https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61225
--- Comment #33 from Uroš Bizjak <ubizjak at gmail dot com> ---
Looking at PR49095, it seems to me that the remaining of this PR is just a
testsuite regression.
PR49095 introduces various peephole2 patterns that generate RMW versions of x86
insns, and fshrink-wrap interferes with these delicate code sequences.
Since the purpose of gcc.target/i386/pr49095.c is to test peepholes, we should
simply add -fno-shrink-wrap to the original testcase, like:
--cut here--
diff --git a/gcc/testsuite/gcc.target/i386/pr49095.c
b/gcc/testsuite/gcc.target/i386/pr49095.c
index 5fde08b..73758f8 100644
--- a/gcc/testsuite/gcc.target/i386/pr49095.c
+++ b/gcc/testsuite/gcc.target/i386/pr49095.c
@@ -1,7 +1,7 @@
/* PR rtl-optimization/49095 */
/* { dg-do compile } */
-/* { dg-options "-Os" } */
-/* { dg-options "-Os -mregparm=2" { target ia32 } } */
+/* { dg-options "-Os -fno-shrink-wrap" } */
+/* { dg-additional-options "-mregparm=2" { target ia32 } } */
void foo (void *);
@@ -70,5 +70,4 @@ G (short)
G (int)
G (long)
-/* See PR61225 for the XFAIL. */
-/* { dg-final { scan-assembler-not "test\[lq\]" { xfail { ia32 } } } } */
+/* { dg-final { scan-assembler-not "test\[lq\]" } } */
--cut here--
