All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Christian König" <ckoenig.leichtzumerken@gmail.com>
To: daniel.vetter@ffwll.ch, airlied@gmail.com
Cc: dri-devel@lists.freedesktop.org
Subject: [PATCH 4/5] drm/radeon: unbind in radeon_ttm_tt_unpopulate()
Date: Wed, 28 Jul 2021 15:05:51 +0200	[thread overview]
Message-ID: <20210728130552.2074-4-christian.koenig@amd.com> (raw)
In-Reply-To: <20210728130552.2074-1-christian.koenig@amd.com>

Doing this in radeon_ttm_tt_destroy() is to late.

It turned out that this is not a good idea at all because it leaves pointers
to freed up system memory pages in the GART tables of the drivers.

Signed-off-by: Christian König <christian.koenig@amd.com>
---
 drivers/gpu/drm/radeon/radeon_ttm.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index a06d4cc2fb1c..ee343b76db54 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -488,9 +488,7 @@ static void radeon_ttm_backend_destroy(struct ttm_device *bdev, struct ttm_tt *t
 {
 	struct radeon_ttm_tt *gtt = (void *)ttm;
 
-	radeon_ttm_backend_unbind(bdev, ttm);
 	ttm_tt_destroy_common(bdev, ttm);
-
 	ttm_tt_fini(&gtt->ttm);
 	kfree(gtt);
 }
@@ -574,6 +572,8 @@ static void radeon_ttm_tt_unpopulate(struct ttm_device *bdev, struct ttm_tt *ttm
 	struct radeon_ttm_tt *gtt = radeon_ttm_tt_to_gtt(rdev, ttm);
 	bool slave = !!(ttm->page_flags & TTM_PAGE_FLAG_SG);
 
+	radeon_ttm_tt_unbind(bdev, ttm);
+
 	if (gtt && gtt->userptr) {
 		kfree(ttm->sg);
 		ttm->page_flags &= ~TTM_PAGE_FLAG_SG;
@@ -651,7 +651,6 @@ static void radeon_ttm_tt_destroy(struct ttm_device *bdev,
 	struct radeon_device *rdev = radeon_get_rdev(bdev);
 
 	if (rdev->flags & RADEON_IS_AGP) {
-		ttm_agp_unbind(ttm);
 		ttm_tt_destroy_common(bdev, ttm);
 		ttm_agp_destroy(ttm);
 		return;
-- 
2.25.1


  parent reply	other threads:[~2021-07-28 13:06 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-28 13:05 [PATCH 1/5] drm/vmwgfx: unbind in vmw_ttm_unpopulate Christian König
2021-07-28 13:05 ` [PATCH 2/5] drm/amdgpu: unbind in amdgpu_ttm_tt_unpopulate Christian König
2021-07-28 13:05 ` [PATCH 3/5] drm/nouveau: unbind in nouveau_ttm_tt_unpopulate Christian König
2021-07-28 13:05 ` Christian König [this message]
2021-07-28 13:05 ` [PATCH 5/5] drm/ttm: remove ttm_tt_destroy_common v2 Christian König
2021-08-23 11:05 ` [PATCH 1/5] drm/vmwgfx: unbind in vmw_ttm_unpopulate Christian König
2021-08-23 11:15   ` Thomas Hellström
2021-08-26  8:49     ` Daniel Vetter
2021-08-26 10:11       ` Christian König
2021-08-26 12:50         ` Daniel Vetter
  -- strict thread matches above, loose matches on Subject: below --
2021-07-22 12:41 Christian König
2021-07-22 12:41 ` [PATCH 4/5] drm/radeon: unbind in radeon_ttm_tt_unpopulate() Christian König
2021-07-22  8:55 [PATCH 1/5] drm/vmwgfx: unbind in vmw_ttm_unpopulate Christian König
2021-07-22  8:55 ` [PATCH 4/5] drm/radeon: unbind in radeon_ttm_tt_unpopulate() 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=20210728130552.2074-4-christian.koenig@amd.com \
    --to=ckoenig.leichtzumerken@gmail.com \
    --cc=airlied@gmail.com \
    --cc=daniel.vetter@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.