mirror/gcc
2
0
Fork 0
mirror of git://gcc.gnu.org/git/gcc.git synced 2025-04-10 18:20:51 +08:00
Code Issues Packages Projects Releases Wiki Activity

[nvptx, libgomp] Update pr85381-{2,4}.c test-cases

Browse Source 
After the fix for "PR tree-optimization/89713 - Assume loop with an exit is
finite" ( r272234 ) empty oacc loops are removed before expand.

Update pr85381-{2,4}.c accordingly.

2019-06-15  Tom de Vries  <tdevries@suse.de>

	PR tree-optimization/89713
	* testsuite/libgomp.oacc-c-c++-common/pr85381-2.c: Expect no bar.sync.
	* testsuite/libgomp.oacc-c-c++-common/pr85381-4.c: Same.

From-SVN: r272324
This commit is contained in:
Tom de Vries 2019-06-15 12:01:30 +00:00 committed by Tom de Vries
parent 9919f5fe87
commit 00908992f2
3 changed files with 8 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
libgomp/ChangeLog
View File

@ -1,3 +1,9 @@
2019-06-15 Tom de Vries <tdevries@suse.de>
PR tree-optimization/89713
* testsuite/libgomp.oacc-c-c++-common/pr85381-2.c: Expect no bar.sync.
* testsuite/libgomp.oacc-c-c++-common/pr85381-4.c: Same.
2019-06-15 Jakub Jelinek <jakub@redhat.com>
PR middle-end/90779

20
libgomp/testsuite/libgomp.oacc-c-c++-common/pr85381-2.c
View File

@ -15,22 +15,4 @@ main (void)
return 0;
}
/* Todo: Boths bar.syncs can be removed.
Atm we generate this dead code inbetween forked and joining:
mov.u32 %r28, %ntid.y;
mov.u32 %r29, %tid.y;
add.u32 %r30, %r29, %r29;
setp.gt.s32 %r31, %r30, 19;
@%r31 bra $L2;
add.u32 %r25, %r28, %r28;
mov.u32 %r24, %r30;
$L3:
add.u32 %r24, %r24, %r25;
setp.le.s32 %r33, %r24, 19;
@%r33 bra $L3;
$L2:
so the loop is not recognized as empty loop (which we detect by seeing if
joining immediately follows forked). */
/* { dg-final { scan-assembler-times "bar.sync" 2 } } */
/* { dg-final { scan-assembler-times "bar.sync" 0 } } */

5
libgomp/testsuite/libgomp.oacc-c-c++-common/pr85381-4.c
View File

@ -21,7 +21,4 @@ main (void)
return 0;
}
/* Atm, %ntid.y is broadcast from one loop to the next, so there are 2 bar.syncs
for that (the other two are there for the same reason as in pr85381-2.c).
Todo: Recompute %ntid.y instead of broadcasting it. */
/* { dg-final { scan-assembler-times "bar.sync" 4 } } */
/* { dg-final { scan-assembler-times "bar.sync" 0 } } */