This revision was automatically updated to reflect the committed changes.
Closed by commit rC350758: [OpenMP] Add flag for preventing the extension to 64
bits for the collapse loop… (authored by gbercea, committed by ).
Changed prior to commit:
https://reviews.llvm.org/D55928?vs=179078&id=18090
gtbercea marked an inline comment as done.
gtbercea added inline comments.
Comment at: docs/OpenMPSupport.rst:120
+compile your program with the `-fopenmp-optimistic-collapse`.
+
+
hfinkel wrote:
> Can you please clarify here what happens when the loop induction
hfinkel added inline comments.
Comment at: docs/OpenMPSupport.rst:120
+compile your program with the `-fopenmp-optimistic-collapse`.
+
+
Can you please clarify here what happens when the loop induction variables are
already 64 bits. If any of them are already 64
gtbercea added inline comments.
Comment at: docs/OpenMPSupport.rst:119
+time. To prevent this conservative choice and use at most 32 bits,
+compile your program with the `-fopenmp-max-32bit-collapse-width`.
+
kkwli0 wrote:
> ABataev wrote:
> > -fopenmp-optimistic
ABataev accepted this revision.
ABataev added a comment.
This revision is now accepted and ready to land.
LG
Repository:
rC Clang
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D55928/new/
https://reviews.llvm.org/D55928
___
cfe-commits ma
gtbercea updated this revision to Diff 179078.
gtbercea added a comment.
- Address comments.
Repository:
rC Clang
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D55928/new/
https://reviews.llvm.org/D55928
Files:
docs/OpenMPSupport.rst
include/clang/Basic/LangOptions.def
include/
ABataev added inline comments.
Comment at: lib/Driver/ToolChains/Clang.cpp:4429
Args.AddAllArgs(CmdArgs, options::OPT_fopenmp_cuda_blocks_per_sm_EQ);
+ Args.AddLastArg(CmdArgs, options::OPT_fopenmp_optimistic_collapse,
+ options::OPT_fno_openmp_op
gtbercea added inline comments.
Comment at: include/clang/Basic/LangOptions.def:210
LANGOPT(OpenMPCUDABlocksPerSM , 32, 0, "Number of blocks per SM for CUDA
devices.")
+LANGOPT(OpenMPMax32BitCollapseWidth , 1, 0, "Use at most 32 bits to represent
the collapsed loop nest coun
gtbercea updated this revision to Diff 179076.
gtbercea marked 4 inline comments as done.
gtbercea edited the summary of this revision.
gtbercea added a comment.
- Address comments.
Repository:
rC Clang
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D55928/new/
https://reviews.llvm.org
kkwli0 added inline comments.
Comment at: docs/OpenMPSupport.rst:119
+time. To prevent this conservative choice and use at most 32 bits,
+compile your program with the `-fopenmp-max-32bit-collapse-width`.
+
ABataev wrote:
> -fopenmp-optimistic-collapse
How about
ABataev added inline comments.
Comment at: docs/OpenMPSupport.rst:119
+time. To prevent this conservative choice and use at most 32 bits,
+compile your program with the `-fopenmp-max-32bit-collapse-width`.
+
-fopenmp-optimistic-collapse
Comment
gtbercea created this revision.
gtbercea added reviewers: ABataev, caomhin.
Herald added subscribers: cfe-commits, guansong.
Introduce a compiler flag for cases when the user knows that the collapsed loop
counter can be safely represented using at most 32 bits. This will prevent the
emission of
12 matches
Mail list logo