SuccessChanges

Summary

  1. [CVP] No-wrap deduction for `shl` (details)
  2. [lit] Remove redundancy from names and comments (details)
  3. AMDGPU: Select basic interp directly from intrinsics (details)
Commit 7cd7f4a83b7a4ab852734ad9188dde14dcfb5072 by lebedev.ri
[CVP] No-wrap deduction for `shl`
Summary: This is the last `OverflowingBinaryOperator` for which we don't
deduce flags. D69217 taught
`ConstantRange::makeGuaranteedNoWrapRegion()` about it.
The effect is better than of the `mul` patch (D69203):
| statistic                              |     old |     new | delta | %
change |
| correlated-value-propagation.NumAddNUW |    7145 |    7144 |    -1 |
-0.0140% |
| correlated-value-propagation.NumAddNW  |   12126 |   12125 |    -1 |
-0.0082% |
| correlated-value-propagation.NumAnd    |     443 |     446 |     3 |
0.6772% |
| correlated-value-propagation.NumNSW    |    5986 |    7158 |  1172 |
19.5790% |
| correlated-value-propagation.NumNUW    |   10512 |   13304 |  2792 |
26.5601% |
| correlated-value-propagation.NumNW     |   16498 |   20462 |  3964 |
24.0272% |
| correlated-value-propagation.NumShlNSW |       0 |    1172 |  1172 | 
       |
| correlated-value-propagation.NumShlNUW |       0 |    2793 |  2793 | 
       |
| correlated-value-propagation.NumShlNW  |       0 |    3965 |  3965 | 
       |
| instcount.NumAShrInst                  |   13824 |   13790 |   -34 |
-0.2459% |
| instcount.NumAddInst                   |  277584 |  277586 |     2 |
0.0007% |
| instcount.NumAndInst                   |   66061 |   66056 |    -5 |
-0.0076% |
| instcount.NumBrInst                    |  709153 |  709147 |    -6 |
-0.0008% |
| instcount.NumICmpInst                  |  483709 |  483708 |    -1 |
-0.0002% |
| instcount.NumSExtInst                  |   79497 |   79496 |    -1 |
-0.0013% |
| instcount.NumShlInst                   |   40691 |   40654 |   -37 |
-0.0909% |
| instcount.NumSubInst                   |   61997 |   61996 |    -1 |
-0.0016% |
| instcount.NumZExtInst                  |   68208 |   68211 |     3 |
0.0044% |
| instcount.TotalBlocks                  |  843916 |  843910 |    -6 |
-0.0007% |
| instcount.TotalInsts                   | 7387528 | 7387448 |   -80 |
-0.0011% |
Reviewers: nikic, reames, sanjoy, timshen
Reviewed By: nikic
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D69277
llvm-svn: 375455
The file was modifiedllvm/test/Transforms/CorrelatedValuePropagation/shl.ll (diff)
The file was modifiedllvm/test/Transforms/CorrelatedValuePropagation/icmp.ll (diff)
The file was modifiedllvm/lib/Transforms/Scalar/CorrelatedValuePropagation.cpp (diff)
Commit 8c6913a07bf5e8e4888380f7dfd6611b1bb1fff2 by jlettner
[lit] Remove redundancy from names and comments
llvm-svn: 375456
The file was modifiedllvm/utils/lit/lit/LitTestCase.py (diff)
The file was modifiedllvm/utils/lit/lit/run.py (diff)
The file was modifiedllvm/utils/lit/lit/worker.py (diff)
Commit ef9a0278f0ac3ccf5eb3bd5f8716a930685402e4 by Matthew.Arsenault
AMDGPU: Select basic interp directly from intrinsics
llvm-svn: 375457
The file was modifiedllvm/lib/Target/AMDGPU/AMDGPUISelLowering.h (diff)
The file was modifiedllvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp (diff)
The file was modifiedllvm/lib/Target/AMDGPU/SIInstructions.td (diff)
The file was modifiedllvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td (diff)
The file was modifiedllvm/lib/Target/AMDGPU/SIISelLowering.cpp (diff)