public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r13-3879] better PHI copy propagation for forwprop
@ 2022-11-10 14:19 Richard Biener
  0 siblings, 0 replies; only message in thread
From: Richard Biener @ 2022-11-10 14:19 UTC (permalink / raw)
  To: gcc-cvs

https://gcc.gnu.org/g:f1b76811f2c3773e8cabcc07932bf13e82e264db

commit r13-3879-gf1b76811f2c3773e8cabcc07932bf13e82e264db
Author: Richard Biener <rguenther@suse.de>
Date:   Thu Nov 10 15:02:37 2022 +0100

    better PHI copy propagation for forwprop
    
    We can handle _1 = PHI <_1, _2> as a copy.
    
            PR tree-optimization/84646
            * tree-ssa-forwprop.cc (pass_forwprop::execute): Improve
            copy propagation across PHIs.

Diff:
---
 gcc/tree-ssa-forwprop.cc | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/gcc/tree-ssa-forwprop.cc b/gcc/tree-ssa-forwprop.cc
index 4b693ef095c..7c7942600ef 100644
--- a/gcc/tree-ssa-forwprop.cc
+++ b/gcc/tree-ssa-forwprop.cc
@@ -3384,7 +3384,12 @@ pass_forwprop::execute (function *fun)
 	  FOR_EACH_PHI_ARG (use_p, phi, it, SSA_OP_USE)
 	    {
 	      tree use = USE_FROM_PTR (use_p);
-	      if (! first)
+	      if (use == res)
+		/* The PHI result can also appear on a backedge, if so
+		   we can ignore this case for the purpose of determining
+		   the singular value.  */
+		;
+	      else if (! first)
 		first = use;
 	      else if (! operand_equal_p (first, use, 0))
 		{

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2022-11-10 14:19 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-11-10 14:19 [gcc r13-3879] better PHI copy propagation for forwprop Richard Biener

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).