From: Pavel Begunkov <[email protected]>
To: Jia-Ju Bai <[email protected]>, [email protected]
Cc: [email protected], [email protected],
TOTE Robot <[email protected]>
Subject: Re: [PATCH] io_uring: Add NULL checks for current->io_uring
Date: Wed, 11 Jan 2023 14:49:34 +0000 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
On 1/11/23 10:19, Jia-Ju Bai wrote:
> As described in a previous commit 998b30c3948e, current->io_uring could
> be NULL, and thus a NULL check is required for this variable.
>
> In the same way, other functions that access current->io_uring also
> require NULL checks of this variable.
io_uring_enter() {
...
ret = io_uring_add_tctx_node(ctx);
if (unlikely(ret))
goto out;
mutex_lock(&ctx->uring_lock);
ret = io_submit_sqes(ctx, to_submit);
}
io_uring_add_tctx_node() should make sure to setup current->io_uring
or fail, so it should be NULL there, and SQPOLL should be fine as well.
Did you see it hitting NULL?
>
> Signed-off-by: Jia-Ju Bai <[email protected]>
> Reported-by: TOTE Robot <[email protected]>
> ---
> io_uring/io_uring.c | 3 ++-
> io_uring/io_uring.h | 3 +++
> io_uring/tctx.c | 9 ++++++++-
> 3 files changed, 13 insertions(+), 2 deletions(-)
>
> diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
> index 2ac1cd8d23ea..8075c0880c7a 100644
> --- a/io_uring/io_uring.c
> +++ b/io_uring/io_uring.c
> @@ -2406,7 +2406,8 @@ int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr)
> /* try again if it submitted nothing and can't allocate a req */
> if (!ret && io_req_cache_empty(ctx))
> ret = -EAGAIN;
> - current->io_uring->cached_refs += left;
> + if (likely(current->io_uring))
> + current->io_uring->cached_refs += left;
> }
>
> io_submit_state_end(ctx);
> diff --git a/io_uring/io_uring.h b/io_uring/io_uring.h
> index ab4b2a1c3b7e..398c7c2ba22b 100644
> --- a/io_uring/io_uring.h
> +++ b/io_uring/io_uring.h
> @@ -362,6 +362,9 @@ static inline void io_get_task_refs(int nr)
> {
> struct io_uring_task *tctx = current->io_uring;
>
> + if (unlikely(!tctx))
> + return;
> +
> tctx->cached_refs -= nr;
> if (unlikely(tctx->cached_refs < 0))
> io_task_refs_refill(tctx);
> diff --git a/io_uring/tctx.c b/io_uring/tctx.c
> index 4324b1cf1f6a..6574bbe82b5d 100644
> --- a/io_uring/tctx.c
> +++ b/io_uring/tctx.c
> @@ -145,7 +145,8 @@ int __io_uring_add_tctx_node_from_submit(struct io_ring_ctx *ctx)
> if (ret)
> return ret;
>
> - current->io_uring->last = ctx;
> + if (likely(current->io_uring))
> + current->io_uring->last = ctx;
> return 0;
> }
>
> @@ -200,6 +201,9 @@ void io_uring_unreg_ringfd(void)
> struct io_uring_task *tctx = current->io_uring;
> int i;
>
> + if (unlikely(!tctx))
> + return;
> +
> for (i = 0; i < IO_RINGFD_REG_MAX; i++) {
> if (tctx->registered_rings[i]) {
> fput(tctx->registered_rings[i]);
> @@ -259,6 +263,9 @@ int io_ringfd_register(struct io_ring_ctx *ctx, void __user *__arg,
> return ret;
>
> tctx = current->io_uring;
> + if (unlikely(!tctx))
> + return -EINVAL;
> +
> for (i = 0; i < nr_args; i++) {
> int start, end;
>
--
Pavel Begunkov
prev parent reply other threads:[~2023-01-11 14:51 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-11 10:19 [PATCH] io_uring: Add NULL checks for current->io_uring Jia-Ju Bai
2023-01-11 14:49 ` Jens Axboe
2023-01-12 2:10 ` Jia-Ju Bai
2023-01-12 2:37 ` Jens Axboe
2023-01-11 14:49 ` Pavel Begunkov [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 \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/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