Re: [PATCH] tpm: Fix the timeout & use ktime

[Jonathan McDowell]

On Wed, Jun 11, 2025 at 04:25:24PM +0000, Orlov, Ivan wrote:
> The current implementation of timeout detection works in the following
> way:
>
> 1. Read completion status. If completed, return the data
> 2. Sleep for some time (usleep_range)
> 3. Check for timeout using current jiffies value. Return an error if
>   timed out
> 4. Goto 1
>
> usleep_range doesn't guarantee it's always going to wake up strictly in
> (min, max) range, so such a situation is possible:
>
> 1. Driver reads completion status. No completion yet
> 2. Process sleeps indefinitely. In the meantime, TPM responds
> 3. We check for timeout without checking for the completion again.
>   Result is lost.

This looks similar to the issue I fixed in 7146dffa875c ('Fix timeout
handling when waiting for TPM status'), I assume you're actually seeing
it in your systems? I think we're starting to see it (rarely) now the
other issues are fixed in our builds. As a similar approach does the
following work?

diff --git a/drivers/char/tpm/tpm-interface.c b/drivers/char/tpm/tpm-interface.c
index 8d7e4da6ed53..18ae0767fa60 100644
--- a/drivers/char/tpm/tpm-interface.c
+++ b/drivers/char/tpm/tpm-interface.c
@@ -127,7 +127,7 @@ static ssize_t tpm_try_transmit(struct tpm_chip *chip, void *buf, size_t bufsiz)
 		goto out_recv;
 
 	stop = jiffies + tpm_calc_ordinal_duration(chip, ordinal);
-	do {
+	while (true) {
 		u8 status = tpm_chip_status(chip);
 		if ((status & chip->ops->req_complete_mask) ==
 		    chip->ops->req_complete_val)
@@ -138,9 +138,12 @@ static ssize_t tpm_try_transmit(struct tpm_chip *chip, void *buf, size_t bufsiz)
 			return -ECANCELED;
 		}
 
+		if (time_after(jiffies, stop))
+			break;
+
 		tpm_msleep(TPM_TIMEOUT_POLL);
 		rmb();
-	} while (time_before(jiffies, stop));
+	}
 
 	tpm_chip_cancel(chip);
 	dev_err(&chip->dev, "Operation Timed out\n");