>The "r->x" alternative results in "vector" decoding on amdfam10. This is
>AMD-speak for microcoded instructions, and AMD optimization manual strongly
>recommends avoiding them. I have CC'd Ganesh, maybe he >can provide more
>relevant data on the performance impact.
Thanks Uros!
Yes, the AMD S
On September 23, 2014 5:33:35 PM CEST, Uros Bizjak wrote:
>On Tue, Sep 23, 2014 at 5:22 PM, Vladimir Makarov
>wrote:
>
You are right constrain_operands is not upto LRA possibilities and
>we should make the following change:
Index: recog.c
==
On Tue, Sep 23, 2014 at 5:22 PM, Vladimir Makarov wrote:
>>> You are right constrain_operands is not upto LRA possibilities and we
>>> should make the following change:
>>>
>>> Index: recog.c
>>> ===
>>> --- recog.c (revision 21
On Tue, Sep 23, 2014 at 5:02 PM, Uros Bizjak wrote:
>>> On Tue, Sep 23, 2014 at 3:26 AM, Vladimir Makarov
>>> wrote:
The previous patch to solve PR61360 fixed the problem in IRA (it was
easier for me to do as I know the code well)
https://gcc.gnu.org/bugzilla/show_bug.cgi?
On 09/23/2014 11:02 AM, Uros Bizjak wrote:
> On Tue, Sep 23, 2014 at 4:52 PM, Vladimir Makarov wrote:
>> On 09/23/2014 02:07 AM, Uros Bizjak wrote:
>>>
>> Uros, my patch does not result in PR60704 (I tested it before submitting
>> the patch).
> No, we didn't understand each other. The fix for PR60
On Tue, Sep 23, 2014 at 4:52 PM, Vladimir Makarov wrote:
> On 09/23/2014 02:07 AM, Uros Bizjak wrote:
>> On Tue, Sep 23, 2014 at 3:26 AM, Vladimir Makarov
>> wrote:
>>> The previous patch to solve PR61360 fixed the problem in IRA (it was
>>> easier for me to do as I know the code well)
>>>
>>>
On 09/23/2014 02:07 AM, Uros Bizjak wrote:
> On Tue, Sep 23, 2014 at 3:26 AM, Vladimir Makarov wrote:
>> The previous patch to solve PR61360 fixed the problem in IRA (it was
>> easier for me to do as I know the code well)
>>
>> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61360
>>
>> Although
On Tue, Sep 23, 2014 at 3:26 AM, Vladimir Makarov wrote:
> The previous patch to solve PR61360 fixed the problem in IRA (it was
> easier for me to do as I know the code well)
>
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61360
>
> Although imo it was an ok fix, Richard expressed concerns wi