linux-kernel - Re: [PATCH v2] mm: kfence: Improve the performance of __kfence_alloc() and __kfence

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] [day] [month] [year] [list]

Date:   Tue, 4 Apr 2023 12:40:23 +0200
From:   Alexander Potapenko <glider@...gle.com>
To:     Peng Zhang <zhangpeng.00@...edance.com>
Cc:     elver@...gle.com, dvyukov@...gle.com, akpm@...ux-foundation.org,
        kasan-dev@...glegroups.com, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] mm: kfence: Improve the performance of
 __kfence_alloc() and __kfence_free()

> >
> >         /* Apply to left of object. */
> > -       for (addr = pageaddr; addr < meta->addr; addr++) {
> > -               if (!fn((u8 *)addr))
> > +       for (; meta->addr - addr >= sizeof(u64); addr += sizeof(u64)) {
> > +               if (unlikely(*((u64 *)addr) != KFENCE_CANARY_PATTERN_U64))
> >                         break;
> >         }
> I am confused. Right now this loop either runs from pageaddr to
> meta_addr if there's no corruption, or breaks at the first corrupted
> byte.
> Regardless of that, we are applying check_canary_byte() to every byte
> of that range in the following loop.
> Shouldn't the two be nested, like in the case of the canary bytes to
> the right of the object?
>
Please disregard my comment. This loop is fine, it just speeds up
finding the first corrupted byte.