If the usual approach of triggering only the missing selftests failed, give the TPM another chance by requesting all selftests once. With the current implementation, this will only help for TPMs that then choose to run the selftests synchronously and return the selftest result. There is no retry loop that again polls a TPM returning TPM_RC_TESTING in this case. Signed-off-by: Alexander Steffen <Alexander.Steffen@xxxxxxxxxxxx> --- Patch depends on https://patchwork.kernel.org/patch/10105483/ drivers/char/tpm/tpm2-cmd.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/char/tpm/tpm2-cmd.c b/drivers/char/tpm/tpm2-cmd.c index c17e753..b5c34d8 100644 --- a/drivers/char/tpm/tpm2-cmd.c +++ b/drivers/char/tpm/tpm2-cmd.c @@ -871,6 +871,14 @@ static int tpm2_do_selftest(struct tpm_chip *chip) tpm_msleep(delay_msec); } + if (rc != TPM2_RC_SUCCESS) { + cmd.header.in = tpm2_selftest_header; + cmd.params.selftest_in.full_test = 1; + + rc = tpm_transmit_cmd(chip, NULL, &cmd, TPM2_SELF_TEST_IN_SIZE, + 0, 0, "full selftest"); + } + return rc; } -- 2.7.4
