| Message ID | 20171211160525.10576-1-Alexander.Steffen@infineon.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Mon, Dec 11, 2017 at 05:05:25PM +0100, Alexander Steffen wrote: > Previously, if the last attempt to execute the selftest command failed with > RC_TESTING, there was still a call to tpm_msleep, even though no further > attempt would be made. This causes an unnecessary delay, therefore ensure > that if the last attempt fails the function is left immediately. > > Also, instead of ensuring that the cumulated runtime of all attempts is > larger than the command duration for TPM2_SelfTest, ensure that there is at > least one attempt for which the delay is larger than the expected command > duration. This allows slow TPMs to execute all their tests in the > background, without slowing down faster TPMs that have finished their tests > earlier. If tests are still not finished even with this long delay, then > something is broken and the TPM is not used. > > Fixes: 125a22105410 ("tpm: React correctly to RC_TESTING from TPM 2.0 self > tests") > > Signed-off-by: Alexander Steffen <Alexander.Steffen@infineon.com> > --- > drivers/char/tpm/tpm2-cmd.c | 12 +++++------- > 1 file changed, 5 insertions(+), 7 deletions(-) > > diff --git a/drivers/char/tpm/tpm2-cmd.c b/drivers/char/tpm/tpm2-cmd.c > index f40d206..c17e753 100644 > --- a/drivers/char/tpm/tpm2-cmd.c > +++ b/drivers/char/tpm/tpm2-cmd.c > @@ -849,28 +849,26 @@ static const struct tpm_input_header tpm2_selftest_header = { > static int tpm2_do_selftest(struct tpm_chip *chip) > { > int rc; > - unsigned int delay_msec = 20; > + unsigned int delay_msec = 10; > long duration; > struct tpm2_cmd cmd; > > duration = jiffies_to_msecs( > tpm2_calc_ordinal_duration(chip, TPM2_CC_SELF_TEST)); > > - while (duration > 0) { > + while (1) { > cmd.header.in = tpm2_selftest_header; > cmd.params.selftest_in.full_test = 0; > > rc = tpm_transmit_cmd(chip, NULL, &cmd, TPM2_SELF_TEST_IN_SIZE, > 0, 0, "continue selftest"); > > - if (rc != TPM2_RC_TESTING) > + if (rc != TPM2_RC_TESTING || delay_msec >= duration) > break; > > - tpm_msleep(delay_msec); > - duration -= delay_msec; > - > - /* wait longer the next round */ > + /* wait longer than before */ > delay_msec *= 2; > + tpm_msleep(delay_msec); > } > > return rc; > -- > 2.7.4 > Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com> /Jarkko
diff --git a/drivers/char/tpm/tpm2-cmd.c b/drivers/char/tpm/tpm2-cmd.c index f40d206..c17e753 100644 --- a/drivers/char/tpm/tpm2-cmd.c +++ b/drivers/char/tpm/tpm2-cmd.c @@ -849,28 +849,26 @@ static const struct tpm_input_header tpm2_selftest_header = { static int tpm2_do_selftest(struct tpm_chip *chip) { int rc; - unsigned int delay_msec = 20; + unsigned int delay_msec = 10; long duration; struct tpm2_cmd cmd; duration = jiffies_to_msecs( tpm2_calc_ordinal_duration(chip, TPM2_CC_SELF_TEST)); - while (duration > 0) { + while (1) { cmd.header.in = tpm2_selftest_header; cmd.params.selftest_in.full_test = 0; rc = tpm_transmit_cmd(chip, NULL, &cmd, TPM2_SELF_TEST_IN_SIZE, 0, 0, "continue selftest"); - if (rc != TPM2_RC_TESTING) + if (rc != TPM2_RC_TESTING || delay_msec >= duration) break; - tpm_msleep(delay_msec); - duration -= delay_msec; - - /* wait longer the next round */ + /* wait longer than before */ delay_msec *= 2; + tpm_msleep(delay_msec); } return rc;
Previously, if the last attempt to execute the selftest command failed with RC_TESTING, there was still a call to tpm_msleep, even though no further attempt would be made. This causes an unnecessary delay, therefore ensure that if the last attempt fails the function is left immediately. Also, instead of ensuring that the cumulated runtime of all attempts is larger than the command duration for TPM2_SelfTest, ensure that there is at least one attempt for which the delay is larger than the expected command duration. This allows slow TPMs to execute all their tests in the background, without slowing down faster TPMs that have finished their tests earlier. If tests are still not finished even with this long delay, then something is broken and the TPM is not used. Fixes: 125a22105410 ("tpm: React correctly to RC_TESTING from TPM 2.0 self tests") Signed-off-by: Alexander Steffen <Alexander.Steffen@infineon.com> --- drivers/char/tpm/tpm2-cmd.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-)