Re: [PATCH v12 16/21] blockjob: protect iostatus field in BlockJob struct

[Kevin Wolf <kwolf@redhat.com>]

Am 26.09.2022 um 11:32 hat Emanuele Giuseppe Esposito geschrieben:
> iostatus is the only field (together with .job) that needs
> protection using the job mutex.
> 
> It is set in the main loop (GLOBAL_STATE functions) but read
> in I/O code (block_job_error_action).
> 
> In order to protect it, change block_job_iostatus_set_err
> to block_job_iostatus_set_err_locked(), always called under
> job lock.
> 
> Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
> Reviewed-by: Kevin Wolf <kwolf@redhat.com>
> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
> ---
>  block/mirror.c | 7 +++++--
>  blockjob.c     | 5 +++--
>  2 files changed, 8 insertions(+), 4 deletions(-)
> 
> diff --git a/block/mirror.c b/block/mirror.c
> index c6bf7f40ce..7e32ee1d31 100644
> --- a/block/mirror.c
> +++ b/block/mirror.c
> @@ -893,7 +893,7 @@ static int coroutine_fn mirror_run(Job *job, Error **errp)
>      MirrorBlockJob *s = container_of(job, MirrorBlockJob, common.job);
>      BlockDriverState *bs = s->mirror_top_bs->backing->bs;
>      BlockDriverState *target_bs = blk_bs(s->target);
> -    bool need_drain = true;
> +    bool need_drain = true, iostatus;

iostatus isn't really a bool, it's BlockDeviceIoStatus.

>      int64_t length;
>      int64_t target_length;
>      BlockDriverInfo bdi;
> @@ -1016,8 +1016,11 @@ static int coroutine_fn mirror_run(Job *job, Error **errp)
>           * We do so every BLKOCK_JOB_SLICE_TIME nanoseconds, or when there is
>           * an error, or when the source is clean, whichever comes first. */
>          delta = qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - s->last_pause_ns;
> +        WITH_JOB_LOCK_GUARD() {
> +            iostatus = s->common.iostatus;
> +        }
>          if (delta < BLOCK_JOB_SLICE_TIME &&
> -            s->common.iostatus == BLOCK_DEVICE_IO_STATUS_OK) {
> +            iostatus == BLOCK_DEVICE_IO_STATUS_OK) {

Your code actually happens to work because the one value that you
compare it against is BLOCK_DEVICE_IO_STATUS_OK, which is 0, so it maps
to false and everything else to true, but... it's still not right. :-)

>              if (s->in_flight >= MAX_IN_FLIGHT || s->buf_free_count == 0 ||
>                  (cnt == 0 && s->in_flight > 0)) {
>                  trace_mirror_yield(s, cnt, s->buf_free_count, s->in_flight);

Kevin