[PR55547] fix alias regression on alpha on misaligned symbols (was: Re: do you have time to review this alpha P1 patch?)

[Uros Bizjak]

On Wed, Jan 16, 2013 at 8:33 AM, Uros Bizjak <ubizjak@gmail.com> wrote:

>>> On 01/15/2013 08:24 AM, Aldy Hernandez wrote:
>>>> Ok, it's really an alias.c bug, but it is Alpha, and aoliva has already
>>>> provided an unreviewed patch...
>>
>>>> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55547
>>
>>> The patch in #C4 is ok.
>>
>> Thanks, I'm checking it in (first patch below), but reviewing the logic
>> that uses negative sizes, I found a number of places that should use the
>> absolute value, and others in which being conservative about negative
>> sizes is unnecessary (e.g., when dealing with CONST_INT addresses).
>> That was implemented and regstrapped on x86_64-linux-gnu.  Uros, would
>> you give the second patch a spin on alpha to make sure it doesn't
>> regress?  Ok to install it?
>
> Thanks, I started a bootstrap/regtest run. If everything goes as
> expected, the results will be available in ~10h from now...

The results looks good [1], no regressions with patch.

[1] http://gcc.gnu.org/ml/gcc-testresults/2013-01/msg01706.html

Uros.