Hi!
In addition to the technical issues pointed out by Jakub for this og12
commit:
On 2022-07-12T15:16:35+0100, Andrew Stubbs wrote:
> This patch [...]
> I will commit a backport to OG12 shortly.
> openmp: fix max_vf setting for amdgcn offloading
> --- a/gcc/omp-general.h
> +++ b/gcc/omp-gene
On Tue, Jul 12, 2022 at 03:16:35PM +0100, Andrew Stubbs wrote:
> --- a/gcc/gimple-loop-versioning.cc
> +++ b/gcc/gimple-loop-versioning.cc
> @@ -555,7 +555,10 @@ loop_versioning::loop_versioning (function *fn)
> unvectorizable code, since it is the largest size that can be
> handled eff
This patch ensures that the maximum vectorization factor used to set the
"safelen" attribute on "omp simd" constructs is suitable for all the
configured offload devices.
Right now it makes the proper adjustment for NVPTX, but otherwise just
uses a value suitable for the host system (always x86
