AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Nirmoy Das <nirmoy.das@amd.com>
To: <amd-gfx@lists.freedesktop.org>
Cc: <Christian.Koenig@amd.com>, <Alexander.Deucher@amd.com>,
	Nirmoy Das <nirmoy.das@amd.com>, zhang <botton_zhang@163.com>
Subject: [PATCH 1/1] drm/amdgpu: fix BO leak after successful move test
Date: Wed, 13 Oct 2021 17:09:02 +0200	[thread overview]
Message-ID: <20211013150902.6646-1-nirmoy.das@amd.com> (raw)
In-Reply-To: <20211012121018.81693-1-nirmoy.das@amd.com>

GTT BO cleanup code is with in the test for loop and
we would skip cleaning up GTT BO on success.

Reported-by: zhang <botton_zhang@163.com>
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_test.c | 25 ++++++++++++------------
 1 file changed, 12 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c
index 909d830b513e..5fe7ff680c29 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c
@@ -35,6 +35,7 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev)
 	struct amdgpu_bo *vram_obj = NULL;
 	struct amdgpu_bo **gtt_obj = NULL;
 	struct amdgpu_bo_param bp;
+	struct dma_fence *fence = NULL;
 	uint64_t gart_addr, vram_addr;
 	unsigned n, size;
 	int i, r;
@@ -82,7 +83,6 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev)
 		void *gtt_map, *vram_map;
 		void **gart_start, **gart_end;
 		void **vram_start, **vram_end;
-		struct dma_fence *fence = NULL;
 
 		bp.domain = AMDGPU_GEM_DOMAIN_GTT;
 		r = amdgpu_bo_create(adev, &bp, gtt_obj + i);
@@ -212,24 +212,23 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev)
 
 		DRM_INFO("Tested GTT->VRAM and VRAM->GTT copy for GTT offset 0x%llx\n",
 			 gart_addr - adev->gmc.gart_start);
-		continue;
+	}
 
+	--i;
 out_lclean_unpin:
-		amdgpu_bo_unpin(gtt_obj[i]);
+	amdgpu_bo_unpin(gtt_obj[i]);
 out_lclean_unres:
-		amdgpu_bo_unreserve(gtt_obj[i]);
+	amdgpu_bo_unreserve(gtt_obj[i]);
 out_lclean_unref:
-		amdgpu_bo_unref(&gtt_obj[i]);
+	amdgpu_bo_unref(&gtt_obj[i]);
 out_lclean:
-		for (--i; i >= 0; --i) {
-			amdgpu_bo_unpin(gtt_obj[i]);
-			amdgpu_bo_unreserve(gtt_obj[i]);
-			amdgpu_bo_unref(&gtt_obj[i]);
-		}
-		if (fence)
-			dma_fence_put(fence);
-		break;
+	for (--i; i >= 0; --i) {
+		amdgpu_bo_unpin(gtt_obj[i]);
+		amdgpu_bo_unreserve(gtt_obj[i]);
+		amdgpu_bo_unref(&gtt_obj[i]);
 	}
+	if (fence)
+		dma_fence_put(fence);
 
 	amdgpu_bo_unpin(vram_obj);
 out_unres:
-- 
2.32.0


  parent reply	other threads:[~2021-10-13 15:09 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-12 12:10 [PATCH 1/1] drm/amdgpu: release gtt bo after each move test Nirmoy Das
     [not found] ` <2be293df-2656-c551-682e-b08955307c0b@163.com>
2021-10-13 10:42   ` Das, Nirmoy
2021-10-13 15:07     ` Das, Nirmoy
2021-10-13 15:04 ` Nirmoy Das
2021-10-13 15:10   ` Das, Nirmoy
2021-10-13 15:09 ` Nirmoy Das [this message]
2021-10-20  8:50   ` [PATCH 1/1] drm/amdgpu: fix BO leak after successful " Das, Nirmoy
2021-10-20 11:50   ` Christian König
2021-10-20 11:51     ` Christian König
2021-10-20 12:55       ` Das, Nirmoy
2021-10-21  6:58         ` Christian König
2021-10-21  2:07       ` zhang
2021-10-21  6:58         ` Christian König

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20211013150902.6646-1-nirmoy.das@amd.com \
    --to=nirmoy.das@amd.com \
    --cc=Alexander.Deucher@amd.com \
    --cc=Christian.Koenig@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=botton_zhang@163.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox