With LAST_UPDATED: "Mon Nov 15 19:30:07 UTC 2004" I get:
gcc.c-torture/execute/simd-4.c execution, -O0
gcc.c-torture/execute/simd-4.c execution, -O1
gcc.c-torture/execute/simd-4.c execution, -O2
gcc.c-torture/execute/simd-4.c execution, -O3 -fomit-frame-pointer
gcc.c-torture/execute/simd-4.c execution, -O3 -g
gcc.c-torture/execute/simd-4.c execution, -Os
With the message in gcc.log being:
*** EXIT code 4242
(a call to abort)
Last known to work on: "Mon Nov 15 13:25:54 UTC 2004".
Earliest known to fail with: "Mon Nov 15 19:30:07 UTC 2004".
Apparently the last change to that test has a bug.
It looks like an assumption that "long" is twice the size
of "long long", which is the same size for mmix-*-*
(and supposedly also some other LP64 targets).
The test needs to be tweaked, perhaps using __attribute__ ((__mode__ (SI))) for
the half-the-long-long types.
--
Summary: mmix-knuth-mmixware testsuite failure: gcc.c-
torture/execute/simd-4.c execution
Product: gcc
Version: 4.0.0
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: target
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: hp at gcc dot gnu dot org
CC: gcc-bugs at gcc dot gnu dot org
GCC host triplet: i686-pc-linux-gnu
GCC target triplet: mmix-knuth-mmixware
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18515
