nvmet-tcp frees CMD buffers in nvmet_tcp_uninit_data_in_cmds(),
and waits the inflight IO requests in nvmet_sq_destroy(). During wait
the inflight IO requests, the callback nvmet_tcp_queue_response()
is called from backend after IO complete, this leads a typical
Use-After-Free issue like this:
BUG: kernel NULL pointer dereference, address: 0000000000000008
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
PGD 0 P4D 0
Oops: 0000 [#1] PREEMPT SMP NOPTI
CPU: 14 PID: 207 Comm: kworker/14:1H Kdump: loaded Tainted: G E 6.0.0-rc2.bm.1-amd64 #12
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
Workqueue: nvmet_tcp_wq nvmet_tcp_io_work [nvmet_tcp]
RIP: 0010:shash_ahash_digest+0x2b/0x110
Code: 1f 44 00 00 41 57 41 56 41 55 41 54 55 48 89 fd 53 48 89 f3 48 83 ec 08 44 8b 67 30 45 85 e4 74 1c 48 8b 57 38 b8 00 10 00 00 <44> 8b 7a 08 44 29 f8 39 42 0c 0f 46 42 0c 41 39 c4 76 43 48 8b 03
RSP: 0018:ffffc900006e3dd8 EFLAGS: 00010206
RAX: 0000000000001000 RBX: ffff888104ac1650 RCX: 0000000000000000
RDX: 0000000000000000 RSI: ffff888104ac1650 RDI: ffff888104ac1600
RBP: ffff888104ac1600 R08: ffff8881073980c8 R09: ffff8881057798b8
R10: 8080808080808080 R11: 0000000000000000 R12: 0000000000001000
R13: 0000000000000000 R14: ffff88810601a1cc R15: ffff888107398000
FS: 0000000000000000(0000) GS:ffff88823fd80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000008 CR3: 000000010a8e8000 CR4: 0000000000350ee0
Call Trace:
<TASK>
nvmet_tcp_io_work+0xa1c/0xb1c [nvmet_tcp]
? __switch_to+0x106/0x420
process_one_work+0x1ae/0x380
? process_one_work+0x380/0x380
worker_thread+0x30/0x360
? process_one_work+0x380/0x380
kthread+0xe6/0x110
? kthread_complete_and_exit+0x20/0x20
ret_from_fork+0x1f/0x30
To fix this issue, free CMD buffers until all the inflight IO
complete.
Signed-off-by: zhenwei pi <[email protected]>
---
drivers/nvme/target/tcp.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/nvme/target/tcp.c b/drivers/nvme/target/tcp.c
index dc3b4dc8fe08..2325246e3b4d 100644
--- a/drivers/nvme/target/tcp.c
+++ b/drivers/nvme/target/tcp.c
@@ -1449,8 +1449,8 @@ static void nvmet_tcp_release_queue_work(struct work_struct *w)
/* stop accepting incoming data */
queue->rcv_state = NVMET_TCP_RECV_ERR;
- nvmet_tcp_uninit_data_in_cmds(queue);
nvmet_sq_destroy(&queue->nvme_sq);
+ nvmet_tcp_uninit_data_in_cmds(queue);
cancel_work_sync(&queue->io_work);
sock_release(queue->sock);
nvmet_tcp_free_cmds(queue);
--
2.20.1
On 8/30/22 11:36, zhenwei pi wrote:
> nvmet-tcp frees CMD buffers in nvmet_tcp_uninit_data_in_cmds(),
> and waits the inflight IO requests in nvmet_sq_destroy(). During wait
> the inflight IO requests, the callback nvmet_tcp_queue_response()
> is called from backend after IO complete, this leads a typical
> Use-After-Free issue like this:
>
> BUG: kernel NULL pointer dereference, address: 0000000000000008
> #PF: supervisor read access in kernel mode
> #PF: error_code(0x0000) - not-present page
> PGD 0 P4D 0
> Oops: 0000 [#1] PREEMPT SMP NOPTI
> CPU: 14 PID: 207 Comm: kworker/14:1H Kdump: loaded Tainted: G E 6.0.0-rc2.bm.1-amd64 #12
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
> Workqueue: nvmet_tcp_wq nvmet_tcp_io_work [nvmet_tcp]
> RIP: 0010:shash_ahash_digest+0x2b/0x110
This data access is specific to data-digest, because at this point there
is no way for the data to be sent down the wire as the socket is shut
down.
> Code: 1f 44 00 00 41 57 41 56 41 55 41 54 55 48 89 fd 53 48 89 f3 48 83 ec 08 44 8b 67 30 45 85 e4 74 1c 48 8b 57 38 b8 00 10 00 00 <44> 8b 7a 08 44 29 f8 39 42 0c 0f 46 42 0c 41 39 c4 76 43 48 8b 03
> RSP: 0018:ffffc900006e3dd8 EFLAGS: 00010206
> RAX: 0000000000001000 RBX: ffff888104ac1650 RCX: 0000000000000000
> RDX: 0000000000000000 RSI: ffff888104ac1650 RDI: ffff888104ac1600
> RBP: ffff888104ac1600 R08: ffff8881073980c8 R09: ffff8881057798b8
> R10: 8080808080808080 R11: 0000000000000000 R12: 0000000000001000
> R13: 0000000000000000 R14: ffff88810601a1cc R15: ffff888107398000
> FS: 0000000000000000(0000) GS:ffff88823fd80000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 0000000000000008 CR3: 000000010a8e8000 CR4: 0000000000350ee0
> Call Trace:
> <TASK>
> nvmet_tcp_io_work+0xa1c/0xb1c [nvmet_tcp]
> ? __switch_to+0x106/0x420
> process_one_work+0x1ae/0x380
> ? process_one_work+0x380/0x380
> worker_thread+0x30/0x360
> ? process_one_work+0x380/0x380
> kthread+0xe6/0x110
> ? kthread_complete_and_exit+0x20/0x20
> ret_from_fork+0x1f/0x30
>
> To fix this issue, free CMD buffers until all the inflight IO
> complete.
>
> Signed-off-by: zhenwei pi <[email protected]>
> ---
> drivers/nvme/target/tcp.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/nvme/target/tcp.c b/drivers/nvme/target/tcp.c
> index dc3b4dc8fe08..2325246e3b4d 100644
> --- a/drivers/nvme/target/tcp.c
> +++ b/drivers/nvme/target/tcp.c
> @@ -1449,8 +1449,8 @@ static void nvmet_tcp_release_queue_work(struct work_struct *w)
> /* stop accepting incoming data */
> queue->rcv_state = NVMET_TCP_RECV_ERR;
>
> - nvmet_tcp_uninit_data_in_cmds(queue);
> nvmet_sq_destroy(&queue->nvme_sq);
> + nvmet_tcp_uninit_data_in_cmds(queue);
We still need to uninit the nvmet request so nvmet_sq_destroy()
can complete. Please separate nvmet_tcp_uninit_data_in_cmds to
uninit and free potions, only the free should go after the sq_destroy.