accel/amdxdna: Improve tracing for job lifecycle and mailbox RX worker

Add more trace coverage to amdxdna job handling and mailbox receive
processing to make driver execution easier to debug.

Extend the xdna_job trace event to record the command opcode in
addition to the job sequence number. Use the enhanced tracepoint in
the job run, sent-to-device, signaled-fence, and job-free paths so
that trace output can be correlated with the command being executed.

Also add debug-point tracing when a command is received through the
submit ioctl path, and add a trace event when the mailbox RX worker
runs.

These changes improve visibility into job lifetime transitions and
mailbox activity, which helps debug command flow and scheduler issues.

Signed-off-by: Max Zhen <max.zhen@amd.com>
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://patch.msgid.link/20260421181502.1970263-1-lizhi.hou@amd.com

diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c
index d37123d925b6ce1a713e38d0ed462e77f6361506..3b0feba448c498a6e13aef4d4d28dbc5aba4147f 100644 (file)
--- a/drivers/accel/amdxdna/aie2_ctx.c
+++ b/drivers/accel/amdxdna/aie2_ctx.c
@@ -64,6 +64,7 @@ static void aie2_job_release(struct kref *ref)
        struct amdxdna_sched_job *job;
 
        job = container_of(ref, struct amdxdna_sched_job, refcnt);
+
        amdxdna_sched_job_cleanup(job);
        atomic64_inc(&job->hwctx->job_free_cnt);
        wake_up(&job->hwctx->priv->job_free_wq);
@@ -195,7 +196,8 @@ aie2_sched_notify(struct amdxdna_sched_job *job)
 {
        struct dma_fence *fence = job->fence;
 
-       trace_xdna_job(&job->base, job->hwctx->name, "signaled fence", job->seq);
+       trace_xdna_job(&job->base, job->hwctx->name, "signaling fence",
+                      job->seq, job->drv_cmd ? job->drv_cmd->opcode : DEFAULT_IO);
 
        aie2_tdr_signal(job->hwctx->client->xdna);
        job->hwctx->priv->completed++;
@@ -366,6 +368,9 @@ aie2_sched_job_run(struct drm_sched_job *sched_job)
        struct dma_fence *fence;
        int ret;
 
+       trace_xdna_job(sched_job, hwctx->name, "job run",
+                      job->seq, job->drv_cmd ? job->drv_cmd->opcode : DEFAULT_IO);
+
        if (!hwctx->priv->mbox_chann)
                return NULL;
 
@@ -409,7 +414,8 @@ out:
        } else {
                aie2_tdr_signal(hwctx->client->xdna);
        }
-       trace_xdna_job(sched_job, hwctx->name, "sent to device", job->seq);
+       trace_xdna_job(sched_job, hwctx->name, "sent to device",
+                      job->seq, job->drv_cmd ? job->drv_cmd->opcode : DEFAULT_IO);
 
        return fence;
 }
@@ -419,7 +425,8 @@ static void aie2_sched_job_free(struct drm_sched_job *sched_job)
        struct amdxdna_sched_job *job = drm_job_to_xdna_job(sched_job);
        struct amdxdna_hwctx *hwctx = job->hwctx;
 
-       trace_xdna_job(sched_job, hwctx->name, "job free", job->seq);
+       trace_xdna_job(sched_job, hwctx->name, "job free",
+                      job->seq, job->drv_cmd ? job->drv_cmd->opcode : DEFAULT_IO);
        if (!job->job_done)
                up(&hwctx->priv->job_sem);
 
@@ -437,7 +444,6 @@ aie2_sched_job_timedout(struct drm_sched_job *sched_job)
        int ret;
 
        xdna = hwctx->client->xdna;
-       trace_xdna_job(sched_job, hwctx->name, "job timedout", job->seq);
 
        guard(mutex)(&xdna->dev_lock);
 

diff --git a/drivers/accel/amdxdna/amdxdna_ctx.c b/drivers/accel/amdxdna/amdxdna_ctx.c
index ff6c3e8e5a15b898dfa2970d219b7f84770d63d0..2c2c21992c87417e7594962037fa01e222c352b3 100644 (file)
--- a/drivers/accel/amdxdna/amdxdna_ctx.c
+++ b/drivers/accel/amdxdna/amdxdna_ctx.c
@@ -514,7 +514,6 @@ int amdxdna_cmd_submit(struct amdxdna_client *client,
                goto unlock_srcu;
        }
 
-
        job->hwctx = hwctx;
        job->mm = current->mm;
 
@@ -612,6 +611,8 @@ int amdxdna_drm_submit_cmd_ioctl(struct drm_device *dev, void *data, struct drm_
        if (args->ext || args->ext_flags)
                return -EINVAL;
 
+       trace_amdxdna_debug_point(current->comm, args->type, "job received");
+
        switch (args->type) {
        case AMDXDNA_CMD_SUBMIT_EXEC_BUF:
                return amdxdna_drm_submit_execbuf(client, args);

diff --git a/drivers/accel/amdxdna/amdxdna_ctx.h b/drivers/accel/amdxdna/amdxdna_ctx.h
index a8557d7e89238d9fa6d367be69881936cbb33f8c..3557986873766019681240c13b539f23c2197ef2 100644 (file)
--- a/drivers/accel/amdxdna/amdxdna_ctx.h
+++ b/drivers/accel/amdxdna/amdxdna_ctx.h
@@ -119,6 +119,7 @@ struct amdxdna_hwctx {
        container_of(j, struct amdxdna_sched_job, base)
 
 enum amdxdna_job_opcode {
+       DEFAULT_IO,
        SYNC_DEBUG_BO,
        ATTACH_DEBUG_BO,
        DETACH_DEBUG_BO,

diff --git a/drivers/accel/amdxdna/amdxdna_mailbox.c b/drivers/accel/amdxdna/amdxdna_mailbox.c
index 37771bdb24a178bc1aa21860163022f94790d3bf..cc8865f4e79c9f62be19e230f68dda60f24cb5f0 100644 (file)
--- a/drivers/accel/amdxdna/amdxdna_mailbox.c
+++ b/drivers/accel/amdxdna/amdxdna_mailbox.c
@@ -361,6 +361,7 @@ static void mailbox_rx_worker(struct work_struct *rx_work)
        int ret;
 
        mb_chann = container_of(rx_work, struct mailbox_channel, rx_work);
+       trace_mbox_rx_worker(MAILBOX_NAME, mb_chann->msix_irq);
 
        if (READ_ONCE(mb_chann->bad_state)) {
                MB_ERR(mb_chann, "Channel in bad state, work aborted");

diff --git a/include/trace/events/amdxdna.h b/include/trace/events/amdxdna.h
index c6cb2da7b706c37f063166eb72d7e8adb54a17f1..71da24267e52ebd203cc07eb9d908cc3cf3c5315 100644 (file)
--- a/include/trace/events/amdxdna.h
+++ b/include/trace/events/amdxdna.h
@@ -30,26 +30,30 @@ TRACE_EVENT(amdxdna_debug_point,
 );
 
 TRACE_EVENT(xdna_job,
-           TP_PROTO(struct drm_sched_job *sched_job, const char *name, const char *str, u64 seq),
+           TP_PROTO(struct drm_sched_job *sched_job, const char *name,
+                    const char *str, u64 seq, u32 op),
 
-           TP_ARGS(sched_job, name, str, seq),
+           TP_ARGS(sched_job, name, str, seq, op),
 
            TP_STRUCT__entry(__string(name, name)
                             __string(str, str)
                             __field(u64, fence_context)
                             __field(u64, fence_seqno)
-                            __field(u64, seq)),
+                            __field(u64, seq)
+                            __field(u32, op)),
 
            TP_fast_assign(__assign_str(name);
                           __assign_str(str);
                           __entry->fence_context = sched_job->s_fence->finished.context;
                           __entry->fence_seqno = sched_job->s_fence->finished.seqno;
-                          __entry->seq = seq;),
+                          __entry->seq = seq;
+                          __entry->op = op;),
 
-           TP_printk("fence=(context:%llu, seqno:%lld), %s seq#:%lld %s",
+           TP_printk("fence=(context:%llu, seqno:%llu), %s seq#:%llu %s, op=%u",
                      __entry->fence_context, __entry->fence_seqno,
                      __get_str(name), __entry->seq,
-                     __get_str(str))
+                     __get_str(str),
+                     __entry->op)
 );
 
 DECLARE_EVENT_CLASS(xdna_mbox_msg,
@@ -81,18 +85,28 @@ DEFINE_EVENT(xdna_mbox_msg, mbox_set_head,
             TP_ARGS(name, chann_id, opcode, id)
 );
 
-TRACE_EVENT(mbox_irq_handle,
-           TP_PROTO(char *name, int irq),
+DECLARE_EVENT_CLASS(xdna_mbox_name_id,
+                   TP_PROTO(char *name, int irq),
 
-           TP_ARGS(name, irq),
+                   TP_ARGS(name, irq),
 
-           TP_STRUCT__entry(__string(name, name)
-                            __field(int, irq)),
+                   TP_STRUCT__entry(__string(name, name)
+                                    __field(int, irq)),
 
-           TP_fast_assign(__assign_str(name);
-                          __entry->irq = irq;),
+                   TP_fast_assign(__assign_str(name);
+                                  __entry->irq = irq;),
+
+                   TP_printk("%s.%d", __get_str(name), __entry->irq)
+);
+
+DEFINE_EVENT(xdna_mbox_name_id, mbox_irq_handle,
+            TP_PROTO(char *name, int irq),
+            TP_ARGS(name, irq)
+);
 
-           TP_printk("%s.%d", __get_str(name), __entry->irq)
+DEFINE_EVENT(xdna_mbox_name_id, mbox_rx_worker,
+            TP_PROTO(char *name, int irq),
+            TP_ARGS(name, irq)
 );
 
 #endif /* !defined(_TRACE_AMDXDNA_H) || defined(TRACE_HEADER_MULTI_READ) */