diff options
author | Tom de Vries <tdevries@suse.de> | 2022-02-10 11:26:16 +0100 |
---|---|---|
committer | Tom de Vries <tdevries@suse.de> | 2022-02-11 08:43:14 +0100 |
commit | fd64b09217fbe8fa33b559e61564071e8aca71e5 (patch) | |
tree | a73ba9f38e02cfbb6485b12afa071ed1738b3c67 /gcc/testsuite/c-c++-common/Warray-bounds-9.c | |
parent | a645583d4d68eecdd9101e1013e7bf01907ea786 (diff) | |
download | gcc-fd64b09217fbe8fa33b559e61564071e8aca71e5.tar.gz |
[nvptx] Handle asm insn in prevent_branch_around_nothing
With GOMP_NVPTX_JIT=-00 and -mptx=3.1, I run into:
...
FAIL: libgomp.oacc-c/../libgomp.oacc-c-c++-common/acc_prof-version-1.c \
-DACC_DEVICE_TYPE_nvidia=1 -DACC_MEM_SHARED=0 -foffload=nvptx-none -O2 \
execution test
...
The problem is that we're generating a diverging branch around nothing:
...
{
.reg.u32 %x;
mov.u32 %x, %tid.x;
setp.ne.u32 %r23, %x, 0;
}
@%r23 bra $L2;
$L2:
...
which the driver JIT has problems with at -O0, so consequently we run into the
nvptx_uniform_warp_check.
Fix this by handling asm ("") and alike in prevent_branch_around_nothing.
Tested on x86_64 with nvptx accelerator.
gcc/ChangeLog:
2022-02-10 Tom de Vries <tdevries@suse.de>
PR target/104456
* config/nvptx/nvptx.cc (prevent_branch_around_nothing): Handle asm
insn.
Diffstat (limited to 'gcc/testsuite/c-c++-common/Warray-bounds-9.c')
0 files changed, 0 insertions, 0 deletions