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> <CANn89iLN758CpQPKcd++NLdj62LS-ekiEUV91VREzMsamLn9bw@mail.gmail.com>
In-Reply-To: <CANn89iLN758CpQPKcd++NLdj62LS-ekiEUV91VREzMsamLn9bw@mail.gmail.com>
From:   Linus Torvalds <torvalds@linux-foundation.org>
Date:   Fri, 8 Nov 2019 10:12:15 -0800
Message-ID: <CAHk-=wg5+N2WS2vO6doxfza1G8zk4DP5RhodiaSeAyf+Ooy7wQ@mail.gmail.com>
Subject: Re: KCSAN: data-race in __alloc_file / __alloc_file
To:     Eric Dumazet <edumazet@google.com>
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 10:02 AM Eric Dumazet <edumazet@google.com> wrote:
>
> Another interesting KCSAN report :
>
> static inline s64 percpu_counter_read_positive(struct percpu_counter *fbc)
> {
>         s64 ret = fbc->count;   // data-race ....

Yeah, I think that's fundamentally broken on 32-bit. It might need a
sequence lock instead of that raw_spinlock_t to be sanely done
properly on 32-bit.

Or we just admit that 32-bit doesn't really matter any more in the
long run, and that this is not a problem in practice because the
32-bit overflow basically never happens on small machines anyway.

               Linus