Re: [PATCH] accel/amdxdna: Fix incorrect command state for timed out job

[Mario Limonciello]

On 10/28/25 12:54 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....@amd.com>
---
  drivers/accel/amdxdna/aie2_ctx.c    | 12 ++++++++++--
  drivers/accel/amdxdna/amdxdna_ctx.h |  1 +
  2 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c
index c6c473c78352..958a64bb5251 100644
--- a/drivers/accel/amdxdna/aie2_ctx.c
+++ b/drivers/accel/amdxdna/aie2_ctx.c
@@ -204,10 +204,12 @@ 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);
                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 +260,11 @@ 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);

Don't you need to set ret here?

+               goto out;
+       }
+
        if (unlikely(!data) || unlikely(size != sizeof(u32) * 3)) {
                amdxdna_cmd_set_state(cmd_abo, ERT_CMD_STATE_ABORT);
                ret = -EINVAL;
@@ -370,6 +377,7 @@ 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;