[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PATCH v2 15/28] tpm-passthrough: workaround a possible rac
From: |
Marc-André Lureau |
Subject: |
[Qemu-devel] [PATCH v2 15/28] tpm-passthrough: workaround a possible race |
Date: |
Mon, 6 Nov 2017 19:39:12 +0100 |
The TPM backend processing thread has common shared variable race
issues. (they should not be so easy to reach since guest interaction
with the device is slow compared to host emulation)
An obvious one is setting op_cancelled from device thread after
calling write(cancel_fd). The backend thread may return before the
device thread has set the variable. Instead set it before
cancellation. Even if the write() failed, the end result is command
get possibly cancelled (even if cancellation came from external
sources it doesn't matter much).
It's worth to consider removing the backend processing thread for now.
Signed-off-by: Marc-André Lureau <address@hidden>
Reviewed-by: Stefan Berger <address@hidden>
---
hw/tpm/tpm_passthrough.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/hw/tpm/tpm_passthrough.c b/hw/tpm/tpm_passthrough.c
index 788be3847d..73554aa792 100644
--- a/hw/tpm/tpm_passthrough.c
+++ b/hw/tpm/tpm_passthrough.c
@@ -89,6 +89,7 @@ static int tpm_passthrough_unix_tx_bufs(TPMPassthruState
*tpm_pt,
bool is_selftest;
const struct tpm_resp_hdr *hdr;
+ /* FIXME: protect shared variables or use other sync mechanism */
tpm_pt->tpm_op_canceled = false;
tpm_pt->tpm_executing = true;
*selftest_done = false;
@@ -178,12 +179,11 @@ static void tpm_passthrough_cancel_cmd(TPMBackend *tb)
*/
if (tpm_pt->tpm_executing) {
if (tpm_pt->cancel_fd >= 0) {
+ tpm_pt->tpm_op_canceled = true;
n = write(tpm_pt->cancel_fd, "-", 1);
if (n != 1) {
error_report("Canceling TPM command failed: %s",
strerror(errno));
- } else {
- tpm_pt->tpm_op_canceled = true;
}
} else {
error_report("Cannot cancel TPM command due to missing "
--
2.15.0.rc0.40.gaefcc5f6f
- [Qemu-devel] [PATCH v2 05/28] tpm-be: call request_completed() out of thread, (continued)
- [Qemu-devel] [PATCH v2 05/28] tpm-be: call request_completed() out of thread, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 07/28] tpm-be: ask model to the TPM interface, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 08/28] tpm: remove unused opened code, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 09/28] tpm-passthrough: don't save guessed cancel_path in options, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 11/28] tpm-passthrough: pass TPMPassthruState to handle_device_opts, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 12/28] tpm-backend: move set 'id' to common code, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 10/28] tpm-be: update optional function pointers, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 13/28] tpm-passthrough: make it safer to destroy after creation, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 14/28] tpm-passthrough: simplify create(), Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 15/28] tpm-passthrough: workaround a possible race,
Marc-André Lureau <=
- [Qemu-devel] [PATCH v2 16/28] tpm-tis: simplify header inclusion, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 17/28] tpm: rename qemu_find_tpm() -> qemu_find_tpm_be(), Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 18/28] tpm: lookup the the TPM interface instead of TIS device, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 19/28] tpm: add TPM interface to lookup TPM version, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 20/28] tpm: add tpm_cmd_get_size() to tpm_util, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 21/28] acpi: change TPM TIS data conditions, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 22/28] tpm-emulator: add a FIXME comment about blocking cancel, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 23/28] tpm-tis: remove redundant 'tpm_tis:' in error messages, Marc-André Lureau, 2017/11/06
- [Qemu-devel] [PATCH v2 24/28] tpm-tis: check that at most one TPM device exists, Marc-André Lureau, 2017/11/06