This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug fortran/80765] New: [8 Regression] 178.galgel in SPEC CPU 2000 fails to run
- From: "hjl.tools at gmail dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Mon, 15 May 2017 15:25:09 +0000
- Subject: [Bug fortran/80765] New: [8 Regression] 178.galgel in SPEC CPU 2000 fails to run
- Auto-submitted: auto-generated
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80765
Bug ID: 80765
Summary: [8 Regression] 178.galgel in SPEC CPU 2000 fails to
run
Product: gcc
Version: 8.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: fortran
Assignee: unassigned at gcc dot gnu.org
Reporter: hjl.tools at gmail dot com
CC: tkoenig at gcc dot gnu.org
Target Milestone: ---
Target: i386
On Linux/i686, r247753 caused 178.galgel in SPEC CPU 2000 fails to run:
...
Solution by global spectral Galerkin method.
________________________________________________________________________________
Parameters:
Aspect ratio (length/height) = 4.0000000
Prandtl number = 0.15000000E-01
Number of basis functions in x-direction = 33
Number of basis functions in z-direction = 11
________________________________________________________________________________
Program received signal SIGSEGV, Segmentation fault.
0xf7ed2a02 in matmul_r8_avx2 (retarray=<optimized out>, a=<optimized out>,
b=<optimized out>, try_blas=<optimized out>, blas_limit=<optimized out>,
gemm=<optimized out>)
at ../../../../src-trunk/libgfortran/generated/matmul_r8.c:911
911 t1[l - ll + 1 + ((i - ii + 1) << 8) - 257] =
(gdb)
=> 0xf7ed2a02 <+2658>: vmovsd %xmm2,(%edi,%ecx,8)
(gdb) p $edi
$5 = 0
(gdb) p/x $ecx
$7 = 0x20000000
(gdb) p/x 0x20000000 * 8
$10 = 0x0