linux-kernel - Re: [PATCH] lib/div64: off by one in shift

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: <20190128173508.GD22143@redhat.com>
Date:   Mon, 28 Jan 2019 18:35:09 +0100
From:   Oleg Nesterov <oleg@...hat.com>
To:     Stanislaw Gruszka <sgruszka@...hat.com>
Cc:     linux-kernel@...r.kernel.org,
        Andrew Morton <akpm@...ux-foundation.org>,
        Brian Behlendorf <behlendorf1@...l.gov>,
        Siarhei Volkau <lis8215@...il.com>
Subject: Re: [PATCH] lib/div64: off by one in shift

On 01/28, Stanislaw Gruszka wrote:
>
> fls counts bits starting from 1 to 32 (returns 0 for zero argument).
> If we add 1 we shift right one bit more and loose precision

I forgot everything about this code, but I think this patch must be correct,

	divisor >> n;

should have MSB == 1 or we loose the precision... Heh, I managed to find the
initial version of this code, see

	https://lore.kernel.org/lkml/20101014121159.GA407@redhat.com/

and note that it uses __fls(), not fls()! I didn't notice the final version
replaced __fls() with fls() which is __fls() + 1 if arg != 0.

Acked-by: Oleg Nesterov <oleg@...hat.com>