linux-kernel - Re: io-uring: KASAN failure, presumably REQ_F

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Sat, 11 Sep 2021 21:43:18 -0700
From:   Nadav Amit <nadav.amit@...il.com>
To:     Jens Axboe <axboe@...nel.dk>,
        Pavel Begunkov <asml.silence@...il.com>
Cc:     io-uring@...r.kernel.org,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: io-uring: KASAN failure, presumably REQ_F_REISSUE issue



> On Sep 11, 2021, at 7:34 PM, Nadav Amit <nadav.amit@...il.com> wrote:
> 
> Hello Jens (& Pavel),
> 
> I hope you are having a nice weekend. I ran into a KASAN failure in io-uring
> which I think is not "my fault".

Small correction of myself (beside the subject):
> 
> I believe the issue is related to the handling of REQ_F_REISSUE and
> specifically to commit 230d50d448acb ("io_uring: move reissue into regular IO
> path"). There seems to be a race between io_write()/io_read()
> and __io_complete_rw()/kiocb_done().
> 
> __io_complete_rw() sets REQ_F_REIUSSE:
> 
>               if ((res == -EAGAIN || res == -EOPNOTSUPP) &&
>                    io_rw_should_reissue(req)) {
>                        req->flags |= REQ_F_REISSUE;
>                        return;
>               }

The race only appears to be with __io_complete_rw(), not kiocb_done().