| Message ID | 20190820122517.2086223-1-stefanb@linux.vnet.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | tpm_tis: Fix interrupt probing | expand |
On Tue, Aug 20, 2019 at 08:25:17AM -0400, Stefan Berger wrote: > From: Stefan Berger <stefanb@linux.ibm.com> > > The interrupt probing of the TPM TIS was broken since we are trying to > run it without an active locality and without the TPM_CHIP_FLAG_IRQ set. > > Signed-off-by: Stefan Berger <stefanb@linux.ibm.com> Should have a fixes tag. I can add it. Thank you. Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com> /Jarkko
On Tue, Aug 20, 2019 at 08:25:17AM -0400, Stefan Berger wrote: > From: Stefan Berger <stefanb@linux.ibm.com> > > The interrupt probing of the TPM TIS was broken since we are trying to > run it without an active locality and without the TPM_CHIP_FLAG_IRQ set. > > Signed-off-by: Stefan Berger <stefanb@linux.ibm.com> Need these: Cc: linux-stable@vger.kernel.org Fixes: a3fbfae82b4c ("tpm: take TPM chip power gating out of tpm_transmit()") Thank you. I'll apply this to my tree. Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com> /Jarkko
On Tue, Aug 27, 2019 at 04:14:00PM +0300, Jarkko Sakkinen wrote: > On Tue, Aug 20, 2019 at 08:25:17AM -0400, Stefan Berger wrote: > > From: Stefan Berger <stefanb@linux.ibm.com> > > > > The interrupt probing of the TPM TIS was broken since we are trying to > > run it without an active locality and without the TPM_CHIP_FLAG_IRQ set. > > > > Signed-off-by: Stefan Berger <stefanb@linux.ibm.com> > > Need these: > > Cc: linux-stable@vger.kernel.org > Fixes: a3fbfae82b4c ("tpm: take TPM chip power gating out of tpm_transmit()") > > Thank you. I'll apply this to my tree. > > Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com> The commit went in the following form: http://git.infradead.org/users/jjs/linux-tpmdd.git/commit/9b558deab2c5d7dc23d5f7a4064892ede482ad32 I refined the long description as they should be written in imperative form. I also changed it to consistently to speak about tpm_tis_core instead of using two differing spellings (tpm_tis and TPM TIS). tpm_tis is a different module than tpm_tis_core. Unfortunately I had to drop the assignment statement because: 1. Generally speaking, two separate bug fixes should never reside in the same commit. They even need their own fixes tags in this case. 2. The commit message did not reason the assignment statement. /Jarkko
On 8/27/19 11:19 AM, Jarkko Sakkinen wrote: > On Tue, Aug 27, 2019 at 04:14:00PM +0300, Jarkko Sakkinen wrote: >> On Tue, Aug 20, 2019 at 08:25:17AM -0400, Stefan Berger wrote: >>> From: Stefan Berger <stefanb@linux.ibm.com> >>> >>> The interrupt probing of the TPM TIS was broken since we are trying to >>> run it without an active locality and without the TPM_CHIP_FLAG_IRQ set. >>> >>> Signed-off-by: Stefan Berger <stefanb@linux.ibm.com> >> Need these: >> >> Cc: linux-stable@vger.kernel.org >> Fixes: a3fbfae82b4c ("tpm: take TPM chip power gating out of tpm_transmit()") >> >> Thank you. I'll apply this to my tree. >> >> Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com> > The commit went in the following form: > > http://git.infradead.org/users/jjs/linux-tpmdd.git/commit/9b558deab2c5d7dc23d5f7a4064892ede482ad32 I saw you dropped the stetting of the IRQ flag - I needed it, otherwise it wouldn't execute certain code paths. Stefan
On Tue, Aug 27, 2019 at 03:34:36PM -0400, Stefan Berger wrote: > On 8/27/19 11:19 AM, Jarkko Sakkinen wrote: > > On Tue, Aug 27, 2019 at 04:14:00PM +0300, Jarkko Sakkinen wrote: > > > On Tue, Aug 20, 2019 at 08:25:17AM -0400, Stefan Berger wrote: > > > > From: Stefan Berger <stefanb@linux.ibm.com> > > > > > > > > The interrupt probing of the TPM TIS was broken since we are trying to > > > > run it without an active locality and without the TPM_CHIP_FLAG_IRQ set. > > > > > > > > Signed-off-by: Stefan Berger <stefanb@linux.ibm.com> > > > Need these: > > > > > > Cc: linux-stable@vger.kernel.org > > > Fixes: a3fbfae82b4c ("tpm: take TPM chip power gating out of tpm_transmit()") > > > > > > Thank you. I'll apply this to my tree. > > > > > > Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com> > > The commit went in the following form: > > > > http://git.infradead.org/users/jjs/linux-tpmdd.git/commit/9b558deab2c5d7dc23d5f7a4064892ede482ad32 > > I saw you dropped the stetting of the IRQ flag - I needed it, otherwise it > wouldn't execute certain code paths. I explained why I removed that part. There was no any reasoning for it. Also, it cannot be in the same commit if it fixes a diffent issue. /Jarkko
On Thu, Aug 29, 2019 at 04:20:21PM +0300, Jarkko Sakkinen wrote: > On Tue, Aug 27, 2019 at 03:34:36PM -0400, Stefan Berger wrote: > > On 8/27/19 11:19 AM, Jarkko Sakkinen wrote: > > > On Tue, Aug 27, 2019 at 04:14:00PM +0300, Jarkko Sakkinen wrote: > > > > On Tue, Aug 20, 2019 at 08:25:17AM -0400, Stefan Berger wrote: > > > > > From: Stefan Berger <stefanb@linux.ibm.com> > > > > > > > > > > The interrupt probing of the TPM TIS was broken since we are trying to > > > > > run it without an active locality and without the TPM_CHIP_FLAG_IRQ set. > > > > > > > > > > Signed-off-by: Stefan Berger <stefanb@linux.ibm.com> > > > > Need these: > > > > > > > > Cc: linux-stable@vger.kernel.org > > > > Fixes: a3fbfae82b4c ("tpm: take TPM chip power gating out of tpm_transmit()") > > > > > > > > Thank you. I'll apply this to my tree. > > > > > > > > Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com> > > > The commit went in the following form: > > > > > > http://git.infradead.org/users/jjs/linux-tpmdd.git/commit/9b558deab2c5d7dc23d5f7a4064892ede482ad32 > > > > I saw you dropped the stetting of the IRQ flag - I needed it, otherwise it > > wouldn't execute certain code paths. > > I explained why I removed that part. There was no any reasoning for > it. Also, it cannot be in the same commit if it fixes a diffent > issue. AFAIK they go with different fixes-tags. /Jarkko
On 8/29/19 12:10 PM, Jarkko Sakkinen wrote: > On Thu, Aug 29, 2019 at 04:20:21PM +0300, Jarkko Sakkinen wrote: >> On Tue, Aug 27, 2019 at 03:34:36PM -0400, Stefan Berger wrote: >>> On 8/27/19 11:19 AM, Jarkko Sakkinen wrote: >>>> On Tue, Aug 27, 2019 at 04:14:00PM +0300, Jarkko Sakkinen wrote: >>>>> On Tue, Aug 20, 2019 at 08:25:17AM -0400, Stefan Berger wrote: >>>>>> From: Stefan Berger <stefanb@linux.ibm.com> >>>>>> >>>>>> The interrupt probing of the TPM TIS was broken since we are trying to >>>>>> run it without an active locality and without the TPM_CHIP_FLAG_IRQ set. >>>>>> >>>>>> Signed-off-by: Stefan Berger <stefanb@linux.ibm.com> >>>>> Need these: >>>>> >>>>> Cc: linux-stable@vger.kernel.org >>>>> Fixes: a3fbfae82b4c ("tpm: take TPM chip power gating out of tpm_transmit()") >>>>> >>>>> Thank you. I'll apply this to my tree. >>>>> >>>>> Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com> >>>> The commit went in the following form: >>>> >>>> http://git.infradead.org/users/jjs/linux-tpmdd.git/commit/9b558deab2c5d7dc23d5f7a4064892ede482ad32 >>> I saw you dropped the stetting of the IRQ flag - I needed it, otherwise it >>> wouldn't execute certain code paths. >> I explained why I removed that part. There was no any reasoning for >> it. Also, it cannot be in the same commit if it fixes a diffent >> issue. > AFAIK they go with different fixes-tags. I sent a separate patch for this. It looks like this bug goes back to when the TPM_CHIP_FLAG_IRQ was introduced in March 2019?! Stefan
On Thu, Aug 29, 2019 at 08:11:27PM -0400, Stefan Berger wrote: > On 8/29/19 12:10 PM, Jarkko Sakkinen wrote: > > On Thu, Aug 29, 2019 at 04:20:21PM +0300, Jarkko Sakkinen wrote: > > > On Tue, Aug 27, 2019 at 03:34:36PM -0400, Stefan Berger wrote: > > > > On 8/27/19 11:19 AM, Jarkko Sakkinen wrote: > > > > > On Tue, Aug 27, 2019 at 04:14:00PM +0300, Jarkko Sakkinen wrote: > > > > > > On Tue, Aug 20, 2019 at 08:25:17AM -0400, Stefan Berger wrote: > > > > > > > From: Stefan Berger <stefanb@linux.ibm.com> > > > > > > > > > > > > > > The interrupt probing of the TPM TIS was broken since we are trying to > > > > > > > run it without an active locality and without the TPM_CHIP_FLAG_IRQ set. > > > > > > > > > > > > > > Signed-off-by: Stefan Berger <stefanb@linux.ibm.com> > > > > > > Need these: > > > > > > > > > > > > Cc: linux-stable@vger.kernel.org > > > > > > Fixes: a3fbfae82b4c ("tpm: take TPM chip power gating out of tpm_transmit()") > > > > > > > > > > > > Thank you. I'll apply this to my tree. > > > > > > > > > > > > Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com> > > > > > The commit went in the following form: > > > > > > > > > > http://git.infradead.org/users/jjs/linux-tpmdd.git/commit/9b558deab2c5d7dc23d5f7a4064892ede482ad32 > > > > I saw you dropped the stetting of the IRQ flag - I needed it, otherwise it > > > > wouldn't execute certain code paths. > > > I explained why I removed that part. There was no any reasoning for > > > it. Also, it cannot be in the same commit if it fixes a diffent > > > issue. > > AFAIK they go with different fixes-tags. > > I sent a separate patch for this. It looks like this bug goes back to when > the TPM_CHIP_FLAG_IRQ was introduced in March 2019?! Thank you! Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com> /Jarkko
diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c index c3181ea9f271..270f43acbb77 100644 --- a/drivers/char/tpm/tpm_tis_core.c +++ b/drivers/char/tpm/tpm_tis_core.c @@ -980,6 +980,8 @@ int tpm_tis_core_init(struct device *dev, struct tpm_tis_data *priv, int irq, goto out_err; } + tpm_chip_start(chip); + chip->flags |= TPM_CHIP_FLAG_IRQ; if (irq) { tpm_tis_probe_irq_single(chip, intmask, IRQF_SHARED, irq); @@ -989,6 +991,7 @@ int tpm_tis_core_init(struct device *dev, struct tpm_tis_data *priv, int irq, } else { tpm_tis_probe_irq(chip, intmask); } + tpm_chip_stop(chip); } rc = tpm_chip_register(chip);