From: Jens Axboe <axboe@kernel.dk>
To: io-uring@vger.kernel.org
Cc: Jens Axboe <axboe@kernel.dk>
Subject: [PATCH 2/2] io_uring/waitid: use io_waitid_remove_wq() consistently
Date: Thu, 9 Oct 2025 11:39:26 -0600 [thread overview]
Message-ID: <20251009174145.2209946-3-axboe@kernel.dk> (raw)
In-Reply-To: <20251009174145.2209946-1-axboe@kernel.dk>
Use it everywhere that the wait_queue_entry is removed from the head,
and be a bit more cautious in zeroing out iw->head whenever the entry is
removed from the list.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
io_uring/waitid.c | 15 +++++++++++----
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/io_uring/waitid.c b/io_uring/waitid.c
index ebe3769c54dc..c5e0d979903a 100644
--- a/io_uring/waitid.c
+++ b/io_uring/waitid.c
@@ -179,18 +179,18 @@ bool io_waitid_remove_all(struct io_ring_ctx *ctx, struct io_uring_task *tctx,
static inline bool io_waitid_drop_issue_ref(struct io_kiocb *req)
{
struct io_waitid *iw = io_kiocb_to_cmd(req, struct io_waitid);
- struct io_waitid_async *iwa = req->async_data;
if (!atomic_sub_return(1, &iw->refs))
return false;
+ io_waitid_remove_wq(req);
+
/*
* Wakeup triggered, racing with us. It was prevented from
* completing because of that, queue up the tw to do that.
*/
req->io_task_work.func = io_waitid_cb;
io_req_task_work_add(req);
- remove_wait_queue(iw->head, &iwa->wo.child_wait);
return true;
}
@@ -245,6 +245,7 @@ static int io_waitid_wait(struct wait_queue_entry *wait, unsigned mode,
return 0;
list_del_init(&wait->entry);
+ iw->head = NULL;
/* cancel is in progress */
if (atomic_fetch_inc(&iw->refs) & IO_WAITID_REF_MASK)
@@ -271,6 +272,7 @@ int io_waitid_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
iw->which = READ_ONCE(sqe->len);
iw->upid = READ_ONCE(sqe->fd);
iw->options = READ_ONCE(sqe->file_index);
+ iw->head = NULL;
iw->infop = u64_to_user_ptr(READ_ONCE(sqe->addr2));
return 0;
}
@@ -301,11 +303,16 @@ int io_waitid(struct io_kiocb *req, unsigned int issue_flags)
* callback.
*/
io_ring_submit_lock(ctx, issue_flags);
+
+ /*
+ * iw->head is valid under the ring lock, and as long as the request
+ * is on the waitid_list where cancelations may find it.
+ */
+ iw->head = ¤t->signal->wait_chldexit;
hlist_add_head(&req->hash_node, &ctx->waitid_list);
init_waitqueue_func_entry(&iwa->wo.child_wait, io_waitid_wait);
iwa->wo.child_wait.private = req->tctx->task;
- iw->head = ¤t->signal->wait_chldexit;
add_wait_queue(iw->head, &iwa->wo.child_wait);
ret = __do_wait(&iwa->wo);
@@ -328,7 +335,7 @@ int io_waitid(struct io_kiocb *req, unsigned int issue_flags)
}
hlist_del_init(&req->hash_node);
- remove_wait_queue(iw->head, &iwa->wo.child_wait);
+ io_waitid_remove_wq(req);
ret = io_waitid_finish(req, ret);
io_ring_submit_unlock(ctx, issue_flags);
--
2.51.0
prev parent reply other threads:[~2025-10-09 17:41 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-09 17:39 [PATCHSET for-next 0/2] io_uring waitid cleanups Jens Axboe
2025-10-09 17:39 ` [PATCH 1/2] io_uring/waitid: have io_waitid_complete() remove wait queue entry Jens Axboe
2025-10-09 17:39 ` Jens Axboe [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20251009174145.2209946-3-axboe@kernel.dk \
--to=axboe@kernel.dk \
--cc=io-uring@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).