[PATCH 1/1] io_uring: add request poisoning

[PATCH 1/1] io_uring: add request poisoning

[Pavel Begunkov]

Poison various request fields on free. __io_req_caches_free() is a slow
path, so can be done unconditionally, but gate it on kasan for
io_req_add_to_cache(). Note that some fields are logically retained
between cache allocations and can't be poisoned in
io_req_add_to_cache().

Ideally, it'd be replaced with KASAN'ed caches, but that can't be
enabled because of some synchronisation nuances.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 include/linux/poison.h |  3 +++
 io_uring/io_uring.c    | 23 +++++++++++++++++++++++
 2 files changed, 26 insertions(+)

diff --git a/include/linux/poison.h b/include/linux/poison.h
index 8ca2235f78d5..299e2dd7da6d 100644
--- a/include/linux/poison.h
+++ b/include/linux/poison.h
@@ -90,4 +90,7 @@
 /********** lib/stackdepot.c **********/
 #define STACK_DEPOT_POISON ((void *)(0xD390 + POISON_POINTER_DELTA))
 
+/********** io_uring/ **********/
+#define IO_URING_PTR_POISON ((void *)(0x1091UL + POISON_POINTER_DELTA))
+
 #endif
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 4ef69dd58734..e6f2353f7460 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -179,6 +179,26 @@ static const struct ctl_table kernel_io_uring_disabled_table[] = {
 };
 #endif
 
+static void io_poison_cached_req(struct io_kiocb *req)
+{
+	req->ctx = IO_URING_PTR_POISON;
+	req->tctx = IO_URING_PTR_POISON;
+	req->file = IO_URING_PTR_POISON;
+	req->apoll = IO_URING_PTR_POISON;
+	req->async_data = IO_URING_PTR_POISON;
+	req->creds = IO_URING_PTR_POISON;
+	req->io_task_work.func = IO_URING_PTR_POISON;
+}
+
+static void io_poison_req(struct io_kiocb *req)
+{
+	io_poison_cached_req(req);
+	req->kbuf = IO_URING_PTR_POISON;
+	req->comp_list.next = IO_URING_PTR_POISON;
+	req->file_node = IO_URING_PTR_POISON;
+	req->link = IO_URING_PTR_POISON;
+}
+
 static inline unsigned int __io_cqring_events(struct io_ring_ctx *ctx)
 {
 	return ctx->cached_cq_tail - READ_ONCE(ctx->rings->cq.head);
@@ -235,6 +255,8 @@ static inline void req_fail_link_node(struct io_kiocb *req, int res)
 
 static inline void io_req_add_to_cache(struct io_kiocb *req, struct io_ring_ctx *ctx)
 {
+	if (IS_ENABLED(CONFIG_KASAN))
+		io_poison_cached_req(req);
 	wq_stack_add_head(&req->comp_list, &ctx->submit_state.free_list);
 }
 
@@ -2766,6 +2788,7 @@ static __cold void __io_req_caches_free(struct io_ring_ctx *ctx)
 
 	while (!io_req_cache_empty(ctx)) {
 		req = io_extract_req(ctx);
+		io_poison_req(req);
 		kmem_cache_free(req_cachep, req);
 		nr++;
 	}
-- 
2.49.0

[syzbot ci] Re: io_uring: add request poisoning

[syzbot ci]

syzbot ci has tested the following series

[v1] io_uring: add request poisoning
https://lore.kernel.org/all/b98edbb8ec4495b053dfb11cb3588f17f5253b6e.1755182071.git.asml.silence@gmail.com
* [PATCH 1/1] io_uring: add request poisoning

and found the following issue:
general protection fault in __io_queue_proc

Full report is available here:
https://ci.syzbot.org/series/f67750f7-0bd0-41d6-a3eb-d1b77b2c9728

***

general protection fault in __io_queue_proc

tree:      torvalds
URL:       https://kernel.googlesource.com/pub/scm/linux/kernel/git/torvalds/linux
base:      dfc0f6373094dd88e1eaf76c44f2ff01b65db851
arch:      amd64
compiler:  Debian clang version 20.1.7 (++20250616065708+6146a88f6049-1~exp1~20250616065826.132), Debian LLD 20.1.7
config:    https://ci.syzbot.org/builds/e0a322a6-324d-4dcd-a2dc-b6ec3bfb0f54/config
C repro:   https://ci.syzbot.org/findings/c93b97f2-efb3-45d6-a6ca-0afeb7311b74/c_repro
syz repro: https://ci.syzbot.org/findings/c93b97f2-efb3-45d6-a6ca-0afeb7311b74/syz_repro

Oops: general protection fault, probably for non-canonical address 0xfbd59c0000000213: 0000 [#1] SMP KASAN PTI
KASAN: maybe wild-memory-access in range [0xdead000000001098-0xdead00000000109f]
CPU: 0 UID: 0 PID: 5985 Comm: syz.0.17 Not tainted 6.17.0-rc1-syzkaller-00036-gdfc0f6373094-dirty #0 PREEMPT(full) 
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.2-debian-1.16.2-1 04/01/2014
RIP: 0010:__io_queue_proc+0x1b8/0x4d0 io_uring/poll.c:475
Code: c1 e8 03 48 89 44 24 20 80 3c 18 00 74 08 4c 89 ef e8 8c 27 63 fd 4d 8b 6d 00 4d 85 ed 74 5a 49 83 c5 08 4c 89 e8 48 c1 e8 03 <80> 3c 18 00 74 08 4c 89 ef e8 6a 27 63 fd 48 8b 44 24 18 49 39 45
RSP: 0018:ffffc9000295f760 EFLAGS: 00010a02
RAX: 1bd5a00000000213 RBX: dffffc0000000000 RCX: 0000000000000000
RDX: ffff888020cd9cc0 RSI: 0000000000000001 RDI: 0000000000000000
RBP: ffffc9000295f900 R08: ffffffff8fa37e37 R09: 1ffffffff1f46fc6
R10: dffffc0000000000 R11: ffffffff84bfdf40 R12: ffff888028da6000
R13: dead000000001099 R14: ffffc9000295f918 R15: 1ffff9200052bf23
FS:  000055558fd42500(0000) GS:ffff8880b861c000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000001b30d63fff CR3: 000000010e6dc000 CR4: 00000000000006f0
Call Trace:
 <TASK>
 poll_wait include/linux/poll.h:45 [inline]
 n_tty_poll+0x9d/0x740 drivers/tty/n_tty.c:2451
 tty_poll+0xbe/0x160 drivers/tty/tty_io.c:2199
 vfs_poll include/linux/poll.h:82 [inline]
 __io_arm_poll_handler+0x372/0xbb0 io_uring/poll.c:581
 io_poll_add+0xcd/0x1f0 io_uring/poll.c:901
 __io_issue_sqe+0x181/0x4b0 io_uring/io_uring.c:1795
 io_issue_sqe+0x165/0xfd0 io_uring/io_uring.c:1818
 io_queue_sqe io_uring/io_uring.c:2047 [inline]
 io_submit_sqe io_uring/io_uring.c:2306 [inline]
 io_submit_sqes+0xa32/0x1e60 io_uring/io_uring.c:2419
 __do_sys_io_uring_enter io_uring/io_uring.c:3487 [inline]
 __se_sys_io_uring_enter+0x2df/0x2b20 io_uring/io_uring.c:3421
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0xfa/0x3b0 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f750a18ebe9
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 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 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffc8b53b938 EFLAGS: 00000246 ORIG_RAX: 00000000000001aa
RAX: ffffffffffffffda RBX: 00007f750a3b5fa0 RCX: 00007f750a18ebe9
RDX: 0000000000000000 RSI: 0000000000000a3d RDI: 0000000000000004
RBP: 00007f750a211e19 R08: 0000000000000000 R09: 000000000000ff39
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007f750a3b5fa0 R14: 00007f750a3b5fa0 R15: 0000000000000006
 </TASK>
Modules linked in:
---[ end trace 0000000000000000 ]---
RIP: 0010:__io_queue_proc+0x1b8/0x4d0 io_uring/poll.c:475
Code: c1 e8 03 48 89 44 24 20 80 3c 18 00 74 08 4c 89 ef e8 8c 27 63 fd 4d 8b 6d 00 4d 85 ed 74 5a 49 83 c5 08 4c 89 e8 48 c1 e8 03 <80> 3c 18 00 74 08 4c 89 ef e8 6a 27 63 fd 48 8b 44 24 18 49 39 45
RSP: 0018:ffffc9000295f760 EFLAGS: 00010a02
RAX: 1bd5a00000000213 RBX: dffffc0000000000 RCX: 0000000000000000
RDX: ffff888020cd9cc0 RSI: 0000000000000001 RDI: 0000000000000000
RBP: ffffc9000295f900 R08: ffffffff8fa37e37 R09: 1ffffffff1f46fc6
R10: dffffc0000000000 R11: ffffffff84bfdf40 R12: ffff888028da6000
R13: dead000000001099 R14: ffffc9000295f918 R15: 1ffff9200052bf23
FS:  000055558fd42500(0000) GS:ffff8880b861c000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000001b30d63fff CR3: 000000010e6dc000 CR4: 00000000000006f0
----------------
Code disassembly (best guess):
   0:	c1 e8 03             	shr    $0x3,%eax
   3:	48 89 44 24 20       	mov    %rax,0x20(%rsp)
   8:	80 3c 18 00          	cmpb   $0x0,(%rax,%rbx,1)
   c:	74 08                	je     0x16
   e:	4c 89 ef             	mov    %r13,%rdi
  11:	e8 8c 27 63 fd       	call   0xfd6327a2
  16:	4d 8b 6d 00          	mov    0x0(%r13),%r13
  1a:	4d 85 ed             	test   %r13,%r13
  1d:	74 5a                	je     0x79
  1f:	49 83 c5 08          	add    $0x8,%r13
  23:	4c 89 e8             	mov    %r13,%rax
  26:	48 c1 e8 03          	shr    $0x3,%rax
* 2a:	80 3c 18 00          	cmpb   $0x0,(%rax,%rbx,1) <-- trapping instruction
  2e:	74 08                	je     0x38
  30:	4c 89 ef             	mov    %r13,%rdi
  33:	e8 6a 27 63 fd       	call   0xfd6327a2
  38:	48 8b 44 24 18       	mov    0x18(%rsp),%rax
  3d:	49                   	rex.WB
  3e:	39                   	.byte 0x39
  3f:	45                   	rex.RB


***

If these findings have caused you to resend the series or submit a
separate fix, please add the following tag to your commit message:
Tested-by: syzbot@syzkaller.appspotmail.com

---
This report is generated by a bot. It may contain errors.
syzbot ci engineers can be reached at syzkaller@googlegroups.com.

Re: [syzbot ci] Re: io_uring: add request poisoning

[Pavel Begunkov]

On 8/16/25 08:30, syzbot ci wrote:
> syzbot ci has tested the following series
> 
> [v1] io_uring: add request poisoning
> https://lore.kernel.org/all/b98edbb8ec4495b053dfb11cb3588f17f5253b6e.1755182071.git.asml.silence@gmail.com
> * [PATCH 1/1] io_uring: add request poisoning
> 
> and found the following issue:
> general protection fault in __io_queue_proc

It dug up a hack poll_add does to initialise ->async_data. Not a
pre-existent bug, but definitely one of the things the patch is
supposed to uncover. I'll update it.

-- 
Pavel Begunkov