diff options
| author | Tom de Vries <tdevries@suse.de> | 2019-01-07 10:01:49 +0000 |
|---|---|---|
| committer | Tom de Vries <vries@gcc.gnu.org> | 2019-01-07 10:01:49 +0000 |
| commit | 6e723923df52bdde7c3aebd6bccaae3c7bff19ee (patch) | |
| tree | 56290fe75d9fead37fbc9cb053d1e010cb3120f1 /gcc/fortran/iresolve.c | |
| parent | d471bdec413c333f2067c0723c61ad045f3589e3 (diff) | |
| download | gcc-6e723923df52bdde7c3aebd6bccaae3c7bff19ee.zip gcc-6e723923df52bdde7c3aebd6bccaae3c7bff19ee.tar.gz gcc-6e723923df52bdde7c3aebd6bccaae3c7bff19ee.tar.bz2 | |
[nvptx] Force vl32 if calling vector-partitionable routines
With PTX_MAX_VECTOR_LENGTH set to larger than PTX_WARP_SIZE, routines can be
called from offloading regions with vector-size set to larger than warp size.
OTOH, vector-partitionable routines assume warp-sized vector length.
Detect if we're calling a vector-partitionable routine from an offloading
region, and if so, fall back to warp-sized vector length in that region.
2019-01-07 Tom de Vries <tdevries@suse.de>
PR target/85486
* config/nvptx/nvptx.c (has_vector_partitionable_routine_calls_p): New
function.
(nvptx_goacc_validate_dims): Force vl32 if calling vector-partitionable
routines.
From-SVN: r267640
Diffstat (limited to 'gcc/fortran/iresolve.c')
0 files changed, 0 insertions, 0 deletions