[syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit:    f5459048c38a Merge tag 'i2c-for-7.0-final' of git://git.ke..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=1028d106580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=6754c86e8d9e4c91
dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1428d106580000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17509036580000

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/530844b89fdc/disk-f5459048.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/551ba51d047c/vmlinux-f5459048.xz
kernel image: https://storage.googleapis.com/syzbot-assets/9e6a44490566/bzImage-f5459048.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+b2e951687503f32f74ce@syzkaller.appspotmail.com

==================================================================
BUG: KASAN: slab-use-after-free in drm_gem_object_release_handle+0x4b/0x1e0 drivers/gpu/drm/drm_gem.c:374
Read of size 8 at addr ffff88806c6ffa88 by task syz.1.18/6089

CPU: 1 UID: 0 PID: 6089 Comm: syz.1.18 Not tainted syzkaller #0 PREEMPT(full) 
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/18/2026
Call Trace:
 <TASK>
 dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120
 print_address_description mm/kasan/report.c:378 [inline]
 print_report+0xba/0x230 mm/kasan/report.c:482
 kasan_report+0x117/0x150 mm/kasan/report.c:595
 drm_gem_object_release_handle+0x4b/0x1e0 drivers/gpu/drm/drm_gem.c:374
 idr_for_each+0x1c6/0x2a0 lib/idr.c:210
 drm_gem_release+0x28/0x40 drivers/gpu/drm/drm_gem.c:1088
 drm_file_free+0x729/0xa00 drivers/gpu/drm/drm_file.c:261
 drm_close_helper drivers/gpu/drm/drm_file.c:290 [inline]
 drm_release+0x2de/0x3f0 drivers/gpu/drm/drm_file.c:438
 __fput+0x44f/0xa70 fs/file_table.c:469
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
 __exit_to_user_mode_loop kernel/entry/common.c:67 [inline]
 exit_to_user_mode_loop+0xed/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:226 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:256 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:325 [inline]
 do_syscall_64+0x32d/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7efc3279c819
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffc28821858 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4
RAX: 0000000000000000 RBX: 00007ffc28821940 RCX: 00007efc3279c819
RDX: 0000000000000000 RSI: 000000000000001e RDI: 0000000000000003
RBP: 000000000001956f R08: 0000000000000001 R09: 0000000000000000
R10: 0000001b2f520000 R11: 0000000000000246 R12: 00007ffc28821980
R13: 00007efc32a1609c R14: 00000000000195b1 R15: 00007efc32a16090
 </TASK>

Allocated by task 6090:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 poison_kmalloc_redzone mm/kasan/common.c:398 [inline]
 __kasan_kmalloc+0x93/0xb0 mm/kasan/common.c:415
 kasan_kmalloc include/linux/kasan.h:263 [inline]
 __kmalloc_cache_noprof+0x31c/0x660 mm/slub.c:5380
 kmalloc_noprof include/linux/slab.h:950 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 __drm_gem_shmem_create+0xc4/0x2e0 drivers/gpu/drm/drm_gem_shmem_helper.c:130
 drm_gem_shmem_create drivers/gpu/drm/drm_gem_shmem_helper.c:157 [inline]
 drm_gem_shmem_create_with_handle drivers/gpu/drm/drm_gem_shmem_helper.c:460 [inline]
 drm_gem_shmem_dumb_create+0x72/0x120 drivers/gpu/drm/drm_gem_shmem_helper.c:549
 drm_mode_create_dumb drivers/gpu/drm/drm_dumb_buffers.c:227 [inline]
 drm_mode_create_dumb_ioctl+0x2bd/0x340 drivers/gpu/drm/drm_dumb_buffers.c:236
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x14d/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 6093:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 kasan_save_free_info+0x46/0x50 mm/kasan/generic.c:584
 poison_slab_object mm/kasan/common.c:253 [inline]
 __kasan_slab_free+0x5c/0x80 mm/kasan/common.c:285
 kasan_slab_free include/linux/kasan.h:235 [inline]
 slab_free_hook mm/slub.c:2685 [inline]
 slab_free mm/slub.c:6165 [inline]
 kfree+0x1c1/0x630 mm/slub.c:6483
 drm_gem_object_release_handle+0xc2/0x1e0 drivers/gpu/drm/drm_gem.c:385
 drm_gem_handle_delete+0x7b/0xb0 drivers/gpu/drm/drm_gem.c:413
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x14d/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

The buggy address belongs to the object at ffff88806c6ff800
 which belongs to the cache kmalloc-1k of size 1024
The buggy address is located 648 bytes inside of
 freed 1024-byte region [ffff88806c6ff800, ffff88806c6ffc00)

The buggy address belongs to the physical page:
page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x6c6f8
head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
flags: 0xfff00000000040(head|node=0|zone=1|lastcpupid=0x7ff)
page_type: f5(slab)
raw: 00fff00000000040 ffff88813fea6dc0 dead000000000100 dead000000000122
raw: 0000000000000000 0000000800100010 00000000f5000000 0000000000000000
head: 00fff00000000040 ffff88813fea6dc0 dead000000000100 dead000000000122
head: 0000000000000000 0000000800100010 00000000f5000000 0000000000000000
head: 00fff00000000003 ffffea0001b1be01 00000000ffffffff 00000000ffffffff
head: ffffffffffffffff 0000000000000000 00000000ffffffff 0000000000000008
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 5954, tgid 5954 (syz-executor), ts 103695420380, free_ts 92416780237
 set_page_owner include/linux/page_owner.h:32 [inline]
 post_alloc_hook+0x231/0x280 mm/page_alloc.c:1889
 prep_new_page mm/page_alloc.c:1897 [inline]
 get_page_from_freelist+0x24dc/0x2580 mm/page_alloc.c:3962
 __alloc_frozen_pages_noprof+0x18d/0x380 mm/page_alloc.c:5250
 alloc_slab_page mm/slub.c:3292 [inline]
 allocate_slab+0x77/0x660 mm/slub.c:3481
 new_slab mm/slub.c:3539 [inline]
 refill_objects+0x331/0x3c0 mm/slub.c:7175
 refill_sheaf mm/slub.c:2812 [inline]
 __pcs_replace_empty_main+0x2e6/0x730 mm/slub.c:4615
 alloc_from_pcs mm/slub.c:4717 [inline]
 slab_alloc_node mm/slub.c:4851 [inline]
 __do_kmalloc_node mm/slub.c:5259 [inline]
 __kmalloc_node_noprof+0x577/0x7c0 mm/slub.c:5266
 kmalloc_node_noprof include/linux/slab.h:1081 [inline]
 qdisc_alloc+0x92/0x9e0 net/sched/sch_generic.c:951
 qdisc_create_dflt+0x8e/0x510 net/sched/sch_generic.c:1006
 attach_one_default_qdisc net/sched/sch_generic.c:1172 [inline]
 netdev_for_each_tx_queue include/linux/netdevice.h:2684 [inline]
 attach_default_qdiscs net/sched/sch_generic.c:1190 [inline]
 dev_activate+0x378/0x1150 net/sched/sch_generic.c:1249
 __dev_open+0x67a/0x830 net/core/dev.c:1711
 __dev_change_flags+0x1f7/0x690 net/core/dev.c:9781
 netif_change_flags+0x88/0x1a0 net/core/dev.c:9844
 do_setlink+0xf82/0x4590 net/core/rtnetlink.c:3180
 rtnl_changelink net/core/rtnetlink.c:3798 [inline]
 __rtnl_newlink net/core/rtnetlink.c:3971 [inline]
 rtnl_newlink+0x15ad/0x1bb0 net/core/rtnetlink.c:4108
 rtnetlink_rcv_msg+0x7d5/0xbe0 net/core/rtnetlink.c:6994
page last free pid 5854 tgid 5854 stack trace:
 reset_page_owner include/linux/page_owner.h:25 [inline]
 __free_pages_prepare mm/page_alloc.c:1433 [inline]
 __free_frozen_pages+0xc2b/0xdb0 mm/page_alloc.c:2978
 vfree+0x25a/0x400 mm/vmalloc.c:3479
 kcov_put kernel/kcov.c:442 [inline]
 kcov_close+0x28/0x50 kernel/kcov.c:543
 __fput+0x44f/0xa70 fs/file_table.c:469
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 exit_task_work include/linux/task_work.h:40 [inline]
 do_exit+0x70f/0x23c0 kernel/exit.c:976
 do_group_exit+0x21b/0x2d0 kernel/exit.c:1118
 get_signal+0x1284/0x1330 kernel/signal.c:3034
 arch_do_signal_or_restart+0xbc/0x830 arch/x86/kernel/signal.c:337
 __exit_to_user_mode_loop kernel/entry/common.c:64 [inline]
 exit_to_user_mode_loop+0x86/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:226 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:256 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:325 [inline]
 do_syscall_64+0x32d/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Memory state around the buggy address:
 ffff88806c6ff980: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88806c6ffa00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff88806c6ffa80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                      ^
 ffff88806c6ffb00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88806c6ffb80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================


---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzkaller@googlegroups.com.

syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title

If you want syzbot to run the reproducer, reply with:
#syz test: git://repo/address.git branch-or-commit-hash
If you attach or paste a git patch, syzbot will apply it before testing.

If you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)

If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-report

If you want to undo deduplication, reply with:
#syz undup

Forwarded: [PATCH] drm/gem: fix use-after-free in drm_gem_release

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: [PATCH] drm/gem: fix use-after-free in drm_gem_release
Author: kartikey406@gmail.com

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master


drm_gem_release() calls idr_for_each() with
drm_gem_object_release_handle() as callback. A concurrent
drm_gem_handle_delete() can free the GEM object between the time
idr_for_each() reads the pointer and drm_gem_object_release_handle()
uses it, causing a use-after-free.

Fix this by splitting the operation into two phases:

1. Under table_lock, walk the IDR using a new collect callback that
   only increments each object's reference count and adds it to a
   local list. Only atomic operations are performed here so holding
   a spinlock is safe.

2. Outside the lock, iterate the local list and call
   drm_gem_object_release_handle() on each object safely. Drop our
   reference afterwards and destroy the IDR only after all objects
   have been properly released.

Holding a reference during phase 1 ensures that a concurrent
drm_gem_handle_delete() cannot free any object while we are still
using it.

Reported-by: syzbot+b2e951687503f32f74ce@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
---
 drivers/gpu/drm/drm_gem.c | 39 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 38 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 891c3bff5ae0..ee9939356de4 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -1073,6 +1073,30 @@ drm_gem_open(struct drm_device *dev, struct drm_file *file_private)
 	spin_lock_init(&file_private->table_lock);
 }
 
+struct drm_gem_object_entry {
+	struct drm_gem_object *obj;
+	struct list_head head;
+};
+
+static int drm_gem_object_collect(int id, void *ptr, void *data)
+{
+	struct drm_gem_object_entry *entry;
+	struct drm_gem_object *obj = ptr;
+	struct list_head *list = data;
+
+	if (!obj)
+		return 0;
+
+	entry = kmalloc_obj(*entry, GFP_ATOMIC);
+	if (!entry)
+		return -ENOMEM;
+
+	drm_gem_object_get(obj);
+	entry->obj = obj;
+	list_add(&entry->head, list);
+	return 0;
+}
+
 /**
  * drm_gem_release - release file-private GEM resources
  * @dev: drm_device which is being closed by userspace
@@ -1085,8 +1109,21 @@ drm_gem_open(struct drm_device *dev, struct drm_file *file_private)
 void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
+	struct drm_gem_object_entry *entry, *tmp;
+	LIST_HEAD(list);
+
+	spin_lock(&file_private->table_lock);
 	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+		     &drm_gem_object_collect, &list);
+	spin_unlock(&file_private->table_lock);
+
+	list_for_each_entry_safe(entry, tmp, &list, head) {
+		list_del(&entry->head);
+		drm_gem_object_release_handle(0, entry->obj, file_private);
+		drm_gem_object_put(entry->obj);
+		kfree(entry);
+	}
+
 	idr_destroy(&file_private->object_idr);
 }
 
-- 
2.43.0

Forwarded: [PATCH] drm/gem: fix use-after-free in drm_gem_release

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: [PATCH] drm/gem: fix use-after-free in drm_gem_release
Author: kartikey406@gmail.com

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master


KASAN reported a use-after-free in drm_gem_object_release_handle
when drm_gem_release() races with drm_gem_handle_delete().

drm_gem_release() calls idr_for_each() which gives out object
pointers, but a concurrent drm_gem_handle_delete() can free the
object before drm_gem_object_release_handle() gets to use it.

Fix this by splitting the operation into two phases:

Phase 1: Walk the IDR under table_lock. For each object, atomically
replace its IDR slot with NULL using idr_replace() so that a
concurrent drm_gem_handle_delete() sees NULL and bails out with
-EINVAL. Grab a reference on the object and collect it into a local
list. Only atomic operations are performed here so holding a
spinlock is safe.

Phase 2: Outside the lock, iterate the local list and call
drm_gem_object_release_handle() on each object safely. Drop our
reference afterwards. Destroy the IDR only after all objects have
been properly released.

The idr_replace(NULL) in phase 1 is the key: it mirrors what
drm_gem_handle_delete() itself does to atomically claim ownership
of an object, ensuring that each object is processed by exactly
one path.

Reported-by: syzbot+b2e951687503f32f74ce@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
---
 drivers/gpu/drm/drm_gem.c | 49 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 48 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 891c3bff5ae0..9ec1f03c6383 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -1073,6 +1073,37 @@ drm_gem_open(struct drm_device *dev, struct drm_file *file_private)
 	spin_lock_init(&file_private->table_lock);
 }
 
+struct drm_gem_object_entry {
+	struct drm_gem_object *obj;
+	struct list_head head;
+};
+
+struct drm_gem_collect_data {
+	struct drm_file    *file_priv;
+	struct list_head   list;
+};
+
+static int drm_gem_object_collect(int id, void *ptr, void *data)
+{
+	struct drm_gem_collect_data *cd = data;
+	struct drm_gem_object_entry *entry;
+	struct drm_gem_object *obj;
+
+	/* Atomically claim the slot → handle_delete sees NULL, bails */
+	obj = idr_replace(&cd->file_priv->object_idr, NULL, id);
+	if (IS_ERR_OR_NULL(obj))
+		return 0;
+
+	entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
+	if (!entry)
+		return -ENOMEM;
+
+	drm_gem_object_get(obj);
+	entry->obj = obj;
+	list_add(&entry->head, &cd->list);
+	return 0;
+}
+
 /**
  * drm_gem_release - release file-private GEM resources
  * @dev: drm_device which is being closed by userspace
@@ -1085,8 +1116,24 @@ drm_gem_open(struct drm_device *dev, struct drm_file *file_private)
 void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
+	struct drm_gem_object_entry *entry, *tmp;
+	struct drm_gem_collect_data cd = {
+		.file_priv = file_private,
+	};
+	INIT_LIST_HEAD(&cd.list);
+
+	spin_lock(&file_private->table_lock);
 	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+		     &drm_gem_object_collect, &cd);
+	spin_unlock(&file_private->table_lock);
+
+	list_for_each_entry_safe(entry, tmp, &cd.list, head) {
+		list_del(&entry->head);
+		drm_gem_object_release_handle(0, entry->obj, file_private);
+		drm_gem_object_put(entry->obj);
+		kfree(entry);
+	}
+
 	idr_destroy(&file_private->object_idr);
 }
 
-- 
2.43.0

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[Hillf Danton]

> Date: Mon, 13 Apr 2026 10:16:36 -0700	[thread overview]
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    f5459048c38a Merge tag 'i2c-for-7.0-final' of git://git.ke..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1028d106580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=6754c86e8d9e4c91
> dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
> compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1428d106580000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17509036580000

#syz test

--- x/drivers/gpu/drm/drm_gem.c
+++ y/drivers/gpu/drm/drm_gem.c
@@ -387,6 +387,23 @@ drm_gem_object_release_handle(int id, vo
 	return 0;
 }
 
+static int
+drm_gem_object_release_idr_handle(int id, void *ptr, void *data)
+{
+	struct drm_file *filp = data;
+	struct drm_gem_object *obj, *tmp = ptr;
+
+	spin_lock(&filp->table_lock);
+	obj = idr_replace(&filp->object_idr, NULL, id);
+	spin_unlock(&filp->table_lock);
+	if (IS_ERR_OR_NULL(obj))
+		return 0;
+	if (obj != tmp)
+		return 0;
+	drm_gem_object_release_handle(id, obj, filp);
+	return 0;
+}
+
 /**
  * drm_gem_handle_delete - deletes the given file-private handle
  * @filp: drm file-private structure to use for the handle look up
@@ -1086,7 +1103,7 @@ void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
 	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+		     &drm_gem_object_release_idr_handle, file_private);
 	idr_destroy(&file_private->object_idr);
 }
 
--

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
KFENCE: use-after-free read in drm_gem_object_release_handle

==================================================================
BUG: KFENCE: use-after-free read in drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374

Use-after-free read at 0xffff88823be98e88 (in kfence-#75):
 drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
 drm_gem_object_release_idr_handle+0x69/0x80 drivers/gpu/drm/drm_gem.c:403
 idr_for_each+0x1c6/0x2a0 lib/idr.c:210
 drm_gem_release+0x28/0x40 drivers/gpu/drm/drm_gem.c:1105
 drm_file_free+0x729/0xa00 drivers/gpu/drm/drm_file.c:261
 drm_close_helper drivers/gpu/drm/drm_file.c:290 [inline]
 drm_release+0x2de/0x3f0 drivers/gpu/drm/drm_file.c:438
 __fput+0x44f/0xa60 fs/file_table.c:510
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
 __exit_to_user_mode_loop kernel/entry/common.c:67 [inline]
 exit_to_user_mode_loop+0xed/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:226 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:256 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:325 [inline]
 do_syscall_64+0x33e/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

kfence-#75: 0xffff88823be98c00-0xffff88823be98ee7, size=744, cache=kmalloc-1k

allocated by task 6583 on cpu 0 at 133.209896s (0.096709s ago):
 kmalloc_noprof include/linux/slab.h:950 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 __drm_gem_shmem_create+0xc4/0x2e0 drivers/gpu/drm/drm_gem_shmem_helper.c:130
 drm_gem_shmem_create drivers/gpu/drm/drm_gem_shmem_helper.c:157 [inline]
 drm_gem_shmem_create_with_handle drivers/gpu/drm/drm_gem_shmem_helper.c:460 [inline]
 drm_gem_shmem_dumb_create+0x72/0x120 drivers/gpu/drm/drm_gem_shmem_helper.c:549
 drm_mode_create_dumb drivers/gpu/drm/drm_dumb_buffers.c:227 [inline]
 drm_mode_create_dumb_ioctl+0x2bd/0x340 drivers/gpu/drm/drm_dumb_buffers.c:236
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

freed by task 6585 on cpu 1 at 133.211010s (0.147010s ago):
 drm_gem_handle_delete+0x7b/0xb0 drivers/gpu/drm/drm_gem.c:430
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

CPU: 1 UID: 0 PID: 6581 Comm: syz.1.67 Not tainted syzkaller #0 PREEMPT(full) 
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/18/2026
RIP: 0010:drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
Code: 5f 6c 62 fc 4d 85 f6 0f 84 9e 00 00 00 4c 8d bb 88 02 00 00 4c 89 f8 48 c1 e8 03 42 80 3c 28 00 74 08 4c 89 ff e8 59 c1 cc fc <4d> 8b 3f 49 83 c7 10 4c 89 f8 48 c1 e8 03 42 80 3c 28 00 74 08 4c
RSP: 0018:ffffc90003f27ad8 EFLAGS: 00010246
RAX: 1ffff110477d31d1 RBX: ffff88823be98c00 RCX: ffff888054fa1e80
RDX: 0000000000000000 RSI: ffff88823be98c00 RDI: 0000000000000080
RBP: 0000000000000080 R08: ffff8880321950e3 R09: 1ffff11006432a1c
R10: dffffc0000000000 R11: ffffed1006432a1d R12: ffff8880321950e0
R13: dffffc0000000000 R14: ffff888032195000 R15: ffff88823be98e88
FS:  000055557167a500(0000) GS:ffff88812554b000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffff88823be98e88 CR3: 000000006dd68000 CR4: 00000000003526f0
Call Trace:
 <TASK>
 drm_gem_object_release_idr_handle+0x69/0x80 drivers/gpu/drm/drm_gem.c:403
 idr_for_each+0x1c6/0x2a0 lib/idr.c:210
 drm_gem_release+0x28/0x40 drivers/gpu/drm/drm_gem.c:1105
 drm_file_free+0x729/0xa00 drivers/gpu/drm/drm_file.c:261
 drm_close_helper drivers/gpu/drm/drm_file.c:290 [inline]
 drm_release+0x2de/0x3f0 drivers/gpu/drm/drm_file.c:438
 __fput+0x44f/0xa60 fs/file_table.c:510
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
 __exit_to_user_mode_loop kernel/entry/common.c:67 [inline]
 exit_to_user_mode_loop+0xed/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:226 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:256 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:325 [inline]
 do_syscall_64+0x33e/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f246c99c819
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffd35630618 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4
RAX: 0000000000000000 RBX: 00007ffd35630700 RCX: 00007f246c99c819
RDX: 0000000000000000 RSI: 000000000000001e RDI: 0000000000000003
RBP: 000000000002082e R08: 0000000000000001 R09: 0000000000000000
R10: 0000001b32820000 R11: 0000000000000246 R12: 00007ffd35630740
R13: 00007f246cc1609c R14: 0000000000020866 R15: 00007f246cc16090
 </TASK>
==================================================================
----------------
Code disassembly (best guess), 3 bytes skipped:
   0:	fc                   	cld
   1:	4d 85 f6             	test   %r14,%r14
   4:	0f 84 9e 00 00 00    	je     0xa8
   a:	4c 8d bb 88 02 00 00 	lea    0x288(%rbx),%r15
  11:	4c 89 f8             	mov    %r15,%rax
  14:	48 c1 e8 03          	shr    $0x3,%rax
  18:	42 80 3c 28 00       	cmpb   $0x0,(%rax,%r13,1)
  1d:	74 08                	je     0x27
  1f:	4c 89 ff             	mov    %r15,%rdi
  22:	e8 59 c1 cc fc       	call   0xfcccc180
* 27:	4d 8b 3f             	mov    (%r15),%r15 <-- trapping instruction
  2a:	49 83 c7 10          	add    $0x10,%r15
  2e:	4c 89 f8             	mov    %r15,%rax
  31:	48 c1 e8 03          	shr    $0x3,%rax
  35:	42 80 3c 28 00       	cmpb   $0x0,(%rax,%r13,1)
  3a:	74 08                	je     0x44
  3c:	4c                   	rex.WR


Tested on:

commit:         d60bc140 Merge tag 'pwrseq-updates-for-v7.1-rc1' of gi..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=157e7cd2580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=85bd86f990609a1
dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
patch:          https://syzkaller.appspot.com/x/patch.diff?x=12673036580000

Forwarded: [PATCH] drm/gem: fix use-after-free in drm_gem_release

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: [PATCH] drm/gem: fix use-after-free in drm_gem_release
Author: kartikey406@gmail.com

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master


drm_gem_release() calls idr_for_each() to iterate over all GEM handles
and release them during file close. However, idr_for_each() does not
hold file_priv->table_lock while reading object pointers from the IDR,
which means it can race with a concurrent drm_gem_handle_delete() from
a GEM_CLOSE ioctl on another thread sharing the same file descriptor.

The race is as follows:

  Thread A (close fd)                Thread B (GEM_CLOSE ioctl)
  drm_gem_release()
    idr_for_each()
      reads obj ptr from IDR
                                     drm_gem_handle_delete()
                                       idr_replace(NULL)
                                       drm_gem_object_release_handle()
                                         handle_put -> kfree(obj)
      drm_gem_object_release_handle()
        obj->funcs->close()  <- UAF!

Fix this by replacing idr_for_each() with a manual iteration loop that
holds table_lock while reading each pointer and uses idr_replace(NULL)
to claim the handle, mirroring the pattern used by
drm_gem_handle_delete(). This ensures mutual exclusion: whichever path
claims the handle first (sets it to NULL) gets to process it, and the
other path safely skips it.

Reported-by: syzbot+b2e951687503f32f74ce@syzkaller.appspotmail.com
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
---
 drivers/gpu/drm/drm_gem.c | 24 ++++++++++++++++++++++--
 1 file changed, 22 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 891c3bff5ae0..b878f82146f1 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -1085,8 +1085,28 @@ drm_gem_open(struct drm_device *dev, struct drm_file *file_private)
 void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
-	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+	struct drm_gem_object *obj;
+	int next_id;
+
+	next_id = 0;
+	while (1) {
+		spin_lock(&file_private->table_lock);
+		obj = idr_get_next(&file_private->object_idr, &next_id);
+		if (!obj) {
+			spin_unlock(&file_private->table_lock);
+			break;
+		}
+		/* Claim the handle — same pattern as drm_gem_handle_delete */
+		idr_replace(&file_private->object_idr, NULL, next_id);
+		spin_unlock(&file_private->table_lock);
+
+		drm_gem_object_release_handle(next_id, obj, file_private);
+
+		spin_lock(&file_private->table_lock);
+		idr_remove(&file_private->object_idr, next_id);
+		spin_unlock(&file_private->table_lock);
+		next_id++;
+	}
 	idr_destroy(&file_private->object_idr);
 }
 
-- 
2.43.0

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[Hillf Danton]

> Date: Mon, 13 Apr 2026 10:16:36 -0700	[thread overview]
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    f5459048c38a Merge tag 'i2c-for-7.0-final' of git://git.ke..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1028d106580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=6754c86e8d9e4c91
> dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
> compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1428d106580000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17509036580000

#syz test

--- x/drivers/gpu/drm/drm_gem.c
+++ y/drivers/gpu/drm/drm_gem.c
@@ -387,6 +387,24 @@ drm_gem_object_release_handle(int id, vo
 	return 0;
 }
 
+static int
+drm_gem_object_release_idr_handle(int id, void *ptr, void *data)
+{
+	struct drm_file *filp = data;
+	struct drm_gem_object *obj, *tmp = ptr;
+
+	spin_lock(&filp->table_lock);
+	obj = idr_replace(&filp->object_idr, NULL, id);
+	idr_remove(&filp->object_idr, id);
+	spin_unlock(&filp->table_lock);
+	if (IS_ERR_OR_NULL(obj))
+		return 0;
+	if (obj != tmp)
+		return 0;
+	drm_gem_object_release_handle(id, obj, filp);
+	return 0;
+}
+
 /**
  * drm_gem_handle_delete - deletes the given file-private handle
  * @filp: drm file-private structure to use for the handle look up
@@ -405,6 +423,7 @@ drm_gem_handle_delete(struct drm_file *f
 
 	/* Check if we currently have a reference on the object */
 	obj = idr_replace(&filp->object_idr, NULL, handle);
+	idr_remove(&filp->object_idr, handle);
 	spin_unlock(&filp->table_lock);
 	if (IS_ERR_OR_NULL(obj))
 		return -EINVAL;
@@ -412,11 +431,6 @@ drm_gem_handle_delete(struct drm_file *f
 	/* Release driver's reference and decrement refcount. */
 	drm_gem_object_release_handle(handle, obj, filp);
 
-	/* And finally make the handle available for future allocations. */
-	spin_lock(&filp->table_lock);
-	idr_remove(&filp->object_idr, handle);
-	spin_unlock(&filp->table_lock);
-
 	return 0;
 }
 EXPORT_SYMBOL(drm_gem_handle_delete);
@@ -1086,7 +1100,7 @@ void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
 	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+		     &drm_gem_object_release_idr_handle, file_private);
 	idr_destroy(&file_private->object_idr);
 }
 
--

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
KASAN: slab-use-after-free Read in drm_gem_object_release_handle

==================================================================
BUG: KASAN: slab-use-after-free in drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
Read of size 8 at addr ffff8880296c6a88 by task syz.2.83/6856

CPU: 1 UID: 0 PID: 6856 Comm: syz.2.83 Not tainted syzkaller #0 PREEMPT(full) 
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/18/2026
Call Trace:
 <TASK>
 dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120
 print_address_description mm/kasan/report.c:378 [inline]
 print_report+0xba/0x230 mm/kasan/report.c:482
 kasan_report+0x117/0x150 mm/kasan/report.c:595
 drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
 drm_gem_object_release_idr_handle+0x83/0xa0 drivers/gpu/drm/drm_gem.c:404
 idr_for_each+0x1c6/0x2a0 lib/idr.c:210
 drm_gem_release+0x28/0x40 drivers/gpu/drm/drm_gem.c:1102
 drm_file_free+0x729/0xa00 drivers/gpu/drm/drm_file.c:261
 drm_close_helper drivers/gpu/drm/drm_file.c:290 [inline]
 drm_release+0x2de/0x3f0 drivers/gpu/drm/drm_file.c:438
 __fput+0x44f/0xa60 fs/file_table.c:510
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
 __exit_to_user_mode_loop kernel/entry/common.c:67 [inline]
 exit_to_user_mode_loop+0xed/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:226 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:256 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:325 [inline]
 do_syscall_64+0x33e/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f736919c819
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fffaf8d5118 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4
RAX: 0000000000000000 RBX: 00007fffaf8d5200 RCX: 00007f736919c819
RDX: 0000000000000000 RSI: 000000000000001e RDI: 0000000000000003
RBP: 000000000001f76e R08: 0000000000000001 R09: 0000000000000000
R10: 0000001b32d20000 R11: 0000000000000246 R12: 00007fffaf8d5240
R13: 00007f736941609c R14: 000000000001f7f0 R15: 00007f7369416090
 </TASK>

Allocated by task 6857:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 poison_kmalloc_redzone mm/kasan/common.c:398 [inline]
 __kasan_kmalloc+0x93/0xb0 mm/kasan/common.c:415
 kasan_kmalloc include/linux/kasan.h:263 [inline]
 __kmalloc_cache_noprof+0x31c/0x660 mm/slub.c:5380
 kmalloc_noprof include/linux/slab.h:950 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 __drm_gem_shmem_create+0xc4/0x2e0 drivers/gpu/drm/drm_gem_shmem_helper.c:130
 drm_gem_shmem_create drivers/gpu/drm/drm_gem_shmem_helper.c:157 [inline]
 drm_gem_shmem_create_with_handle drivers/gpu/drm/drm_gem_shmem_helper.c:460 [inline]
 drm_gem_shmem_dumb_create+0x72/0x120 drivers/gpu/drm/drm_gem_shmem_helper.c:549
 drm_mode_create_dumb drivers/gpu/drm/drm_dumb_buffers.c:227 [inline]
 drm_mode_create_dumb_ioctl+0x2bd/0x340 drivers/gpu/drm/drm_dumb_buffers.c:236
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 6864:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 kasan_save_free_info+0x46/0x50 mm/kasan/generic.c:584
 poison_slab_object mm/kasan/common.c:253 [inline]
 __kasan_slab_free+0x5c/0x80 mm/kasan/common.c:285
 kasan_slab_free include/linux/kasan.h:235 [inline]
 slab_free_hook mm/slub.c:2685 [inline]
 slab_free mm/slub.c:6165 [inline]
 kfree+0x1c1/0x630 mm/slub.c:6483
 drm_gem_handle_delete+0x90/0xb0 drivers/gpu/drm/drm_gem.c:432
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

The buggy address belongs to the object at ffff8880296c6800
 which belongs to the cache kmalloc-1k of size 1024
The buggy address is located 648 bytes inside of
 freed 1024-byte region [ffff8880296c6800, ffff8880296c6c00)

The buggy address belongs to the physical page:
page: refcount:0 mapcount:0 mapping:0000000000000000 index:0xffff8880296c1800 pfn:0x296c0
head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
flags: 0xfff00000000240(workingset|head|node=0|zone=1|lastcpupid=0x7ff)
page_type: f5(slab)
raw: 00fff00000000240 ffff88813fea7dc0 ffffea0001f5ea10 ffffea0000a31410
raw: ffff8880296c1800 000000080010000e 00000000f5000000 0000000000000000
head: 00fff00000000240 ffff88813fea7dc0 ffffea0001f5ea10 ffffea0000a31410
head: ffff8880296c1800 000000080010000e 00000000f5000000 0000000000000000
head: 00fff00000000003 ffffea0000a5b001 00000000ffffffff 00000000ffffffff
head: ffffffffffffffff 0000000000000000 00000000ffffffff 0000000000000008
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd2000(__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 619, tgid 619 (kworker/u8:7), ts 7851559667, free_ts 0
 set_page_owner include/linux/page_owner.h:32 [inline]
 post_alloc_hook+0x231/0x280 mm/page_alloc.c:1889
 prep_new_page mm/page_alloc.c:1897 [inline]
 get_page_from_freelist+0x24dc/0x2580 mm/page_alloc.c:3962
 __alloc_frozen_pages_noprof+0x18d/0x380 mm/page_alloc.c:5250
 alloc_slab_page mm/slub.c:3292 [inline]
 allocate_slab+0x77/0x660 mm/slub.c:3481
 new_slab mm/slub.c:3539 [inline]
 refill_objects+0x331/0x3c0 mm/slub.c:7175
 refill_sheaf mm/slub.c:2812 [inline]
 __pcs_replace_empty_main+0x2e6/0x730 mm/slub.c:4615
 alloc_from_pcs mm/slub.c:4717 [inline]
 slab_alloc_node mm/slub.c:4851 [inline]
 __kmalloc_cache_node_noprof+0x465/0x6b0 mm/slub.c:5388
 kmalloc_node_noprof include/linux/slab.h:1077 [inline]
 blk_mq_alloc_hctx block/blk-mq.c:4054 [inline]
 blk_mq_alloc_and_init_hctx+0x17e/0xc50 block/blk-mq.c:4554
 __blk_mq_realloc_hw_ctxs+0x2bd/0x670 block/blk-mq.c:4605
 blk_mq_realloc_hw_ctxs block/blk-mq.c:4642 [inline]
 blk_mq_init_allocated_queue+0x365/0x13e0 block/blk-mq.c:4672
 blk_mq_alloc_queue+0x1ba/0x2e0 block/blk-mq.c:4454
 scsi_alloc_sdev+0x7c0/0xc80 drivers/scsi/scsi_scan.c:339
 scsi_probe_and_add_lun+0x200/0x4830 drivers/scsi/scsi_scan.c:1215
 __scsi_scan_target+0x1f0/0xe10 drivers/scsi/scsi_scan.c:1786
 scsi_scan_channel drivers/scsi/scsi_scan.c:1874 [inline]
 scsi_scan_host_selected+0x372/0x690 drivers/scsi/scsi_scan.c:1903
 do_scsi_scan_host drivers/scsi/scsi_scan.c:2042 [inline]
 do_scan_async+0x124/0x760 drivers/scsi/scsi_scan.c:2052
page_owner free stack trace missing

Memory state around the buggy address:
 ffff8880296c6980: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff8880296c6a00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff8880296c6a80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                      ^
 ffff8880296c6b00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff8880296c6b80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================


Tested on:

commit:         d60bc140 Merge tag 'pwrseq-updates-for-v7.1-rc1' of gi..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=142d04ce580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=85bd86f990609a1
dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
patch:          https://syzkaller.appspot.com/x/patch.diff?x=1284b036580000

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[Hillf Danton]

> Date: Mon, 13 Apr 2026 10:16:36 -0700	[thread overview]
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    f5459048c38a Merge tag 'i2c-for-7.0-final' of git://git.ke..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1028d106580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=6754c86e8d9e4c91
> dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
> compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1428d106580000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17509036580000

#syz test

--- x/drivers/gpu/drm/drm_gem.c
+++ y/drivers/gpu/drm/drm_gem.c
@@ -387,6 +387,26 @@ drm_gem_object_release_handle(int id, vo
 	return 0;
 }
 
+static int
+drm_gem_object_release_idr_handle(int id, void *ptr, void *data)
+{
+	struct drm_file *filp = data;
+	struct drm_gem_object *obj, *tmp = ptr;
+
+	spin_lock(&filp->table_lock);
+	obj = idr_find(&filp->object_idr, id);
+	if (obj != tmp) {
+		spin_unlock(&filp->table_lock);
+		return 0;
+	}
+	idr_remove(&filp->object_idr, id);
+	spin_unlock(&filp->table_lock);
+	if (IS_ERR_OR_NULL(obj))
+		return 0;
+	drm_gem_object_release_handle(id, obj, filp);
+	return 0;
+}
+
 /**
  * drm_gem_handle_delete - deletes the given file-private handle
  * @filp: drm file-private structure to use for the handle look up
@@ -404,7 +424,8 @@ drm_gem_handle_delete(struct drm_file *f
 	spin_lock(&filp->table_lock);
 
 	/* Check if we currently have a reference on the object */
-	obj = idr_replace(&filp->object_idr, NULL, handle);
+	obj = idr_find(&filp->object_idr, handle);
+	idr_remove(&filp->object_idr, handle);
 	spin_unlock(&filp->table_lock);
 	if (IS_ERR_OR_NULL(obj))
 		return -EINVAL;
@@ -412,11 +433,6 @@ drm_gem_handle_delete(struct drm_file *f
 	/* Release driver's reference and decrement refcount. */
 	drm_gem_object_release_handle(handle, obj, filp);
 
-	/* And finally make the handle available for future allocations. */
-	spin_lock(&filp->table_lock);
-	idr_remove(&filp->object_idr, handle);
-	spin_unlock(&filp->table_lock);
-
 	return 0;
 }
 EXPORT_SYMBOL(drm_gem_handle_delete);
@@ -1086,7 +1102,7 @@ void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
 	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+		     &drm_gem_object_release_idr_handle, file_private);
 	idr_destroy(&file_private->object_idr);
 }
 
--

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
KASAN: slab-use-after-free Read in drm_gem_object_release_handle

==================================================================
BUG: KASAN: slab-use-after-free in drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
Read of size 8 at addr ffff888058ae0288 by task syz.1.26/6493

CPU: 1 UID: 0 PID: 6493 Comm: syz.1.26 Not tainted syzkaller #0 PREEMPT(full) 
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/18/2026
Call Trace:
 <TASK>
 dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120
 print_address_description mm/kasan/report.c:378 [inline]
 print_report+0xba/0x230 mm/kasan/report.c:482
 kasan_report+0x117/0x150 mm/kasan/report.c:595
 drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
 drm_gem_object_release_idr_handle+0x99/0xb0 drivers/gpu/drm/drm_gem.c:406
 idr_for_each+0x1c6/0x2a0 lib/idr.c:210
 drm_gem_release+0x28/0x40 drivers/gpu/drm/drm_gem.c:1104
 drm_file_free+0x729/0xa00 drivers/gpu/drm/drm_file.c:261
 drm_close_helper drivers/gpu/drm/drm_file.c:290 [inline]
 drm_release+0x2de/0x3f0 drivers/gpu/drm/drm_file.c:438
 __fput+0x44f/0xa60 fs/file_table.c:510
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
 __exit_to_user_mode_loop kernel/entry/common.c:67 [inline]
 exit_to_user_mode_loop+0xed/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:226 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:256 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:325 [inline]
 do_syscall_64+0x33e/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f384cb9c819
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffea5f94ac8 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4
RAX: 0000000000000000 RBX: 00007ffea5f94bb0 RCX: 00007f384cb9c819
RDX: 0000000000000000 RSI: 000000000000001e RDI: 0000000000000003
RBP: 000000000001ea55 R08: 0000000000000001 R09: 0000000000000000
R10: 0000001b30e20000 R11: 0000000000000246 R12: 00007ffea5f94bf0
R13: 00007f384ce1609c R14: 000000000001ea8b R15: 00007f384ce16090
 </TASK>

Allocated by task 6494:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 poison_kmalloc_redzone mm/kasan/common.c:398 [inline]
 __kasan_kmalloc+0x93/0xb0 mm/kasan/common.c:415
 kasan_kmalloc include/linux/kasan.h:263 [inline]
 __kmalloc_cache_noprof+0x31c/0x660 mm/slub.c:5380
 kmalloc_noprof include/linux/slab.h:950 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 __drm_gem_shmem_create+0xc4/0x2e0 drivers/gpu/drm/drm_gem_shmem_helper.c:130
 drm_gem_shmem_create drivers/gpu/drm/drm_gem_shmem_helper.c:157 [inline]
 drm_gem_shmem_create_with_handle drivers/gpu/drm/drm_gem_shmem_helper.c:460 [inline]
 drm_gem_shmem_dumb_create+0x72/0x120 drivers/gpu/drm/drm_gem_shmem_helper.c:549
 drm_mode_create_dumb drivers/gpu/drm/drm_dumb_buffers.c:227 [inline]
 drm_mode_create_dumb_ioctl+0x2bd/0x340 drivers/gpu/drm/drm_dumb_buffers.c:236
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 6495:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 kasan_save_free_info+0x46/0x50 mm/kasan/generic.c:584
 poison_slab_object mm/kasan/common.c:253 [inline]
 __kasan_slab_free+0x5c/0x80 mm/kasan/common.c:285
 kasan_slab_free include/linux/kasan.h:235 [inline]
 slab_free_hook mm/slub.c:2685 [inline]
 slab_free mm/slub.c:6165 [inline]
 kfree+0x1c1/0x630 mm/slub.c:6483
 drm_gem_handle_delete+0x84/0xa0 drivers/gpu/drm/drm_gem.c:434
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

The buggy address belongs to the object at ffff888058ae0000
 which belongs to the cache kmalloc-1k of size 1024
The buggy address is located 648 bytes inside of
 freed 1024-byte region [ffff888058ae0000, ffff888058ae0400)

The buggy address belongs to the physical page:
page: refcount:0 mapcount:0 mapping:0000000000000000 index:0xffff888058ae1800 pfn:0x58ae0
head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
flags: 0xfff00000000240(workingset|head|node=0|zone=1|lastcpupid=0x7ff)
page_type: f5(slab)
raw: 00fff00000000240 ffff88813fea7dc0 ffffea0000a1de10 ffff88813fea6888
raw: ffff888058ae1800 0000000800100003 00000000f5000000 0000000000000000
head: 00fff00000000240 ffff88813fea7dc0 ffffea0000a1de10 ffff88813fea6888
head: ffff888058ae1800 0000000800100003 00000000f5000000 0000000000000000
head: 00fff00000000003 ffffea000162b801 00000000ffffffff 00000000ffffffff
head: ffffffffffffffff 0000000000000000 00000000ffffffff 0000000000000008
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd2820(GFP_ATOMIC|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 13, tgid 13 (kworker/u8:1), ts 125536208154, free_ts 123699282996
 set_page_owner include/linux/page_owner.h:32 [inline]
 post_alloc_hook+0x231/0x280 mm/page_alloc.c:1889
 prep_new_page mm/page_alloc.c:1897 [inline]
 get_page_from_freelist+0x24dc/0x2580 mm/page_alloc.c:3962
 __alloc_frozen_pages_noprof+0x18d/0x380 mm/page_alloc.c:5250
 alloc_slab_page mm/slub.c:3292 [inline]
 allocate_slab+0x77/0x660 mm/slub.c:3481
 new_slab mm/slub.c:3539 [inline]
 refill_objects+0x331/0x3c0 mm/slub.c:7175
 refill_sheaf mm/slub.c:2812 [inline]
 __pcs_replace_empty_main+0x2e6/0x730 mm/slub.c:4615
 alloc_from_pcs mm/slub.c:4717 [inline]
 slab_alloc_node mm/slub.c:4851 [inline]
 __do_kmalloc_node mm/slub.c:5259 [inline]
 __kmalloc_noprof+0x474/0x760 mm/slub.c:5272
 kmalloc_noprof include/linux/slab.h:954 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 ieee802_11_parse_elems_full+0x159/0x2ab0 net/mac80211/parse.c:1051
 ieee802_11_parse_elems net/mac80211/ieee80211_i.h:2480 [inline]
 ieee80211_inform_bss+0x161/0x1160 net/mac80211/scan.c:79
 rdev_inform_bss net/wireless/rdev-ops.h:418 [inline]
 cfg80211_inform_single_bss_data+0xd08/0x1b70 net/wireless/scan.c:2372
 cfg80211_inform_bss_data+0x266/0x3c40 net/wireless/scan.c:3226
 cfg80211_inform_bss_frame_data+0x3c7/0x760 net/wireless/scan.c:3317
 ieee80211_bss_info_update+0x794/0xa40 net/mac80211/scan.c:230
 ieee80211_rx_bss_info net/mac80211/ibss.c:1094 [inline]
 ieee80211_rx_mgmt_probe_beacon net/mac80211/ibss.c:1575 [inline]
 ieee80211_ibss_rx_queued_mgmt+0x1901/0x2cd0 net/mac80211/ibss.c:1602
 ieee80211_iface_process_skb net/mac80211/iface.c:1748 [inline]
 ieee80211_iface_work+0x84e/0x1340 net/mac80211/iface.c:1802
 cfg80211_wiphy_work+0x2ab/0x4a0 net/wireless/core.c:440
page last free pid 6361 tgid 6361 stack trace:
 reset_page_owner include/linux/page_owner.h:25 [inline]
 __free_pages_prepare mm/page_alloc.c:1433 [inline]
 __free_frozen_pages+0xc2b/0xdb0 mm/page_alloc.c:2978
 __slab_free+0x263/0x2b0 mm/slub.c:5573
 qlink_free mm/kasan/quarantine.c:163 [inline]
 qlist_free_all+0x97/0x100 mm/kasan/quarantine.c:179
 kasan_quarantine_reduce+0x148/0x160 mm/kasan/quarantine.c:286
 __kasan_slab_alloc+0x22/0x80 mm/kasan/common.c:350
 kasan_slab_alloc include/linux/kasan.h:253 [inline]
 slab_post_alloc_hook mm/slub.c:4538 [inline]
 slab_alloc_node mm/slub.c:4866 [inline]
 __do_kmalloc_node mm/slub.c:5259 [inline]
 __kmalloc_noprof+0x316/0x760 mm/slub.c:5272
 kmalloc_noprof include/linux/slab.h:954 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 fib6_info_alloc+0x30/0xf0 net/ipv6/ip6_fib.c:155
 ip6_route_info_create+0x142/0x860 net/ipv6/route.c:3818
 ip6_route_add+0x49/0x1b0 net/ipv6/route.c:3947
 addrconf_prefix_route+0x3a2/0x480 net/ipv6/addrconf.c:2487
 add_v4_addrs+0x5a6/0x7e0 net/ipv6/addrconf.c:3264
 addrconf_sit_config net/ipv6/addrconf.c:3520 [inline]
 addrconf_init_auto_addrs+0x248/0xa50 net/ipv6/addrconf.c:3558
 addrconf_notify+0xb1e/0x1050 net/ipv6/addrconf.c:3744
 notifier_call_chain+0x1be/0x400 kernel/notifier.c:85
 call_netdevice_notifiers_extack net/core/dev.c:2287 [inline]
 call_netdevice_notifiers net/core/dev.c:2301 [inline]
 __dev_notify_flags+0x1a9/0x310 net/core/dev.c:9820
 netif_change_flags+0xe8/0x1a0 net/core/dev.c:9849

Memory state around the buggy address:
 ffff888058ae0180: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff888058ae0200: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff888058ae0280: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                      ^
 ffff888058ae0300: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff888058ae0380: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================


Tested on:

commit:         d60bc140 Merge tag 'pwrseq-updates-for-v7.1-rc1' of gi..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=13d304ce580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=85bd86f990609a1
dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
patch:          https://syzkaller.appspot.com/x/patch.diff?x=11597b02580000

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[Hillf Danton]

> Date: Mon, 13 Apr 2026 10:16:36 -0700	[thread overview]
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    f5459048c38a Merge tag 'i2c-for-7.0-final' of git://git.ke..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1028d106580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=6754c86e8d9e4c91
> dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
> compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1428d106580000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17509036580000

#syz test

--- x/drivers/gpu/drm/drm_gem.c
+++ y/drivers/gpu/drm/drm_gem.c
@@ -404,7 +404,8 @@ drm_gem_handle_delete(struct drm_file *f
 	spin_lock(&filp->table_lock);
 
 	/* Check if we currently have a reference on the object */
-	obj = idr_replace(&filp->object_idr, NULL, handle);
+	obj = idr_find(&filp->object_idr, handle);
+	idr_remove(&filp->object_idr, handle);
 	spin_unlock(&filp->table_lock);
 	if (IS_ERR_OR_NULL(obj))
 		return -EINVAL;
@@ -412,11 +413,6 @@ drm_gem_handle_delete(struct drm_file *f
 	/* Release driver's reference and decrement refcount. */
 	drm_gem_object_release_handle(handle, obj, filp);
 
-	/* And finally make the handle available for future allocations. */
-	spin_lock(&filp->table_lock);
-	idr_remove(&filp->object_idr, handle);
-	spin_unlock(&filp->table_lock);
-
 	return 0;
 }
 EXPORT_SYMBOL(drm_gem_handle_delete);
@@ -1085,8 +1081,20 @@ drm_gem_open(struct drm_device *dev, str
 void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
-	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+	struct drm_gem_object *obj;
+	int id;
+
+	for (;;) {
+		id = 0;
+		spin_lock(&file_private->table_lock);
+		obj = idr_get_next(&file_private->object_idr, &id);
+		if (obj)
+			idr_remove(&file_private->object_idr, id);
+		spin_unlock(&file_private->table_lock);
+		if (!obj)
+			break;
+		drm_gem_object_release_handle(id, obj, file_private);
+	}
 	idr_destroy(&file_private->object_idr);
 }
 
--

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
KASAN: slab-use-after-free Read in drm_gem_object_release_handle

==================================================================
BUG: KASAN: slab-use-after-free in drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
Read of size 8 at addr ffff88802868ba78 by task syz.4.82/6783

CPU: 0 UID: 0 PID: 6783 Comm: syz.4.82 Not tainted syzkaller #0 PREEMPT(full) 
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/18/2026
Call Trace:
 <TASK>
 dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120
 print_address_description mm/kasan/report.c:378 [inline]
 print_report+0xba/0x230 mm/kasan/report.c:482
 kasan_report+0x117/0x150 mm/kasan/report.c:595
 drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
 drm_gem_release+0xd2/0x170 drivers/gpu/drm/drm_gem.c:1096
 drm_file_free+0x729/0xa00 drivers/gpu/drm/drm_file.c:261
 drm_close_helper drivers/gpu/drm/drm_file.c:290 [inline]
 drm_release+0x2de/0x3f0 drivers/gpu/drm/drm_file.c:438
 __fput+0x44f/0xa60 fs/file_table.c:510
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
 __exit_to_user_mode_loop kernel/entry/common.c:67 [inline]
 exit_to_user_mode_loop+0xed/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:207 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:238 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:328 [inline]
 do_syscall_64+0x33e/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f7a5259c819
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fff2b485db8 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4
RAX: 0000000000000000 RBX: 00007f7a52817da0 RCX: 00007f7a5259c819
RDX: 0000000000000000 RSI: 000000000000001e RDI: 0000000000000003
RBP: 00007f7a52817da0 R08: 00007f7a52816038 R09: 0000000000000000
R10: 000000000003fdc0 R11: 0000000000000246 R12: 000000000002456b
R13: 00007f7a5281609c R14: 000000000002426f R15: 00007f7a52816090
 </TASK>

Allocated by task 6785:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 poison_kmalloc_redzone mm/kasan/common.c:398 [inline]
 __kasan_kmalloc+0x93/0xb0 mm/kasan/common.c:415
 kasan_kmalloc include/linux/kasan.h:263 [inline]
 __kmalloc_cache_noprof+0x31c/0x660 mm/slub.c:5380
 kmalloc_noprof include/linux/slab.h:950 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 __drm_gem_shmem_create+0xc4/0x2e0 drivers/gpu/drm/drm_gem_shmem_helper.c:130
 drm_gem_shmem_create drivers/gpu/drm/drm_gem_shmem_helper.c:157 [inline]
 drm_gem_shmem_create_with_handle drivers/gpu/drm/drm_gem_shmem_helper.c:460 [inline]
 drm_gem_shmem_dumb_create+0x72/0x120 drivers/gpu/drm/drm_gem_shmem_helper.c:549
 drm_mode_create_dumb drivers/gpu/drm/drm_dumb_buffers.c:227 [inline]
 drm_mode_create_dumb_ioctl+0x2bd/0x340 drivers/gpu/drm/drm_dumb_buffers.c:236
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 6785:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 kasan_save_free_info+0x46/0x50 mm/kasan/generic.c:584
 poison_slab_object mm/kasan/common.c:253 [inline]
 __kasan_slab_free+0x5c/0x80 mm/kasan/common.c:285
 kasan_slab_free include/linux/kasan.h:235 [inline]
 slab_free_hook mm/slub.c:2685 [inline]
 slab_free mm/slub.c:6165 [inline]
 kfree+0x1c1/0x620 mm/slub.c:6483
 drm_gem_object_free drivers/gpu/drm/drm_gem.c:1138 [inline]
 kref_put include/linux/kref.h:65 [inline]
 __drm_gem_object_put include/drm/drm_gem.h:565 [inline]
 drm_gem_object_put include/drm/drm_gem.h:578 [inline]
 drm_gem_change_handle_ioctl+0x392/0x4d0 drivers/gpu/drm/drm_gem.c:1052
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

The buggy address belongs to the object at ffff88802868b800
 which belongs to the cache kmalloc-1k of size 1024
The buggy address is located 632 bytes inside of
 freed 1024-byte region [ffff88802868b800, ffff88802868bc00)

The buggy address belongs to the physical page:
page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x28688
head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
flags: 0xfff00000000040(head|node=0|zone=1|lastcpupid=0x7ff)
page_type: f5(slab)
raw: 00fff00000000040 ffff88813fe2cdc0 dead000000000100 dead000000000122
raw: 0000000000000000 0000000800100010 00000000f5000000 0000000000000000
head: 00fff00000000040 ffff88813fe2cdc0 dead000000000100 dead000000000122
head: 0000000000000000 0000000800100010 00000000f5000000 0000000000000000
head: 00fff00000000003 ffffea0000a1a201 00000000ffffffff 00000000ffffffff
head: ffffffffffffffff 0000000000000000 00000000ffffffff 0000000000000008
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd2000(__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 69, tgid 69 (kworker/u8:4), ts 7935238695, free_ts 0
 set_page_owner include/linux/page_owner.h:32 [inline]
 post_alloc_hook+0x231/0x280 mm/page_alloc.c:1889
 prep_new_page mm/page_alloc.c:1897 [inline]
 get_page_from_freelist+0x24dc/0x2580 mm/page_alloc.c:3962
 __alloc_frozen_pages_noprof+0x18d/0x380 mm/page_alloc.c:5250
 alloc_slab_page mm/slub.c:3292 [inline]
 allocate_slab+0x77/0x660 mm/slub.c:3481
 new_slab mm/slub.c:3539 [inline]
 refill_objects+0x331/0x3c0 mm/slub.c:7175
 refill_sheaf mm/slub.c:2812 [inline]
 __pcs_replace_empty_main+0x2e6/0x730 mm/slub.c:4615
 alloc_from_pcs mm/slub.c:4717 [inline]
 slab_alloc_node mm/slub.c:4851 [inline]
 __do_kmalloc_node mm/slub.c:5259 [inline]
 __kmalloc_node_noprof+0x577/0x7c0 mm/slub.c:5266
 kmalloc_node_noprof include/linux/slab.h:1081 [inline]
 blk_alloc_flush_queue+0xe2/0x230 block/blk-flush.c:492
 blk_mq_init_hctx block/blk-mq.c:4019 [inline]
 blk_mq_alloc_and_init_hctx+0x79e/0xc50 block/blk-mq.c:4558
 __blk_mq_realloc_hw_ctxs+0x2bd/0x670 block/blk-mq.c:4605
 blk_mq_realloc_hw_ctxs block/blk-mq.c:4642 [inline]
 blk_mq_init_allocated_queue+0x365/0x13e0 block/blk-mq.c:4672
 blk_mq_alloc_queue+0x1ba/0x2e0 block/blk-mq.c:4454
 scsi_alloc_sdev+0x7c0/0xc80 drivers/scsi/scsi_scan.c:339
 scsi_probe_and_add_lun+0x200/0x4830 drivers/scsi/scsi_scan.c:1215
 __scsi_scan_target+0x1f0/0xe10 drivers/scsi/scsi_scan.c:1786
 scsi_scan_channel drivers/scsi/scsi_scan.c:1874 [inline]
 scsi_scan_host_selected+0x372/0x690 drivers/scsi/scsi_scan.c:1903
page_owner free stack trace missing

Memory state around the buggy address:
 ffff88802868b900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88802868b980: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff88802868ba00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                                                                ^
 ffff88802868ba80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88802868bb00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================


Tested on:

commit:         91a4855d Merge tag 'net-next-7.1' of git://git.kernel...
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=1079fb02580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=95729ed00549063a
dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
patch:          https://syzkaller.appspot.com/x/patch.diff?x=112238ce580000

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[Hillf Danton]

> Date: Mon, 13 Apr 2026 10:16:36 -0700	[thread overview]
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    f5459048c38a Merge tag 'i2c-for-7.0-final' of git://git.ke..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1028d106580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=6754c86e8d9e4c91
> dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
> compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1428d106580000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17509036580000

#syz test

--- x/drivers/gpu/drm/drm_gem.c
+++ y/drivers/gpu/drm/drm_gem.c
@@ -404,7 +404,8 @@ drm_gem_handle_delete(struct drm_file *f
 	spin_lock(&filp->table_lock);
 
 	/* Check if we currently have a reference on the object */
-	obj = idr_replace(&filp->object_idr, NULL, handle);
+	obj = idr_find(&filp->object_idr, handle);
+	idr_remove(&filp->object_idr, handle);
 	spin_unlock(&filp->table_lock);
 	if (IS_ERR_OR_NULL(obj))
 		return -EINVAL;
@@ -412,11 +413,6 @@ drm_gem_handle_delete(struct drm_file *f
 	/* Release driver's reference and decrement refcount. */
 	drm_gem_object_release_handle(handle, obj, filp);
 
-	/* And finally make the handle available for future allocations. */
-	spin_lock(&filp->table_lock);
-	idr_remove(&filp->object_idr, handle);
-	spin_unlock(&filp->table_lock);
-
 	return 0;
 }
 EXPORT_SYMBOL(drm_gem_handle_delete);
@@ -1085,9 +1081,23 @@ drm_gem_open(struct drm_device *dev, str
 void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
-	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+	struct drm_gem_object *obj;
+	int id;
+
+	mutex_lock(&file_private->prime.lock);
+	for (;;) {
+		id = 0;
+		spin_lock(&file_private->table_lock);
+		obj = idr_get_next(&file_private->object_idr, &id);
+		if (obj)
+			idr_remove(&file_private->object_idr, id);
+		spin_unlock(&file_private->table_lock);
+		if (!obj)
+			break;
+		drm_gem_object_release_handle(id, obj, file_private);
+	}
 	idr_destroy(&file_private->object_idr);
+	mutex_unlock(&file_private->prime.lock);
 }
 
 /**
--

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
possible deadlock in drm_gem_object_release_handle

============================================
WARNING: possible recursive locking detected
syzkaller #0 Not tainted
--------------------------------------------
syz.2.19/6521 is trying to acquire lock:
ffff88801d349388 (&prime_fpriv->lock){+.+.}-{4:4}, at: drm_gem_object_release_handle+0x98/0x1e0 drivers/gpu/drm/drm_gem.c:377

but task is already holding lock:
ffff88801d349388 (&prime_fpriv->lock){+.+.}-{4:4}, at: drm_gem_release+0x90/0x190 drivers/gpu/drm/drm_gem.c:1087

other info that might help us debug this:
 Possible unsafe locking scenario:

       CPU0
       ----
  lock(&prime_fpriv->lock);
  lock(&prime_fpriv->lock);

 *** DEADLOCK ***

 May be due to missing lock nesting notation

1 lock held by syz.2.19/6521:
 #0: ffff88801d349388 (&prime_fpriv->lock){+.+.}-{4:4}, at: drm_gem_release+0x90/0x190 drivers/gpu/drm/drm_gem.c:1087

stack backtrace:
CPU: 1 UID: 0 PID: 6521 Comm: syz.2.19 Not tainted syzkaller #0 PREEMPT(full) 
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/18/2026
Call Trace:
 <TASK>
 dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120
 print_deadlock_bug+0x279/0x290 kernel/locking/lockdep.c:3041
 check_deadlock kernel/locking/lockdep.c:3093 [inline]
 validate_chain kernel/locking/lockdep.c:3895 [inline]
 __lock_acquire+0x253f/0x2cf0 kernel/locking/lockdep.c:5237
 lock_acquire+0x106/0x350 kernel/locking/lockdep.c:5868
 __mutex_lock_common kernel/locking/mutex.c:632 [inline]
 __mutex_lock+0x1a3/0x1550 kernel/locking/mutex.c:806
 drm_gem_object_release_handle+0x98/0x1e0 drivers/gpu/drm/drm_gem.c:377
 drm_gem_release+0xe5/0x190 drivers/gpu/drm/drm_gem.c:1097
 drm_file_free+0x729/0xa00 drivers/gpu/drm/drm_file.c:261
 drm_close_helper drivers/gpu/drm/drm_file.c:290 [inline]
 drm_release+0x2de/0x3f0 drivers/gpu/drm/drm_file.c:438
 __fput+0x44f/0xa60 fs/file_table.c:510
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
 __exit_to_user_mode_loop kernel/entry/common.c:67 [inline]
 exit_to_user_mode_loop+0xed/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:207 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:238 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:328 [inline]
 do_syscall_64+0x33e/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f141259c819
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffcebeea6c8 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4
RAX: 0000000000000000 RBX: 00007ffcebeea7b0 RCX: 00007f141259c819
RDX: 0000000000000000 RSI: 000000000000001e RDI: 0000000000000003
RBP: 000000000001fd63 R08: 0000000000000001 R09: 0000000000000000
R10: 0000001b31120000 R11: 0000000000000246 R12: 00007ffcebeea7f0
R13: 00007f141281609c R14: 000000000001fdc6 R15: 00007f1412816090
 </TASK>


Tested on:

commit:         1f5ffc67 Fix mismerge of the arm64 / timer-core interr..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=145ebb16580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=95729ed00549063a
dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
patch:          https://syzkaller.appspot.com/x/patch.diff?x=150dfcd2580000

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[Hillf Danton]

> Date: Mon, 13 Apr 2026 10:16:36 -0700	[thread overview]
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    f5459048c38a Merge tag 'i2c-for-7.0-final' of git://git.ke..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1028d106580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=6754c86e8d9e4c91
> dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
> compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1428d106580000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17509036580000

#syz test

--- x/drivers/gpu/drm/drm_gem.c
+++ y/drivers/gpu/drm/drm_gem.c
@@ -404,7 +404,8 @@ drm_gem_handle_delete(struct drm_file *f
 	spin_lock(&filp->table_lock);
 
 	/* Check if we currently have a reference on the object */
-	obj = idr_replace(&filp->object_idr, NULL, handle);
+	obj = idr_find(&filp->object_idr, handle);
+	idr_remove(&filp->object_idr, handle);
 	spin_unlock(&filp->table_lock);
 	if (IS_ERR_OR_NULL(obj))
 		return -EINVAL;
@@ -412,11 +413,6 @@ drm_gem_handle_delete(struct drm_file *f
 	/* Release driver's reference and decrement refcount. */
 	drm_gem_object_release_handle(handle, obj, filp);
 
-	/* And finally make the handle available for future allocations. */
-	spin_lock(&filp->table_lock);
-	idr_remove(&filp->object_idr, handle);
-	spin_unlock(&filp->table_lock);
-
 	return 0;
 }
 EXPORT_SYMBOL(drm_gem_handle_delete);
@@ -1085,8 +1081,22 @@ drm_gem_open(struct drm_device *dev, str
 void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
-	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+	struct drm_gem_object *obj;
+	int id;
+
+	for (;;) {
+		id = 0;
+		mutex_lock(&file_private->prime.lock);
+		spin_lock(&file_private->table_lock);
+		obj = idr_get_next(&file_private->object_idr, &id);
+		if (obj)
+			idr_remove(&file_private->object_idr, id);
+		spin_unlock(&file_private->table_lock);
+		mutex_unlock(&file_private->prime.lock);
+		if (!obj)
+			break;
+		drm_gem_object_release_handle(id, obj, file_private);
+	}
 	idr_destroy(&file_private->object_idr);
 }
 
--

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
KASAN: slab-use-after-free Read in drm_gem_object_release_handle

==================================================================
BUG: KASAN: slab-use-after-free in drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
Read of size 8 at addr ffff88802afb9278 by task syz.1.182/7005

CPU: 1 UID: 0 PID: 7005 Comm: syz.1.182 Not tainted syzkaller #0 PREEMPT(full) 
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/18/2026
Call Trace:
 <TASK>
 dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120
 print_address_description mm/kasan/report.c:378 [inline]
 print_report+0xba/0x230 mm/kasan/report.c:482
 kasan_report+0x117/0x150 mm/kasan/report.c:595
 drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
 drm_gem_release+0xe8/0x190 drivers/gpu/drm/drm_gem.c:1098
 drm_file_free+0x729/0xa00 drivers/gpu/drm/drm_file.c:261
 drm_close_helper drivers/gpu/drm/drm_file.c:290 [inline]
 drm_release+0x2de/0x3f0 drivers/gpu/drm/drm_file.c:438
 __fput+0x44f/0xa60 fs/file_table.c:510
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
 __exit_to_user_mode_loop kernel/entry/common.c:67 [inline]
 exit_to_user_mode_loop+0xed/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:207 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:238 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:328 [inline]
 do_syscall_64+0x33e/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7faf5959c819
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffdb67de818 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4
RAX: 0000000000000000 RBX: 00007ffdb67de900 RCX: 00007faf5959c819
RDX: 0000000000000000 RSI: 000000000000001e RDI: 0000000000000003
RBP: 000000000001ff8e R08: 0000000000000001 R09: 0000000000000000
R10: 0000001b2c220000 R11: 0000000000000246 R12: 00007ffdb67de940
R13: 00007faf5981609c R14: 000000000001ffcf R15: 00007faf59816090
 </TASK>

Allocated by task 7007:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 poison_kmalloc_redzone mm/kasan/common.c:398 [inline]
 __kasan_kmalloc+0x93/0xb0 mm/kasan/common.c:415
 kasan_kmalloc include/linux/kasan.h:263 [inline]
 __kmalloc_cache_noprof+0x31c/0x660 mm/slub.c:5380
 kmalloc_noprof include/linux/slab.h:950 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 __drm_gem_shmem_create+0xc4/0x2e0 drivers/gpu/drm/drm_gem_shmem_helper.c:130
 drm_gem_shmem_create drivers/gpu/drm/drm_gem_shmem_helper.c:157 [inline]
 drm_gem_shmem_create_with_handle drivers/gpu/drm/drm_gem_shmem_helper.c:460 [inline]
 drm_gem_shmem_dumb_create+0x72/0x120 drivers/gpu/drm/drm_gem_shmem_helper.c:549
 drm_mode_create_dumb drivers/gpu/drm/drm_dumb_buffers.c:227 [inline]
 drm_mode_create_dumb_ioctl+0x2bd/0x340 drivers/gpu/drm/drm_dumb_buffers.c:236
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 7008:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 kasan_save_free_info+0x46/0x50 mm/kasan/generic.c:584
 poison_slab_object mm/kasan/common.c:253 [inline]
 __kasan_slab_free+0x5c/0x80 mm/kasan/common.c:285
 kasan_slab_free include/linux/kasan.h:235 [inline]
 slab_free_hook mm/slub.c:2685 [inline]
 slab_free mm/slub.c:6165 [inline]
 kfree+0x1c1/0x620 mm/slub.c:6483
 drm_gem_handle_delete+0x84/0xa0 drivers/gpu/drm/drm_gem.c:414
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

The buggy address belongs to the object at ffff88802afb9000
 which belongs to the cache kmalloc-1k of size 1024
The buggy address is located 632 bytes inside of
 freed 1024-byte region [ffff88802afb9000, ffff88802afb9400)

The buggy address belongs to the physical page:
page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x2afb8
head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
flags: 0xfff00000000040(head|node=0|zone=1|lastcpupid=0x7ff)
page_type: f5(slab)
raw: 00fff00000000040 ffff88813fe2cdc0 dead000000000100 dead000000000122
raw: 0000000000000000 0000000800100010 00000000f5000000 0000000000000000
head: 00fff00000000040 ffff88813fe2cdc0 dead000000000100 dead000000000122
head: 0000000000000000 0000000800100010 00000000f5000000 0000000000000000
head: 00fff00000000003 ffffea0000abee01 00000000ffffffff 00000000ffffffff
head: ffffffffffffffff 0000000000000000 00000000ffffffff 0000000000000008
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd2820(GFP_ATOMIC|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 12, tgid 12 (kworker/u8:0), ts 131001296136, free_ts 130963840270
 set_page_owner include/linux/page_owner.h:32 [inline]
 post_alloc_hook+0x231/0x280 mm/page_alloc.c:1889
 prep_new_page mm/page_alloc.c:1897 [inline]
 get_page_from_freelist+0x24dc/0x2580 mm/page_alloc.c:3962
 __alloc_frozen_pages_noprof+0x18d/0x380 mm/page_alloc.c:5250
 alloc_slab_page mm/slub.c:3292 [inline]
 allocate_slab+0x77/0x660 mm/slub.c:3481
 new_slab mm/slub.c:3539 [inline]
 refill_objects+0x331/0x3c0 mm/slub.c:7175
 refill_sheaf mm/slub.c:2812 [inline]
 __pcs_replace_empty_main+0x2e6/0x730 mm/slub.c:4615
 alloc_from_pcs mm/slub.c:4717 [inline]
 slab_alloc_node mm/slub.c:4851 [inline]
 __do_kmalloc_node mm/slub.c:5259 [inline]
 __kmalloc_noprof+0x474/0x760 mm/slub.c:5272
 kmalloc_noprof include/linux/slab.h:954 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 ieee802_11_parse_elems_full+0x159/0x2ab0 net/mac80211/parse.c:1051
 ieee802_11_parse_elems net/mac80211/ieee80211_i.h:2518 [inline]
 ieee80211_inform_bss+0x161/0x1160 net/mac80211/scan.c:79
 rdev_inform_bss net/wireless/rdev-ops.h:418 [inline]
 cfg80211_inform_single_bss_data+0xcf9/0x1af0 net/wireless/scan.c:2372
 cfg80211_inform_bss_data+0x266/0x3c40 net/wireless/scan.c:3226
 cfg80211_inform_bss_frame_data+0x3c7/0x730 net/wireless/scan.c:3317
 ieee80211_bss_info_update+0x794/0xa40 net/mac80211/scan.c:230
 ieee80211_rx_bss_info net/mac80211/ibss.c:1088 [inline]
 ieee80211_rx_mgmt_probe_beacon net/mac80211/ibss.c:1569 [inline]
 ieee80211_ibss_rx_queued_mgmt+0x1901/0x2cb0 net/mac80211/ibss.c:1596
 ieee80211_iface_process_skb net/mac80211/iface.c:1795 [inline]
 ieee80211_iface_work+0x845/0x1380 net/mac80211/iface.c:1849
 cfg80211_wiphy_work+0x2cf/0x460 net/wireless/core.c:513
page last free pid 6998 tgid 6998 stack trace:
 reset_page_owner include/linux/page_owner.h:25 [inline]
 __free_pages_prepare mm/page_alloc.c:1433 [inline]
 __free_frozen_pages+0xc2b/0xdb0 mm/page_alloc.c:2978
 __slab_free+0x263/0x2b0 mm/slub.c:5573
 qlink_free mm/kasan/quarantine.c:163 [inline]
 qlist_free_all+0x97/0x100 mm/kasan/quarantine.c:179
 kasan_quarantine_reduce+0x148/0x160 mm/kasan/quarantine.c:286
 __kasan_slab_alloc+0x22/0x80 mm/kasan/common.c:350
 kasan_slab_alloc include/linux/kasan.h:253 [inline]
 slab_post_alloc_hook mm/slub.c:4538 [inline]
 slab_alloc_node mm/slub.c:4866 [inline]
 __do_kmalloc_node mm/slub.c:5259 [inline]
 __kmalloc_noprof+0x316/0x760 mm/slub.c:5272
 kmalloc_noprof include/linux/slab.h:954 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 tomoyo_encode2 security/tomoyo/realpath.c:45 [inline]
 tomoyo_encode+0x28b/0x550 security/tomoyo/realpath.c:80
 tomoyo_path_perm+0x318/0x560 security/tomoyo/file.c:836
 tomoyo_path_symlink+0xab/0xf0 security/tomoyo/tomoyo.c:212
 security_path_symlink+0x16f/0x360 security/security.c:1502
 filename_symlinkat+0x134/0x410 fs/namei.c:5666
 __do_sys_symlinkat fs/namei.c:5688 [inline]
 __se_sys_symlinkat+0x4e/0x2b0 fs/namei.c:5683
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Memory state around the buggy address:
 ffff88802afb9100: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88802afb9180: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff88802afb9200: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                                                                ^
 ffff88802afb9280: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88802afb9300: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================


Tested on:

commit:         1f5ffc67 Fix mismerge of the arm64 / timer-core interr..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=13a4e906580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=95729ed00549063a
dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
patch:          https://syzkaller.appspot.com/x/patch.diff?x=143538ce580000

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[Hillf Danton]

> Date: Mon, 13 Apr 2026 10:16:36 -0700	[thread overview]
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    f5459048c38a Merge tag 'i2c-for-7.0-final' of git://git.ke..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1028d106580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=6754c86e8d9e4c91
> dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
> compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1428d106580000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17509036580000

#syz test

--- x/drivers/gpu/drm/drm_gem.c
+++ y/drivers/gpu/drm/drm_gem.c
@@ -401,22 +401,20 @@ drm_gem_handle_delete(struct drm_file *f
 {
 	struct drm_gem_object *obj;
 
+	mutex_lock(&filp->prime.lock);
 	spin_lock(&filp->table_lock);
 
 	/* Check if we currently have a reference on the object */
-	obj = idr_replace(&filp->object_idr, NULL, handle);
+	obj = idr_find(&filp->object_idr, handle);
+	idr_remove(&filp->object_idr, handle);
 	spin_unlock(&filp->table_lock);
+	mutex_unlock(&filp->prime.lock);
+
 	if (IS_ERR_OR_NULL(obj))
 		return -EINVAL;
-
 	/* Release driver's reference and decrement refcount. */
 	drm_gem_object_release_handle(handle, obj, filp);
 
-	/* And finally make the handle available for future allocations. */
-	spin_lock(&filp->table_lock);
-	idr_remove(&filp->object_idr, handle);
-	spin_unlock(&filp->table_lock);
-
 	return 0;
 }
 EXPORT_SYMBOL(drm_gem_handle_delete);
@@ -1085,8 +1083,22 @@ drm_gem_open(struct drm_device *dev, str
 void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
-	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+	struct drm_gem_object *obj;
+	int id;
+
+	for (;;) {
+		id = 0;
+		mutex_lock(&file_private->prime.lock);
+		spin_lock(&file_private->table_lock);
+		obj = idr_get_next(&file_private->object_idr, &id);
+		if (obj)
+			idr_remove(&file_private->object_idr, id);
+		spin_unlock(&file_private->table_lock);
+		mutex_unlock(&file_private->prime.lock);
+		if (!obj)
+			break;
+		drm_gem_object_release_handle(id, obj, file_private);
+	}
 	idr_destroy(&file_private->object_idr);
 }
 
--

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[Edward Adam Davis]

#syz test

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 891c3bff5ae0..63a8d7e980b5 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -374,14 +374,8 @@ drm_gem_object_release_handle(int id, void *ptr, void *data)
 	if (obj->funcs->close)
 		obj->funcs->close(obj, file_priv);
 
-	mutex_lock(&file_priv->prime.lock);
-
 	drm_prime_remove_buf_handle(&file_priv->prime, id);
-
-	mutex_unlock(&file_priv->prime.lock);
-
 	drm_vma_node_revoke(&obj->vma_node, file_priv);
-
 	drm_gem_object_handle_put_unlocked(obj);
 
 	return 0;
@@ -401,13 +395,16 @@ drm_gem_handle_delete(struct drm_file *filp, u32 handle)
 {
 	struct drm_gem_object *obj;
 
+	mutex_lock(&filp->prime.lock);
 	spin_lock(&filp->table_lock);
 
 	/* Check if we currently have a reference on the object */
 	obj = idr_replace(&filp->object_idr, NULL, handle);
 	spin_unlock(&filp->table_lock);
-	if (IS_ERR_OR_NULL(obj))
+	if (IS_ERR_OR_NULL(obj)) {
+		mutex_unlock(&filp->prime.lock);
 		return -EINVAL;
+	}
 
 	/* Release driver's reference and decrement refcount. */
 	drm_gem_object_release_handle(handle, obj, filp);
@@ -416,6 +413,7 @@ drm_gem_handle_delete(struct drm_file *filp, u32 handle)
 	spin_lock(&filp->table_lock);
 	idr_remove(&filp->object_idr, handle);
 	spin_unlock(&filp->table_lock);
+	mutex_unlock(&filp->prime.lock);
 
 	return 0;
 }
@@ -1012,17 +1010,18 @@ int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data,
 		return -EINVAL;
 	handle = args->new_handle;
 
+	mutex_lock(&file_priv->prime.lock);
 	obj = drm_gem_object_lookup(file_priv, args->handle);
-	if (!obj)
+	if (!obj) {
+		mutex_unlock(&file_priv->prime.lock);
 		return -ENOENT;
+	}
 
 	if (args->handle == handle) {
 		ret = 0;
-		goto out;
+		goto out_unlock;
 	}
 
-	mutex_lock(&file_priv->prime.lock);
-
 	spin_lock(&file_priv->table_lock);
 	ret = idr_alloc(&file_priv->object_idr, obj, handle, handle + 1,
 			GFP_NOWAIT);
@@ -1051,9 +1050,8 @@ int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data,
 	spin_unlock(&file_priv->table_lock);
 
 out_unlock:
-	mutex_unlock(&file_priv->prime.lock);
-out:
 	drm_gem_object_put(obj);
+	mutex_unlock(&file_priv->prime.lock);
 
 	return ret;
 }
@@ -1085,8 +1083,10 @@ drm_gem_open(struct drm_device *dev, struct drm_file *file_private)
 void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
+	mutex_lock(&file_private->prime.lock);
 	idr_for_each(&file_private->object_idr,
 		     &drm_gem_object_release_handle, file_private);
+	mutex_unlock(&file_private->prime.lock);
 	idr_destroy(&file_private->object_idr);
 }
 

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
KASAN: slab-use-after-free Read in drm_gem_object_release_handle

==================================================================
BUG: KASAN: slab-use-after-free in drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
Read of size 8 at addr ffff888035e4c278 by task syz.1.169/7028

CPU: 1 UID: 0 PID: 7028 Comm: syz.1.169 Not tainted syzkaller #0 PREEMPT(full) 
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/18/2026
Call Trace:
 <TASK>
 dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120
 print_address_description+0x55/0x1e0 mm/kasan/report.c:378
 print_report+0x58/0x70 mm/kasan/report.c:482
 kasan_report+0x117/0x150 mm/kasan/report.c:595
 drm_gem_object_release_handle+0x47/0x1e0 drivers/gpu/drm/drm_gem.c:374
 drm_gem_release+0xe8/0x190 drivers/gpu/drm/drm_gem.c:1118
 drm_file_free+0x729/0xa00 drivers/gpu/drm/drm_file.c:261
 drm_close_helper drivers/gpu/drm/drm_file.c:290 [inline]
 drm_release+0x2de/0x3f0 drivers/gpu/drm/drm_file.c:438
 __fput+0x44f/0xa60 fs/file_table.c:510
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
 __exit_to_user_mode_loop kernel/entry/common.c:67 [inline]
 exit_to_user_mode_loop+0xed/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:207 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:238 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:328 [inline]
 do_syscall_64+0x33e/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f4a7e79c819
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffd28753448 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4
RAX: 0000000000000000 RBX: 00007ffd28753530 RCX: 00007f4a7e79c819
RDX: 0000000000000000 RSI: 000000000000001e RDI: 0000000000000003
RBP: 000000000001f80f R08: 0000000000000001 R09: 0000000000000000
R10: 0000001b30a20000 R11: 0000000000000246 R12: 00007ffd28753570
R13: 00007f4a7ea1609c R14: 000000000001f863 R15: 00007f4a7ea16090
 </TASK>

Allocated by task 7032:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 poison_kmalloc_redzone mm/kasan/common.c:398 [inline]
 __kasan_kmalloc+0x93/0xb0 mm/kasan/common.c:415
 kasan_kmalloc include/linux/kasan.h:263 [inline]
 __kmalloc_cache_noprof+0x31c/0x660 mm/slub.c:5415
 kmalloc_noprof include/linux/slab.h:950 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 __drm_gem_shmem_create+0xc4/0x2e0 drivers/gpu/drm/drm_gem_shmem_helper.c:130
 drm_gem_shmem_create drivers/gpu/drm/drm_gem_shmem_helper.c:157 [inline]
 drm_gem_shmem_create_with_handle drivers/gpu/drm/drm_gem_shmem_helper.c:464 [inline]
 drm_gem_shmem_dumb_create+0x72/0x120 drivers/gpu/drm/drm_gem_shmem_helper.c:553
 drm_mode_create_dumb drivers/gpu/drm/drm_dumb_buffers.c:227 [inline]
 drm_mode_create_dumb_ioctl+0x2bd/0x340 drivers/gpu/drm/drm_dumb_buffers.c:236
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 7033:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 kasan_save_free_info+0x46/0x50 mm/kasan/generic.c:584
 poison_slab_object mm/kasan/common.c:253 [inline]
 __kasan_slab_free+0x5c/0x80 mm/kasan/common.c:285
 kasan_slab_free include/linux/kasan.h:235 [inline]
 slab_free_hook mm/slub.c:2689 [inline]
 slab_free mm/slub.c:6246 [inline]
 kfree+0x1c5/0x640 mm/slub.c:6561
 drm_gem_handle_delete+0xa3/0xc0 drivers/gpu/drm/drm_gem.c:416
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804
 drm_ioctl+0x6ba/0xb80 drivers/gpu/drm/drm_ioctl.c:901
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:597 [inline]
 __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:583
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

The buggy address belongs to the object at ffff888035e4c000
 which belongs to the cache kmalloc-1k of size 1024
The buggy address is located 632 bytes inside of
 freed 1024-byte region [ffff888035e4c000, ffff888035e4c400)

The buggy address belongs to the physical page:
page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x35e48
head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
flags: 0xfff00000000040(head|node=0|zone=1|lastcpupid=0x7ff)
page_type: f5(slab)
raw: 00fff00000000040 ffff88813fe29dc0 dead000000000100 dead000000000122
raw: 0000000000000000 0000000800100010 00000000f5000000 0000000000000000
head: 00fff00000000040 ffff88813fe29dc0 dead000000000100 dead000000000122
head: 0000000000000000 0000000800100010 00000000f5000000 0000000000000000
head: 00fff00000000003 fffffffffffffe01 00000000ffffffff 00000000ffffffff
head: ffffffffffffffff 0000000000000000 00000000ffffffff 0000000000000008
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd2820(GFP_ATOMIC|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 129, tgid 129 (kworker/u8:6), ts 129037001160, free_ts 129034880394
 set_page_owner include/linux/page_owner.h:32 [inline]
 post_alloc_hook+0x231/0x280 mm/page_alloc.c:1860
 prep_new_page mm/page_alloc.c:1868 [inline]
 get_page_from_freelist+0x24ba/0x2540 mm/page_alloc.c:3948
 __alloc_frozen_pages_noprof+0x18d/0x380 mm/page_alloc.c:5228
 alloc_slab_page mm/slub.c:3278 [inline]
 allocate_slab+0x77/0x660 mm/slub.c:3467
 new_slab mm/slub.c:3525 [inline]
 refill_objects+0x339/0x3d0 mm/slub.c:7251
 refill_sheaf mm/slub.c:2816 [inline]
 __pcs_replace_empty_main+0x321/0x720 mm/slub.c:4651
 alloc_from_pcs mm/slub.c:4749 [inline]
 slab_alloc_node mm/slub.c:4883 [inline]
 __do_kmalloc_node mm/slub.c:5294 [inline]
 __kmalloc_noprof+0x474/0x760 mm/slub.c:5307
 kmalloc_noprof include/linux/slab.h:954 [inline]
 kzalloc_noprof include/linux/slab.h:1188 [inline]
 ieee802_11_parse_elems_full+0x159/0x2ab0 net/mac80211/parse.c:1051
 ieee802_11_parse_elems net/mac80211/ieee80211_i.h:2518 [inline]
 ieee80211_inform_bss+0x161/0x1160 net/mac80211/scan.c:79
 rdev_inform_bss net/wireless/rdev-ops.h:418 [inline]
 cfg80211_inform_single_bss_data+0xcf9/0x1af0 net/wireless/scan.c:2372
 cfg80211_inform_bss_data+0x266/0x3c40 net/wireless/scan.c:3226
 cfg80211_inform_bss_frame_data+0x3c7/0x730 net/wireless/scan.c:3317
 ieee80211_bss_info_update+0x794/0xa40 net/mac80211/scan.c:230
 ieee80211_rx_bss_info net/mac80211/ibss.c:1088 [inline]
 ieee80211_rx_mgmt_probe_beacon net/mac80211/ibss.c:1569 [inline]
 ieee80211_ibss_rx_queued_mgmt+0x1901/0x2cb0 net/mac80211/ibss.c:1596
 ieee80211_iface_process_skb net/mac80211/iface.c:1795 [inline]
 ieee80211_iface_work+0x845/0x1380 net/mac80211/iface.c:1849
 cfg80211_wiphy_work+0x2cf/0x460 net/wireless/core.c:513
page last free pid 7008 tgid 7008 stack trace:
 reset_page_owner include/linux/page_owner.h:25 [inline]
 __free_pages_prepare mm/page_alloc.c:1404 [inline]
 __free_frozen_pages+0xbc7/0xd30 mm/page_alloc.c:2945
 __slab_free+0x274/0x2c0 mm/slub.c:5608
 qlink_free mm/kasan/quarantine.c:163 [inline]
 qlist_free_all+0x99/0x100 mm/kasan/quarantine.c:179
 kasan_quarantine_reduce+0x148/0x160 mm/kasan/quarantine.c:286
 __kasan_slab_alloc+0x22/0x80 mm/kasan/common.c:350
 kasan_slab_alloc include/linux/kasan.h:253 [inline]
 slab_post_alloc_hook mm/slub.c:4569 [inline]
 slab_alloc_node mm/slub.c:4898 [inline]
 kmem_cache_alloc_noprof+0x2bc/0x650 mm/slub.c:4905
 alloc_filename fs/namei.c:142 [inline]
 do_getname+0x2e/0x250 fs/namei.c:182
 getname include/linux/fs.h:2533 [inline]
 getname_maybe_null include/linux/fs.h:2540 [inline]
 class_filename_maybe_null_constructor include/linux/fs.h:2564 [inline]
 vfs_fstatat+0x45/0x170 fs/stat.c:368
 __do_sys_newfstatat fs/stat.c:538 [inline]
 __se_sys_newfstatat fs/stat.c:532 [inline]
 __x64_sys_newfstatat+0x151/0x200 fs/stat.c:532
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Memory state around the buggy address:
 ffff888035e4c100: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff888035e4c180: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff888035e4c200: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                                                                ^
 ffff888035e4c280: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff888035e4c300: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================


Tested on:

commit:         1d51b370 Merge tag 'jfs-7.1' of github.com:kleikamp/li..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=125881ba580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=769ed9499e4e3d7d
dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
patch:          https://syzkaller.appspot.com/x/patch.diff?x=17f344ce580000

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot]

Hello,

syzbot has tested the proposed patch and the reproducer did not trigger any issue:

Reported-by: syzbot+b2e951687503f32f74ce@syzkaller.appspotmail.com
Tested-by: syzbot+b2e951687503f32f74ce@syzkaller.appspotmail.com

Tested on:

commit:         1d51b370 Merge tag 'jfs-7.1' of github.com:kleikamp/li..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=157881ba580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=769ed9499e4e3d7d
dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
patch:          https://syzkaller.appspot.com/x/patch.diff?x=16181a6a580000

Note: testing is done by a robot and is best-effort only.

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[Hillf Danton]

> Date: Mon, 13 Apr 2026 10:16:36 -0700	[thread overview]
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    f5459048c38a Merge tag 'i2c-for-7.0-final' of git://git.ke..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1028d106580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=6754c86e8d9e4c91
> dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
> compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1428d106580000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17509036580000

#syz test

--- x/drivers/gpu/drm/drm_gem.c
+++ y/drivers/gpu/drm/drm_gem.c
@@ -401,22 +401,20 @@ drm_gem_handle_delete(struct drm_file *f
 {
 	struct drm_gem_object *obj;
 
+	mutex_lock(&filp->prime.lock);
 	spin_lock(&filp->table_lock);
 
 	/* Check if we currently have a reference on the object */
-	obj = idr_replace(&filp->object_idr, NULL, handle);
+	obj = idr_find(&filp->object_idr, handle);
+	idr_remove(&filp->object_idr, handle);
 	spin_unlock(&filp->table_lock);
+	mutex_unlock(&filp->prime.lock);
+
 	if (IS_ERR_OR_NULL(obj))
 		return -EINVAL;
-
 	/* Release driver's reference and decrement refcount. */
 	drm_gem_object_release_handle(handle, obj, filp);
 
-	/* And finally make the handle available for future allocations. */
-	spin_lock(&filp->table_lock);
-	idr_remove(&filp->object_idr, handle);
-	spin_unlock(&filp->table_lock);
-
 	return 0;
 }
 EXPORT_SYMBOL(drm_gem_handle_delete);
@@ -1012,17 +1010,20 @@ int drm_gem_change_handle_ioctl(struct d
 		return -EINVAL;
 	handle = args->new_handle;
 
+	mutex_lock(&file_priv->prime.lock);
+
 	obj = drm_gem_object_lookup(file_priv, args->handle);
-	if (!obj)
+	if (!obj) {
+		mutex_unlock(&file_priv->prime.lock);
 		return -ENOENT;
+	}
 
 	if (args->handle == handle) {
+		mutex_unlock(&file_priv->prime.lock);
 		ret = 0;
 		goto out;
 	}
 
-	mutex_lock(&file_priv->prime.lock);
-
 	spin_lock(&file_priv->table_lock);
 	ret = idr_alloc(&file_priv->object_idr, obj, handle, handle + 1,
 			GFP_NOWAIT);
@@ -1085,8 +1086,22 @@ drm_gem_open(struct drm_device *dev, str
 void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
-	idr_for_each(&file_private->object_idr,
-		     &drm_gem_object_release_handle, file_private);
+	struct drm_gem_object *obj;
+	int id;
+
+	for (;;) {
+		id = 0;
+		mutex_lock(&file_private->prime.lock);
+		spin_lock(&file_private->table_lock);
+		obj = idr_get_next(&file_private->object_idr, &id);
+		if (obj)
+			idr_remove(&file_private->object_idr, id);
+		spin_unlock(&file_private->table_lock);
+		mutex_unlock(&file_private->prime.lock);
+		if (!obj)
+			break;
+		drm_gem_object_release_handle(id, obj, file_private);
+	}
 	idr_destroy(&file_private->object_idr);
 }
 
--

[PATCH] drm: Avoid the chaotic interleaving of change and delete handle

[Edward Adam Davis]

First, let's take a look at how GEM object is freed:
CPU0					CPU1
====					====
drm_gem_change_handle_ioctl()
 drm_gem_object_lookup() // got GEM obj and refcnt is 2
 ... blocks on prime.lock
					drm_gem_handle_delete()
	                                 drm_gem_object_release_handle()
					  ... acquires prime.lock
					  drm_prime_remove_buf_handle()
					  ... unlock prime.lock
					  drm_gem_object_handle_put_unlocked()
					   drm_gem_object_put() // obj refcnt is 1
 ... acquires prime.lock
 drm_gem_object_put() // obj refcnt is 0
  drm_gem_shmem_free() // obj is freed

After a GEM object has been freed, a Use-After-Free vulnerability [1]
is triggered when closes the DRM file, as the drm_gem_release() function
attempts to access the already-freed GEM object.

Adjust the change handle ioctl and handle delete ioctl to be atomic
operations; this prevents simultaneous change and delete operations
on the same GEM object from interfering with the release of the GEM
object during the closing of the DRM file.

[1]
BUG: KASAN: slab-use-after-free in drm_gem_object_release_handle+0x4b/0x1e0 drivers/gpu/drm/drm_gem.c:374
Call Trace:
 drm_gem_object_release_handle+0x4b/0x1e0 drivers/gpu/drm/drm_gem.c:374
 idr_for_each+0x1c6/0x2a0 lib/idr.c:210
 drm_gem_release+0x28/0x40 drivers/gpu/drm/drm_gem.c:1088
 drm_file_free+0x729/0xa00 drivers/gpu/drm/drm_file.c:261
 drm_close_helper drivers/gpu/drm/drm_file.c:290 [inline]
 drm_release+0x2de/0x3f0 drivers/gpu/drm/drm_file.c:438

Allocated by task 6090:
 __drm_gem_shmem_create+0xc4/0x2e0 drivers/gpu/drm/drm_gem_shmem_helper.c:130
 drm_gem_shmem_create drivers/gpu/drm/drm_gem_shmem_helper.c:157 [inline]
 drm_gem_shmem_create_with_handle drivers/gpu/drm/drm_gem_shmem_helper.c:460 [inline]
 drm_gem_shmem_dumb_create+0x72/0x120 drivers/gpu/drm/drm_gem_shmem_helper.c:549
 
Freed by task 6093:
 drm_gem_object_release_handle+0xc2/0x1e0 drivers/gpu/drm/drm_gem.c:385
 drm_gem_handle_delete+0x7b/0xb0 drivers/gpu/drm/drm_gem.c:413
 drm_ioctl_kernel+0x2df/0x3b0 drivers/gpu/drm/drm_ioctl.c:804

Fixes: 53096728b891 ("drm: Add DRM prime interface to reassign GEM handle")
Reported-by: syzbot+b2e951687503f32f74ce@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
Tested-by: syzbot+b2e951687503f32f74ce@syzkaller.appspotmail.com
Signed-off-by: Edward Adam Davis <eadavis@qq.com>
---
 drivers/gpu/drm/drm_gem.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 891c3bff5ae0..63a8d7e980b5 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -374,14 +374,8 @@ drm_gem_object_release_handle(int id, void *ptr, void *data)
 	if (obj->funcs->close)
 		obj->funcs->close(obj, file_priv);
 
-	mutex_lock(&file_priv->prime.lock);
-
 	drm_prime_remove_buf_handle(&file_priv->prime, id);
-
-	mutex_unlock(&file_priv->prime.lock);
-
 	drm_vma_node_revoke(&obj->vma_node, file_priv);
-
 	drm_gem_object_handle_put_unlocked(obj);
 
 	return 0;
@@ -401,13 +395,16 @@ drm_gem_handle_delete(struct drm_file *filp, u32 handle)
 {
 	struct drm_gem_object *obj;
 
+	mutex_lock(&filp->prime.lock);
 	spin_lock(&filp->table_lock);
 
 	/* Check if we currently have a reference on the object */
 	obj = idr_replace(&filp->object_idr, NULL, handle);
 	spin_unlock(&filp->table_lock);
-	if (IS_ERR_OR_NULL(obj))
+	if (IS_ERR_OR_NULL(obj)) {
+		mutex_unlock(&filp->prime.lock);
 		return -EINVAL;
+	}
 
 	/* Release driver's reference and decrement refcount. */
 	drm_gem_object_release_handle(handle, obj, filp);
@@ -416,6 +413,7 @@ drm_gem_handle_delete(struct drm_file *filp, u32 handle)
 	spin_lock(&filp->table_lock);
 	idr_remove(&filp->object_idr, handle);
 	spin_unlock(&filp->table_lock);
+	mutex_unlock(&filp->prime.lock);
 
 	return 0;
 }
@@ -1012,17 +1010,18 @@ int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data,
 		return -EINVAL;
 	handle = args->new_handle;
 
+	mutex_lock(&file_priv->prime.lock);
 	obj = drm_gem_object_lookup(file_priv, args->handle);
-	if (!obj)
+	if (!obj) {
+		mutex_unlock(&file_priv->prime.lock);
 		return -ENOENT;
+	}
 
 	if (args->handle == handle) {
 		ret = 0;
-		goto out;
+		goto out_unlock;
 	}
 
-	mutex_lock(&file_priv->prime.lock);
-
 	spin_lock(&file_priv->table_lock);
 	ret = idr_alloc(&file_priv->object_idr, obj, handle, handle + 1,
 			GFP_NOWAIT);
@@ -1051,9 +1050,8 @@ int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data,
 	spin_unlock(&file_priv->table_lock);
 
 out_unlock:
-	mutex_unlock(&file_priv->prime.lock);
-out:
 	drm_gem_object_put(obj);
+	mutex_unlock(&file_priv->prime.lock);
 
 	return ret;
 }
@@ -1085,8 +1083,10 @@ drm_gem_open(struct drm_device *dev, struct drm_file *file_private)
 void
 drm_gem_release(struct drm_device *dev, struct drm_file *file_private)
 {
+	mutex_lock(&file_private->prime.lock);
 	idr_for_each(&file_private->object_idr,
 		     &drm_gem_object_release_handle, file_private);
+	mutex_unlock(&file_private->prime.lock);
 	idr_destroy(&file_private->object_idr);
 }
 
-- 
2.43.0

Re: [syzbot] [dri?] KASAN: slab-use-after-free Read in drm_gem_object_release_handle

[syzbot]

Hello,

syzbot has tested the proposed patch and the reproducer did not trigger any issue:

Reported-by: syzbot+b2e951687503f32f74ce@syzkaller.appspotmail.com
Tested-by: syzbot+b2e951687503f32f74ce@syzkaller.appspotmail.com

Tested on:

commit:         1d51b370 Merge tag 'jfs-7.1' of github.com:kleikamp/li..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=10b5e906580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=769ed9499e4e3d7d
dashboard link: https://syzkaller.appspot.com/bug?extid=b2e951687503f32f74ce
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
patch:          https://syzkaller.appspot.com/x/patch.diff?x=154398ce580000

Note: testing is done by a robot and is best-effort only.