I noticed that as of commit 7cb252f60c1f600066300ede2781a2c9e3808adb
"libgomp: Add by-device capabilities for OpenMP 'omp requires' handling",
'libgomp.fortran/reverse-offload-5.f90' no longer XFAILs NVIDIA GPU offloading
execution, no matter whether non-USM or USM single-GPU system. That was
suspicious, given that this commit shouldn't cause any such difference in
behavior, per my understanding. Well, turns out, this commit indeed didn't
magically fix the failure, but it's just another case of non-existing static
type checking of flag values...
libgomp/
* target.c (gomp_target_init): Fix OpenMP reverse offload
capability checking.
---
libgomp/target.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libgomp/target.c b/libgomp/target.c
index 637601dc816..2abb364244e 100644
--- a/libgomp/target.c
+++ b/libgomp/target.c
@@ -6503,7 +6503,7 @@ gomp_target_init (void)
if (((omp_requires_mask & GOMP_REQUIRES_UNIFIED_ADDRESS)
&& !(dev_caps & GOMP_OFFLOAD_CAP_UNIFIED_ADDR))
|| ((omp_requires_mask & GOMP_REQUIRES_REVERSE_OFFLOAD)
- && !(dev_caps & GOMP_REQUIRES_REVERSE_OFFLOAD)))
+ && !(dev_caps & GOMP_OFFLOAD_CAP_REV_OFFLOAD)))
continue;
if (omp_requires_mask & (GOMP_REQUIRES_UNIFIED_SHARED_MEMORY
--
2.34.1
