Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <000000000000c422a80596d595ee@google.com> <6bddae34-93df-6820-0390-ac18dcbf0927@gmail.com>
 <CAHk-=whh5bcxCecEL5Fy4XvQjgBTJ9uqvyp7dW=CLU6VNxS9iA@mail.gmail.com>
 <CANn89iK9mTJ4BN-X3MeSx5LGXGYafXkhZyaUpdXDjVivTwA6Jg@mail.gmail.com>
 <CAHk-=whNBL63qmO176qOQpkY16xvomog5ocvM=9K55hUgAgOPA@mail.gmail.com>
 <CANn89iJJiB6avNtZ1qQNTeJwyjW32Pxk_2CwvEJxgQ==kgY0fA@mail.gmail.com>
 <CANn89i+RrngUr11_iOYDuqDvAZnPfG3ieJR025M78uhiwEPuvQ@mail.gmail.com> <CAHk-=wi-aTQx5-gD51QC6UWJYxQv1p1CnrPpfbn4X1S4AC7G-g@mail.gmail.com>
In-Reply-To: <CAHk-=wi-aTQx5-gD51QC6UWJYxQv1p1CnrPpfbn4X1S4AC7G-g@mail.gmail.com>
From:   Eric Dumazet <edumazet@google.com>
Date:   Fri, 8 Nov 2019 12:53:38 -0800
Message-ID: <CANn89iJh-WcvZYQEfdK=RGswQX8e1rp=CR27a6kWQkgK996P7g@mail.gmail.com>
Subject: Re: KCSAN: data-race in __alloc_file / __alloc_file
To:     Linus Torvalds <torvalds@linux-foundation.org>
Cc:     Eric Dumazet <eric.dumazet@gmail.com>,
        syzbot <syzbot+3ef049d50587836c0606@syzkaller.appspotmail.com>,
        Marco Elver <elver@google.com>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        syzkaller-bugs <syzkaller-bugs@googlegroups.com>,
        Al Viro <viro@zeniv.linux.org.uk>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Fri, Nov 8, 2019 at 12:30 PM Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
> On Fri, Nov 8, 2019 at 9:56 AM Eric Dumazet <edumazet@google.com> wrote:
> >
> > BTW, I would love an efficient ADD_ONCE(variable, value)
> >
> > Using WRITE_ONCE(variable, variable + value) is not good, since it can
> > not use the optimized instructions operating directly on memory.
>
> So I'm having a hard time seeing how this could possibly ever be valid.
>
> Is this a "writer is locked, readers are unlocked" case or something?

per cpu SNMP counters mostly, with no IRQ safety requirements.

Note that this could be implemented using local{64}_add() on arches like x86_64,
while others might have to fallback to WRITE_ONCE(variable, variable + add)

>
> Because we don't really have any sane way to do that any more
> efficiently, unless we'd have to add new architecture-specific
> functions for it (like we do have fo the percpu ops).
>
> Anyway, if you have a really hot case you care about, maybe you could
> convince the gcc people to just add it as a peephole optimization?
> Right now, gcc ends up doing some strange things with volatiles, and
> basically disables a lot of stuff over them. But with a test-case,
> maybe you can convince somebody that certain optimizations are still
> fine. A "read+add+write" really does the exact same accesses as an
> add-to-memory instruction, but gcc has some logic to disable that
> instruction fusion.
>
>           Linus