This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug tree-optimization/52666] [4.8 Regression] FAIL: scal-to-vec1.c
- From: "gjl at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Thu, 22 Mar 2012 13:00:53 +0000
- Subject: [Bug tree-optimization/52666] [4.8 Regression] FAIL: scal-to-vec1.c
- Auto-submitted: auto-generated
- References: <bug-52666-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52666
--- Comment #2 from Georg-Johann Lay <gjl at gcc dot gnu.org> 2012-03-22 13:00:53 UTC ---
Created attachment 26954
--> http://gcc.gnu.org/bugzilla/attachment.cgi?id=26954
scal-to-vec1.c.149t.optimized
Dump created with
$ avr-gcc -mmcu=atmega128 -Os -fdump-tree-optimized
In the source from attachment 26953 there is
int main (int argc, char *argv[])
{
vector(8, short) v0 = {one, 1, 2, 3, 4, 5, 6, 7};
vector(8, short) v1;
v1 = 2 & v0; check (short, 8, v0, v1, 2, &, l);
return 0;
}
But the .optimized dump has:
<bb 51>:
v1 = { 0, 0, 0, 0, 0, 0, 0, 0 };
shouldn't there be non-zero values for the vector components that are 2 mod 4
or 3 mod 4?