Re: [PATCH 2/2 v3] tpm: separate cmd_ready/go_idle from runtime_pm

[Jarkko Sakkinen]

On Mon, Mar 05, 2018 at 08:08:31PM +0200, Jarkko Sakkinen wrote:
> On Tue, Feb 20, 2018 at 04:13:30PM +0200, Jarkko Sakkinen wrote:
> > On Mon, 2018-02-19 at 13:52 +0200, Jarkko Sakkinen wrote:
> > > On Wed, Feb 14, 2018 at 03:43:19PM +0200, Tomas Winkler wrote:
> > > > We cannot use go_idle cmd_ready commands via runtime_pm handles
> > > > as with the introduction of localities this is no longer an
> > > > optional
> > > > feature, while runtime pm can be not enabled.
> > > > Though cmd_ready/go_idle provides power saving feature, it's also
> > > > part of
> > > > TPM2 protocol and should be called explicitly.
> > > > This patch exposes cmd_read/go_idle via tpm class ops and removes
> > > > runtime pm support as it is not used by any driver.
> > > > 
> > > > Signed-off-by: Tomas Winkler 
> > > > ---
> > > > V2: resent
> > > > V3: resent
> > > 
> > > Only the cover letter needs a backlog but the code change looks
> > > good to me.
> > > 
> > > Reviewed-by: Jarkko Sakkinen 
> > 
> > Tested-by: Jarkko Sakkinen 
> > 
> > /Jarkko
> 
> Applied.
> 
> /Jarkko

This commit breaks my smoke tests on Kabylake [1] so I need to drop it.
The bug fix for locality works without any issues.

[1]
$ python -m unittest -v tpm2_smoke
test_seal_with_auth (tpm2_smoke.SmokeTest) ... ok
test_seal_with_policy (tpm2_smoke.SmokeTest) ... ok
test_seal_with_too_long_auth (tpm2_smoke.SmokeTest) ... ok
test_too_short_cmd (tpm2_smoke.SmokeTest) ... ok
test_unseal_with_wrong_auth (tpm2_smoke.SmokeTest) ... ok
test_unseal_with_wrong_policy (tpm2_smoke.SmokeTest) ... ok
test_flush_context (tpm2_smoke.SpaceTest) ... ERROR
test_get_handles (tpm2_smoke.SpaceTest) ... ERROR
test_invalid_cc (tpm2_smoke.SpaceTest) ... ok
test_make_two_spaces (tpm2_smoke.SpaceTest) ... ERROR

==
ERROR: test_flush_context (tpm2_smoke.SpaceTest)
--
Traceback (most recent call last):
  File "tpm2_smoke.py", line 185, in test_flush_context
space1.flush_context(root1)
  File "tpm2.py", line 517, in flush_context
self.send_cmd(cmd)
  File "tpm2.py", line 378, in send_cmd
raise ProtocolError(cc, rc)
ProtocolError: TPM_RC_UNKNOWN: cc=0x0165, rc=0x0165

==
ERROR: test_get_handles (tpm2_smoke.SpaceTest)
--
Traceback (most recent call last):
  File "tpm2_smoke.py", line 195, in test_get_handles
space2.create_root_key()
  File "tpm2.py", line 562, in create_root_key
return struct.unpack('>I', self.send_cmd(cmd)[10:14])[0]
  File "tpm2.py", line 378, in send_cmd
raise ProtocolError(cc, rc)
ProtocolError: TPM_RC_UNBALANCED: cc=0x0131, rc=0x0131

==
ERROR: test_make_two_spaces (tpm2_smoke.SpaceTest)
--
Traceback (most recent call last):
  File "tpm2_smoke.py", line 171, in test_make_two_spaces
root3 = space2.create_root_key()
  File "tpm2.py", line 562, in create_root_key
return struct.unpack('>I', self.send_cmd(cmd)[10:14])[0]
  File "tpm2.py", line 378, in send_cmd
raise ProtocolError(cc, rc)
ProtocolError: TPM_RC_UNBALANCED: cc=0x0131, rc=0x0131

--
Ran 10 tests in 44.545s

FAILED (errors=3)

/Jarkko

Re: [PATCH 2/2 v3] tpm: separate cmd_ready/go_idle from runtime_pm

[Jarkko Sakkinen]

On Mon, Mar 05, 2018 at 08:08:31PM +0200, Jarkko Sakkinen wrote:
> On Tue, Feb 20, 2018 at 04:13:30PM +0200, Jarkko Sakkinen wrote:
> > On Mon, 2018-02-19 at 13:52 +0200, Jarkko Sakkinen wrote:
> > > On Wed, Feb 14, 2018 at 03:43:19PM +0200, Tomas Winkler wrote:
> > > > We cannot use go_idle cmd_ready commands via runtime_pm handles
> > > > as with the introduction of localities this is no longer an
> > > > optional
> > > > feature, while runtime pm can be not enabled.
> > > > Though cmd_ready/go_idle provides power saving feature, it's also
> > > > part of
> > > > TPM2 protocol and should be called explicitly.
> > > > This patch exposes cmd_read/go_idle via tpm class ops and removes
> > > > runtime pm support as it is not used by any driver.
> > > > 
> > > > Signed-off-by: Tomas Winkler 
> > > > ---
> > > > V2: resent
> > > > V3: resent
> > > 
> > > Only the cover letter needs a backlog but the code change looks
> > > good to me.
> > > 
> > > Reviewed-by: Jarkko Sakkinen 
> > 
> > Tested-by: Jarkko Sakkinen 
> > 
> > /Jarkko
> 
> Applied.
> 
> /Jarkko

This commit breaks my smoke tests on Kabylake [1] so I need to drop it.
The bug fix for locality works without any issues.

[1]
$ python -m unittest -v tpm2_smoke
test_seal_with_auth (tpm2_smoke.SmokeTest) ... ok
test_seal_with_policy (tpm2_smoke.SmokeTest) ... ok
test_seal_with_too_long_auth (tpm2_smoke.SmokeTest) ... ok
test_too_short_cmd (tpm2_smoke.SmokeTest) ... ok
test_unseal_with_wrong_auth (tpm2_smoke.SmokeTest) ... ok
test_unseal_with_wrong_policy (tpm2_smoke.SmokeTest) ... ok
test_flush_context (tpm2_smoke.SpaceTest) ... ERROR
test_get_handles (tpm2_smoke.SpaceTest) ... ERROR
test_invalid_cc (tpm2_smoke.SpaceTest) ... ok
test_make_two_spaces (tpm2_smoke.SpaceTest) ... ERROR

==
ERROR: test_flush_context (tpm2_smoke.SpaceTest)
--
Traceback (most recent call last):
  File "tpm2_smoke.py", line 185, in test_flush_context
space1.flush_context(root1)
  File "tpm2.py", line 517, in flush_context
self.send_cmd(cmd)
  File "tpm2.py", line 378, in send_cmd
raise ProtocolError(cc, rc)
ProtocolError: TPM_RC_UNKNOWN: cc=0x0165, rc=0x0165

==
ERROR: test_get_handles (tpm2_smoke.SpaceTest)
--
Traceback (most recent call last):
  File "tpm2_smoke.py", line 195, in test_get_handles
space2.create_root_key()
  File "tpm2.py", line 562, in create_root_key
return struct.unpack('>I', self.send_cmd(cmd)[10:14])[0]
  File "tpm2.py", line 378, in send_cmd
raise ProtocolError(cc, rc)
ProtocolError: TPM_RC_UNBALANCED: cc=0x0131, rc=0x0131

==
ERROR: test_make_two_spaces (tpm2_smoke.SpaceTest)
--
Traceback (most recent call last):
  File "tpm2_smoke.py", line 171, in test_make_two_spaces
root3 = space2.create_root_key()
  File "tpm2.py", line 562, in create_root_key
return struct.unpack('>I', self.send_cmd(cmd)[10:14])[0]
  File "tpm2.py", line 378, in send_cmd
raise ProtocolError(cc, rc)
ProtocolError: TPM_RC_UNBALANCED: cc=0x0131, rc=0x0131

--
Ran 10 tests in 44.545s

FAILED (errors=3)

/Jarkko

Re: [PATCH 2/2 v3] tpm: separate cmd_ready/go_idle from runtime_pm

[Jarkko Sakkinen]

On Tue, Feb 20, 2018 at 04:13:30PM +0200, Jarkko Sakkinen wrote:
> On Mon, 2018-02-19 at 13:52 +0200, Jarkko Sakkinen wrote:
> > On Wed, Feb 14, 2018 at 03:43:19PM +0200, Tomas Winkler wrote:
> > > We cannot use go_idle cmd_ready commands via runtime_pm handles
> > > as with the introduction of localities this is no longer an
> > > optional
> > > feature, while runtime pm can be not enabled.
> > > Though cmd_ready/go_idle provides power saving feature, it's also
> > > part of
> > > TPM2 protocol and should be called explicitly.
> > > This patch exposes cmd_read/go_idle via tpm class ops and removes
> > > runtime pm support as it is not used by any driver.
> > > 
> > > Signed-off-by: Tomas Winkler 
> > > ---
> > > V2: resent
> > > V3: resent
> > 
> > Only the cover letter needs a backlog but the code change looks
> > good to me.
> > 
> > Reviewed-by: Jarkko Sakkinen 
> 
> Tested-by: Jarkko Sakkinen 
> 
> /Jarkko

Applied.

/Jarkko

Re: [PATCH 2/2 v3] tpm: separate cmd_ready/go_idle from runtime_pm

[Jarkko Sakkinen]

On Tue, Feb 20, 2018 at 04:13:30PM +0200, Jarkko Sakkinen wrote:
> On Mon, 2018-02-19 at 13:52 +0200, Jarkko Sakkinen wrote:
> > On Wed, Feb 14, 2018 at 03:43:19PM +0200, Tomas Winkler wrote:
> > > We cannot use go_idle cmd_ready commands via runtime_pm handles
> > > as with the introduction of localities this is no longer an
> > > optional
> > > feature, while runtime pm can be not enabled.
> > > Though cmd_ready/go_idle provides power saving feature, it's also
> > > part of
> > > TPM2 protocol and should be called explicitly.
> > > This patch exposes cmd_read/go_idle via tpm class ops and removes
> > > runtime pm support as it is not used by any driver.
> > > 
> > > Signed-off-by: Tomas Winkler 
> > > ---
> > > V2: resent
> > > V3: resent
> > 
> > Only the cover letter needs a backlog but the code change looks
> > good to me.
> > 
> > Reviewed-by: Jarkko Sakkinen 
> 
> Tested-by: Jarkko Sakkinen 
> 
> /Jarkko

Applied.

/Jarkko

Re: [PATCH 2/2 v3] tpm: separate cmd_ready/go_idle from runtime_pm

[Jarkko Sakkinen]

On Mon, 2018-02-19 at 13:52 +0200, Jarkko Sakkinen wrote:
> On Wed, Feb 14, 2018 at 03:43:19PM +0200, Tomas Winkler wrote:
> > We cannot use go_idle cmd_ready commands via runtime_pm handles
> > as with the introduction of localities this is no longer an
> > optional
> > feature, while runtime pm can be not enabled.
> > Though cmd_ready/go_idle provides power saving feature, it's also
> > part of
> > TPM2 protocol and should be called explicitly.
> > This patch exposes cmd_read/go_idle via tpm class ops and removes
> > runtime pm support as it is not used by any driver.
> > 
> > Signed-off-by: Tomas Winkler 
> > ---
> > V2: resent
> > V3: resent
> 
> Only the cover letter needs a backlog but the code change looks
> good to me.
> 
> Reviewed-by: Jarkko Sakkinen 

Tested-by: Jarkko Sakkinen 

/Jarkko

Re: [PATCH 2/2 v3] tpm: separate cmd_ready/go_idle from runtime_pm

[Jarkko Sakkinen]

On Mon, 2018-02-19 at 13:52 +0200, Jarkko Sakkinen wrote:
> On Wed, Feb 14, 2018 at 03:43:19PM +0200, Tomas Winkler wrote:
> > We cannot use go_idle cmd_ready commands via runtime_pm handles
> > as with the introduction of localities this is no longer an
> > optional
> > feature, while runtime pm can be not enabled.
> > Though cmd_ready/go_idle provides power saving feature, it's also
> > part of
> > TPM2 protocol and should be called explicitly.
> > This patch exposes cmd_read/go_idle via tpm class ops and removes
> > runtime pm support as it is not used by any driver.
> > 
> > Signed-off-by: Tomas Winkler 
> > ---
> > V2: resent
> > V3: resent
> 
> Only the cover letter needs a backlog but the code change looks
> good to me.
> 
> Reviewed-by: Jarkko Sakkinen 

Tested-by: Jarkko Sakkinen 

/Jarkko

Re: [PATCH 2/2 v3] tpm: separate cmd_ready/go_idle from runtime_pm

[Jarkko Sakkinen]

On Wed, Feb 14, 2018 at 03:43:19PM +0200, Tomas Winkler wrote:
> We cannot use go_idle cmd_ready commands via runtime_pm handles
> as with the introduction of localities this is no longer an optional
> feature, while runtime pm can be not enabled.
> Though cmd_ready/go_idle provides power saving feature, it's also part of
> TPM2 protocol and should be called explicitly.
> This patch exposes cmd_read/go_idle via tpm class ops and removes
> runtime pm support as it is not used by any driver.
> 
> Signed-off-by: Tomas Winkler 
> ---
> V2: resent
> V3: resent

Only the cover letter needs a backlog but the code change looks
good to me.

Reviewed-by: Jarkko Sakkinen 

/Jarkko

Re: [PATCH 2/2 v3] tpm: separate cmd_ready/go_idle from runtime_pm

[Jarkko Sakkinen]

On Wed, Feb 14, 2018 at 03:43:19PM +0200, Tomas Winkler wrote:
> We cannot use go_idle cmd_ready commands via runtime_pm handles
> as with the introduction of localities this is no longer an optional
> feature, while runtime pm can be not enabled.
> Though cmd_ready/go_idle provides power saving feature, it's also part of
> TPM2 protocol and should be called explicitly.
> This patch exposes cmd_read/go_idle via tpm class ops and removes
> runtime pm support as it is not used by any driver.
> 
> Signed-off-by: Tomas Winkler 
> ---
> V2: resent
> V3: resent

Only the cover letter needs a backlog but the code change looks
good to me.

Reviewed-by: Jarkko Sakkinen 

/Jarkko