tianshilei1992 added inline comments.
================
Comment at: openmp/libomptarget/test/api/omp_target_memcpy_async1.c:2
+// RUN: %libomptarget-compile-run-and-check-nvptx64-nvidia-cuda
+// REQUIRES: nvptx64-nvidia-cuda
+
----------------
Does it work on AMDGPU and other targets? Why does it require Nvidia here?
================
Comment at: openmp/libomptarget/test/api/omp_target_memcpy_rect_async3.c:21
+
+void timer_start(int i) {
+ if (i < 0 || i > num_timers)
----------------
If I read this test correctly, this test is more for performance evaluation
instead of correctness. We don't have any performance test cases yet, and I'm
not sure we need them right now.
================
Comment at: openmp/libomptarget/test/api/omp_target_memcpy_rect_async3.c:181
+ {
+#ifdef VERBOSE_TIME
+ t2 = omp_get_wtime();
----------------
If we don't profile the time, I don't see a point in keeping them.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D136103/new/
https://reviews.llvm.org/D136103
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
