[debbugs-tracker] bug#16535: closed (Compiler bug ?)

[GNU bug Tracking System]

Your message dated Fri, 24 Jan 2014 15:35:21 -0800
with message-id <address@hidden>
and subject line Re: Compiler bug ?
has caused the debbugs.gnu.org bug report #16535,
regarding Compiler bug ?
to be marked as done.

(If you believe you have received this mail in error, please contact
address@hidden)


-- 
16535: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=16535
GNU Bug Tracking System
Contact address@hidden with problems

--- Begin Message --- Subject: Compiler bug ? Date: Fri, 24 Jan 2014 16:03:17 +0100

I hit a bug while compiling Gnu Emacs trunk development.

The following test :

 passed    17/1152  bool-vector-count-population-1-nil

fails unless I apply the following patch :

 === modified file 'src/data.c'

--- src/data.c  2014-01-03 06:47:27 +0000

+++ src/data.c  2014-01-24 14:59:16 +0000

@@ -3012,8 +3012,10 @@

     {

       int i = 0, count = 0;

       while (count += count_one_bits_ll (w),

-            BITS_PER_BITS_WORD <= (i += BITS_PER_ULL))

+            BITS_PER_BITS_WORD <= i) {

+       i += BITS_PER_ULL;

        w = shift_right_ull (w);

+      }

       return count;

     }

 }

This is with :

$ gcc --version

gcc.exe (Rev6, Built by MSYS2 project) 4.8.2

Copyright (C) 2013 Free Software Foundation, Inc.

This is free software; see the source for copying conditions.  There is NO

warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Given the nature of the patch, I don't think it is specific of th MinGW64 gcc version.

Regards,

Fabrice Popineau 

--- End Message ---

> --- Begin Message ---
>
>
>
> Subject: 
>
> Re: Compiler bug ?
>
>
>
>
> Date: 
>
> Fri, 24 Jan 2014 15:35:21 -0800
>
>
>
>
> User-agent: 
>
> Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0
>
>
>
>
> On 01/24/2014 03:06 PM, Fabrice Popineau wrote:
> > Actually, afaik BITS_PER_ULL == BITS_PER_BITS_WORD
>
> Ah, OK, thanks for following up. It looks like the negated-comparison 
> bug caused undefined behavior due to integer overflow in 
> count_one_bits_word; once that happens it's anybody's guess what 
> compilers will do; maybe you hit the jackpot. Anyway, the Emacs bug is 
> fixed (and there doesn't seem to have been a compiler bug) so I'll close 
> this report.
>
>
> --- End Message ---