This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] Fix crashes on invalid input in IBM gconv modules [BZ #17325]
- From: Allan McRae <allan at archlinux dot org>
- To: Florian Weimer <fweimer at redhat dot com>, Andreas Schwab <schwab at linux-m68k dot org>
- Cc: GNU C Library <libc-alpha at sourceware dot org>, azanella at linux dot vnet dot ibm dot com, Roland McGrath <roland at hack dot frob dot com>
- Date: Mon, 01 Sep 2014 10:33:19 +1000
- Subject: Re: [PATCH] Fix crashes on invalid input in IBM gconv modules [BZ #17325]
- Authentication-results: sourceware.org; auth=none
- References: <54006E57 dot 8030800 at redhat dot com> <87y4u7o1yr dot fsf at igel dot home> <54008EF8 dot 7030507 at redhat dot com>
On 30/08/14 00:32, Florian Weimer wrote:
> On 08/29/2014 03:33 PM, Andreas Schwab wrote:
>> Florian Weimer <fweimer@redhat.com> writes:
>>
>>> + if (__builtin_expect (rp2->start == 0xffff, 0) \
>>
>> Please use either this
>>
>>> + if (__builtin_expect (ch == 0xffff, 0) \
>>
>> or this consistently.
>
> Oops, you are right. I went with the first variant, as in the original
> fix.
>
> Retested successfully.
>
> Allan, is this okay for master at this stage?
>
I am OK with this, once there is an ack to use __builtin_expect instead
of __glibc_likely.
I agree with you that the cleanup of all the __builtin_expect should
happen in a following patch so the change is minimal at this stage and
easy to backport.
Allan