Skip to contentSkip to conversation

drm/i915: Free wa_batchbuffer when freeing error state

PATCH Applied Mika Kuoppala 4 participants 6 messages May 4, 2015 – May 6, 2015

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 drm/i915: Free wa_batchbuffer when freeing error state

wa_batchbuffer is part of some error states. Make sure it is freed.

---
 drivers/gpu/drm/i915/i915_gpu_error.c | 1 +
 1 file changed, 1 insertion(+)
drivers/gpu/drm/i915/i915_gpu_error.c +1 -0
index ac22614..a3e330d 100644
@@ -554,6 +554,7 @@ static void i915_error_state_free(struct kref *error_ref)
 
 	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
 		i915_error_object_free(error->ring[i].batchbuffer);
+		i915_error_object_free(error->ring[i].wa_batchbuffer);
 		i915_error_object_free(error->ring[i].ringbuffer);
 		i915_error_object_free(error->ring[i].hws_page);
 		i915_error_object_free(error->ring[i].ctx);
Signature · 6 lines
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#1 lore raw

PATCH 2/2 drm/i915: Record the object matching actual head to error state

If we have chained batch, the request only contains the batch buffer that branched the execution into chained batch. We can try to find object for actual head and if it is different than the object from request, record and print its state also.

---
 drivers/gpu/drm/i915/i915_drv.h       |  3 ++-
 drivers/gpu/drm/i915/i915_gpu_error.c | 49 ++++++++++++++++++++++++++++++++---
 2 files changed, 47 insertions(+), 5 deletions(-)
drivers/gpu/drm/i915/i915_drv.h +2 -1
index 136d42a..d23ca8e 100644
@@ -480,7 +480,8 @@ struct drm_i915_error_state {
 			int page_count;
 			u32 gtt_offset;
 			u32 *pages[0];
-		} *ringbuffer, *batchbuffer, *wa_batchbuffer, *ctx, *hws_page;
+		} *ringbuffer, *batchbuffer, *wa_batchbuffer,
+			*active_batchbuffer, *ctx, *hws_page;
 
 		struct drm_i915_error_request {
 			long jiffies;
drivers/gpu/drm/i915/i915_gpu_error.c +45 -4

96 lines not on this page · raw message

Signature · 6 lines
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#2 lore raw

Re: [PATCH 2/2] drm/i915: Record the object matching actual head to error state

drivers/gpu/drm/i915/i915_gpu_error.c static void i915_gem_record_rings(struct drm_device *dev, · line 1064

Quoted 11 lines · Mika Kuoppala
drivers/gpu/drm/i915/i915_gpu_error.c +11 -0
+		obj = find_obj_for_addr(dev, error->ring[i].acthd);
+		if (request && request->batch_obj == obj)
+			obj = NULL;
+
+		if (obj) {
+			if (vm && !i915_gem_obj_bound(obj, vm))
+				vm = NULL;
+
+			error->ring[i].active_batchbuffer =
+				i915_error_object_create(dev_priv, obj, vm);
+		}

I was ok up to here. But after seeing this, I really want to defer this until after i915_error_object_create() is fixed to just take the vma.

http://cgit.freedesktop.org/~ickle/linux-2.6/diff/drivers/gpu/drm/i915/i915_gpu_error.c?h=nightly&id=cc584ccfc109e5b13813548eb0ef1fd8f5751d16
which in about patch 150/200 that Daniel promised to review ;). -Chris

Signature · 5 lines
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#3 lore raw

Re: [PATCH 2/2] drm/i915: Record the object matching actual head to error state

In reply to Mika Kuoppala
Quoted 5 lines · Mika Kuoppala

If we have chained batch, the request only contains the batch buffer that branched the execution into chained batch. We can try to find object for actual head and if it is different than the object from request, record and print its state also.

I was also going to say if we want to add even more large objects (and chasing a random pointer can mean we see very, very large objects), we want to compress the output:

http://cgit.freedesktop.org/~ickle/linux-2.6/commit/?h=nightly&id=435e640e6396d6fa51fc09e035f2f4703a86a8bd
-Chris

Signature · 5 lines
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Tested-by on v1 2/2
#4 lore raw

Re: [PATCH 2/2] drm/i915: Record the object matching actual head to error state

In reply to Mika Kuoppala

Task id: 6313 -------------------------------------Summary-------------------------------------
Platform Delta drm-intel-nightly Series Applied

PNV                                  276/276              276/276
ILK                                  302/302              302/302
SNB                                  316/316              316/316
IVB                                  342/342              342/342
BYT                                  286/286              286/286
BDW                                  321/321              321/321

-------------------------------------Detailed-------------------------------------
Platform Test drm-intel-nightly Series Applied
Note: You need to pay more attention to line start with '*'
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#5 lore raw
In reply to Mika Kuoppala
Quoted 4 lines · Mika Kuoppala

wa_batchbuffer is part of some error states. Make sure it is freed.

Queued for -next, thanks for the patch.
-Daniel

Quoted 23 lines
---
 drivers/gpu/drm/i915/i915_gpu_error.c | 1 +
 1 file changed, 1 insertion(+)
drivers/gpu/drm/i915/i915_gpu_error.c +1 -0
index ac22614..a3e330d 100644
@@ -554,6 +554,7 @@ static void i915_error_state_free(struct kref *error_ref)
 
 	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
 		i915_error_object_free(error->ring[i].batchbuffer);
+		i915_error_object_free(error->ring[i].wa_batchbuffer);
 		i915_error_object_free(error->ring[i].ringbuffer);
 		i915_error_object_free(error->ring[i].hws_page);
 		i915_error_object_free(error->ring[i].ctx);
Signature · 6 lines
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 7 lines
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx