[PATCH] tpm: Replace WARN_ONCE() with dev_err_once() in tpm_tis_status()

[PATCH] tpm: Replace WARN_ONCE() with dev_err_once() in tpm_tis_status()

[Jarkko Sakkinen]

Do not torn down the system when getting invalid status from a TPM chip.
This can happen when panic-on-warn is used.

In addition, print out the value of TPM_STS.x instead of "invalid
status". In order to get the earlier benefits for forensics, also call
dump_stack().

Link: https://lore.kernel.org/keyrings/YKzlTR1AzUigShtZ@kroah.com/
Fixes: 55707d531af6 ("tpm_tis: Add a check for invalid status")
Cc: stable@vger.kernel.org
Cc: Hans de Goede <hdegoede@redhat.com>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
---
 drivers/char/tpm/tpm_tis_core.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
index 55b9d3965ae1..514a481829c9 100644
--- a/drivers/char/tpm/tpm_tis_core.c
+++ b/drivers/char/tpm/tpm_tis_core.c
@@ -202,7 +202,16 @@ static u8 tpm_tis_status(struct tpm_chip *chip)
 		 * acquired.  Usually because tpm_try_get_ops() hasn't
 		 * been called before doing a TPM operation.
 		 */
-		WARN_ONCE(1, "TPM returned invalid status\n");
+		dev_err_once(&chip->dev, "invalid TPM_STS.x 0x%02x, dumping stack for forensics\n",
+			     status);
+
+		/*
+		 * Dump stack for forensics, as invalid TPM_STS.x could be
+		 * potentially triggered by impaired tpm_try_get_ops() or
+		 * tpm_find_get_ops().
+		 */
+		dump_stack();
+
 		return 0;
 	}
 
-- 
2.31.1

Re: [PATCH] tpm: Replace WARN_ONCE() with dev_err_once() in tpm_tis_status()

[Greg KH]

On Mon, May 31, 2021 at 07:51:31AM +0300, Jarkko Sakkinen wrote:
> Do not torn down the system when getting invalid status from a TPM chip.
> This can happen when panic-on-warn is used.
> 
> In addition, print out the value of TPM_STS.x instead of "invalid
> status". In order to get the earlier benefits for forensics, also call
> dump_stack().
> 
> Link: https://lore.kernel.org/keyrings/YKzlTR1AzUigShtZ@kroah.com/
> Fixes: 55707d531af6 ("tpm_tis: Add a check for invalid status")
> Cc: stable@vger.kernel.org
> Cc: Hans de Goede <hdegoede@redhat.com>
> Cc: Greg KH <greg@kroah.com>
> Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
> ---
>  drivers/char/tpm/tpm_tis_core.c | 11 ++++++++++-
>  1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
> index 55b9d3965ae1..514a481829c9 100644
> --- a/drivers/char/tpm/tpm_tis_core.c
> +++ b/drivers/char/tpm/tpm_tis_core.c
> @@ -202,7 +202,16 @@ static u8 tpm_tis_status(struct tpm_chip *chip)
>  		 * acquired.  Usually because tpm_try_get_ops() hasn't
>  		 * been called before doing a TPM operation.
>  		 */
> -		WARN_ONCE(1, "TPM returned invalid status\n");
> +		dev_err_once(&chip->dev, "invalid TPM_STS.x 0x%02x, dumping stack for forensics\n",
> +			     status);
> +
> +		/*
> +		 * Dump stack for forensics, as invalid TPM_STS.x could be
> +		 * potentially triggered by impaired tpm_try_get_ops() or
> +		 * tpm_find_get_ops().
> +		 */
> +		dump_stack();
> +
>  		return 0;
>  	}

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Re: [PATCH] tpm: Replace WARN_ONCE() with dev_err_once() in tpm_tis_status()

[Paul Menzel]

Dear Jarkko,


Am 31.05.21 um 06:51 schrieb Jarkko Sakkinen:
> Do not torn down the system when getting invalid status from a TPM chip.

Nit: Do not *tear* down …?

> This can happen when panic-on-warn is used.

Hmm, I’d say it works as expected then? Shouldn’t an invalid status of 
an important device like TPM considered a warning?

> In addition, print out the value of TPM_STS.x instead of "invalid
> status". In order to get the earlier benefits for forensics, also call
> dump_stack().
> 
> Link: https://lore.kernel.org/keyrings/YKzlTR1AzUigShtZ@kroah.com/
> Fixes: 55707d531af6 ("tpm_tis: Add a check for invalid status")
> Cc: stable@vger.kernel.org
> Cc: Hans de Goede <hdegoede@redhat.com>
> Cc: Greg KH <greg@kroah.com>
> Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
> ---
>   drivers/char/tpm/tpm_tis_core.c | 11 ++++++++++-
>   1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
> index 55b9d3965ae1..514a481829c9 100644
> --- a/drivers/char/tpm/tpm_tis_core.c
> +++ b/drivers/char/tpm/tpm_tis_core.c
> @@ -202,7 +202,16 @@ static u8 tpm_tis_status(struct tpm_chip *chip)
>   		 * acquired.  Usually because tpm_try_get_ops() hasn't
>   		 * been called before doing a TPM operation.
>   		 */
> -		WARN_ONCE(1, "TPM returned invalid status\n");
> +		dev_err_once(&chip->dev, "invalid TPM_STS.x 0x%02x, dumping stack for forensics\n",
> +			     status);
> +
> +		/*
> +		 * Dump stack for forensics, as invalid TPM_STS.x could be
> +		 * potentially triggered by impaired tpm_try_get_ops() or
> +		 * tpm_find_get_ops().
> +		 */
> +		dump_stack();
> +
>   		return 0;
>   	}


Kind regards,

Paul

Re: [PATCH] tpm: Replace WARN_ONCE() with dev_err_once() in tpm_tis_status()

[Jarkko Sakkinen]

On Mon, May 31, 2021 at 07:33:44AM +0200, Paul Menzel wrote:
> Dear Jarkko,
> 
> 
> Am 31.05.21 um 06:51 schrieb Jarkko Sakkinen:
> > Do not torn down the system when getting invalid status from a TPM chip.
> 
> Nit: Do not *tear* down …?

Oops, a typo, thank you.

> > This can happen when panic-on-warn is used.
> 
> Hmm, I’d say it works as expected then? Shouldn’t an invalid status of an
> important device like TPM considered a warning?

By warning do you mean WARN() or pr_warn()?

/Jarkko