linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Roberto Sassu <roberto.sassu@huawei.com>
To: <maarten.lankhorst@linux.intel.com>, <mripard@kernel.org>,
	<tzimmermann@suse.de>, <airlied@linux.ie>, <daniel@ffwll.ch>
Cc: <dri-devel@lists.freedesktop.org>, <linux-kernel@vger.kernel.org>,
	Roberto Sassu <roberto.sassu@huawei.com>,
	<stable@vger.kernel.org>,
	<syzbot+c8ae65286134dd1b800d@syzkaller.appspotmail.com>
Subject: [PATCH] drm: Fix gem obj imbalance due to calling drm_gem_object_put() twice
Date: Tue, 21 Dec 2021 12:38:37 +0100	[thread overview]
Message-ID: <20211221113837.1607448-1-roberto.sassu@huawei.com> (raw)

After commit 9786b65bc61ac ("drm/ttm: fix mmap refcounting"),
drm_gem_mmap_obj() takes a reference of the passed drm_gem_object at the
beginning of the function to safely dereference the mmap offset pointer,
and releases it at the end, if an error occurred. However, the cma and
shmem helpers are also releasing that reference in case of an error,
which causes the imbalance of the reference counter and the panic
reported by syzbot.

Don't release the reference in drm_gem_mmap_obj() if the mmap method was
called and it returned an error, and uniformly apply the same behavior of
the cma and shmem helpers to the ttm helper (release the reference in the
helper, not in the caller, when an error occurs).

Cc: stable@vger.kernel.org
Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
Reported-by: syzbot+c8ae65286134dd1b800d@syzkaller.appspotmail.com
Fixes: 9786b65bc61ac ("drm/ttm: fix mmap refcounting")
---
 drivers/gpu/drm/drm_gem.c            | 3 ++-
 drivers/gpu/drm/drm_gem_ttm_helper.c | 4 +---
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 4dcdec6487bb..7264a1a7a8d2 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -1049,8 +1049,9 @@ int drm_gem_mmap_obj(struct drm_gem_object *obj, unsigned long obj_size,
 
 	if (obj->funcs->mmap) {
 		ret = obj->funcs->mmap(obj, vma);
+		/* All helpers call drm_gem_object_put() */
 		if (ret)
-			goto err_drm_gem_object_put;
+			return ret;
 		WARN_ON(!(vma->vm_flags & VM_DONTEXPAND));
 	} else {
 		if (!vma->vm_ops) {
diff --git a/drivers/gpu/drm/drm_gem_ttm_helper.c b/drivers/gpu/drm/drm_gem_ttm_helper.c
index ecf3d2a54a98..c44bfdbb722d 100644
--- a/drivers/gpu/drm/drm_gem_ttm_helper.c
+++ b/drivers/gpu/drm/drm_gem_ttm_helper.c
@@ -101,8 +101,6 @@ int drm_gem_ttm_mmap(struct drm_gem_object *gem,
 	int ret;
 
 	ret = ttm_bo_mmap_obj(vma, bo);
-	if (ret < 0)
-		return ret;
 
 	/*
 	 * ttm has its own object refcounting, so drop gem reference
@@ -110,7 +108,7 @@ int drm_gem_ttm_mmap(struct drm_gem_object *gem,
 	 */
 	drm_gem_object_put(gem);
 
-	return 0;
+	return ret;
 }
 EXPORT_SYMBOL(drm_gem_ttm_mmap);
 
-- 
2.32.0


             reply	other threads:[~2021-12-21 11:40 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-21 11:38 Roberto Sassu [this message]
2022-01-07 10:34 ` [PATCH] drm: Fix gem obj imbalance due to calling drm_gem_object_put() twice Roberto Sassu

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=20211221113837.1607448-1-roberto.sassu@huawei.com \
    --to=roberto.sassu@huawei.com \
    --cc=airlied@linux.ie \
    --cc=daniel@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=mripard@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=syzbot+c8ae65286134dd1b800d@syzkaller.appspotmail.com \
    --cc=tzimmermann@suse.de \
    /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;
as well as URLs for NNTP newsgroup(s).