[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <8324911f-ff49-457c-9021-a860d2d4c0c7@kernel.org>
Date: Wed, 29 Oct 2025 14:38:51 -0500
From: Mario Limonciello <superm1@...nel.org>
To: Lizhi Hou <lizhi.hou@....com>, ogabbay@...nel.org,
quic_jhugo@...cinc.com, maciej.falkowski@...ux.intel.com,
dri-devel@...ts.freedesktop.org
Cc: linux-kernel@...r.kernel.org, max.zhen@....com, sonal.santan@....com
Subject: Re: [PATCH V2] accel/amdxdna: Fix incorrect command state for timed
out job
On 10/29/25 2:34 PM, Lizhi Hou wrote:
> When a command times out, mark it as ERT_CMD_STATE_TIMEOUT. Any other
> commands that are canceled due to this timeout should be marked as
> ERT_CMD_STATE_ABORT.
>
> Fixes: aac243092b70 ("accel/amdxdna: Add command execution")
> Signed-off-by: Lizhi Hou <lizhi.hou@....com>
Reviewed-by: Mario Limonciello (AMD) <superm1@...nel.org>> ---
> drivers/accel/amdxdna/aie2_ctx.c | 15 +++++++++++++--
> drivers/accel/amdxdna/amdxdna_ctx.h | 1 +
> 2 files changed, 14 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c
> index c6c473c78352..289a2aaf4cae 100644
> --- a/drivers/accel/amdxdna/aie2_ctx.c
> +++ b/drivers/accel/amdxdna/aie2_ctx.c
> @@ -204,10 +204,13 @@ aie2_sched_resp_handler(void *handle, void __iomem *data, size_t size)
>
> cmd_abo = job->cmd_bo;
>
> - if (unlikely(!data))
> + if (unlikely(job->job_timeout)) {
> + amdxdna_cmd_set_state(cmd_abo, ERT_CMD_STATE_TIMEOUT);
> + ret = -EINVAL;
> goto out;
> + }
>
> - if (unlikely(size != sizeof(u32))) {
> + if (unlikely(!data) || unlikely(size != sizeof(u32))) {
> amdxdna_cmd_set_state(cmd_abo, ERT_CMD_STATE_ABORT);
> ret = -EINVAL;
> goto out;
> @@ -258,6 +261,13 @@ aie2_sched_cmdlist_resp_handler(void *handle, void __iomem *data, size_t size)
> int ret = 0;
>
> cmd_abo = job->cmd_bo;
> +
> + if (unlikely(job->job_timeout)) {
> + amdxdna_cmd_set_state(cmd_abo, ERT_CMD_STATE_TIMEOUT);
> + ret = -EINVAL;
> + goto out;
> + }
> +
> if (unlikely(!data) || unlikely(size != sizeof(u32) * 3)) {
> amdxdna_cmd_set_state(cmd_abo, ERT_CMD_STATE_ABORT);
> ret = -EINVAL;
> @@ -370,6 +380,7 @@ aie2_sched_job_timedout(struct drm_sched_job *sched_job)
>
> xdna = hwctx->client->xdna;
> trace_xdna_job(sched_job, hwctx->name, "job timedout", job->seq);
> + job->job_timeout = true;
> mutex_lock(&xdna->dev_lock);
> aie2_hwctx_stop(xdna, hwctx, sched_job);
>
> diff --git a/drivers/accel/amdxdna/amdxdna_ctx.h b/drivers/accel/amdxdna/amdxdna_ctx.h
> index cbe60efbe60b..919c654dfea6 100644
> --- a/drivers/accel/amdxdna/amdxdna_ctx.h
> +++ b/drivers/accel/amdxdna/amdxdna_ctx.h
> @@ -116,6 +116,7 @@ struct amdxdna_sched_job {
> /* user can wait on this fence */
> struct dma_fence *out_fence;
> bool job_done;
> + bool job_timeout;
> u64 seq;
> struct amdxdna_drv_cmd *drv_cmd;
> struct amdxdna_gem_obj *cmd_bo;
Powered by blists - more mailing lists