This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: RFA: patch to solve PR37535


On Tue, Sep 23, 2008 at 1:35 PM, Jeff Law <law@redhat.com> wrote:
> H.J. Lu wrote:
>>
>> On Tue, Sep 16, 2008 at 6:20 PM, Vladimir Makarov <vmakarov@redhat.com>
>> wrote:
>>
>>>
>>> The following patch solves the PR37535.  The analysis of the problem can
>>> be
>>> found on
>>>
>>> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37535
>>>
>>> The patch was successfully bootstraped on x86_64.
>>>
>>>
>>>
>>
>> Hi Vladimir,
>>
>> I tried your patch on ira-merge branch. It failed to boostrap
>> on Linux/ia32. I got
>>
>> Comparing stages 2 and 3
>> warning: ./cc1-checksum.o differs
>> warning: ./cc1plus-checksum.o differs
>> warning: ./cc1obj-checksum.o differs
>> Bootstrap comparison failure!
>> /fold-const.o differs
>> make[4]: *** [compare] Error 1
>>
>>
>>
>
> But weren't you seeing comparison failures prior to the patch for 37535?  If

No, I saw comparison failure after Vladamir's fix for PR 37535. It was OK
without the fix for PR 37535.

> we haven't fixed those (and it sounds like removing the uses of doubles from
> IRA didn't), then we shouldn't really be looking at 37535 as the root cause
> for the comparison failures.
>
> Jeff
>



-- 
H.J.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]