On 30/12/2020 21:34, Pavel Begunkov wrote:
> For cancelling io_uring requests it needs either to be able to run
> currently enqueued task_wokrs or having it shut down by that moment.
> Otherwise io_uring_cancel_files() may be waiting for requests that won't
> ever complete.
>
> Go with the first way and do cancellations before setting PF_EXITING and
> so before putting the task_work infrastructure into a transition state
> where task_work_run() would better not be called.
It won't do in the end, because after cancellation we don't have files
NULLed yet and SQPOLL task may use find and use them. I guess can't be
dropped because of rc, I'll think whether to revert or somehow shut
sqpoll.
> Cc: stable@xxxxxxxxxxxxxxx # 5.5+
> Signed-off-by: Pavel Begunkov <asml.silence@xxxxxxxxx>
> ---
> fs/file.c | 2 --
> kernel/exit.c | 2 ++
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/fs/file.c b/fs/file.c
> index c0b60961c672..dab120b71e44 100644
> --- a/fs/file.c
> +++ b/fs/file.c
> @@ -21,7 +21,6 @@
> #include <linux/rcupdate.h>
> #include <linux/close_range.h>
> #include <net/sock.h>
> -#include <linux/io_uring.h>
>
> unsigned int sysctl_nr_open __read_mostly = 1024*1024;
> unsigned int sysctl_nr_open_min = BITS_PER_LONG;
> @@ -428,7 +427,6 @@ void exit_files(struct task_struct *tsk)
> struct files_struct * files = tsk->files;
>
> if (files) {
> - io_uring_files_cancel(files);
> task_lock(tsk);
> tsk->files = NULL;
> task_unlock(tsk);
> diff --git a/kernel/exit.c b/kernel/exit.c
> index 3594291a8542..04029e35e69a 100644
> --- a/kernel/exit.c
> +++ b/kernel/exit.c
> @@ -63,6 +63,7 @@
> #include <linux/random.h>
> #include <linux/rcuwait.h>
> #include <linux/compat.h>
> +#include <linux/io_uring.h>
>
> #include <linux/uaccess.h>
> #include <asm/unistd.h>
> @@ -776,6 +777,7 @@ void __noreturn do_exit(long code)
> schedule();
> }
>
> + io_uring_files_cancel(tsk->files);
> exit_signals(tsk); /* sets PF_EXITING */
>
> /* sync mm's RSS info before statistics gathering */
>
--
Pavel Begunkov
