This revision was automatically updated to reflect the committed changes.
Closed by commit rL345509: [OpenMP][NVPTX] Use single loops when generating
code for distribute parallel… (authored by gbercea, committed by ).
Herald added a subscriber: llvm-commits.
Changed prior to commit:
https://rev
gtbercea updated this revision to Diff 171499.
gtbercea added a comment.
Fix comment.
Repository:
rC Clang
https://reviews.llvm.org/D53448
Files:
include/clang/AST/StmtOpenMP.h
lib/AST/StmtOpenMP.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CGOpenMPRuntime.h
lib/CodeGen/CGOpenMP
ABataev accepted this revision.
ABataev added a comment.
This revision is now accepted and ready to land.
LG, with a nit
Comment at: lib/Sema/SemaOpenMP.cpp:5308
+
+// Build IV <= PrevEUB to be used in parallel for is in combination with
+// a distribute directive with
gtbercea updated this revision to Diff 171180.
gtbercea added a comment.
Change tests.
Repository:
rC Clang
https://reviews.llvm.org/D53448
Files:
include/clang/AST/StmtOpenMP.h
lib/AST/StmtOpenMP.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CGOpenMPRuntime.h
lib/CodeGen/CGOpenM
ABataev added inline comments.
Comment at: test/OpenMP/distribute_parallel_for_codegen.cpp:410
// LAMBDA-DAG: [[OMP_IV:%.omp.iv]] = alloca
+ // LAMBDA-DAG: [[OMP_CAPT_EXPR:%.capture_expr.1]] = alloca
// LAMBDA-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
gtbercea updated this revision to Diff 171170.
gtbercea added a comment.
Add test for collapse.
Repository:
rC Clang
https://reviews.llvm.org/D53448
Files:
include/clang/AST/StmtOpenMP.h
lib/AST/StmtOpenMP.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CGOpenMPRuntime.h
lib/CodeGe
gtbercea updated this revision to Diff 171127.
gtbercea added a comment.
Simplify code.
Repository:
rC Clang
https://reviews.llvm.org/D53448
Files:
include/clang/AST/StmtOpenMP.h
lib/AST/StmtOpenMP.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CGOpenMPRuntime.h
lib/CodeGen/CGOp
ABataev added a comment.
What about collapsed loops?
Comment at: lib/CodeGen/CGStmtOpenMP.cpp:3390
// UB = min(UB, GlobalUB);
-EmitIgnoredExpr(isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+EmitIgnoredExpr(StaticChunked ||
+
gtbercea updated this revision to Diff 171116.
gtbercea marked an inline comment as done.
gtbercea added a comment.
Refactor static chunk schedules. Fix tests.
Repository:
rC Clang
https://reviews.llvm.org/D53448
Files:
include/clang/AST/StmtOpenMP.h
lib/AST/StmtOpenMP.cpp
lib/CodeGe
ABataev added inline comments.
Comment at: lib/CodeGen/CGOpenMPRuntime.h:904
+ ///
+ virtual bool isStaticChunked(OpenMPDistScheduleClauseKind ScheduleKind,
+ bool Chunked) const;
gtbercea wrote:
> ABataev wrote:
> > I'd rename thi
gtbercea marked 3 inline comments as done.
gtbercea added inline comments.
Comment at: lib/CodeGen/CGOpenMPRuntime.h:904
+ ///
+ virtual bool isStaticChunked(OpenMPDistScheduleClauseKind ScheduleKind,
+ bool Chunked) const;
ABataev
gtbercea updated this revision to Diff 171084.
gtbercea marked an inline comment as done.
gtbercea added a comment.
Use NumIterations.
Repository:
rC Clang
https://reviews.llvm.org/D53448
Files:
include/clang/AST/StmtOpenMP.h
lib/AST/StmtOpenMP.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib
gtbercea updated this revision to Diff 171081.
gtbercea added a comment.
Refactor chunk one checking.
Repository:
rC Clang
https://reviews.llvm.org/D53448
Files:
include/clang/AST/StmtOpenMP.h
lib/AST/StmtOpenMP.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CGOpenMPRuntime.h
li
ABataev added inline comments.
Comment at: lib/CodeGen/CGOpenMPRuntime.h:904
+ ///
+ virtual bool isStaticChunked(OpenMPDistScheduleClauseKind ScheduleKind,
+ bool Chunked) const;
I'd rename this into `isDistStaticChunked`
==
gtbercea updated this revision to Diff 170241.
gtbercea added a comment.
Rebase.
Repository:
rC Clang
https://reviews.llvm.org/D53448
Files:
include/clang/AST/StmtOpenMP.h
include/clang/Basic/OpenMPKinds.h
lib/AST/StmtOpenMP.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CGOpenM
gtbercea created this revision.
gtbercea added reviewers: ABataev, caomhin.
Herald added subscribers: cfe-commits, guansong, jholewinski.
This patch adds a new code generation path for bound sharing directives
containing distribute parallel for. The new code generation scheme applies to
chunked
16 matches
Mail list logo