Hahnfeld accepted this revision.
Hahnfeld added a comment.
In https://reviews.llvm.org/D52434#1248032, @gtbercea wrote:
> That is the intention. I just took out that part from here to not confuse
> things since I wanted to have that as a separate patch. :)
Ok, perfect. I was probably confused
gtbercea added a comment.
In https://reviews.llvm.org/D52434#1248012, @Hahnfeld wrote:
> Should we also change the default `schedule` to `static, 1`? I know that's
> not really needed for `teams distribute parallel for` (because the new
> default `dist_schedule` only leaves one iteration per
Hahnfeld added a comment.
Should we also change the default `schedule` to `static, 1`? I know that's not
really needed for `teams distribute parallel for` (because the new default
`dist_schedule` only leaves one iteration per thread), but this doesn't happen
for `target parallel for`.
ABataev accepted this revision.
ABataev added a comment.
This revision is now accepted and ready to land.
LG
Repository:
rC Clang
https://reviews.llvm.org/D52434
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
gtbercea updated this revision to Diff 167326.
gtbercea added a comment.
Fix type of chunk size.
Repository:
rC Clang
https://reviews.llvm.org/D52434
Files:
lib/CodeGen/CGOpenMPRuntime.h
lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
lib/CodeGen/CGOpenMPRuntimeNVPTX.h
ABataev added inline comments.
Comment at: lib/CodeGen/CGOpenMPRuntime.cpp:9199
+ OpenMPDistScheduleClauseKind *ScheduleKind, llvm::Value *) const {
+ return;
+}
Remove `return;`, it is not required
Comment at:
gtbercea updated this revision to Diff 167172.
gtbercea edited the summary of this revision.
gtbercea added a comment.
Only change default schedule for distribute directive.
Repository:
rC Clang
https://reviews.llvm.org/D52434
Files:
lib/CodeGen/CGOpenMPRuntime.cpp
ABataev added inline comments.
Comment at: lib/CodeGen/CGStmtOpenMP.cpp:2304
+ const auto *C = S.getSingleClause();
+ if (C) {
+// If schedule clause is present.
Restore original code here
Comment at:
gtbercea created this revision.
gtbercea added reviewers: ABataev, caomhin.
Herald added subscribers: cfe-commits, guansong, jholewinski.
For the OpenMP NVPTX toolchain choose default schedules which ensure coalescing
on the GPU when in SPMD mode. This significantly increases the performance of
