linux-kernel - Re: [PATCH] block: switch to atomic

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:   Tue, 7 Dec 2021 09:45:29 -0800
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Christoph Hellwig <hch@...radead.org>,
        Jens Axboe <axboe@...nel.dk>,
        "linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
        Kees Cook <keescook@...omium.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] block: switch to atomic_t for request references

On Tue, Dec 7, 2021 at 9:43 AM Linus Torvalds
<torvalds@...ux-foundation.org> wrote:
>
> In the case of "dec_and_test" the "decrement to zero" case may not be
> hugely exceptional, but if you do the same for "increment with
> overflow protection" you do end up having the two different "zero vs
> too big", so it would actually be more consistent, I think..

Hmm.. Of course, that "increment with overflow protection" might not
actually want two targets in the first place, so maybe this argument
is BS.

Particularly if you can do the "zero or overflow" case with just one
test, maybe you really do want both cases to be one single error case.

And then your "refcount_dec_and_test()" with a return value (for
decrement to zero) and a separate error case (for errors) looks fine.

               Linus