* Fix cost of integer loads for generic model
@ 2018-01-02 9:37 Jan Hubicka
0 siblings, 0 replies; only message in thread
From: Jan Hubicka @ 2018-01-02 9:37 UTC (permalink / raw)
To: gcc-patches
Hi,
this patch fixes register allocation issue in spec2006 lbm benchmark improving it by
about 13%.
Bootstrapped/regtested x86_64-linux, will commit it soon.
Honza
* x86-tune-costs.h (generic_cost): Increase cost of integer load to match
other memory operations.
diff --git a/gcc/config/i386/x86-tune-costs.h b/gcc/config/i386/x86-tune-costs.h
index eff7f7f6bbf..4f1795ad8d3 100644
--- a/gcc/config/i386/x86-tune-costs.h
+++ b/gcc/config/i386/x86-tune-costs.h
@@ -2259,8 +2260,8 @@ struct processor_costs generic_cost = {
/* All move costs are relative to integer->integer move times 2 and thus
they are latency*2. */
- 4, /* cost for loading QImode using movzbl */
- {4, 4, 4}, /* cost of loading integer registers
+ 6, /* cost for loading QImode using movzbl */
+ {6, 6, 6}, /* cost of loading integer registers
in QImode, HImode and SImode.
Relative to reg-reg move (2). */
{6, 6, 6}, /* cost of storing integer registers */
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2018-01-02 9:37 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-01-02 9:37 Fix cost of integer loads for generic model Jan Hubicka
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).