>>>>>spec.
>>>>>>>>
>>>>>>>>/Jarkko
>>>>>>>0x81 is saying the access register status is valid, and the locality
>>>>>>>is not active. That first bit means A Dynamic OS has not 
>>>>>>>been previously
>>>>>>>established on the platform. Normally we would see 0xa1, which would
>>>>>>>mean valid register status, and the locality is active.
>>>>>>I think the important thing to note here is that STS has bits set that
>>>>>>should never be set. So we can conclude that TPM might be either
>>>>>>
>>>>>>1. Powered off
>>>>>>2. In some transition state?
>>>>>>
>>>>>>/Jarkko
>>>>>If it was powered off would we be getting a valid read from the access
>>>>>register?
>>>>I think there is no universal answer to that :-)
>>>>
>>>>Maybe adding a extra delay would be next test to make? If for random
>>>>reason it is in-between states...
>>>Any more ideas?
>>>
>>>The chip is definitely in a weird state :/ I tried several ways to 
>>>reset the chip (windows, tpm-tools,...).
>>>
>>>I've been able to reset the chip via the bios (which now shows 
>>>unowned) but chip is still locked apparently.
>>>
>>>But still with < 4.12 I'm able to get /some/ information Public 
>>>EK, PCR,... out of the chip so it's not completely broken...
>>
>>OK correction the TPM is now unlocked (I let the computer running 
>>for more than 24h with nothing accessing the TPM) and with 4.9 I've 
>>been able to take the ownership again.
>>
>>Under 4.12 I still have the same errors as mentioned originally
>
>Hi Laurent,
>
>Would it be possible for you to run ftrace from boot with the following kernel parameters:
>
>ftrace=function_graph ftrace_filter=tpm*
>

actually 'ftrace=function_graph ftrace_filter=tpm*,*locality,wait_for_tpm_stat' would be better

>and then send me the results of 'cat /sys/kernel/debug/tracing/trace' ?

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: multipart/mixed; boundary="------------D033D66A59EE10C39792CEF7", Size: 11188 bytes --]

Le 10/11/17 a 08:07, Jerry Snitselaar a ecrit :
> On Thu Nov 09 17, Jerry Snitselaar wrote:
>> On Thu Nov 09 17, Laurent Bigonville wrote:
>>> Le 09/11/17 a 01:04, Laurent Bigonville a ecrit :
>>>>
>>>>
>>>> Le 24/10/17 a 18:07, Jarkko Sakkinen a ecrit :
>>>>> On Tue, Oct 24, 2017 at 07:57:06AM -0700, Jerry Snitselaar wrote:
>>>>>> On Tue Oct 24 17, Jarkko Sakkinen wrote:
>>>>>>> On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
>>>>>>>> On Mon Oct 23 17, Jarkko Sakkinen wrote:
>>>>>>>>> On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville 
>>>>>>>>> wrote:
>>>>>>>>>> Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>>>>>>>>>>> On Wed Sep 06 17, Jarkko Sakkinen wrote:
>>>>>>>>>>>> On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent 
>>>>>>>>>>>> Bigonville wrote:
>>>>>>>>>>>>> Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>>>>>>>>>>>>> On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>>>>>>>>>>>>>>> Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>>>>>>>>>>>>>>>> Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a 
>>>>>>>>>>>>>>>>> ecrit :
>>>>>>>>>>>>>>>>>>> An idea how to troubleshoot this?
>>>>>>>>>>>>>>>>>> Can you run git bisect on the changes between 4.11 and
>>>>>>>>>>>>> 4.12, so that
>>>>>>>>>>>>>>>>>> we find the offending commit? It is probably sufficient
>>>>>>>>>>>>> to limit the
>>>>>>>>>>>>>>>>>> search to commits that touch something in 
>>>>>>>>>>>>>>>>>> drivers/char/tpm.
>>>>>>>>>>>>>>>>> I'll try and keep you posted.
>>>>>>>>>>>>>>>> OK I've been able to bisect the problem and the bad 
>>>>>>>>>>>>>>>> commit is:
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first 
>>>>>>>>>>>>>>>> bad commit
>>>>>>>>>>>>>>>> commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>>>>>>>>>>>>>>> Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>>>> Date:   Mon Mar 27 08:46:04 2017 -0700
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>       tpm_tis: convert to using locality callbacks
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>       This patch converts tpm_tis to use of the new tpm 
>>>>>>>>>>>>>>>> class ops
>>>>>>>>>>>>>>>>       request_locality, and relinquish_locality.
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>       With the move to using the callbacks, 
>>>>>>>>>>>>>>>> release_locality is
>>>>>>>>>>>>>>>> changed so
>>>>>>>>>>>>>>>>       that we now release the locality even if there is no
>>>>>>>>>>>>>>>> request pending.
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>       This required some changes to the tpm_tis_core_init
>>>>>>>>>>>>> code path to
>>>>>>>>>>>>>>>>       make sure locality is requested when needed:
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>         - tpm2_probe code path will end up calling
>>>>>>>>>>>>>>>> request/release through
>>>>>>>>>>>>>>>>           callbacks, so request_locality prior to
>>>>>>>>>>>>> tpm2_probe not needed.
>>>>>>>>>>>>>>>>         - probe_itpm makes calls to tpm_tis_send_data 
>>>>>>>>>>>>>>>> which no
>>>>>>>>>>>>>>>> longer calls
>>>>>>>>>>>>>>>>           request_locality, so add request_locality 
>>>>>>>>>>>>>>>> prior to
>>>>>>>>>>>>>>>> tpm_tis_send_data
>>>>>>>>>>>>>>>>           calls. Also drop release_locality call in 
>>>>>>>>>>>>>>>> middleof
>>>>>>>>>>>>>>>> probe_itpm, and
>>>>>>>>>>>>>>>>           keep locality until release_locality called 
>>>>>>>>>>>>>>>> at end of
>>>>>>>>>>>>>>>> probe_itpm.
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>       Cc: Peter Huewe <peterhuewe@gmx.de>
>>>>>>>>>>>>>>>>       Cc: Jarkko Sakkinen 
>>>>>>>>>>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>>       Cc: Jason Gunthorpe 
>>>>>>>>>>>>>>>> <jgunthorpe@obsidianresearch.com>
>>>>>>>>>>>>>>>>       Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>>>>>>>>>>>>>>>       Signed-off-by: Jerry Snitselaar 
>>>>>>>>>>>>>>>> <jsnitsel@redhat.com>
>>>>>>>>>>>>>>>>       Reviewed-by: Jarkko Sakkinen
>>>>>>>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>> Tested-by: Jarkko Sakkinen 
>>>>>>>>>>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>>       Signed-off-by: Jarkko Sakkinen
>>>>>>>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>> :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>>>>>>>>>>>>>>> 72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> I've looked again at the code in question, but could not 
>>>>>>>>>>>>>>> find
>>>>>>>>>>>>>>> anything that is obviously wrong there. Locality is now
>>>>>>>>>>>>>>> requested/released at slightly different points in the 
>>>>>>>>>>>>>>> process than
>>>>>>>>>>>>>>> before, but that's it. It does not seem to cause 
>>>>>>>>>>>>>>> problems with the
>>>>>>>>>>>>>>> majority of TPMs, since you are the first to report any, so
>>>>>>>>>>>>> maybe it
>>>>>>>>>>>>>>> is a quirk that only affects this device.
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> Perhaps Jerry can help, since this is his change?
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> Alexander
>>>>>>>>>>>>>> Getting some caffeine in me, and starting to take a look. 
>>>>>>>>>>>>>> Adding
>>>>>>>>>>>>>> Jarkko as well since this might involve the general 
>>>>>>>>>>>>>> locality changes.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> Laurent, if I send you a patch with some debugging code 
>>>>>>>>>>>>>> added, would
>>>>>>>>>>>>>> you be able to run it on that system? I wasn't running 
>>>>>>>>>>>>>> into issues
>>>>>>>>>>>>>> on the system I had with a 1.2 device, but I no longer 
>>>>>>>>>>>>>> have access
>>>>>>>>>>>>>> to it. I'll see if I can find one in our labs and 
>>>>>>>>>>>>>> reproduce it there.
>>>>>>>>>>>>> Yes I should be able to do that
>>>>>>>>>>>> Any findings?
>>>>>>>>>>>>
>>>>>>>>>>>> /Jarkko
>>>>>>>>>>> Okay, finally getting back to this. Looking at the code it 
>>>>>>>>>>> isn't clear
>>>>>>>>>>> to me
>>>>>>>>>>> why the change is causing this. So while I stare at this 
>>>>>>>>>>> some more
>>>>>>>>>>> Laurent
>>>>>>>>>>> could you reproduce it with this patch so I can see what the 
>>>>>>>>>>> status and
>>>>>>>>>>> access registers look like? Does anyone else on here happen 
>>>>>>>>>>> to have a
>>>>>>>>>>> Sinosun
>>>>>>>>>>> tpm device? The systems I have access to with TPM1.2 devices 
>>>>>>>>>>> don't have
>>>>>>>>>>> this
>>>>>>>>>>> issue.
>>>>>>>>>>>
>>>>>>>>>>> --8<--
>>>>>>>>>>>
>>>>>>>>>>> diff --git a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>> b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>> index fdde971bc810..7d60a7e4b50a 100644
>>>>>>>>>>> --- a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>> +++ b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>> @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct 
>>>>>>>>>>> tpm_chip *chip,
>>>>>>>>>>> const u8 *buf, size_t len)
>>>>>>>>>>>      int rc, status, burstcnt;
>>>>>>>>>>>      size_t count = 0;
>>>>>>>>>>>      bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>>>>>>>>>>> +    u8 access;
>>>>>>>>>>>
>>>>>>>>>>>      status = tpm_tis_status(chip);
>>>>>>>>>>>      if ((status & TPM_STS_COMMAND_READY) == 0) {
>>>>>>>>>>> @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct 
>>>>>>>>>>> tpm_chip *chip,
>>>>>>>>>>> const u8 *buf, size_t len)
>>>>>>>>>>>          }
>>>>>>>>>>>          status = tpm_tis_status(chip);
>>>>>>>>>>>          if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>>>>>>>>>>> +            rc = tpm_tis_read8(priv, 
>>>>>>>>>>> TPM_ACCESS(priv->locality),
>>>>>>>>>>> &access);
>>>>>>>>>>> +            if (rc < 0)
>>>>>>>>>>> +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT 
>>>>>>>>>>> == 0: read
>>>>>>>>>>> failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>>> +            else
>>>>>>>>>>> +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT 
>>>>>>>>>>> == 0:
>>>>>>>>>>> locality: %d status: %x access: %x\n", priv->locality, 
>>>>>>>>>>> status, access);
>>>>>>>>>>>              rc = -EIO;
>>>>>>>>>>>              goto out_err;
>>>>>>>>>>>          }
>>>>>>>>>>> @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct 
>>>>>>>>>>> tpm_chip *chip,
>>>>>>>>>>> const u8 *buf, size_t len)
>>>>>>>>>>>      }
>>>>>>>>>>>      status = tpm_tis_status(chip);
>>>>>>>>>>>      if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>>>>>>>>>>> +        rc = tpm_tis_read8(priv, 
>>>>>>>>>>> TPM_ACCESS(priv->locality), &access);
>>>>>>>>>>> +        if (rc < 0)
>>>>>>>>>>> +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: 
>>>>>>>>>>> read
>>>>>>>>>>> failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>>> +        else
>>>>>>>>>>> +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: 
>>>>>>>>>>> locality:
>>>>>>>>>>> %d status: %x access: %x\n", priv->locality, status, access);
>>>>>>>>>>>          rc = -EIO;
>>>>>>>>>>>          goto out_err;
>>>>>>>>>>>      }
>>>>>>>>>> Please find here the dmesg output of the patched kernel
>>>>>>>>> At least 0xff is corrupted value in senseful way. CPU fills 
>>>>>>>>> the read
>>>>>>>>> with ones for example for unaligned bus read. See table 19 in 
>>>>>>>>> PC client
>>>>>>>>> spec. This can happen when you do unaligned read for example.
>>>>>>>>>
>>>>>>>>> Maybe TPM is unreachable i.e. powered off. Bit busy with stuff 
>>>>>>>>> ATM but
>>>>>>>>> would probably make sense to compare that 0x81 to table 18 in 
>>>>>>>>> the same
>>>>>>>>> spec.
>>>>>>>>>
>>>>>>>>> /Jarkko
>>>>>>>> 0x81 is saying the access register status is valid, and the 
>>>>>>>> locality
>>>>>>>> is not active. That first bit means A Dynamic OS has not been 
>>>>>>>> previously
>>>>>>>> established on the platform. Normally we would see 0xa1, which 
>>>>>>>> would
>>>>>>>> mean valid register status, and the locality is active.
>>>>>>> I think the important thing to note here is that STS has bits 
>>>>>>> set that
>>>>>>> should never be set. So we can conclude that TPM might be either
>>>>>>>
>>>>>>> 1. Powered off
>>>>>>> 2. In some transition state?
>>>>>>>
>>>>>>> /Jarkko
>>>>>> If it was powered off would we be getting a valid read from the 
>>>>>> access
>>>>>> register?
>>>>> I think there is no universal answer to that :-)
>>>>>
>>>>> Maybe adding a extra delay would be next test to make? If for random
>>>>> reason it is in-between states...
>>>> Any more ideas?
>>>>
>>>> The chip is definitely in a weird state :/ I tried several ways to 
>>>> reset the chip (windows, tpm-tools,...).
>>>>
>>>> I've been able to reset the chip via the bios (which now shows 
>>>> unowned) but chip is still locked apparently.
>>>>
>>>> But still with < 4.12 I'm able to get /some/ information Public EK, 
>>>> PCR,... out of the chip so it's not completely broken...
>>>
>>> OK correction the TPM is now unlocked (I let the computer running 
>>> for more than 24h with nothing accessing the TPM) and with 4.9 I've 
>>> been able to take the ownership again.
>>>
>>> Under 4.12 I still have the same errors as mentioned originally
>>
>> Hi Laurent,
>>
>> Would it be possible for you to run ftrace from boot with the 
>> following kernel parameters:
>>
>> ftrace=function_graph ftrace_filter=tpm*
>>
>
> actually 'ftrace=function_graph 
> ftrace_filter=tpm*,*locality,wait_for_tpm_stat' would be better
>
>> and then send me the results of 'cat /sys/kernel/debug/tracing/trace' ?
Here you are.


    [ Part 2, Text/PLAIN (Name: "trace.txt") ~8 KB. ]
    [ Unable to print this part. ]

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Fri Nov 10 17, Laurent Bigonville wrote:
>Le 10/11/17 a 08:07, Jerry Snitselaar a ecrit :
>>On Thu Nov 09 17, Jerry Snitselaar wrote:
>>>On Thu Nov 09 17, Laurent Bigonville wrote:
>>>>Le 09/11/17 a 01:04, Laurent Bigonville a ecrit :
>>>>>
>>>>>
>>>>>Le 24/10/17 a 18:07, Jarkko Sakkinen a ecrit :
>>>>>>On Tue, Oct 24, 2017 at 07:57:06AM -0700, Jerry Snitselaar wrote:
>>>>>>>On Tue Oct 24 17, Jarkko Sakkinen wrote:
>>>>>>>>On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
>>>>>>>>>On Mon Oct 23 17, Jarkko Sakkinen wrote:
>>>>>>>>>>On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent 
>>>>>>>>>>Bigonville wrote:
>>>>>>>>>>>Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>>>>>>>>>>>>On Wed Sep 06 17, Jarkko Sakkinen wrote:
>>>>>>>>>>>>>On Fri, Sep 01, 2017 at 02:10:18PM +0200, 
>>>>>>>>>>>>>Laurent Bigonville wrote:
>>>>>>>>>>>>>>Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>>>>>>>>>>>>>>On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>>>>>>>>>>>>>>>>Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>>>>>>>>>>>>>>>>>Le 29/08/17 a 18:00, 
>>>>>>>>>>>>>>>>>>Alexander.Steffen@infineon.com a 
>>>>>>>>>>>>>>>>>>ecrit :
>>>>>>>>>>>>>>>>>>>>An idea how to troubleshoot this?
>>>>>>>>>>>>>>>>>>>Can you run git bisect on the changes between 4.11 and
>>>>>>>>>>>>>>4.12, so that
>>>>>>>>>>>>>>>>>>>we find the offending commit? It is probably sufficient
>>>>>>>>>>>>>>to limit the
>>>>>>>>>>>>>>>>>>>search to commits that touch 
>>>>>>>>>>>>>>>>>>>something in drivers/char/tpm.
>>>>>>>>>>>>>>>>>>I'll try and keep you posted.
>>>>>>>>>>>>>>>>>OK I've been able to bisect the 
>>>>>>>>>>>>>>>>>problem and the bad commit is:
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 
>>>>>>>>>>>>>>>>>is the first bad commit
>>>>>>>>>>>>>>>>>commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>>>>>>>>>>>>>>>>Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>>>>>>>>Date:   Mon Mar 27 08:46:04 2017 -0700
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>      tpm_tis: convert to using locality callbacks
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>      This patch converts tpm_tis to 
>>>>>>>>>>>>>>>>>use of the new tpm class ops
>>>>>>>>>>>>>>>>>      request_locality, and relinquish_locality.
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>      With the move to using the 
>>>>>>>>>>>>>>>>>callbacks, release_locality is
>>>>>>>>>>>>>>>>>changed so
>>>>>>>>>>>>>>>>>      that we now release the locality even if there is no
>>>>>>>>>>>>>>>>>request pending.
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>      This required some changes to the tpm_tis_core_init
>>>>>>>>>>>>>>code path to
>>>>>>>>>>>>>>>>>      make sure locality is requested when needed:
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>        - tpm2_probe code path will end up calling
>>>>>>>>>>>>>>>>>request/release through
>>>>>>>>>>>>>>>>>          callbacks, so request_locality prior to
>>>>>>>>>>>>>>tpm2_probe not needed.
>>>>>>>>>>>>>>>>>        - probe_itpm makes calls to 
>>>>>>>>>>>>>>>>>tpm_tis_send_data which no
>>>>>>>>>>>>>>>>>longer calls
>>>>>>>>>>>>>>>>>          request_locality, so add 
>>>>>>>>>>>>>>>>>request_locality prior to
>>>>>>>>>>>>>>>>>tpm_tis_send_data
>>>>>>>>>>>>>>>>>          calls. Also drop 
>>>>>>>>>>>>>>>>>release_locality call in middleof
>>>>>>>>>>>>>>>>>probe_itpm, and
>>>>>>>>>>>>>>>>>          keep locality until 
>>>>>>>>>>>>>>>>>release_locality called at end of
>>>>>>>>>>>>>>>>>probe_itpm.
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>      Cc: Peter Huewe <peterhuewe@gmx.de>
>>>>>>>>>>>>>>>>>      Cc: Jarkko Sakkinen 
>>>>>>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>>>      Cc: Jason Gunthorpe 
>>>>>>>>>>>>>>>>><jgunthorpe@obsidianresearch.com>
>>>>>>>>>>>>>>>>>      Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>>>>>>>>>>>>>>>>      Signed-off-by: Jerry Snitselaar 
>>>>>>>>>>>>>>>>><jsnitsel@redhat.com>
>>>>>>>>>>>>>>>>>      Reviewed-by: Jarkko Sakkinen
>>>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>>>Tested-by: Jarkko Sakkinen 
>>>>>>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>>>      Signed-off-by: Jarkko Sakkinen
>>>>>>>>>>>>>><jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>>>>>>>>:040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>>>>>>>>>>>>>>>>72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>I've looked again at the code in 
>>>>>>>>>>>>>>>>question, but could not find
>>>>>>>>>>>>>>>>anything that is obviously wrong there. Locality is now
>>>>>>>>>>>>>>>>requested/released at slightly different 
>>>>>>>>>>>>>>>>points in the process than
>>>>>>>>>>>>>>>>before, but that's it. It does not seem 
>>>>>>>>>>>>>>>>to cause problems with the
>>>>>>>>>>>>>>>>majority of TPMs, since you are the first to report any, so
>>>>>>>>>>>>>>maybe it
>>>>>>>>>>>>>>>>is a quirk that only affects this device.
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>Perhaps Jerry can help, since this is his change?
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>Alexander
>>>>>>>>>>>>>>>Getting some caffeine in me, and starting 
>>>>>>>>>>>>>>>to take a look. Adding
>>>>>>>>>>>>>>>Jarkko as well since this might involve 
>>>>>>>>>>>>>>>the general locality changes.
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>Laurent, if I send you a patch with some 
>>>>>>>>>>>>>>>debugging code added, would
>>>>>>>>>>>>>>>you be able to run it on that system? I 
>>>>>>>>>>>>>>>wasn't running into issues
>>>>>>>>>>>>>>>on the system I had with a 1.2 device, but 
>>>>>>>>>>>>>>>I no longer have access
>>>>>>>>>>>>>>>to it. I'll see if I can find one in our 
>>>>>>>>>>>>>>>labs and reproduce it there.
>>>>>>>>>>>>>>Yes I should be able to do that
>>>>>>>>>>>>>Any findings?
>>>>>>>>>>>>>
>>>>>>>>>>>>>/Jarkko
>>>>>>>>>>>>Okay, finally getting back to this. Looking at 
>>>>>>>>>>>>the code it isn't clear
>>>>>>>>>>>>to me
>>>>>>>>>>>>why the change is causing this. So while I stare 
>>>>>>>>>>>>at this some more
>>>>>>>>>>>>Laurent
>>>>>>>>>>>>could you reproduce it with this patch so I can 
>>>>>>>>>>>>see what the status and
>>>>>>>>>>>>access registers look like? Does anyone else on 
>>>>>>>>>>>>here happen to have a
>>>>>>>>>>>>Sinosun
>>>>>>>>>>>>tpm device? The systems I have access to with 
>>>>>>>>>>>>TPM1.2 devices don't have
>>>>>>>>>>>>this
>>>>>>>>>>>>issue.
>>>>>>>>>>>>
>>>>>>>>>>>>--8<--
>>>>>>>>>>>>
>>>>>>>>>>>>diff --git a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>>>b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>>>index fdde971bc810..7d60a7e4b50a 100644
>>>>>>>>>>>>--- a/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>>>+++ b/drivers/char/tpm/tpm_tis_core.c
>>>>>>>>>>>>@@ -258,6 +258,7 @@ static int 
>>>>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>>>>     int rc, status, burstcnt;
>>>>>>>>>>>>     size_t count = 0;
>>>>>>>>>>>>     bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>>>>>>>>>>>>+    u8 access;
>>>>>>>>>>>>
>>>>>>>>>>>>     status = tpm_tis_status(chip);
>>>>>>>>>>>>     if ((status & TPM_STS_COMMAND_READY) == 0) {
>>>>>>>>>>>>@@ -292,6 +293,11 @@ static int 
>>>>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>>>>         }
>>>>>>>>>>>>         status = tpm_tis_status(chip);
>>>>>>>>>>>>         if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>>>>>>>>>>>>+            rc = tpm_tis_read8(priv, 
>>>>>>>>>>>>TPM_ACCESS(priv->locality),
>>>>>>>>>>>>&access);
>>>>>>>>>>>>+            if (rc < 0)
>>>>>>>>>>>>+                dev_info(&chip->dev, 
>>>>>>>>>>>>"TPM_STS_DATA_EXPECT == 0: read
>>>>>>>>>>>>failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>>>>+            else
>>>>>>>>>>>>+                dev_info(&chip->dev, 
>>>>>>>>>>>>"TPM_STS_DATA_EXPECT == 0:
>>>>>>>>>>>>locality: %d status: %x access: %x\n", 
>>>>>>>>>>>>priv->locality, status, access);
>>>>>>>>>>>>             rc = -EIO;
>>>>>>>>>>>>             goto out_err;
>>>>>>>>>>>>         }
>>>>>>>>>>>>@@ -309,6 +315,11 @@ static int 
>>>>>>>>>>>>tpm_tis_send_data(struct tpm_chip *chip,
>>>>>>>>>>>>const u8 *buf, size_t len)
>>>>>>>>>>>>     }
>>>>>>>>>>>>     status = tpm_tis_status(chip);
>>>>>>>>>>>>     if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>>>>>>>>>>>>+        rc = tpm_tis_read8(priv, 
>>>>>>>>>>>>TPM_ACCESS(priv->locality), &access);
>>>>>>>>>>>>+        if (rc < 0)
>>>>>>>>>>>>+            dev_info(&chip->dev, 
>>>>>>>>>>>>"TPM_STS_DATA_EXPECT != 0: read
>>>>>>>>>>>>failure TPM_ACCESS(%d)\n", priv->locality);
>>>>>>>>>>>>+        else
>>>>>>>>>>>>+            dev_info(&chip->dev, 
>>>>>>>>>>>>"TPM_STS_DATA_EXPECT != 0: locality:
>>>>>>>>>>>>%d status: %x access: %x\n", priv->locality, status, access);
>>>>>>>>>>>>         rc = -EIO;
>>>>>>>>>>>>         goto out_err;
>>>>>>>>>>>>     }
>>>>>>>>>>>Please find here the dmesg output of the patched kernel
>>>>>>>>>>At least 0xff is corrupted value in senseful way. 
>>>>>>>>>>CPU fills the read
>>>>>>>>>>with ones for example for unaligned bus read. See 
>>>>>>>>>>table 19 in PC client
>>>>>>>>>>spec. This can happen when you do unaligned read for example.
>>>>>>>>>>
>>>>>>>>>>Maybe TPM is unreachable i.e. powered off. Bit busy 
>>>>>>>>>>with stuff ATM but
>>>>>>>>>>would probably make sense to compare that 0x81 to 
>>>>>>>>>>table 18 in the same
>>>>>>>>>>spec.
>>>>>>>>>>
>>>>>>>>>>/Jarkko
>>>>>>>>>0x81 is saying the access register status is valid, 
>>>>>>>>>and the locality
>>>>>>>>>is not active. That first bit means A Dynamic OS has 
>>>>>>>>>not been previously
>>>>>>>>>established on the platform. Normally we would see 
>>>>>>>>>0xa1, which would
>>>>>>>>>mean valid register status, and the locality is active.
>>>>>>>>I think the important thing to note here is that STS has 
>>>>>>>>bits set that
>>>>>>>>should never be set. So we can conclude that TPM might be either
>>>>>>>>
>>>>>>>>1. Powered off
>>>>>>>>2. In some transition state?
>>>>>>>>
>>>>>>>>/Jarkko
>>>>>>>If it was powered off would we be getting a valid read 
>>>>>>>from the access
>>>>>>>register?
>>>>>>I think there is no universal answer to that :-)
>>>>>>
>>>>>>Maybe adding a extra delay would be next test to make? If for random
>>>>>>reason it is in-between states...
>>>>>Any more ideas?
>>>>>
>>>>>The chip is definitely in a weird state :/ I tried several 
>>>>>ways to reset the chip (windows, tpm-tools,...).
>>>>>
>>>>>I've been able to reset the chip via the bios (which now shows 
>>>>>unowned) but chip is still locked apparently.
>>>>>
>>>>>But still with < 4.12 I'm able to get /some/ information 
>>>>>Public EK, PCR,... out of the chip so it's not completely 
>>>>>broken...
>>>>
>>>>OK correction the TPM is now unlocked (I let the computer 
>>>>running for more than 24h with nothing accessing the TPM) and 
>>>>with 4.9 I've been able to take the ownership again.
>>>>
>>>>Under 4.12 I still have the same errors as mentioned originally
>>>
>>>Hi Laurent,
>>>
>>>Would it be possible for you to run ftrace from boot with the 
>>>following kernel parameters:
>>>
>>>ftrace=function_graph ftrace_filter=tpm*
>>>
>>
>>actually 'ftrace=function_graph 
>>ftrace_filter=tpm*,*locality,wait_for_tpm_stat' would be better
>>
>>>and then send me the results of 'cat /sys/kernel/debug/tracing/trace' ?
>Here you are.

Thank you Laurent.

A few interesting things here:

># tracer: function_graph
>#
># CPU  DURATION                  FUNCTION CALLS
># |     |   |                     |   |   |   |
>  1)   4.510 us    |  tpm_init();
>  1)               |  tpm_tis_pnp_init() {
>  1)               |    tpm_tis_init.part.3() {
>  1)               |      tpm_tis_core_init() {
>  1)               |        tpmm_chip_alloc() {
>  1)   3.935 us    |          tpm_chip_alloc();
>  1)   4.787 us    |        }
>  1)   0.199 us    |        tpm_tcg_read_bytes();
>  1)   0.217 us    |        tpm_tcg_read32();
>  1)   0.117 us    |        tpm_tcg_write32();
>  1)               |        tpm2_probe() {
>  1)               |          tpm_transmit_cmd() {
>  1)               |            tpm_transmit() {
>  1)               |              request_locality() {
>  1)               |                check_locality() {
>  1)   0.045 us    |                  tpm_tcg_read_bytes();
>  1)   1.708 us    |                }
>  1)   2.077 us    |              }


So it goes request locality, does check to see if the locality
is already active and it is, so doesn't go through the process
of requesting the locality.


>  1)   0.168 us    |              tpm2_prepare_space();
>  1)               |              tpm_tis_send() {
>  1)               |                tpm_tis_send_main() {
>  1)               |                  tpm_tis_send_data() {
>  1)   0.045 us    |                    tpm_tcg_read_bytes();
>  1)   0.044 us    |                    tpm_tcg_read32();
>  1)   1.723 us    |                    tpm_tcg_write_bytes();
>  1)               |                    wait_for_tpm_stat() {
>  1)               |                      tpm_tis_status() {
>  1)   0.042 us    |                        tpm_tcg_read_bytes();
>  1)   5.163 us    |                      }
>  1)               |                      tpm_tis_status() {
>  1)   0.167 us    |                        tpm_tcg_read_bytes();
>  1)   2.182 us    |                      }
>  1)               |                      tpm_tis_status() {
>  1)   0.194 us    |                        tpm_tcg_read_bytes();
>  1)   2.452 us    |                      }
>  1) * 30406.24 us |                    }
>  1)   0.088 us    |                    tpm_tcg_read_bytes();
>  1)   0.123 us    |                    tpm_tcg_write_bytes();
>  1)               |                    wait_for_tpm_stat() {
>  1)               |                      tpm_tis_status() {
>  1)   0.084 us    |                        tpm_tcg_read_bytes();
>  1)   1.940 us    |                      }
>  1)               |                      tpm_tis_status() {
>  1)   0.170 us    |                        tpm_tcg_read_bytes();
>  1)   2.217 us    |                      }
>  1) * 16000.74 us |                    }
>  1)   0.084 us    |                    tpm_tcg_read_bytes();
>  1) * 46425.66 us |                  }
>  1)   0.115 us    |                  tpm_tcg_write_bytes();
>  1) * 46426.85 us |                }
>  1) * 46427.35 us |              }
>  1)               |              tpm_tis_status() {
>  1)   1.389 us    |                tpm_tcg_read_bytes();
>  1)   1.909 us    |              }
>  1)   0.095 us    |              tpm_tis_req_canceled();
>  1)               |              tpm_tis_status() {
>  1)   0.187 us    |                tpm_tcg_read_bytes();
>  1)   2.110 us    |              }
>  1)               |              tpm_tis_recv() {
>  1)               |                wait_for_tpm_stat() {
>  1)               |                  tpm_tis_status() {
>  1)   0.082 us    |                    tpm_tcg_read_bytes();
>  1)   1.898 us    |                  }
>  1)   2.425 us    |                }
>  1)   0.102 us    |                tpm_tcg_read32();
>  1)   5.453 us    |                tpm_tcg_read_bytes();
>  1)               |                wait_for_tpm_stat() {
>  1)               |                  tpm_tis_status() {
>  1)   0.092 us    |                    tpm_tcg_read_bytes();
>  1)   1.858 us    |                  }
>  1)   2.332 us    |                }
>  1)   0.077 us    |                tpm_tcg_read_bytes();
>  1)   0.112 us    |                tpm_tcg_write_bytes();
>  1) + 25.466 us   |              }
>  1)   0.152 us    |              tpm2_commit_space();
>  1)               |              release_locality() {
>  1)   0.080 us    |                tpm_tcg_write_bytes();
>  1)   0.672 us    |              }

Releases the locality as it comes back out of tpm_transmit_cmd

>  1) * 62349.91 us |            }
>  1) * 62350.38 us |          }
>  1) * 62350.87 us |        }
>  1)   0.070 us    |        tpm_tcg_read32();
>  1)   0.073 us    |        tpm_tcg_read_bytes();
>  1)   0.096 us    |        tpm_tcg_read16();
>  1)   0.042 us    |        tpm_tcg_read32();
>  1)               |        tpm_chip_register() {
>  1)               |          tpm1_auto_startup() {
>  1)               |            tpm_get_timeouts() {
>  1)               |              tpm_get_timeouts.part.6() {
>  1)               |                tpm_getcap() {
>  1)               |                  tpm_transmit_cmd() {
>  1)               |                    tpm_transmit() {
>  1)               |                      request_locality() {
>  1)               |                        check_locality() {
>  1)   0.091 us    |                          tpm_tcg_read_bytes();
>  1)   1.680 us    |                        }
>  1)   1.973 us    |                      }

Goes to request the locality again, but once again the check sees
that the locality is already active, so doesn't go through process
of requesting the locality. Very odd.

>  1)   0.072 us    |                      tpm2_prepare_space();
>  1)               |                      tpm_tis_send() {
>  1)               |                        tpm_tis_send_main() {
>  1)               |                          tpm_tis_send_data() {
>  1)   0.043 us    |                            tpm_tcg_read_bytes();

tpm_tis_stat inlined?

>  1)   0.071 us    |                            tpm_tcg_write_bytes();

tpm_tis_ready inlined?

>  1)               |                            wait_for_tpm_stat() {
>  1)               |                              tpm_tis_status() {
>  1)   0.042 us    |                                tpm_tcg_read_bytes();
>  1)   1.634 us    |                              }
>  1)               |                              tpm_tis_status() {
>  1)   0.183 us    |                                tpm_tcg_read_bytes();
>  1)   2.121 us    |                              }
>  1) * 15942.66 us |                            }

The wait_for_tpm_stat does its loop twice checking the status register
so at this point apparently it isn't reading 0xff.

>  1)   0.108 us    |                            tpm_tcg_read32();

This read is get_burstcount.

>  1)   1.773 us    |                            tpm_tcg_write_bytes();
>  1)               |                            wait_for_tpm_stat() {
>  1)               |                              tpm_tis_status() {
>  1)   0.082 us    |                                tpm_tcg_read_bytes();
>  1)   5.011 us    |                              }
>  1)   5.618 us    |                            }
>  1)   0.077 us    |                            tpm_tcg_read_bytes();

tpm_tis_status inlined?

>  1)   0.090 us    |                            tpm_tcg_write_bytes();

Come out of loop and write last byte

>  1)               |                            wait_for_tpm_stat() {
>  1)               |                              tpm_tis_status() {
>  1)   0.080 us    |                                tpm_tcg_read_bytes();
>  1)   1.921 us    |                              }
>  1)   2.376 us    |                            }
>  1)   0.077 us    |                            tpm_tcg_read_bytes();

inlined tpm_tis_status again? Here is where it sees data is still expected and errors out.
With that debugging patch at this point it was showing that the locality was not active.
Also didn't show that the tpm had been seized by another locality.

>  1)   0.080 us    |                            tpm_tcg_write_bytes();

tpm_tis_ready inlined? in out_err

>  1) * 15970.97 us |                          }
>  1) * 15971.39 us |                        }
>  1) * 15971.81 us |                      }
>  1)               |                      release_locality() {
>  1)   0.103 us    |                        tpm_tcg_write_bytes();
>  1)   0.535 us    |                      }
>  1) * 16023.04 us |                    }
>  1) * 16023.35 us |                  }
>  1) * 16024.79 us |                }
>  1) * 16067.18 us |              }
>  1) * 16067.67 us |            }
>  1) * 16068.10 us |          }
>  1) * 16068.47 us |        }
>  1) * 78448.69 us |      }
>  1) * 78455.45 us |    }
>  1) * 78456.71 us |  }
>  1)   0.564 us    |  tpm_dev_release();
>  0)               |  tpm_pcr_read() {
>  0)   0.225 us    |    tpm_chip_find_get();
>  0)   0.734 us    |  }

I wonder if it is possible that the release locality from the probe
isn't completing on the chip until after the request/check that
happens at the beginning of the tpm_get_timeouts.  Perhaps we need
something like wait_for_tpm_stat for the access register, and
verifying that locality was released before returning out of
release_locality?

Any thoughts?

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jason Gunthorpe]

On Fri, Nov 10, 2017 at 01:53:00PM -0700, Jerry Snitselaar wrote:

> I wonder if it is possible that the release locality from the probe
> isn't completing on the chip until after the request/check that
> happens at the beginning of the tpm_get_timeouts.  Perhaps we need
> something like wait_for_tpm_stat for the access register, and
> verifying that locality was released before returning out of
> release_locality?

This is not a bad guess.. Adding a largish delay after
release_locality might be interesting too.

Are we sure our tis request/release process is even right? Another
options is that the 'check if already in locality' doesn't work on
this chip, or isn't coded right...

Jason

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Sat Nov 11 17, Jason Gunthorpe wrote:
>On Fri, Nov 10, 2017 at 01:53:00PM -0700, Jerry Snitselaar wrote:
>
>> I wonder if it is possible that the release locality from the probe
>> isn't completing on the chip until after the request/check that
>> happens at the beginning of the tpm_get_timeouts.  Perhaps we need
>> something like wait_for_tpm_stat for the access register, and
>> verifying that locality was released before returning out of
>> release_locality?
>
>This is not a bad guess.. Adding a largish delay after
>release_locality might be interesting too.
>
>Are we sure our tis request/release process is even right? Another

At this point, no? :) The tpm1.2 and tpm2.0 tis devices I have
access to seem to behave properly, and the traces look like I'd
expect, but that doesn't prove anything.

I think the check_locality code is okay since it is looking at the
access register for the desired locality and seeing if the active
locality and valid register bits are set. According to the fifo
interface locality usage per register chart, the access register
should always return correct values unlike the status register.

For request_locality, currently it sets the request use bit in the
access register, and then does check_locality potentially up until
timeout a. Looking at the spec, it could take longer than timeout a
if there is another active locality. It can take up until timeout a
from the point where that locality relinquishes the tpm. So that
could possibly be improved.

Before the release_locality code would only actually release the
locality if the request use bit was set. So after it grabbed the
locality during probe it probably never released it. The idea with the
new code was to release it when it was no longer needed so another
requester would be able to take the tpm without having to wait for it
to be released. With the old code I think it would have to wait either
until the next time release_locality was called, or attempt to seize
the tpm with the seize bit in the access register. I need to read
through the spec some more, but does the tpm ever force a change when
the request use bit is set, or does it leave it up to the software
to deal with it and only gets involved in the case where the seize
bit has been set?

/me goes back to reading specs

>options is that the 'check if already in locality' doesn't work on
>this chip, or isn't coded right...
>
>Jason

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jason Gunthorpe]

On Sat, Nov 11, 2017 at 12:12:57PM -0700, Jerry Snitselaar wrote:

> Before the release_locality code would only actually release the
> locality if the request use bit was set. So after it grabbed the
> locality during probe it probably never released it. The idea with the
> new code was to release it when it was no longer needed so another
> requester would be able to take the tpm without having to wait for it
> to be released.

If I recall, this was so that system level things outside linux could
access the TPM properly??

> With the old code I think it would have to wait either
> until the next time release_locality was called, or attempt to seize
> the tpm with the seize bit in the access register. I need to read
> through the spec some more, but does the tpm ever force a change when
> the request use bit is set, or does it leave it up to the software
> to deal with it and only gets involved in the case where the seize
> bit has been set?

Do we handle these cases? Maybe something like that has happened..

Jason

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Sat Nov 11 17, Jason Gunthorpe wrote:
>On Sat, Nov 11, 2017 at 12:12:57PM -0700, Jerry Snitselaar wrote:
>
>> Before the release_locality code would only actually release the
>> locality if the request use bit was set. So after it grabbed the
>> locality during probe it probably never released it. The idea with the
>> new code was to release it when it was no longer needed so another
>> requester would be able to take the tpm without having to wait for it
>> to be released.
>
>If I recall, this was so that system level things outside linux could
>access the TPM properly??
>

Yes, that is what drove this initially. I believe Jarkko was also
thinking of the possibility in the future where something like a vm
could request a locality as well, but that is just a hazy recollection
of emails from back then.

>> With the old code I think it would have to wait either
>> until the next time release_locality was called, or attempt to seize
>> the tpm with the seize bit in the access register. I need to read
>> through the spec some more, but does the tpm ever force a change when
>> the request use bit is set, or does it leave it up to the software
>> to deal with it and only gets involved in the case where the seize
>> bit has been set?
>
>Do we handle these cases? Maybe something like that has happened..
>
>Jason

If that is what happened in this case we should see the beenSeized bit
set in the access register (assuming the chip is doing things properly),
but it only had the tpmRegValidSts and tpmEstablishment bits set.

There is code in the interrupt handling that notices if the locality
changes if the chip has that capability, but I don't think there is
anything that deals with the seize bit. Another thing to be looked
at.

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

Le 11/11/17 a 21:31, Jerry Snitselaar a ecrit :
> On Sat Nov 11 17, Jason Gunthorpe wrote:
>> On Sat, Nov 11, 2017 at 12:12:57PM -0700, Jerry Snitselaar wrote:
>>
>>> Before the release_locality code would only actually release the
>>> locality if the request use bit was set. So after it grabbed the
>>> locality during probe it probably never released it. The idea with the
>>> new code was to release it when it was no longer needed so another
>>> requester would be able to take the tpm without having to wait for it
>>> to be released.
>>
>> If I recall, this was so that system level things outside linux could
>> access the TPM properly??
>>
>
> Yes, that is what drove this initially. I believe Jarkko was also
> thinking of the possibility in the future where something like a vm
> could request a locality as well, but that is just a hazy recollection
> of emails from back then.
>
>>> With the old code I think it would have to wait either
>>> until the next time release_locality was called, or attempt to seize
>>> the tpm with the seize bit in the access register. I need to read
>>> through the spec some more, but does the tpm ever force a change when
>>> the request use bit is set, or does it leave it up to the software
>>> to deal with it and only gets involved in the case where the seize
>>> bit has been set?
>>
>> Do we handle these cases? Maybe something like that has happened..
>>
>> Jason
>
> If that is what happened in this case we should see the beenSeized bit
> set in the access register (assuming the chip is doing things properly),
> but it only had the tpmRegValidSts and tpmEstablishment bits set.
>
> There is code in the interrupt handling that notices if the locality
> changes if the chip has that capability, but I don't think there is
> anything that deals with the seize bit. Another thing to be looked
> at.

I don't know if it's relevant, I was looking at the different tpm_* 
tools and tpm_nvinfo is giving me errors with 4.9:

$ sudo tpm_nvinfo  -l debug
Tspi_Context_Create success
Tspi_Context_Connect success
Tspi_Context_GetTpmObject success
Tspi_TPM_GetCapability success
NVRAM index   : 0x00000000 (0)
Tspi_TPM_GetCapability failed: 0x00000002 - layer=tpm, code=0002 (2), Bad memory index

NVRAM index   : 0x10000001 (268435457)
Tspi_TPM_GetCapability failed: 0x00000002 - layer=tpm, code=0002 (2), Bad memory index

NVRAM index   : 0xffffffff (4294967295)
Tspi_TPM_GetCapability failed: 0x00000002 - layer=tpm, code=0002 (2), Bad memory index

Tspi_Context_FreeMemory success
Tspi_Context_Close success

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jason Gunthorpe]

On Tue, Nov 14, 2017 at 01:26:16AM +0100, Laurent Bigonville wrote:

> I don't know if it's relevant, I was looking at the different tpm_* tools
> and tpm_nvinfo is giving me errors with 4.9:

This looks like good communication with the chip to me, just the chip
refusing to run the command for whatever reason?

Jason

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Sat, Nov 11, 2017 at 01:31:32PM -0700, Jerry Snitselaar wrote:
> On Sat Nov 11 17, Jason Gunthorpe wrote:
> > On Sat, Nov 11, 2017 at 12:12:57PM -0700, Jerry Snitselaar wrote:
> > 
> > > Before the release_locality code would only actually release the
> > > locality if the request use bit was set. So after it grabbed the
> > > locality during probe it probably never released it. The idea with the
> > > new code was to release it when it was no longer needed so another
> > > requester would be able to take the tpm without having to wait for it
> > > to be released.
> > 
> > If I recall, this was so that system level things outside linux could
> > access the TPM properly??
> > 
> 
> Yes, that is what drove this initially. I believe Jarkko was also
> thinking of the possibility in the future where something like a vm
> could request a locality as well, but that is just a hazy recollection
> of emails from back then.

This was something I recall discussing in LPC 2016 in the hallway at
least :-) A tidbit but it could make sense to tie it to VMM, not VM.

> 
> > > With the old code I think it would have to wait either
> > > until the next time release_locality was called, or attempt to seize
> > > the tpm with the seize bit in the access register. I need to read
> > > through the spec some more, but does the tpm ever force a change when
> > > the request use bit is set, or does it leave it up to the software
> > > to deal with it and only gets involved in the case where the seize
> > > bit has been set?
> > 
> > Do we handle these cases? Maybe something like that has happened..
> > 
> > Jason
> 
> If that is what happened in this case we should see the beenSeized bit
> set in the access register (assuming the chip is doing things properly),
> but it only had the tpmRegValidSts and tpmEstablishment bits set.
> 
> There is code in the interrupt handling that notices if the locality
> changes if the chip has that capability, but I don't think there is
> anything that deals with the seize bit. Another thing to be looked
> at.

For me it is hard to understand who is the 3rd party who would be using
other locality and accessing TPM after OS handover in the regression in
question.

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[James Bottomley]

On Tue, 2017-11-14 at 16:59 +0200, Jarkko Sakkinen wrote:
> On Sat, Nov 11, 2017 at 01:31:32PM -0700, Jerry Snitselaar wrote:
> > 
> > On Sat Nov 11 17, Jason Gunthorpe wrote:
> > > 
> > > On Sat, Nov 11, 2017 at 12:12:57PM -0700, Jerry Snitselaar wrote:
> > > 
> > > > 
> > > > Before the release_locality code would only actually release
> > > > the locality if the request use bit was set. So after it
> > > > grabbed the locality during probe it probably never released
> > > > it. The idea with the new code was to release it when it was no
> > > > longer needed so another requester would be able to take the
> > > > tpm without having to wait for it to be released.
> > > 
> > > If I recall, this was so that system level things outside linux
> > > could access the TPM properly??
> > > 
> > 
> > Yes, that is what drove this initially. I believe Jarkko was also
> > thinking of the possibility in the future where something like a vm
> > could request a locality as well, but that is just a hazy
> > recollection of emails from back then.
> 
> This was something I recall discussing in LPC 2016 in the hallway at
> least :-) A tidbit but it could make sense to tie it to VMM, not VM.

I think we should be extremely wary of different localities before we
have a cast iron definition of what they mean.  All the TPM PC spec
says is that locality 4 is reserved for firmware (meaning the kernel
should have no access) and it implies there's a privilege hierarchy,
making 4 the most privileged and 0 the least but leaves all the
definition to the OS.  Since we only have four other localities to play
with, we need a global definition of what they mean in Linux (and who
protects them) otherwise we'll get conflicting uses.  What does Windows
use them for?

James

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Tue, Nov 14, 2017 at 07:17:11AM -0800, James Bottomley wrote:
> On Tue, 2017-11-14 at 16:59 +0200, Jarkko Sakkinen wrote:
> > On Sat, Nov 11, 2017 at 01:31:32PM -0700, Jerry Snitselaar wrote:
> > > 
> > > On Sat Nov 11 17, Jason Gunthorpe wrote:
> > > > 
> > > > On Sat, Nov 11, 2017 at 12:12:57PM -0700, Jerry Snitselaar wrote:
> > > > 
> > > > > 
> > > > > Before the release_locality code would only actually release
> > > > > the locality if the request use bit was set. So after it
> > > > > grabbed the locality during probe it probably never released
> > > > > it. The idea with the new code was to release it when it was no
> > > > > longer needed so another requester would be able to take the
> > > > > tpm without having to wait for it to be released.
> > > > 
> > > > If I recall, this was so that system level things outside linux
> > > > could access the TPM properly??
> > > > 
> > > 
> > > Yes, that is what drove this initially. I believe Jarkko was also
> > > thinking of the possibility in the future where something like a vm
> > > could request a locality as well, but that is just a hazy
> > > recollection of emails from back then.
> > 
> > This was something I recall discussing in LPC 2016 in the hallway at
> > least :-) A tidbit but it could make sense to tie it to VMM, not VM.
> 
> I think we should be extremely wary of different localities before we
> have a cast iron definition of what they mean.  All the TPM PC spec
> says is that locality 4 is reserved for firmware (meaning the kernel
> should have no access) and it implies there's a privilege hierarchy,
> making 4 the most privileged and 0 the least but leaves all the
> definition to the OS.  Since we only have four other localities to play
> with, we need a global definition of what they mean in Linux (and who
> protects them) otherwise we'll get conflicting uses.  What does Windows
> use them for?
> 
> James

No idea. If I had to guess, they use only one locality for OS as this
what PTT/fTPM had when it didn't have localities. At least their
implementation works with only one locality.

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

Le 17/11/17 a 14:16, Jarkko Sakkinen a ecrit :
> On Tue, Nov 14, 2017 at 07:17:11AM -0800, James Bottomley wrote:
>> On Tue, 2017-11-14 at 16:59 +0200, Jarkko Sakkinen wrote:
>>> On Sat, Nov 11, 2017 at 01:31:32PM -0700, Jerry Snitselaar wrote:
>>>> On Sat Nov 11 17, Jason Gunthorpe wrote:
>>>>> On Sat, Nov 11, 2017 at 12:12:57PM -0700, Jerry Snitselaar wrote:
>>>>>
>>>>>> Before the release_locality code would only actually release
>>>>>> the locality if the request use bit was set. So after it
>>>>>> grabbed the locality during probe it probably never released
>>>>>> it. The idea with the new code was to release it when it was no
>>>>>> longer needed so another requester would be able to take the
>>>>>> tpm without having to wait for it to be released.
>>>>> If I recall, this was so that system level things outside linux
>>>>> could access the TPM properly??
>>>>>
>>>> Yes, that is what drove this initially. I believe Jarkko was also
>>>> thinking of the possibility in the future where something like a vm
>>>> could request a locality as well, but that is just a hazy
>>>> recollection of emails from back then.
>>> This was something I recall discussing in LPC 2016 in the hallway at
>>> least :-) A tidbit but it could make sense to tie it to VMM, not VM.
>> I think we should be extremely wary of different localities before we
>> have a cast iron definition of what they mean.  All the TPM PC spec
>> says is that locality 4 is reserved for firmware (meaning the kernel
>> should have no access) and it implies there's a privilege hierarchy,
>> making 4 the most privileged and 0 the least but leaves all the
>> definition to the OS.  Since we only have four other localities to play
>> with, we need a global definition of what they mean in Linux (and who
>> protects them) otherwise we'll get conflicting uses.  What does Windows
>> use them for?
>>
>> James
> No idea. If I had to guess, they use only one locality for OS as this
> what PTT/fTPM had when it didn't have localities. At least their
> implementation works with only one locality.
>

No more idea then? :(

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Wed Jan 03 18, Laurent Bigonville wrote:
>Le 17/11/17 a 14:16, Jarkko Sakkinen a ecrit :
>>On Tue, Nov 14, 2017 at 07:17:11AM -0800, James Bottomley wrote:
>>>On Tue, 2017-11-14 at 16:59 +0200, Jarkko Sakkinen wrote:
>>>>On Sat, Nov 11, 2017 at 01:31:32PM -0700, Jerry Snitselaar wrote:
>>>>>On Sat Nov 11 17, Jason Gunthorpe wrote:
>>>>>>On Sat, Nov 11, 2017 at 12:12:57PM -0700, Jerry Snitselaar wrote:
>>>>>>
>>>>>>>Before the release_locality code would only actually release
>>>>>>>the locality if the request use bit was set. So after it
>>>>>>>grabbed the locality during probe it probably never released
>>>>>>>it. The idea with the new code was to release it when it was no
>>>>>>>longer needed so another requester would be able to take the
>>>>>>>tpm without having to wait for it to be released.
>>>>>>If I recall, this was so that system level things outside linux
>>>>>>could access the TPM properly??
>>>>>>
>>>>>Yes, that is what drove this initially. I believe Jarkko was also
>>>>>thinking of the possibility in the future where something like a vm
>>>>>could request a locality as well, but that is just a hazy
>>>>>recollection of emails from back then.
>>>>This was something I recall discussing in LPC 2016 in the hallway at
>>>>least :-) A tidbit but it could make sense to tie it to VMM, not VM.
>>>I think we should be extremely wary of different localities before we
>>>have a cast iron definition of what they mean.  All the TPM PC spec
>>>says is that locality 4 is reserved for firmware (meaning the kernel
>>>should have no access) and it implies there's a privilege hierarchy,
>>>making 4 the most privileged and 0 the least but leaves all the
>>>definition to the OS.  Since we only have four other localities to play
>>>with, we need a global definition of what they mean in Linux (and who
>>>protects them) otherwise we'll get conflicting uses.  What does Windows
>>>use them for?
>>>
>>>James
>>No idea. If I had to guess, they use only one locality for OS as this
>>what PTT/fTPM had when it didn't have localities. At least their
>>implementation works with only one locality.
>>
>
>No more idea then? :(

Hi Laurent,

Can you try the following debug patch (earlier idea of adding a sleep to allow
tpm to complete state transition):

--8<--

diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
index fdde971bc810..6a9325b02059 100644
--- a/drivers/char/tpm/tpm_tis_core.c
+++ b/drivers/char/tpm/tpm_tis_core.c
@@ -80,6 +80,7 @@ static void release_locality(struct tpm_chip *chip, int l)
 	struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
 
 	tpm_tis_write8(priv, TPM_ACCESS(l), TPM_ACCESS_ACTIVE_LOCALITY);
+	tpm_msleep(200);
 }
 
 static int request_locality(struct tpm_chip *chip, int l)
-- 
2.15.0

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

Le 03/01/18 a 01:33, Jerry Snitselaar a ecrit :
> On Wed Jan 03 18, Laurent Bigonville wrote:
>> Le 17/11/17 a 14:16, Jarkko Sakkinen a ecrit :
>>> On Tue, Nov 14, 2017 at 07:17:11AM -0800, James Bottomley wrote:
>>>> On Tue, 2017-11-14 at 16:59 +0200, Jarkko Sakkinen wrote:
>>>>> On Sat, Nov 11, 2017 at 01:31:32PM -0700, Jerry Snitselaar wrote:
>>>>>> On Sat Nov 11 17, Jason Gunthorpe wrote:
>>>>>>> On Sat, Nov 11, 2017 at 12:12:57PM -0700, Jerry Snitselaar wrote:
>>>>>>>
>>>>>>>> Before the release_locality code would only actually release
>>>>>>>> the locality if the request use bit was set. So after it
>>>>>>>> grabbed the locality during probe it probably never released
>>>>>>>> it. The idea with the new code was to release it when it was no
>>>>>>>> longer needed so another requester would be able to take the
>>>>>>>> tpm without having to wait for it to be released.
>>>>>>> If I recall, this was so that system level things outside linux
>>>>>>> could access the TPM properly??
>>>>>>>
>>>>>> Yes, that is what drove this initially. I believe Jarkko was also
>>>>>> thinking of the possibility in the future where something like a vm
>>>>>> could request a locality as well, but that is just a hazy
>>>>>> recollection of emails from back then.
>>>>> This was something I recall discussing in LPC 2016 in the hallway at
>>>>> least :-) A tidbit but it could make sense to tie it to VMM, not VM.
>>>> I think we should be extremely wary of different localities before we
>>>> have a cast iron definition of what they mean.  All the TPM PC spec
>>>> says is that locality 4 is reserved for firmware (meaning the kernel
>>>> should have no access) and it implies there's a privilege hierarchy,
>>>> making 4 the most privileged and 0 the least but leaves all the
>>>> definition to the OS.  Since we only have four other localities to 
>>>> play
>>>> with, we need a global definition of what they mean in Linux (and who
>>>> protects them) otherwise we'll get conflicting uses.  What does 
>>>> Windows
>>>> use them for?
>>>>
>>>> James
>>> No idea. If I had to guess, they use only one locality for OS as this
>>> what PTT/fTPM had when it didn't have localities. At least their
>>> implementation works with only one locality.
>>>
>>
>> No more idea then? :(
>
> Hi Laurent,
>
> Can you try the following debug patch (earlier idea of adding a sleep 
> to allow
> tpm to complete state transition):
>
> --8<--
>
> diff --git a/drivers/char/tpm/tpm_tis_core.c 
> b/drivers/char/tpm/tpm_tis_core.c
> index fdde971bc810..6a9325b02059 100644
> --- a/drivers/char/tpm/tpm_tis_core.c
> +++ b/drivers/char/tpm/tpm_tis_core.c
> @@ -80,6 +80,7 @@ static void release_locality(struct tpm_chip *chip, 
> int l)
>     struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
>
>     tpm_tis_write8(priv, TPM_ACCESS(l), TPM_ACCESS_ACTIVE_LOCALITY);
> +    tpm_msleep(200);
> }
>
> static int request_locality(struct tpm_chip *chip, int l)

Thanks,

With that patch the device is showing up again in /dev and I 
tpm_sealdata is working as well

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

I don't remember if I replied to this, re-posting to be sure:

Le 03/01/18 a 01:33, Jerry Snitselaar a ecrit :
> Hi Laurent,
>
> Can you try the following debug patch (earlier idea of adding a sleep 
> to allow
> tpm to complete state transition):
>
> --8<--
>
> diff --git a/drivers/char/tpm/tpm_tis_core.c 
> b/drivers/char/tpm/tpm_tis_core.c
> index fdde971bc810..6a9325b02059 100644
> --- a/drivers/char/tpm/tpm_tis_core.c
> +++ b/drivers/char/tpm/tpm_tis_core.c
> @@ -80,6 +80,7 @@ static void release_locality(struct tpm_chip *chip, 
> int l)
>     struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
>
>     tpm_tis_write8(priv, TPM_ACCESS(l), TPM_ACCESS_ACTIVE_LOCALITY);
> +    tpm_msleep(200);
> }
>
> static int request_locality(struct tpm_chip *chip, int l)

I tried the patch and this is working.

Would that be a viable solution?

Kind regards,

Laurent Bigonville

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Fri, Feb 09, 2018 at 11:53:55AM +0100, Laurent Bigonville wrote:
> I don't remember if I replied to this, re-posting to be sure:
>
> Le 03/01/18 a 01:33, Jerry Snitselaar a ecrit :
> > Hi Laurent,
> >
> > Can you try the following debug patch (earlier idea of adding a sleep to
> > allow
> > tpm to complete state transition):
> >
> > --8<--
> >
> > diff --git a/drivers/char/tpm/tpm_tis_core.c
> > b/drivers/char/tpm/tpm_tis_core.c
> > index fdde971bc810..6a9325b02059 100644
> > --- a/drivers/char/tpm/tpm_tis_core.c
> > +++ b/drivers/char/tpm/tpm_tis_core.c
> > @@ -80,6 +80,7 @@ static void release_locality(struct tpm_chip *chip,
> > int l)
> >     struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
> >
> >     tpm_tis_write8(priv, TPM_ACCESS(l), TPM_ACCESS_ACTIVE_LOCALITY);
> > +    tpm_msleep(200);
> > }
> >
> > static int request_locality(struct tpm_chip *chip, int l)
>
> I tried the patch and this is working.
>
> Would that be a viable solution?
>
> Kind regards,
>
> Laurent Bigonville

According to the PC client specification in the section 5.5.2.3:

"For commands indicated as short or medium duration (i.e., those that do
not cause key  generation), the TPM SHALL respond to an abort within
TIMEOUT_A. For commands indicated as lon g duration or those that cause
key generation, the TPM SHALL respond to a request to abort the command
within TIMEOUT B"

The section 5.5.2.4 describing the access register does not give much
more light to this i.e. what the expected duration maximum is when there
is no command executing.

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

Le 14/02/18 a 12:44, Jarkko Sakkinen a ecrit :
> On Fri, Feb 09, 2018 at 11:53:55AM +0100, Laurent Bigonville wrote:
>> I don't remember if I replied to this, re-posting to be sure:
>>
>> Le 03/01/18 a 01:33, Jerry Snitselaar a ecrit :
>>> Hi Laurent,
>>>
>>> Can you try the following debug patch (earlier idea of adding a sleep to
>>> allow
>>> tpm to complete state transition):
>>>
>>> --8<--
>>>
>>> diff --git a/drivers/char/tpm/tpm_tis_core.c
>>> b/drivers/char/tpm/tpm_tis_core.c
>>> index fdde971bc810..6a9325b02059 100644
>>> --- a/drivers/char/tpm/tpm_tis_core.c
>>> +++ b/drivers/char/tpm/tpm_tis_core.c
>>> @@ -80,6 +80,7 @@ static void release_locality(struct tpm_chip *chip,
>>> int l)
>>>      struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
>>>
>>>      tpm_tis_write8(priv, TPM_ACCESS(l), TPM_ACCESS_ACTIVE_LOCALITY);
>>> +    tpm_msleep(200);
>>> }
>>>
>>> static int request_locality(struct tpm_chip *chip, int l)
>> I tried the patch and this is working.
>>
>> Would that be a viable solution?
>>
>> Kind regards,
>>
>> Laurent Bigonville
> According to the PC client specification in the section 5.5.2.3:
>
> "For commands indicated as short or medium duration (i.e., those that do
> not cause key  generation), the TPM SHALL respond to an abort within
> TIMEOUT_A. For commands indicated as lon g duration or those that cause
> key generation, the TPM SHALL respond to a request to abort the command
> within TIMEOUT B"
>
> The section 5.5.2.4 describing the access register does not give much
> more light to this i.e. what the expected duration maximum is when there
> is no command executing.
The duration that that was in your patch seems to work, cannot this be 
implemented?

I'm quite surprised I'm the only one impacted by this...

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Fri, 2018-03-09 at 18:24 +0100, Laurent Bigonville wrote:
> The duration that that was in your patch seems to work, cannot this be
> implemented?
>
> I'm quite surprised I'm the only one impacted by this...

Sorry it took so long time response but I had forgotten so too many
details.

After re-reading (PHEW!) the whole thread I'm thinking if we could
use section 5.5.2.4 as a reference to select a timeout of TIMEOUT_A
(albeit it is for request locality) and use Jerry's suggestion to
put wait_for_tpm_stat() there? Opinions?

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

Le 15/03/18 a 17:24, Jarkko Sakkinen a ecrit :
> On Fri, 2018-03-09 at 18:24 +0100, Laurent Bigonville wrote:
>> The duration that that was in your patch seems to work, cannot this be
>> implemented?
>>
>> I'm quite surprised I'm the only one impacted by this...
> Sorry it took so long time response but I had forgotten so too many
> details.
>
> After re-reading (PHEW!) the whole thread I'm thinking if we could
> use section 5.5.2.4 as a reference to select a timeout of TIMEOUT_A
> (albeit it is for request locality) and use Jerry's suggestion to
> put wait_for_tpm_stat() there? Opinions?
Hey,

Sorry to bother you again, but are there any progress on getting this 
mainlined?

Kind regards,

Laurent Bigonville

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Thu May 03 18, Laurent Bigonville wrote:
>Le 15/03/18 a 17:24, Jarkko Sakkinen a ecrit :
>>On Fri, 2018-03-09 at 18:24 +0100, Laurent Bigonville wrote:
>>>The duration that that was in your patch seems to work, cannot this be
>>>implemented?
>>>
>>>I'm quite surprised I'm the only one impacted by this...
>>Sorry it took so long time response but I had forgotten so too many
>>details.
>>
>>After re-reading (PHEW!) the whole thread I'm thinking if we could
>>use section 5.5.2.4 as a reference to select a timeout of TIMEOUT_A
>>(albeit it is for request locality) and use Jerry's suggestion to
>>put wait_for_tpm_stat() there? Opinions?
>Hey,
>
>Sorry to bother you again, but are there any progress on getting this 
>mainlined?
>
>Kind regards,
>
>Laurent Bigonville

Jarkko would something like this work? Building it to test on tpm2.0 system right now.
I don't have access to a system with a tpm1.2 tis device at the moment. I tested a
patch similar to this based on the slightly older code that is currently in the
rhel kernel and it seemed work fine. release_locality was still a void function
back then, so might have to think more about the return values here and checking
them. Also wondering if release_locality, check_locality, and wait_for_tpm_stat
can be refactored since they would all have basically the same chunk of code.

Jerry

diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
index 5a1f47b43947..31ee154450eb 100644
--- a/drivers/char/tpm/tpm_tis_core.c
+++ b/drivers/char/tpm/tpm_tis_core.c
@@ -143,12 +143,56 @@ static bool check_locality(struct tpm_chip *chip, int l)
        return false;
 }
 
+static bool locality_inactive(struct tpm_chip *chip, int l)
+{
+       struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
+       int rc;
+       u8 access;
+
+       rc = tpm_tis_read8(priv, TPM_ACCESS(l), &access);
+       if (rc < 0)
+               return false;
+
+       if ((access & (TPM_ACCESS_VALID | TPM_ACCESS_ACTIVE_LOCALITY)) == TPM_ACCESS_VALID)
+               return true;
+
+       return false;
+}
+
 static int release_locality(struct tpm_chip *chip, int l)
 {
        struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
+       unsigned long stop, timeout;
+       long rc;
 
        tpm_tis_write8(priv, TPM_ACCESS(l), TPM_ACCESS_ACTIVE_LOCALITY);
 
+       stop = jiffies + chip->timeout_a;
+
+       if (chip->flags & TPM_CHIP_FLAG_IRQ) {
+again:
+               timeout = stop - jiffies;
+               if ((long)timeout <= 0)
+                       return -1;
+
+               rc = wait_event_interruptible_timeout(priv->int_queue,
+                                                     (locality_inactive(chip, l)),
+                                                     timeout);
+
+               if (rc > 0)
+                       return rc;
+
+               if (rc == -ERESTARTSYS && freezing(current)) {
+                       clear_thread_flag(TIF_SIGPENDING);
+                       goto again;
+               }
+       } else {
+               do {
+                       if (locality_inactive(chip, l))
+                               return 0;
+                       tpm_msleep(TPM_TIMEOUT);
+               } while (time_before(jiffies, stop));
+       }
        return 0;
 }

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Thu, May 03, 2018 at 10:43:11AM -0700, Jerry Snitselaar wrote:
> On Thu May 03 18, Laurent Bigonville wrote:
> > Le 15/03/18 a 17:24, Jarkko Sakkinen a ecrit :
> > > On Fri, 2018-03-09 at 18:24 +0100, Laurent Bigonville wrote:
> > > > The duration that that was in your patch seems to work, cannot this be
> > > > implemented?
> > > > 
> > > > I'm quite surprised I'm the only one impacted by this...
> > > Sorry it took so long time response but I had forgotten so too many
> > > details.
> > > 
> > > After re-reading (PHEW!) the whole thread I'm thinking if we could
> > > use section 5.5.2.4 as a reference to select a timeout of TIMEOUT_A
> > > (albeit it is for request locality) and use Jerry's suggestion to
> > > put wait_for_tpm_stat() there? Opinions?
> > Hey,
> > 
> > Sorry to bother you again, but are there any progress on getting this
> > mainlined?
> > 
> > Kind regards,
> > 
> > Laurent Bigonville
> 
> Jarkko would something like this work? Building it to test on tpm2.0 system right now.
> I don't have access to a system with a tpm1.2 tis device at the moment. I tested a
> patch similar to this based on the slightly older code that is currently in the
> rhel kernel and it seemed work fine. release_locality was still a void function
> back then, so might have to think more about the return values here and checking
> them. Also wondering if release_locality, check_locality, and wait_for_tpm_stat
> can be refactored since they would all have basically the same chunk of code.
> 
> Jerry
> 
> diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
> index 5a1f47b43947..31ee154450eb 100644
> --- a/drivers/char/tpm/tpm_tis_core.c
> +++ b/drivers/char/tpm/tpm_tis_core.c
> @@ -143,12 +143,56 @@ static bool check_locality(struct tpm_chip *chip, int l)
>        return false;
> }
> 
> +static bool locality_inactive(struct tpm_chip *chip, int l)
> +{
> +       struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
> +       int rc;
> +       u8 access;
> +
> +       rc = tpm_tis_read8(priv, TPM_ACCESS(l), &access);
> +       if (rc < 0)
> +               return false;
> +
> +       if ((access & (TPM_ACCESS_VALID | TPM_ACCESS_ACTIVE_LOCALITY)) == TPM_ACCESS_VALID)
> +               return true;
> +
> +       return false;
> +}
> +
> static int release_locality(struct tpm_chip *chip, int l)
> {
>        struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
> +       unsigned long stop, timeout;
> +       long rc;
> 
>        tpm_tis_write8(priv, TPM_ACCESS(l), TPM_ACCESS_ACTIVE_LOCALITY);
> 
> +       stop = jiffies + chip->timeout_a;
> +
> +       if (chip->flags & TPM_CHIP_FLAG_IRQ) {
> +again:
> +               timeout = stop - jiffies;
> +               if ((long)timeout <= 0)
> +                       return -1;
> +
> +               rc = wait_event_interruptible_timeout(priv->int_queue,
> +                                                     (locality_inactive(chip, l)),
> +                                                     timeout);
> +
> +               if (rc > 0)
> +                       return rc;
> +
> +               if (rc == -ERESTARTSYS && freezing(current)) {
> +                       clear_thread_flag(TIF_SIGPENDING);
> +                       goto again;
> +               }
> +       } else {
> +               do {
> +                       if (locality_inactive(chip, l))
> +                               return 0;
> +                       tpm_msleep(TPM_TIMEOUT);
> +               } while (time_before(jiffies, stop));
> +       }

Maybe have the second branch first with a return statement and
convert the thing going on in the first branch as a loop?

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Thu, May 03, 2018 at 01:38:17PM +0200, Laurent Bigonville wrote:
> Le 15/03/18 a 17:24, Jarkko Sakkinen a ecrit :
> > On Fri, 2018-03-09 at 18:24 +0100, Laurent Bigonville wrote:
> > > The duration that that was in your patch seems to work, cannot this be
> > > implemented?
> > > 
> > > I'm quite surprised I'm the only one impacted by this...
> > Sorry it took so long time response but I had forgotten so too many
> > details.
> > 
> > After re-reading (PHEW!) the whole thread I'm thinking if we could
> > use section 5.5.2.4 as a reference to select a timeout of TIMEOUT_A
> > (albeit it is for request locality) and use Jerry's suggestion to
> > put wait_for_tpm_stat() there? Opinions?
> Hey,
> 
> Sorry to bother you again, but are there any progress on getting this
> mainlined?
> 
> Kind regards,
> 
> Laurent Bigonville

Sorry for not doing anything. Jerry, do you want to send patch?

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jerry Snitselaar]

On Fri May 04 18, Jarkko Sakkinen wrote:
>On Thu, May 03, 2018 at 01:38:17PM +0200, Laurent Bigonville wrote:
>> Le 15/03/18 a 17:24, Jarkko Sakkinen a ecrit :
>> > On Fri, 2018-03-09 at 18:24 +0100, Laurent Bigonville wrote:
>> > > The duration that that was in your patch seems to work, cannot this be
>> > > implemented?
>> > >
>> > > I'm quite surprised I'm the only one impacted by this...
>> > Sorry it took so long time response but I had forgotten so too many
>> > details.
>> >
>> > After re-reading (PHEW!) the whole thread I'm thinking if we could
>> > use section 5.5.2.4 as a reference to select a timeout of TIMEOUT_A
>> > (albeit it is for request locality) and use Jerry's suggestion to
>> > put wait_for_tpm_stat() there? Opinions?
>> Hey,
>>
>> Sorry to bother you again, but are there any progress on getting this
>> mainlined?
>>
>> Kind regards,
>>
>> Laurent Bigonville
>
>Sorry for not doing anything. Jerry, do you want to send patch?
>
>/Jarkko

Yes. I'll send a patch today.

Jerry

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Sat, Nov 11, 2017 at 12:46:47PM -0700, Jason Gunthorpe wrote:
> On Sat, Nov 11, 2017 at 12:12:57PM -0700, Jerry Snitselaar wrote:
> 
> > Before the release_locality code would only actually release the
> > locality if the request use bit was set. So after it grabbed the
> > locality during probe it probably never released it. The idea with the
> > new code was to release it when it was no longer needed so another
> > requester would be able to take the tpm without having to wait for it
> > to be released.
> 
> If I recall, this was so that system level things outside linux could
> access the TPM properly??
> 
> > With the old code I think it would have to wait either
> > until the next time release_locality was called, or attempt to seize
> > the tpm with the seize bit in the access register. I need to read
> > through the spec some more, but does the tpm ever force a change when
> > the request use bit is set, or does it leave it up to the software
> > to deal with it and only gets involved in the case where the seize
> > bit has been set?
> 
> Do we handle these cases? Maybe something like that has happened..
> 
> Jason

At least the driver works with TXT properly that uses locality 2.

The locality stuff for tpm_crb has bee made to work sanely with TXT.

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Jarkko Sakkinen]

On Fri, Nov 10, 2017 at 01:53:00PM -0700, Jerry Snitselaar wrote:
> On Fri Nov 10 17, Laurent Bigonville wrote:
> > Le 10/11/17 a 08:07, Jerry Snitselaar a ecrit :
> > > On Thu Nov 09 17, Jerry Snitselaar wrote:
> > > > On Thu Nov 09 17, Laurent Bigonville wrote:
> > > > > Le 09/11/17 a 01:04, Laurent Bigonville a ecrit :
> > > > > > 
> > > > > > 
> > > > > > Le 24/10/17 a 18:07, Jarkko Sakkinen a ecrit :
> > > > > > > On Tue, Oct 24, 2017 at 07:57:06AM -0700, Jerry Snitselaar wrote:
> > > > > > > > On Tue Oct 24 17, Jarkko Sakkinen wrote:
> > > > > > > > > On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
> > > > > > > > > > On Mon Oct 23 17, Jarkko Sakkinen wrote:
> > > > > > > > > > > On Sat, Oct 21, 2017 at 10:53:55AM
> > > > > > > > > > > +0200, Laurent Bigonville wrote:
> > > > > > > > > > > > Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
> > > > > > > > > > > > > On Wed Sep 06 17, Jarkko Sakkinen wrote:
> > > > > > > > > > > > > > On Fri, Sep 01, 2017 at
> > > > > > > > > > > > > > 02:10:18PM +0200,
> > > > > > > > > > > > > > Laurent Bigonville
> > > > > > > > > > > > > > wrote:
> > > > > > > > > > > > > > > Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
> > > > > > > > > > > > > > > > On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
> > > > > > > > > > > > > > > > > > Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
> > > > > > > > > > > > > > > > > > > Le
> > > > > > > > > > > > > > > > > > > 29/08/17
> > > > > > > > > > > > > > > > > > > a
> > > > > > > > > > > > > > > > > > > 18:00,
> > > > > > > > > > > > > > > > > > > Alexander.Steffen@infineon.com
> > > > > > > > > > > > > > > > > > > a
> > > > > > > > > > > > > > > > > > > ecrit :
> > > > > > > > > > > > > > > > > > > > > An idea how to troubleshoot this?
> > > > > > > > > > > > > > > > > > > > Can you run git bisect on the changes between 4.11 and
> > > > > > > > > > > > > > > 4.12, so that
> > > > > > > > > > > > > > > > > > > > we find the offending commit? It is probably sufficient
> > > > > > > > > > > > > > > to limit the
> > > > > > > > > > > > > > > > > > > > search
> > > > > > > > > > > > > > > > > > > > to
> > > > > > > > > > > > > > > > > > > > commits
> > > > > > > > > > > > > > > > > > > > that
> > > > > > > > > > > > > > > > > > > > touch
> > > > > > > > > > > > > > > > > > > > something
> > > > > > > > > > > > > > > > > > > > in
> > > > > > > > > > > > > > > > > > > > drivers/char/tpm.
> > > > > > > > > > > > > > > > > > > I'll try and keep you posted.
> > > > > > > > > > > > > > > > > > OK I've
> > > > > > > > > > > > > > > > > > been
> > > > > > > > > > > > > > > > > > able to
> > > > > > > > > > > > > > > > > > bisect
> > > > > > > > > > > > > > > > > > the
> > > > > > > > > > > > > > > > > > problem
> > > > > > > > > > > > > > > > > > and the
> > > > > > > > > > > > > > > > > > bad
> > > > > > > > > > > > > > > > > > commit
> > > > > > > > > > > > > > > > > > is:
> > > > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > > > e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
> > > > > > > > > > > > > > > > > > is the
> > > > > > > > > > > > > > > > > > first
> > > > > > > > > > > > > > > > > > bad
> > > > > > > > > > > > > > > > > > commit
> > > > > > > > > > > > > > > > > > commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
> > > > > > > > > > > > > > > > > > Author: Jerry Snitselaar <jsnitsel@redhat.com>
> > > > > > > > > > > > > > > > > > Date:   Mon Mar 27 08:46:04 2017 -0700
> > > > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > > >       tpm_tis: convert to using locality callbacks
> > > > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > > >      
> > > > > > > > > > > > > > > > > > This
> > > > > > > > > > > > > > > > > > patch
> > > > > > > > > > > > > > > > > > converts
> > > > > > > > > > > > > > > > > > tpm_tis
> > > > > > > > > > > > > > > > > > to use
> > > > > > > > > > > > > > > > > > of the
> > > > > > > > > > > > > > > > > > new tpm
> > > > > > > > > > > > > > > > > > class
> > > > > > > > > > > > > > > > > > ops
> > > > > > > > > > > > > > > > > >       request_locality, and relinquish_locality.
> > > > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > > >      
> > > > > > > > > > > > > > > > > > With the
> > > > > > > > > > > > > > > > > > move to
> > > > > > > > > > > > > > > > > > using
> > > > > > > > > > > > > > > > > > the
> > > > > > > > > > > > > > > > > > callbacks,
> > > > > > > > > > > > > > > > > > release_locality
> > > > > > > > > > > > > > > > > > is
> > > > > > > > > > > > > > > > > > changed so
> > > > > > > > > > > > > > > > > >       that we now release the locality even if there is no
> > > > > > > > > > > > > > > > > > request pending.
> > > > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > > >       This required some changes to the tpm_tis_core_init
> > > > > > > > > > > > > > > code path to
> > > > > > > > > > > > > > > > > >       make sure locality is requested when needed:
> > > > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > > >         - tpm2_probe code path will end up calling
> > > > > > > > > > > > > > > > > > request/release through
> > > > > > > > > > > > > > > > > >           callbacks, so request_locality prior to
> > > > > > > > > > > > > > > tpm2_probe not needed.
> > > > > > > > > > > > > > > > > >        
> > > > > > > > > > > > > > > > > > -
> > > > > > > > > > > > > > > > > > probe_itpm
> > > > > > > > > > > > > > > > > > makes
> > > > > > > > > > > > > > > > > > calls to
> > > > > > > > > > > > > > > > > > tpm_tis_send_data
> > > > > > > > > > > > > > > > > > which no
> > > > > > > > > > > > > > > > > > longer calls
> > > > > > > > > > > > > > > > > >          
> > > > > > > > > > > > > > > > > > request_locality,
> > > > > > > > > > > > > > > > > > so add
> > > > > > > > > > > > > > > > > > request_locality
> > > > > > > > > > > > > > > > > > prior to
> > > > > > > > > > > > > > > > > > tpm_tis_send_data
> > > > > > > > > > > > > > > > > >          
> > > > > > > > > > > > > > > > > > calls.
> > > > > > > > > > > > > > > > > > Also
> > > > > > > > > > > > > > > > > > drop
> > > > > > > > > > > > > > > > > > release_locality
> > > > > > > > > > > > > > > > > > call in
> > > > > > > > > > > > > > > > > > middleof
> > > > > > > > > > > > > > > > > > probe_itpm, and
> > > > > > > > > > > > > > > > > >          
> > > > > > > > > > > > > > > > > > keep
> > > > > > > > > > > > > > > > > > locality
> > > > > > > > > > > > > > > > > > until
> > > > > > > > > > > > > > > > > > release_locality
> > > > > > > > > > > > > > > > > > called
> > > > > > > > > > > > > > > > > > at end
> > > > > > > > > > > > > > > > > > of
> > > > > > > > > > > > > > > > > > probe_itpm.
> > > > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > > >       Cc: Peter Huewe <peterhuewe@gmx.de>
> > > > > > > > > > > > > > > > > >      
> > > > > > > > > > > > > > > > > > Cc:
> > > > > > > > > > > > > > > > > > Jarkko
> > > > > > > > > > > > > > > > > > Sakkinen
> > > > > > > > > > > > > > > > > > <jarkko.sakkinen@linux.intel.com>
> > > > > > > > > > > > > > > > > >      
> > > > > > > > > > > > > > > > > > Cc:
> > > > > > > > > > > > > > > > > > Jason
> > > > > > > > > > > > > > > > > > Gunthorpe
> > > > > > > > > > > > > > > > > > <jgunthorpe@obsidianresearch.com>
> > > > > > > > > > > > > > > > > >       Cc: Marcel Selhorst <tpmdd@selhorst.net>
> > > > > > > > > > > > > > > > > >      
> > > > > > > > > > > > > > > > > > Signed-off-by:
> > > > > > > > > > > > > > > > > > Jerry
> > > > > > > > > > > > > > > > > > Snitselaar
> > > > > > > > > > > > > > > > > > <jsnitsel@redhat.com>
> > > > > > > > > > > > > > > > > >       Reviewed-by: Jarkko Sakkinen
> > > > > > > > > > > > > > > <jarkko.sakkinen@linux.intel.com>
> > > > > > > > > > > > > > > > > > Tested-by:
> > > > > > > > > > > > > > > > > > Jarkko
> > > > > > > > > > > > > > > > > > Sakkinen
> > > > > > > > > > > > > > > > > > <jarkko.sakkinen@linux.intel.com>
> > > > > > > > > > > > > > > > > >       Signed-off-by: Jarkko Sakkinen
> > > > > > > > > > > > > > > <jarkko.sakkinen@linux.intel.com>
> > > > > > > > > > > > > > > > > > :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
> > > > > > > > > > > > > > > > > > 72f21b446e45ea1003de75902b0553deb99157fd M drivers
> > > > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > > I've looked
> > > > > > > > > > > > > > > > > again at the
> > > > > > > > > > > > > > > > > code in
> > > > > > > > > > > > > > > > > question,
> > > > > > > > > > > > > > > > > but could
> > > > > > > > > > > > > > > > > not find
> > > > > > > > > > > > > > > > > anything that is obviously wrong there. Locality is now
> > > > > > > > > > > > > > > > > requested/released
> > > > > > > > > > > > > > > > > at slightly
> > > > > > > > > > > > > > > > > different
> > > > > > > > > > > > > > > > > points in
> > > > > > > > > > > > > > > > > the process
> > > > > > > > > > > > > > > > > than
> > > > > > > > > > > > > > > > > before, but
> > > > > > > > > > > > > > > > > that's it.
> > > > > > > > > > > > > > > > > It does not
> > > > > > > > > > > > > > > > > seem to
> > > > > > > > > > > > > > > > > cause
> > > > > > > > > > > > > > > > > problems
> > > > > > > > > > > > > > > > > with the
> > > > > > > > > > > > > > > > > majority of TPMs, since you are the first to report any, so
> > > > > > > > > > > > > > > maybe it
> > > > > > > > > > > > > > > > > is a quirk that only affects this device.
> > > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > > Perhaps Jerry can help, since this is his change?
> > > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > > Alexander
> > > > > > > > > > > > > > > > Getting some
> > > > > > > > > > > > > > > > caffeine in me,
> > > > > > > > > > > > > > > > and starting to
> > > > > > > > > > > > > > > > take a look.
> > > > > > > > > > > > > > > > Adding
> > > > > > > > > > > > > > > > Jarkko as well
> > > > > > > > > > > > > > > > since this might
> > > > > > > > > > > > > > > > involve the
> > > > > > > > > > > > > > > > general locality
> > > > > > > > > > > > > > > > changes.
> > > > > > > > > > > > > > > > 
> > > > > > > > > > > > > > > > Laurent, if I
> > > > > > > > > > > > > > > > send you a patch
> > > > > > > > > > > > > > > > with some
> > > > > > > > > > > > > > > > debugging code
> > > > > > > > > > > > > > > > added, would
> > > > > > > > > > > > > > > > you be able to
> > > > > > > > > > > > > > > > run it on that
> > > > > > > > > > > > > > > > system? I wasn't
> > > > > > > > > > > > > > > > running into
> > > > > > > > > > > > > > > > issues
> > > > > > > > > > > > > > > > on the system I
> > > > > > > > > > > > > > > > had with a 1.2
> > > > > > > > > > > > > > > > device, but I no
> > > > > > > > > > > > > > > > longer have
> > > > > > > > > > > > > > > > access
> > > > > > > > > > > > > > > > to it. I'll see
> > > > > > > > > > > > > > > > if I can find
> > > > > > > > > > > > > > > > one in our labs
> > > > > > > > > > > > > > > > and reproduce it
> > > > > > > > > > > > > > > > there.
> > > > > > > > > > > > > > > Yes I should be able to do that
> > > > > > > > > > > > > > Any findings?
> > > > > > > > > > > > > > 
> > > > > > > > > > > > > > /Jarkko
> > > > > > > > > > > > > Okay, finally getting back
> > > > > > > > > > > > > to this. Looking at the code
> > > > > > > > > > > > > it isn't clear
> > > > > > > > > > > > > to me
> > > > > > > > > > > > > why the change is causing
> > > > > > > > > > > > > this. So while I stare at
> > > > > > > > > > > > > this some more
> > > > > > > > > > > > > Laurent
> > > > > > > > > > > > > could you reproduce it with
> > > > > > > > > > > > > this patch so I can see what
> > > > > > > > > > > > > the status and
> > > > > > > > > > > > > access registers look like?
> > > > > > > > > > > > > Does anyone else on here
> > > > > > > > > > > > > happen to have a
> > > > > > > > > > > > > Sinosun
> > > > > > > > > > > > > tpm device? The systems I
> > > > > > > > > > > > > have access to with TPM1.2
> > > > > > > > > > > > > devices don't have
> > > > > > > > > > > > > this
> > > > > > > > > > > > > issue.
> > > > > > > > > > > > > 
> > > > > > > > > > > > > --8<--
> > > > > > > > > > > > > 
> > > > > > > > > > > > > diff --git a/drivers/char/tpm/tpm_tis_core.c
> > > > > > > > > > > > > b/drivers/char/tpm/tpm_tis_core.c
> > > > > > > > > > > > > index fdde971bc810..7d60a7e4b50a 100644
> > > > > > > > > > > > > --- a/drivers/char/tpm/tpm_tis_core.c
> > > > > > > > > > > > > +++ b/drivers/char/tpm/tpm_tis_core.c
> > > > > > > > > > > > > @@ -258,6 +258,7 @@ static
> > > > > > > > > > > > > int tpm_tis_send_data(struct
> > > > > > > > > > > > > tpm_chip *chip,
> > > > > > > > > > > > > const u8 *buf, size_t len)
> > > > > > > > > > > > >      int rc, status, burstcnt;
> > > > > > > > > > > > >      size_t count = 0;
> > > > > > > > > > > > >      bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
> > > > > > > > > > > > > +    u8 access;
> > > > > > > > > > > > > 
> > > > > > > > > > > > >      status = tpm_tis_status(chip);
> > > > > > > > > > > > >      if ((status & TPM_STS_COMMAND_READY) == 0) {
> > > > > > > > > > > > > @@ -292,6 +293,11 @@ static
> > > > > > > > > > > > > int tpm_tis_send_data(struct
> > > > > > > > > > > > > tpm_chip *chip,
> > > > > > > > > > > > > const u8 *buf, size_t len)
> > > > > > > > > > > > >          }
> > > > > > > > > > > > >          status = tpm_tis_status(chip);
> > > > > > > > > > > > >          if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
> > > > > > > > > > > > > +            rc =
> > > > > > > > > > > > > tpm_tis_read8(priv,
> > > > > > > > > > > > > TPM_ACCESS(priv->locality),
> > > > > > > > > > > > > &access);
> > > > > > > > > > > > > +            if (rc < 0)
> > > > > > > > > > > > > +               
> > > > > > > > > > > > > dev_info(&chip->dev,
> > > > > > > > > > > > > "TPM_STS_DATA_EXPECT == 0:
> > > > > > > > > > > > > read
> > > > > > > > > > > > > failure TPM_ACCESS(%d)\n", priv->locality);
> > > > > > > > > > > > > +            else
> > > > > > > > > > > > > +               
> > > > > > > > > > > > > dev_info(&chip->dev,
> > > > > > > > > > > > > "TPM_STS_DATA_EXPECT == 0:
> > > > > > > > > > > > > locality: %d status: %x
> > > > > > > > > > > > > access: %x\n",
> > > > > > > > > > > > > priv->locality, status,
> > > > > > > > > > > > > access);
> > > > > > > > > > > > >              rc = -EIO;
> > > > > > > > > > > > >              goto out_err;
> > > > > > > > > > > > >          }
> > > > > > > > > > > > > @@ -309,6 +315,11 @@ static
> > > > > > > > > > > > > int tpm_tis_send_data(struct
> > > > > > > > > > > > > tpm_chip *chip,
> > > > > > > > > > > > > const u8 *buf, size_t len)
> > > > > > > > > > > > >      }
> > > > > > > > > > > > >      status = tpm_tis_status(chip);
> > > > > > > > > > > > >      if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
> > > > > > > > > > > > > +        rc =
> > > > > > > > > > > > > tpm_tis_read8(priv,
> > > > > > > > > > > > > TPM_ACCESS(priv->locality),
> > > > > > > > > > > > > &access);
> > > > > > > > > > > > > +        if (rc < 0)
> > > > > > > > > > > > > +           
> > > > > > > > > > > > > dev_info(&chip->dev,
> > > > > > > > > > > > > "TPM_STS_DATA_EXPECT != 0:
> > > > > > > > > > > > > read
> > > > > > > > > > > > > failure TPM_ACCESS(%d)\n", priv->locality);
> > > > > > > > > > > > > +        else
> > > > > > > > > > > > > +           
> > > > > > > > > > > > > dev_info(&chip->dev,
> > > > > > > > > > > > > "TPM_STS_DATA_EXPECT != 0:
> > > > > > > > > > > > > locality:
> > > > > > > > > > > > > %d status: %x access: %x\n", priv->locality, status, access);
> > > > > > > > > > > > >          rc = -EIO;
> > > > > > > > > > > > >          goto out_err;
> > > > > > > > > > > > >      }
> > > > > > > > > > > > Please find here the dmesg output of the patched kernel
> > > > > > > > > > > At least 0xff is corrupted value in
> > > > > > > > > > > senseful way. CPU fills the read
> > > > > > > > > > > with ones for example for unaligned
> > > > > > > > > > > bus read. See table 19 in PC client
> > > > > > > > > > > spec. This can happen when you do unaligned read for example.
> > > > > > > > > > > 
> > > > > > > > > > > Maybe TPM is unreachable i.e.
> > > > > > > > > > > powered off. Bit busy with stuff ATM
> > > > > > > > > > > but
> > > > > > > > > > > would probably make sense to compare
> > > > > > > > > > > that 0x81 to table 18 in the same
> > > > > > > > > > > spec.
> > > > > > > > > > > 
> > > > > > > > > > > /Jarkko
> > > > > > > > > > 0x81 is saying the access register
> > > > > > > > > > status is valid, and the locality
> > > > > > > > > > is not active. That first bit means A
> > > > > > > > > > Dynamic OS has not been previously
> > > > > > > > > > established on the platform. Normally we
> > > > > > > > > > would see 0xa1, which would
> > > > > > > > > > mean valid register status, and the locality is active.
> > > > > > > > > I think the important thing to note here is
> > > > > > > > > that STS has bits set that
> > > > > > > > > should never be set. So we can conclude that TPM might be either
> > > > > > > > > 
> > > > > > > > > 1. Powered off
> > > > > > > > > 2. In some transition state?
> > > > > > > > > 
> > > > > > > > > /Jarkko
> > > > > > > > If it was powered off would we be getting a
> > > > > > > > valid read from the access
> > > > > > > > register?
> > > > > > > I think there is no universal answer to that :-)
> > > > > > > 
> > > > > > > Maybe adding a extra delay would be next test to make? If for random
> > > > > > > reason it is in-between states...
> > > > > > Any more ideas?
> > > > > > 
> > > > > > The chip is definitely in a weird state :/ I tried
> > > > > > several ways to reset the chip (windows, tpm-tools,...).
> > > > > > 
> > > > > > I've been able to reset the chip via the bios (which now
> > > > > > shows unowned) but chip is still locked apparently.
> > > > > > 
> > > > > > But still with < 4.12 I'm able to get /some/ information
> > > > > > Public EK, PCR,... out of the chip so it's not
> > > > > > completely broken...
> > > > > 
> > > > > OK correction the TPM is now unlocked (I let the computer
> > > > > running for more than 24h with nothing accessing the TPM)
> > > > > and with 4.9 I've been able to take the ownership again.
> > > > > 
> > > > > Under 4.12 I still have the same errors as mentioned originally
> > > > 
> > > > Hi Laurent,
> > > > 
> > > > Would it be possible for you to run ftrace from boot with the
> > > > following kernel parameters:
> > > > 
> > > > ftrace=function_graph ftrace_filter=tpm*
> > > > 
> > > 
> > > actually 'ftrace=function_graph
> > > ftrace_filter=tpm*,*locality,wait_for_tpm_stat' would be better
> > > 
> > > > and then send me the results of 'cat /sys/kernel/debug/tracing/trace' ?
> > Here you are.
> 
> Thank you Laurent.
> 
> A few interesting things here:
> 
> > # tracer: function_graph
> > #
> > # CPU  DURATION                  FUNCTION CALLS
> > # |     |   |                     |   |   |   |
> >  1)   4.510 us    |  tpm_init();
> >  1)               |  tpm_tis_pnp_init() {
> >  1)               |    tpm_tis_init.part.3() {
> >  1)               |      tpm_tis_core_init() {
> >  1)               |        tpmm_chip_alloc() {
> >  1)   3.935 us    |          tpm_chip_alloc();
> >  1)   4.787 us    |        }
> >  1)   0.199 us    |        tpm_tcg_read_bytes();
> >  1)   0.217 us    |        tpm_tcg_read32();
> >  1)   0.117 us    |        tpm_tcg_write32();
> >  1)               |        tpm2_probe() {
> >  1)               |          tpm_transmit_cmd() {
> >  1)               |            tpm_transmit() {
> >  1)               |              request_locality() {
> >  1)               |                check_locality() {
> >  1)   0.045 us    |                  tpm_tcg_read_bytes();
> >  1)   1.708 us    |                }
> >  1)   2.077 us    |              }
> 
> 
> So it goes request locality, does check to see if the locality
> is already active and it is, so doesn't go through the process
> of requesting the locality.
> 
> 
> >  1)   0.168 us    |              tpm2_prepare_space();
> >  1)               |              tpm_tis_send() {
> >  1)               |                tpm_tis_send_main() {
> >  1)               |                  tpm_tis_send_data() {
> >  1)   0.045 us    |                    tpm_tcg_read_bytes();
> >  1)   0.044 us    |                    tpm_tcg_read32();
> >  1)   1.723 us    |                    tpm_tcg_write_bytes();
> >  1)               |                    wait_for_tpm_stat() {
> >  1)               |                      tpm_tis_status() {
> >  1)   0.042 us    |                        tpm_tcg_read_bytes();
> >  1)   5.163 us    |                      }
> >  1)               |                      tpm_tis_status() {
> >  1)   0.167 us    |                        tpm_tcg_read_bytes();
> >  1)   2.182 us    |                      }
> >  1)               |                      tpm_tis_status() {
> >  1)   0.194 us    |                        tpm_tcg_read_bytes();
> >  1)   2.452 us    |                      }
> >  1) * 30406.24 us |                    }
> >  1)   0.088 us    |                    tpm_tcg_read_bytes();
> >  1)   0.123 us    |                    tpm_tcg_write_bytes();
> >  1)               |                    wait_for_tpm_stat() {
> >  1)               |                      tpm_tis_status() {
> >  1)   0.084 us    |                        tpm_tcg_read_bytes();
> >  1)   1.940 us    |                      }
> >  1)               |                      tpm_tis_status() {
> >  1)   0.170 us    |                        tpm_tcg_read_bytes();
> >  1)   2.217 us    |                      }
> >  1) * 16000.74 us |                    }
> >  1)   0.084 us    |                    tpm_tcg_read_bytes();
> >  1) * 46425.66 us |                  }
> >  1)   0.115 us    |                  tpm_tcg_write_bytes();
> >  1) * 46426.85 us |                }
> >  1) * 46427.35 us |              }
> >  1)               |              tpm_tis_status() {
> >  1)   1.389 us    |                tpm_tcg_read_bytes();
> >  1)   1.909 us    |              }
> >  1)   0.095 us    |              tpm_tis_req_canceled();
> >  1)               |              tpm_tis_status() {
> >  1)   0.187 us    |                tpm_tcg_read_bytes();
> >  1)   2.110 us    |              }
> >  1)               |              tpm_tis_recv() {
> >  1)               |                wait_for_tpm_stat() {
> >  1)               |                  tpm_tis_status() {
> >  1)   0.082 us    |                    tpm_tcg_read_bytes();
> >  1)   1.898 us    |                  }
> >  1)   2.425 us    |                }
> >  1)   0.102 us    |                tpm_tcg_read32();
> >  1)   5.453 us    |                tpm_tcg_read_bytes();
> >  1)               |                wait_for_tpm_stat() {
> >  1)               |                  tpm_tis_status() {
> >  1)   0.092 us    |                    tpm_tcg_read_bytes();
> >  1)   1.858 us    |                  }
> >  1)   2.332 us    |                }
> >  1)   0.077 us    |                tpm_tcg_read_bytes();
> >  1)   0.112 us    |                tpm_tcg_write_bytes();
> >  1) + 25.466 us   |              }
> >  1)   0.152 us    |              tpm2_commit_space();
> >  1)               |              release_locality() {
> >  1)   0.080 us    |                tpm_tcg_write_bytes();
> >  1)   0.672 us    |              }
> 
> Releases the locality as it comes back out of tpm_transmit_cmd
> 
> >  1) * 62349.91 us |            }
> >  1) * 62350.38 us |          }
> >  1) * 62350.87 us |        }
> >  1)   0.070 us    |        tpm_tcg_read32();
> >  1)   0.073 us    |        tpm_tcg_read_bytes();
> >  1)   0.096 us    |        tpm_tcg_read16();
> >  1)   0.042 us    |        tpm_tcg_read32();
> >  1)               |        tpm_chip_register() {
> >  1)               |          tpm1_auto_startup() {
> >  1)               |            tpm_get_timeouts() {
> >  1)               |              tpm_get_timeouts.part.6() {
> >  1)               |                tpm_getcap() {
> >  1)               |                  tpm_transmit_cmd() {
> >  1)               |                    tpm_transmit() {
> >  1)               |                      request_locality() {
> >  1)               |                        check_locality() {
> >  1)   0.091 us    |                          tpm_tcg_read_bytes();
> >  1)   1.680 us    |                        }
> >  1)   1.973 us    |                      }
> 
> Goes to request the locality again, but once again the check sees
> that the locality is already active, so doesn't go through process
> of requesting the locality. Very odd.
> 
> >  1)   0.072 us    |                      tpm2_prepare_space();
> >  1)               |                      tpm_tis_send() {
> >  1)               |                        tpm_tis_send_main() {
> >  1)               |                          tpm_tis_send_data() {
> >  1)   0.043 us    |                            tpm_tcg_read_bytes();
> 
> tpm_tis_stat inlined?
> 
> >  1)   0.071 us    |                            tpm_tcg_write_bytes();
> 
> tpm_tis_ready inlined?
> 
> >  1)               |                            wait_for_tpm_stat() {
> >  1)               |                              tpm_tis_status() {
> >  1)   0.042 us    |                                tpm_tcg_read_bytes();
> >  1)   1.634 us    |                              }
> >  1)               |                              tpm_tis_status() {
> >  1)   0.183 us    |                                tpm_tcg_read_bytes();
> >  1)   2.121 us    |                              }
> >  1) * 15942.66 us |                            }
> 
> The wait_for_tpm_stat does its loop twice checking the status register
> so at this point apparently it isn't reading 0xff.
> 
> >  1)   0.108 us    |                            tpm_tcg_read32();
> 
> This read is get_burstcount.
> 
> >  1)   1.773 us    |                            tpm_tcg_write_bytes();
> >  1)               |                            wait_for_tpm_stat() {
> >  1)               |                              tpm_tis_status() {
> >  1)   0.082 us    |                                tpm_tcg_read_bytes();
> >  1)   5.011 us    |                              }
> >  1)   5.618 us    |                            }
> >  1)   0.077 us    |                            tpm_tcg_read_bytes();
> 
> tpm_tis_status inlined?
> 
> >  1)   0.090 us    |                            tpm_tcg_write_bytes();
> 
> Come out of loop and write last byte
> 
> >  1)               |                            wait_for_tpm_stat() {
> >  1)               |                              tpm_tis_status() {
> >  1)   0.080 us    |                                tpm_tcg_read_bytes();
> >  1)   1.921 us    |                              }
> >  1)   2.376 us    |                            }
> >  1)   0.077 us    |                            tpm_tcg_read_bytes();
> 
> inlined tpm_tis_status again? Here is where it sees data is still expected and errors out.
> With that debugging patch at this point it was showing that the locality was not active.
> Also didn't show that the tpm had been seized by another locality.
> 
> >  1)   0.080 us    |                            tpm_tcg_write_bytes();
> 
> tpm_tis_ready inlined? in out_err
> 
> >  1) * 15970.97 us |                          }
> >  1) * 15971.39 us |                        }
> >  1) * 15971.81 us |                      }
> >  1)               |                      release_locality() {
> >  1)   0.103 us    |                        tpm_tcg_write_bytes();
> >  1)   0.535 us    |                      }
> >  1) * 16023.04 us |                    }
> >  1) * 16023.35 us |                  }
> >  1) * 16024.79 us |                }
> >  1) * 16067.18 us |              }
> >  1) * 16067.67 us |            }
> >  1) * 16068.10 us |          }
> >  1) * 16068.47 us |        }
> >  1) * 78448.69 us |      }
> >  1) * 78455.45 us |    }
> >  1) * 78456.71 us |  }
> >  1)   0.564 us    |  tpm_dev_release();
> >  0)               |  tpm_pcr_read() {
> >  0)   0.225 us    |    tpm_chip_find_get();
> >  0)   0.734 us    |  }
> 
> I wonder if it is possible that the release locality from the probe
> isn't completing on the chip until after the request/check that
> happens at the beginning of the tpm_get_timeouts.  Perhaps we need
> something like wait_for_tpm_stat for the access register, and
> verifying that locality was released before returning out of
> release_locality?
> 
> Any thoughts?

Sorry for late response.

Would make sense if that is the root cause. For just testing the
hypothesis (not something to be merged) adding a delay to release
would probably make sense (like 200 ms delay to make sure that the
TPM has done the state transition).

/Jarkko

Re: [tpmdd-devel] tpm device not showing up in /dev anymore

[Laurent Bigonville]

Le 24/10/17 a 15:51, Jarkko Sakkinen a ecrit :
> On Mon, Oct 23, 2017 at 06:45:15AM -0700, Jerry Snitselaar wrote:
>> On Mon Oct 23 17, Jarkko Sakkinen wrote:
>>> On Sat, Oct 21, 2017 at 10:53:55AM +0200, Laurent Bigonville wrote:
>>>> Le 14/10/17 a 10:13, Jerry Snitselaar a ecrit :
>>>>> On Wed Sep 06 17, Jarkko Sakkinen wrote:
>>>>>> On Fri, Sep 01, 2017 at 02:10:18PM +0200, Laurent Bigonville wrote:
>>>>>>> Le 31/08/17 a 18:40, Jerry Snitselaar a ecrit :
>>>>>>>> On Thu Aug 31 17, Alexander.Steffen@infineon.com wrote:
>>>>>>>>>> Le 29/08/17 a 18:35, Laurent Bigonville a ecrit :
>>>>>>>>>>> Le 29/08/17 a 18:00, Alexander.Steffen@infineon.com a ecrit :
>>>>>>>>>>>>> An idea how to troubleshoot this?
>>>>>>>>>>>> Can you run git bisect on the changes between 4.11 and
>>>>>>> 4.12, so that
>>>>>>>>>>>> we find the offending commit? It is probably sufficient
>>>>>>> to limit the
>>>>>>>>>>>> search to commits that touch something in drivers/char/tpm.
>>>>>>>>>>> I'll try and keep you posted.
>>>>>>>>>> OK I've been able to bisect the problem and the bad commit is:
>>>>>>>>>>
>>>>>>>>>> e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3 is the first bad commit
>>>>>>>>>> commit e6aef069b6e97790cb127d5eeb86ae9ff0b7b0e3
>>>>>>>>>> Author: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>> Date:   Mon Mar 27 08:46:04 2017 -0700
>>>>>>>>>>
>>>>>>>>>>       tpm_tis: convert to using locality callbacks
>>>>>>>>>>
>>>>>>>>>>       This patch converts tpm_tis to use of the new tpm class ops
>>>>>>>>>>       request_locality, and relinquish_locality.
>>>>>>>>>>
>>>>>>>>>>       With the move to using the callbacks, release_locality is
>>>>>>>>>> changed so
>>>>>>>>>>       that we now release the locality even if there is no
>>>>>>>>>> request pending.
>>>>>>>>>>
>>>>>>>>>>       This required some changes to the tpm_tis_core_init
>>>>>>> code path to
>>>>>>>>>>       make sure locality is requested when needed:
>>>>>>>>>>
>>>>>>>>>>         - tpm2_probe code path will end up calling
>>>>>>>>>> request/release through
>>>>>>>>>>           callbacks, so request_locality prior to
>>>>>>> tpm2_probe not needed.
>>>>>>>>>>         - probe_itpm makes calls to tpm_tis_send_data which no
>>>>>>>>>> longer calls
>>>>>>>>>>           request_locality, so add request_locality prior to
>>>>>>>>>> tpm_tis_send_data
>>>>>>>>>>           calls. Also drop release_locality call in middleof
>>>>>>>>>> probe_itpm, and
>>>>>>>>>>           keep locality until release_locality called at end of
>>>>>>>>>> probe_itpm.
>>>>>>>>>>
>>>>>>>>>>       Cc: Peter Huewe <peterhuewe@gmx.de>
>>>>>>>>>>       Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>       Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>>>>>>>>>>       Cc: Marcel Selhorst <tpmdd@selhorst.net>
>>>>>>>>>>       Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
>>>>>>>>>>       Reviewed-by: Jarkko Sakkinen
>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>       Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>>       Signed-off-by: Jarkko Sakkinen
>>>>>>> <jarkko.sakkinen@linux.intel.com>
>>>>>>>>>> :040000 040000 70234365da69959d47076ebb40c8d17f520c3e44
>>>>>>>>>> 72f21b446e45ea1003de75902b0553deb99157fd M drivers
>>>>>>>>>>
>>>>>>>>> I've looked again at the code in question, but could not find
>>>>>>>>> anything that is obviously wrong there. Locality is now
>>>>>>>>> requested/released at slightly different points in the process than
>>>>>>>>> before, but that's it. It does not seem to cause problems with the
>>>>>>>>> majority of TPMs, since you are the first to report any, so
>>>>>>> maybe it
>>>>>>>>> is a quirk that only affects this device.
>>>>>>>>>
>>>>>>>>> Perhaps Jerry can help, since this is his change?
>>>>>>>>>
>>>>>>>>> Alexander
>>>>>>>> Getting some caffeine in me, and starting to take a look. Adding
>>>>>>>> Jarkko as well since this might involve the general locality changes.
>>>>>>>>
>>>>>>>> Laurent, if I send you a patch with some debugging code added, would
>>>>>>>> you be able to run it on that system? I wasn't running into issues
>>>>>>>> on the system I had with a 1.2 device, but I no longer have access
>>>>>>>> to it. I'll see if I can find one in our labs and reproduce it there.
>>>>>>> Yes I should be able to do that
>>>>>> Any findings?
>>>>>>
>>>>>> /Jarkko
>>>>> Okay, finally getting back to this. Looking at the code it isn't clear
>>>>> to me
>>>>> why the change is causing this. So while I stare at this some more
>>>>> Laurent
>>>>> could you reproduce it with this patch so I can see what the status and
>>>>> access registers look like? Does anyone else on here happen to have a
>>>>> Sinosun
>>>>> tpm device? The systems I have access to with TPM1.2 devices don't have
>>>>> this
>>>>> issue.
>>>>>
>>>>> --8<--
>>>>>
>>>>> diff --git a/drivers/char/tpm/tpm_tis_core.c
>>>>> b/drivers/char/tpm/tpm_tis_core.c
>>>>> index fdde971bc810..7d60a7e4b50a 100644
>>>>> --- a/drivers/char/tpm/tpm_tis_core.c
>>>>> +++ b/drivers/char/tpm/tpm_tis_core.c
>>>>> @@ -258,6 +258,7 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>>>>> const u8 *buf, size_t len)
>>>>>      int rc, status, burstcnt;
>>>>>      size_t count = 0;
>>>>>      bool itpm = priv->flags & TPM_TIS_ITPM_WORKAROUND;
>>>>> +    u8 access;
>>>>>
>>>>>      status = tpm_tis_status(chip);
>>>>>      if ((status & TPM_STS_COMMAND_READY) == 0) {
>>>>> @@ -292,6 +293,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>>>>> const u8 *buf, size_t len)
>>>>>          }
>>>>>          status = tpm_tis_status(chip);
>>>>>          if (!itpm && (status & TPM_STS_DATA_EXPECT) == 0) {
>>>>> +            rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality),
>>>>> &access);
>>>>> +            if (rc < 0)
>>>>> +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0: read
>>>>> failure TPM_ACCESS(%d)\n", priv->locality);
>>>>> +            else
>>>>> +                dev_info(&chip->dev, "TPM_STS_DATA_EXPECT == 0:
>>>>> locality: %d status: %x access: %x\n", priv->locality, status, access);
>>>>>              rc = -EIO;
>>>>>              goto out_err;
>>>>>          }
>>>>> @@ -309,6 +315,11 @@ static int tpm_tis_send_data(struct tpm_chip *chip,
>>>>> const u8 *buf, size_t len)
>>>>>      }
>>>>>      status = tpm_tis_status(chip);
>>>>>      if (!itpm && (status & TPM_STS_DATA_EXPECT) != 0) {
>>>>> +        rc = tpm_tis_read8(priv, TPM_ACCESS(priv->locality), &access);
>>>>> +        if (rc < 0)
>>>>> +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: read
>>>>> failure TPM_ACCESS(%d)\n", priv->locality);
>>>>> +        else
>>>>> +            dev_info(&chip->dev, "TPM_STS_DATA_EXPECT != 0: locality:
>>>>> %d status: %x access: %x\n", priv->locality, status, access);
>>>>>          rc = -EIO;
>>>>>          goto out_err;
>>>>>      }
>>>> Please find here the dmesg output of the patched kernel
>>> At least 0xff is corrupted value in senseful way. CPU fills the read
>>> with ones for example for unaligned bus read. See table 19 in PC client
>>> spec. This can happen when you do unaligned read for example.
>>>
>>> Maybe TPM is unreachable i.e. powered off. Bit busy with stuff ATM but
>>> would probably make sense to compare that 0x81 to table 18 in the same
>>> spec.
>>>
>>> /Jarkko
>> 0x81 is saying the access register status is valid, and the locality
>> is not active. That first bit means A Dynamic OS has not been previously
>> established on the platform. Normally we would see 0xa1, which would
>> mean valid register status, and the locality is active.
> I think the important thing to note here is that STS has bits set that
> should never be set. So we can conclude that TPM might be either
>
> 1. Powered off
> 2. In some transition state?
MMhhhh It's actually possible that windows has blocked the chip after 
trying to enter automatically a wrong owner passphrase. But that doesn't 
explain device was showing up with previous kernel versions