This is a note to let you know that I've just added the patch titled
io_uring: ensure io_lockdep_assert_cq_locked() handles disabled rings
to the 6.5-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
io_uring-ensure-io_lockdep_assert_cq_locked-handles-disabled-rings.patch
and it can be found in the queue-6.5 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From 1658633c04653578429ff5dfc62fdc159203a8f2 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@xxxxxxxxx>
Date: Mon, 2 Oct 2023 19:51:38 -0600
Subject: io_uring: ensure io_lockdep_assert_cq_locked() handles disabled rings
From: Jens Axboe <axboe@xxxxxxxxx>
commit 1658633c04653578429ff5dfc62fdc159203a8f2 upstream.
io_lockdep_assert_cq_locked() checks that locking is correctly done when
a CQE is posted. If the ring is setup in a disabled state with
IORING_SETUP_R_DISABLED, then ctx->submitter_task isn't assigned until
the ring is later enabled. We generally don't post CQEs in this state,
as no SQEs can be submitted. However it is possible to generate a CQE
if tagged resources are being updated. If this happens and PROVE_LOCKING
is enabled, then the locking check helper will dereference
ctx->submitter_task, which hasn't been set yet.
Fixup io_lockdep_assert_cq_locked() to handle this case correctly. While
at it, convert it to a static inline as well, so that generated line
offsets will actually reflect which condition failed, rather than just
the line offset for io_lockdep_assert_cq_locked() itself.
Reported-and-tested-by: syzbot+efc45d4e7ba6ab4ef1eb@xxxxxxxxxxxxxxxxxxxxxxxxx
Fixes: f26cc9593581 ("io_uring: lockdep annotate CQ locking")
Cc: stable@xxxxxxxxxxxxxxx
Signed-off-by: Jens Axboe <axboe@xxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
io_uring/io_uring.h | 41 +++++++++++++++++++++++++++--------------
1 file changed, 27 insertions(+), 14 deletions(-)
--- a/io_uring/io_uring.h
+++ b/io_uring/io_uring.h
@@ -87,20 +87,33 @@ bool __io_alloc_req_refill(struct io_rin
bool io_match_task_safe(struct io_kiocb *head, struct task_struct *task,
bool cancel_all);
-#define io_lockdep_assert_cq_locked(ctx) \
- do { \
- lockdep_assert(in_task()); \
- \
- if (ctx->flags & IORING_SETUP_IOPOLL) { \
- lockdep_assert_held(&ctx->uring_lock); \
- } else if (!ctx->task_complete) { \
- lockdep_assert_held(&ctx->completion_lock); \
- } else if (ctx->submitter_task->flags & PF_EXITING) { \
- lockdep_assert(current_work()); \
- } else { \
- lockdep_assert(current == ctx->submitter_task); \
- } \
- } while (0)
+#if defined(CONFIG_PROVE_LOCKING)
+static inline void io_lockdep_assert_cq_locked(struct io_ring_ctx *ctx)
+{
+ lockdep_assert(in_task());
+
+ if (ctx->flags & IORING_SETUP_IOPOLL) {
+ lockdep_assert_held(&ctx->uring_lock);
+ } else if (!ctx->task_complete) {
+ lockdep_assert_held(&ctx->completion_lock);
+ } else if (ctx->submitter_task) {
+ /*
+ * ->submitter_task may be NULL and we can still post a CQE,
+ * if the ring has been setup with IORING_SETUP_R_DISABLED.
+ * Not from an SQE, as those cannot be submitted, but via
+ * updating tagged resources.
+ */
+ if (ctx->submitter_task->flags & PF_EXITING)
+ lockdep_assert(current_work());
+ else
+ lockdep_assert(current == ctx->submitter_task);
+ }
+}
+#else
+static inline void io_lockdep_assert_cq_locked(struct io_ring_ctx *ctx)
+{
+}
+#endif
static inline void io_req_task_work_add(struct io_kiocb *req)
{
Patches currently in stable-queue which might be from axboe@xxxxxxxxx are
queue-6.5/io_uring-don-t-allow-ioring_setup_no_mmap-rings-on-highmem-pages.patch
queue-6.5/io_uring-ensure-io_lockdep_assert_cq_locked-handles-disabled-rings.patch
queue-6.5/io_uring-kbuf-don-t-allow-registered-buffer-rings-on-highmem-pages.patch
