* Why is REG_ALLOC_ORDER not defined on Aarch64
@ 2018-05-25 22:36 Steve Ellcey
2018-05-25 22:41 ` Andrew Pinski
0 siblings, 1 reply; 3+ messages in thread
From: Steve Ellcey @ 2018-05-25 22:36 UTC (permalink / raw)
To: gcc
I was curious if there was any reason that REG_ALLOC_ORDER is not
defined for Aarch64.  Has anyone tried this to see if it could help
performance? Â It is defined for many other platforms.
Steve Ellcey
sellcey@cavium.com
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Why is REG_ALLOC_ORDER not defined on Aarch64
2018-05-25 22:36 Why is REG_ALLOC_ORDER not defined on Aarch64 Steve Ellcey
@ 2018-05-25 22:41 ` Andrew Pinski
2018-05-26 9:39 ` Richard Sandiford
0 siblings, 1 reply; 3+ messages in thread
From: Andrew Pinski @ 2018-05-25 22:41 UTC (permalink / raw)
To: sellcey; +Cc: gcc
On Fri, May 25, 2018 at 3:35 PM, Steve Ellcey <sellcey@cavium.com> wrote:
> I was curious if there was any reason that REG_ALLOC_ORDER is not
> defined for Aarch64. Has anyone tried this to see if it could help
> performance? It is defined for many other platforms.
https://gcc.gnu.org/ml/gcc-patches/2015-07/msg01815.html
https://gcc.gnu.org/ml/gcc-patches/2015-07/msg01822.html
>
> Steve Ellcey
> sellcey@cavium.com
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Why is REG_ALLOC_ORDER not defined on Aarch64
2018-05-25 22:41 ` Andrew Pinski
@ 2018-05-26 9:39 ` Richard Sandiford
0 siblings, 0 replies; 3+ messages in thread
From: Richard Sandiford @ 2018-05-26 9:39 UTC (permalink / raw)
To: Andrew Pinski; +Cc: sellcey, gcc
Andrew Pinski <pinskia@gmail.com> writes:
> On Fri, May 25, 2018 at 3:35 PM, Steve Ellcey <sellcey@cavium.com> wrote:
>> I was curious if there was any reason that REG_ALLOC_ORDER is not
>> defined for Aarch64. Has anyone tried this to see if it could help
>> performance? It is defined for many other platforms.
>
> https://gcc.gnu.org/ml/gcc-patches/2015-07/msg01815.html
> https://gcc.gnu.org/ml/gcc-patches/2015-07/msg01822.html
It looks like the immediate reason for reverting was the effect of
listing the argument registers in reverse order. I wonder how much that
actually helps with IRA and LRA? They track per-register costs, and
would be able to increase the cost of a pseudo that conflicts with a
hard-register call argument.
It just felt like it might have been a "best practice" idea passed down
from the old local.c and global.c days.
Thanks,
Richard
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2018-05-26 9:39 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-05-25 22:36 Why is REG_ALLOC_ORDER not defined on Aarch64 Steve Ellcey
2018-05-25 22:41 ` Andrew Pinski
2018-05-26 9:39 ` Richard Sandiford
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).