linux-kernel - Re: [RFC] splice/tee: len=0 fast path after validity check

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]

Message-ID: <29de6592-8d62-5b5d-9859-d7adcc58759b@kernel.dk>
Date:   Sat, 9 May 2020 08:42:56 -0600
From:   Jens Axboe <axboe@...nel.dk>
To:     Pavel Begunkov <asml.silence@...il.com>,
        Alexander Viro <viro@...iv.linux.org.uk>,
        io-uring@...r.kernel.org, linux-fsdevel@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [RFC] splice/tee: len=0 fast path after validity check

On 5/9/20 2:46 AM, Pavel Begunkov wrote:
> When len=0, splice() and tee() return 0 even if specified fds are
> invalid, hiding errors from users. Move len=0 optimisation later after
> basic validity checks.
> 
> before:
> splice(len=0, fd_in=-1, ...) == 0;
> 
> after:
> splice(len=0, fd_in=-1, ...) == -EBADF;

I'm not sure what the purpose of this would be. It probably should have
been done that way from the beginning, but it wasn't.  While there's
very little risk of breaking any applications due to this change, it
also seems like a pointless exercise at this point.

So my suggestion would be to just leave it alone.

-- 
Jens Axboe