The patch below does not apply to the 5.12-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to stable@vger.kernel.org.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From 447c19f3b5074409c794b350b10306e1da1ef4ba Mon Sep 17 00:00:00 2001
From: Pavel Begunkov asml.silence@gmail.com Date: Fri, 14 May 2021 12:02:50 +0100 Subject: [PATCH] io_uring: fix ltout double free on completion race
Always remove linked timeout on io_link_timeout_fn() from the master request link list, otherwise we may get use-after-free when first io_link_timeout_fn() puts linked timeout in the fail path, and then will be found and put on master's free.
Cc: stable@vger.kernel.org # 5.10+ Fixes: 90cd7e424969d ("io_uring: track link timeout's master explicitly") Reported-and-tested-by: syzbot+5a864149dd970b546223@syzkaller.appspotmail.com Signed-off-by: Pavel Begunkov asml.silence@gmail.com Link: https://lore.kernel.org/r/69c46bf6ce37fec4fdcd98f0882e18eb07ce693a.162099012... Signed-off-by: Jens Axboe axboe@kernel.dk
diff --git a/fs/io_uring.c b/fs/io_uring.c index 9ac5e278a91e..599102cc6dfc 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -6354,10 +6354,11 @@ static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer) * We don't expect the list to be empty, that will only happen if we * race with the completion of the linked work. */ - if (prev && req_ref_inc_not_zero(prev)) + if (prev) { io_remove_next_linked(prev); - else - prev = NULL; + if (!req_ref_inc_not_zero(prev)) + prev = NULL; + } spin_unlock_irqrestore(&ctx->completion_lock, flags);
if (prev) {