diff mbox

[PATCH/AARCH64] Improve ThunderX2 T99 address cost model

Message ID CA+=Sn1mUG3V3AO0dQ=mmwxgP9NcZdxv2AA3Uo3T6yPMxoHB5=Q@mail.gmail.com
State New
Headers show

Commit Message

Andrew Pinski Jan. 23, 2017, 11:35 p.m. UTC 
Hi,
  On ThunderX 2, the logical shifts that are part of the address cause
a 1 cycle extra and this is not modeled correctly.  For induction
variables we don't want to do the shift.  When I change the cost for
the shift of the addresses, I get a 12% improvement on HMMER; all over
benchmarks in SPEC CPU 2006 were neutral.

OK? Bootstrapped and tested on aarch64-linux-gnu with no regressions.

Thanks,
Andrew Pinski

ChangeLog:
* config/aarch64/aarch64.c (thunderx2t99_addrcost_table): Improve cost table.

Comments

Richard Earnshaw Jan. 23, 2017, 11:59 p.m. UTC | #1 
On 23/01/17 23:35, Andrew Pinski wrote:
> Hi,
>   On ThunderX 2, the logical shifts that are part of the address cause
> a 1 cycle extra and this is not modeled correctly.  For induction
> variables we don't want to do the shift.  When I change the cost for
> the shift of the addresses, I get a 12% improvement on HMMER; all over
> benchmarks in SPEC CPU 2006 were neutral.
> 
> OK? Bootstrapped and tested on aarch64-linux-gnu with no regressions.
> 
> Thanks,
> Andrew Pinski
> 
> ChangeLog:
> * config/aarch64/aarch64.c (thunderx2t99_addrcost_table): Improve cost table.
> 

OK.

R.
diff mbox

Patch

Index: gcc/config/aarch64/aarch64.c
===================================================================
--- gcc/config/aarch64/aarch64.c	(revision 244839)
+++ gcc/config/aarch64/aarch64.c	(working copy)
@@ -273,9 +273,9 @@  static const struct cpu_addrcost_table q
 static const struct cpu_addrcost_table thunderx2t99_addrcost_table =
 {
     {
-      0, /* hi  */
-      0, /* si  */
-      0, /* di  */
+      1, /* hi  */
+      1, /* si  */
+      1, /* di  */
       2, /* ti  */
     },
   0, /* pre_modify  */