drm/xe: Move lrc snapshot capturing to xe_lrc.c

This allows the dumping of HWSP and HW Context without exporting more
functions.

Changes since v1:
- GFP_KERNEL -> GFP_NOWAIT. (Souza)

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240227131248.92910-1-maarten.lankhorst@linux.intel.com

drivers/gpu/drm/xe/xe_guc_submit.c

@@ -1820,21 +1820,14 @@ xe_guc_exec_queue_snapshot_capture(struct xe_sched_job *job)
 	snapshot->sched_props.preempt_timeout_us =
 		q->sched_props.preempt_timeout_us;
 
-	snapshot->lrc = kmalloc_array(q->width, sizeof(struct lrc_snapshot),
+	snapshot->lrc = kmalloc_array(q->width, sizeof(struct xe_lrc_snapshot *),
 				      GFP_ATOMIC);
 
 	if (snapshot->lrc) {
 		for (i = 0; i < q->width; ++i) {
 			struct xe_lrc *lrc = q->lrc + i;
 
-			snapshot->lrc[i].context_desc =
-				lower_32_bits(xe_lrc_ggtt_addr(lrc));
-			snapshot->lrc[i].head = xe_lrc_ring_head(lrc);
-			snapshot->lrc[i].tail.internal = lrc->ring.tail;
-			snapshot->lrc[i].tail.memory =
-				xe_lrc_read_ctx_reg(lrc, CTX_RING_TAIL);
-			snapshot->lrc[i].start_seqno = xe_lrc_start_seqno(lrc);
-			snapshot->lrc[i].seqno = xe_lrc_seqno(lrc);
+			snapshot->lrc[i] = xe_lrc_snapshot_capture(lrc);
 		}
 	}
 
@@ -1900,18 +1893,9 @@ xe_guc_exec_queue_snapshot_print(struct xe_guc_submit_exec_queue_snapshot *snaps
 	drm_printf(p, "\tPreempt timeout: %u (us)\n",
 		   snapshot->sched_props.preempt_timeout_us);
 
-	for (i = 0; snapshot->lrc && i < snapshot->width; ++i) {
-		drm_printf(p, "\tHW Context Desc: 0x%08x\n",
-			   snapshot->lrc[i].context_desc);
-		drm_printf(p, "\tLRC Head: (memory) %u\n",
-			   snapshot->lrc[i].head);
-		drm_printf(p, "\tLRC Tail: (internal) %u, (memory) %u\n",
-			   snapshot->lrc[i].tail.internal,
-			   snapshot->lrc[i].tail.memory);
-		drm_printf(p, "\tStart seqno: (memory) %d\n",
-			   snapshot->lrc[i].start_seqno);
-		drm_printf(p, "\tSeqno: (memory) %d\n", snapshot->lrc[i].seqno);
-	}
+	for (i = 0; snapshot->lrc && i < snapshot->width; ++i)
+		xe_lrc_snapshot_print(snapshot->lrc[i], p);
+
 	drm_printf(p, "\tSchedule State: 0x%x\n", snapshot->schedule_state);
 	drm_printf(p, "\tFlags: 0x%lx\n", snapshot->exec_queue_flags);
 
@@ -1936,10 +1920,15 @@ xe_guc_exec_queue_snapshot_print(struct xe_guc_submit_exec_queue_snapshot *snaps
  */
 void xe_guc_exec_queue_snapshot_free(struct xe_guc_submit_exec_queue_snapshot *snapshot)
 {
+	int i;
 	if (!snapshot)
 		return;
 
-	kfree(snapshot->lrc);
+	if (snapshot->lrc) {
+		for (i = 0; i < snapshot->width; i++)
+			xe_lrc_snapshot_free(snapshot->lrc[i]);
+		kfree(snapshot->lrc);
+	}
 	kfree(snapshot->pending_list);
 	kfree(snapshot);
 }

drivers/gpu/drm/xe/xe_guc_submit_types.h

@@ -61,17 +61,6 @@ struct guc_submit_parallel_scratch {
 	u32 wq[WQ_SIZE / sizeof(u32)];
 };
 
-struct lrc_snapshot {
-	u32 context_desc;
-	u32 head;
-	struct {
-		u32 internal;
-		u32 memory;
-	} tail;
-	u32 start_seqno;
-	u32 seqno;
-};
-
 struct pending_list_snapshot {
 	u32 seqno;
 	bool fence;
@@ -109,7 +98,7 @@ struct xe_guc_submit_exec_queue_snapshot {
 	} sched_props;
 
 	/** @lrc: LRC Snapshot */
-	struct lrc_snapshot *lrc;
+	struct xe_lrc_snapshot **lrc;
 
 	/** @schedule_state: Schedule State at the moment of Crash */
 	u32 schedule_state;

drivers/gpu/drm/xe/xe_lrc.c

@@ -32,6 +32,17 @@
 #define ENGINE_CLASS_SHIFT			61
 #define ENGINE_INSTANCE_SHIFT			48
 
+struct xe_lrc_snapshot {
+	u32 context_desc;
+	u32 head;
+	struct {
+		u32 internal;
+		u32 memory;
+	} tail;
+	u32 start_seqno;
+	u32 seqno;
+};
+
 static struct xe_device *
 lrc_to_xe(struct xe_lrc *lrc)
 {
@@ -1331,3 +1342,37 @@ void xe_lrc_emit_hwe_state_instructions(struct xe_exec_queue *q, struct xe_bb *b
 		bb->len += num_dw;
 	}
 }
+
+struct xe_lrc_snapshot *xe_lrc_snapshot_capture(struct xe_lrc *lrc)
+{
+	struct xe_lrc_snapshot *snapshot = kmalloc(sizeof(*snapshot), GFP_NOWAIT);
+
+	if (!snapshot)
+		return NULL;
+
+	snapshot->context_desc = lower_32_bits(xe_lrc_ggtt_addr(lrc));
+	snapshot->head = xe_lrc_ring_head(lrc);
+	snapshot->tail.internal = lrc->ring.tail;
+	snapshot->tail.memory = xe_lrc_read_ctx_reg(lrc, CTX_RING_TAIL);
+	snapshot->start_seqno = xe_lrc_start_seqno(lrc);
+	snapshot->seqno = xe_lrc_seqno(lrc);
+	return snapshot;
+}
+
+void xe_lrc_snapshot_print(struct xe_lrc_snapshot *snapshot, struct drm_printer *p)
+{
+	if (!snapshot)
+		return;
+
+	drm_printf(p, "\tHW Context Desc: 0x%08x\n", snapshot->context_desc);
+	drm_printf(p, "\tLRC Head: (memory) %u\n", snapshot->head);
+	drm_printf(p, "\tLRC Tail: (internal) %u, (memory) %u\n",
+		   snapshot->tail.internal, snapshot->tail.memory);
+	drm_printf(p, "\tStart seqno: (memory) %d\n", snapshot->start_seqno);
+	drm_printf(p, "\tSeqno: (memory) %d\n", snapshot->seqno);
+}
+
+void xe_lrc_snapshot_free(struct xe_lrc_snapshot *snapshot)
+{
+	kfree(snapshot);
+}

drivers/gpu/drm/xe/xe_lrc.h

@@ -55,4 +55,8 @@ void xe_lrc_dump_default(struct drm_printer *p,
 
 void xe_lrc_emit_hwe_state_instructions(struct xe_exec_queue *q, struct xe_bb *bb);
 
+struct xe_lrc_snapshot *xe_lrc_snapshot_capture(struct xe_lrc *lrc);
+void xe_lrc_snapshot_print(struct xe_lrc_snapshot *snapshot, struct drm_printer *p);
+void xe_lrc_snapshot_free(struct xe_lrc_snapshot *snapshot);
+
 #endif

drivers/gpu/drm/xe/xe_lrc_types.h

@@ -43,4 +43,6 @@ struct xe_lrc {
 	struct xe_hw_fence_ctx fence_ctx;
 };
 
+struct xe_lrc_snapshot;
+
 #endif