User Details
User Details
- User Since
- Jan 31 2013, 4:40 AM (241 w, 6 d)
Recent Activity
Recent Activity
Today
Today
[OPENMP] Support for re-declarations when checking captured variables.
Yesterday
Yesterday
ABataev accepted D38049: [OpenMP] fix seg-faults printing diagnostics with invalid ordered(n) values.
LG
ABataev added a comment to D27846: [SLP] Support for horizontal min/max reduction.
ABataev updated the diff for D27846: [SLP] Support for horizontal min/max reduction.
Update after test update
Mon, Sep 18
Mon, Sep 18
ABataev updated the diff for D27846: [SLP] Support for horizontal min/max reduction.
Update after fixing PR34635
[SLP] Add a test for PR34635, NFC.
Wed, Sep 13
Wed, Sep 13
[OPENMP] Fix types for the target specific parameters in debug mode.
[OPENMP] Allow all classes as mappable types.
Tue, Sep 12
Tue, Sep 12
ABataev updated the diff for D29641: [SLP] Fix for PR31847: Assertion failed: (isLoopInvariant(Operands[i], L) && "SCEVAddRecExpr operand is not loop-invariant!").
Update after review
[SLP] Test with mutiple uses of conditional op and wrong parent.
[SLP] Fix for PHINode during horizontal reduction scanning, NFC.
Fri, Sep 8
Fri, Sep 8
[SLP] Fix the warning about paths not returning the value, NFC.
[SLP] Support for horizontal min/max reduction.
Wed, Sep 6
Wed, Sep 6
[OPENMP] Fix for PR33922: New ident_t flags for
ABataev committed rL312638: [OPENMP] Fix for PR34445: Reduction initializer segfaults at runtime in.
[OPENMP] Fix for PR34445: Reduction initializer segfaults at runtime in
Thu, Aug 31
Thu, Aug 31
ABataev committed rL312296: [OPENMP] Fix the test, NFC..
[OPENMP] Fix the test, NFC.
[OPENMP] Fix for PR34398: assert with random access iterator if the
Tue, Aug 29
Tue, Aug 29
ABataev committed rL312030: [SimplifyCFG] Fix for PR34219: Preserve alignment after merging conditional….
[SimplifyCFG] Fix for PR34219: Preserve alignment after merging conditional…
[OPENMP] Capture global variables in all target executable regions.
ABataev added inline comments to D36841: [SimplifyCFG] Fix for PR34219: Preserve alignment after merging conditional stores..
ABataev updated the diff for D36841: [SimplifyCFG] Fix for PR34219: Preserve alignment after merging conditional stores..
Update after review.
[SimplifyCFG] Update initial test for better testing of the fix for
Mon, Aug 28
Mon, Aug 28
ABataev added inline comments to D36841: [SimplifyCFG] Fix for PR34219: Preserve alignment after merging conditional stores..
[OPENMP] Remove unused header files, NFC.
ABataev updated the diff for D36841: [SimplifyCFG] Fix for PR34219: Preserve alignment after merging conditional stores..
Update after review
Fri, Aug 25
Fri, Aug 25
ABataev added inline comments to D36841: [SimplifyCFG] Fix for PR34219: Preserve alignment after merging conditional stores..
ABataev added inline comments to D36841: [SimplifyCFG] Fix for PR34219: Preserve alignment after merging conditional stores..
ABataev committed rL311777: [OPENMP] Fix for PR34321: ustom OpenMP reduction in C++ template causes.
[OPENMP] Fix for PR34321: ustom OpenMP reduction in C++ template causes
ABataev added inline comments to D36841: [SimplifyCFG] Fix for PR34219: Preserve alignment after merging conditional stores..
ABataev added a comment to D37129: [OpenMP] Capture global variables with combined/composite pragmas containing target.
Test?
Wed, Aug 23
Wed, Aug 23
Tue, Aug 22
Tue, Aug 22
ABataev committed rL311479: [OPENMP] Fix for PR34014: OpenMP 4.5: Target construct in static method.
[OPENMP] Fix for PR34014: OpenMP 4.5: Target construct in static method
Aug 17 2017
Aug 17 2017
ABataev added inline comments to D36837: [SLP] Fix for PR34219, part 1: Use minimal alignment for vectorized loads..
[SimplifyCFG] Add a test for preserve store alignment, NFC.
Aug 16 2017
Aug 16 2017
[OPENMP] Fix for PR28581: OpenMP linear clause - wrong results.
[OPRNMP] Fix for PR33445: ICE: OpenMP target containing ordered for.
Aug 15 2017
Aug 15 2017
[OPENMP] Fix compiler crash on argument translate for NVPTX.
Aug 14 2017
Aug 14 2017
ABataev updated the diff for D27846: [SLP] Support for horizontal min/max reduction.
Update after review
[OPENMP] Fix for PR33922: New ident_t flags for
ABataev committed rL310850: [OPENMP][DEBUG] Fix for PR33676: Debug info for OpenMP region is broken..
[OPENMP][DEBUG] Fix for PR33676: Debug info for OpenMP region is broken.
[OPENMP] Generalization of calls of the outlined functions.
Aug 10 2017
Aug 10 2017
ABataev added a comment to D30200: [SLP] Fix for PR31880: shuffle and vectorize repeated scalar ops on extracted elements.
Hi Dorit,
With this patch I tried to simulate what InstCombiner expects, i.e. I tried to synchronize the behavior of SLP vectorizer and InstCombiner. So, if you have some troubles with the cost, it means that it must be fixed for your target.
I don't think this is the right solution to calculate the gather cost here because InstCombiner does not treat these ops as gather. Just like I said, it is better to fix the cost model.
Aug 9 2017
Aug 9 2017
[OPENMP] Emit non-debug version of outlined functions with original
Aug 8 2017
Aug 8 2017
ABataev updated the diff for D27846: [SLP] Support for horizontal min/max reduction.
Update after review.
[OPENMP][DEBUG] Set proper address space info if required by target.
ABataev committed rL310379: Revert "[OPENMP][DEBUG] Set proper address space info if required by target.".
Revert "[OPENMP][DEBUG] Set proper address space info if required by target."
[OPENMP][DEBUG] Set proper address space info if required by target.
ABataev committed rL310364: Revert "[OPENMP][DEBUG] Set proper address space info if required by target.".
Revert "[OPENMP][DEBUG] Set proper address space info if required by target."
[OPENMP][DEBUG] Set proper address space info if required by target.
Aug 7 2017
Aug 7 2017
[SLP] General improvements of SLP vectorization process.
Revert "[SLP] General improvements of SLP vectorization process."
ABataev updated the diff for D27846: [SLP] Support for horizontal min/max reduction.
Update after review
[SLP] General improvements of SLP vectorization process.
Aug 4 2017
Aug 4 2017
ABataev committed rL310135: Revert "[OPENMP][DEBUG] Set proper address space info if required by target.".
Revert "[OPENMP][DEBUG] Set proper address space info if required by target."
Revert "[OPENMP] Fix for pacify buildbots, NFC."
[OPENMP] Fix for pacify buildbots, NFC.
[OPENMP][DEBUG] Set proper address space info if required by target.
[OPENMP] Unify generation of outlined function calls.
Aug 3 2017
Aug 3 2017
ABataev updated the diff for D29826: [SLP] General improvements of SLP vectorization process..
Update after review
ABataev updated the diff for D27846: [SLP] Support for horizontal min/max reduction.
Update to latest revision.
ABataev added a comment to D29641: [SLP] Fix for PR31847: Assertion failed: (isLoopInvariant(Operands[i], L) && "SCEVAddRecExpr operand is not loop-invariant!").
ABataev added inline comments to D29826: [SLP] General improvements of SLP vectorization process..
Aug 2 2017
Aug 2 2017
ABataev updated the diff for D29826: [SLP] General improvements of SLP vectorization process..
Update after review.
[SLPVectorizer] Generalize interface of functions, NFC.
[SLPVectorizer] Test update, NFC.
ABataev added a dependent revision for D29826: [SLP] General improvements of SLP vectorization process.: D27846: [SLP] Support for horizontal min/max reduction.
ABataev committed rL309812: [SLP] Fix for PR31880: shuffle and vectorize repeated scalar ops on extracted….
[SLP] Fix for PR31880: shuffle and vectorize repeated scalar ops on extracted…
Jul 31 2017
Jul 31 2017
ABataev committed rL309575: [OPENMP] Change the name of outer non-debug function in debug mode, NFC..
[OPENMP] Change the name of outer non-debug function in debug mode, NFC.
ABataev committed rL309566: [SLP] Initial rework for min/max horizontal reduction vectorization, NFC..
[SLP] Initial rework for min/max horizontal reduction vectorization, NFC.
[Cost] Rename getReductionCost() to getArithmeticReductionCost(), NFC.
Jul 28 2017
Jul 28 2017
ABataev committed rL309425: [SLP] Allow vectorization of the instruction from the same basic blocks only….
[SLP] Allow vectorization of the instruction from the same basic blocks only…
ABataev updated the diff for D34881: [SLP] Allow vectorization of the instruction from the same basic blocks only, NFC..
Update after review
ABataev added inline comments to D34881: [SLP] Allow vectorization of the instruction from the same basic blocks only, NFC..
ABataev updated the diff for D30200: [SLP] Fix for PR31880: shuffle and vectorize repeated scalar ops on extracted elements.
Update after review
ABataev updated the diff for D30200: [SLP] Fix for PR31880: shuffle and vectorize repeated scalar ops on extracted elements.
Update after review
ABataev added inline comments to D30200: [SLP] Fix for PR31880: shuffle and vectorize repeated scalar ops on extracted elements.
Jul 27 2017
Jul 27 2017
[OPENMP] Allow all lvalues in 'depend' clause.
ABataev updated the diff for D30200: [SLP] Fix for PR31880: shuffle and vectorize repeated scalar ops on extracted elements.
Update after review
[SLP] Outline code for the check that instruction users are part of
ABataev added inline comments to D30200: [SLP] Fix for PR31880: shuffle and vectorize repeated scalar ops on extracted elements.
ABataev updated the diff for D29402: [SLP] Initial rework for min/max horizontal reduction vectorization, NFC..
Update after review
ABataev added inline comments to D29402: [SLP] Initial rework for min/max horizontal reduction vectorization, NFC..