From: Amir Mizinski <[email protected]>
tpm_tis_req_canceled() function is used to check if the caller requested
to abort the current operation. It was found that in some cases
tpm_tis_req_canceled() wrongly returned true.
Since a cancel request sets the TPM_STS.commandReady field to TRUE, the
tpm_tis_req_canceled() function should check only the TPM_STS.commandReady
field value.
The case for TPM_VID_WINBOND is wrong and was therefore removed.
Also, the default comparison is wrong. Only cmdReady bit needs to be
compared instead of the full lower status register byte.
Signed-off-by: Amir Mizinski <[email protected]>
---
drivers/char/tpm/tpm_tis_core.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
index f833f35..6ff8b44 100644
--- a/drivers/char/tpm/tpm_tis_core.c
+++ b/drivers/char/tpm/tpm_tis_core.c
@@ -661,13 +661,10 @@ static bool tpm_tis_req_canceled(struct tpm_chip *chip, u8 status)
struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
switch (priv->manufacturer_id) {
- case TPM_VID_WINBOND:
- return ((status == TPM_STS_VALID) ||
- (status == (TPM_STS_VALID | TPM_STS_COMMAND_READY)));
case TPM_VID_STM:
return (status == (TPM_STS_VALID | TPM_STS_COMMAND_READY));
default:
- return (status == TPM_STS_COMMAND_READY);
+ return (status & TPM_STS_COMMAND_READY) == TPM_STS_COMMAND_READY;
}
}
--
2.7.4
On Thu, 2021-11-04 at 16:02 +0200, [email protected] wrote:
> From: Amir Mizinski <[email protected]>
>
> tpm_tis_req_canceled() function is used to check if the caller requested
> to abort the current operation. It was found that in some cases
> tpm_tis_req_canceled() wrongly returned true.
Please, bring some context, i.e. please describe the scenarios.
> Since a cancel request sets the TPM_STS.commandReady field to TRUE, the
~~~~~~~~~~~~~~
What the heck is this? Please refer the exact things.
> tpm_tis_req_canceled() function should check only the TPM_STS.commandReady
> field value.
Why?
> The case for TPM_VID_WINBOND is wrong and was therefore removed.
Why?
It is not removed in the existing mainline, so it definitely *was not*
removed.
> Also, the default comparison is wrong. Only cmdReady bit needs to be
> compared instead of the full lower status register byte.
You should split this into two patches, if it fixes two different
issues.
/Jarkko