AbortedChanges

Summary

  1. Merging r311554: ------------------------------------------------------------------------ r311554 | mcrosier | 2017-08-23 07:10:06 -0700 (Wed, 23 Aug 2017) | 10 lines [Reassociate] Don't canonicalize x + (-Constant * y) -> x - (Constant * y).. ..if the resulting subtract will be broken up later. This can cause us to get into an infinite loop. x + (-5.0 * y) -> x - (5.0 * y) ; Canonicalize neg const x - (5.0 * y) -> x + (0 - (5.0 * y)) ; Break up subtract x + (0 - (5.0 * y)) -> x + (-5.0 * y) ; Replace 0-X with X*-1. PR34078 ------------------------------------------------------------------------
  2. Merging r311565: ------------------------------------------------------------------------ r311565 | hans | 2017-08-23 08:43:28 -0700 (Wed, 23 Aug 2017) | 8 lines LowerAtomic: Don't skip optnone functions; atomic still need lowering (PR34020) The lowering isn't really an optimization, so optnone shouldn't make a difference. ARM relies on the pass running when using "-mthread-model single", because in that mode, it doesn't run AtomicExpand. See bug for more details. Differential Revision: https://reviews.llvm.org/D37040 ------------------------------------------------------------------------
Revision 311603 by hans:
Merging r311554:
------------------------------------------------------------------------
r311554 | mcrosier | 2017-08-23 07:10:06 -0700 (Wed, 23 Aug 2017) | 10 lines

[Reassociate] Don't canonicalize x + (-Constant * y) -> x - (Constant * y)..

..if the resulting subtract will be broken up later.  This can cause us to get
into an infinite loop.

x + (-5.0 * y)      -> x - (5.0 * y)       ; Canonicalize neg const
x - (5.0 * y)       -> x + (0 - (5.0 * y)) ; Break up subtract
x + (0 - (5.0 * y)) -> x + (-5.0 * y)      ; Replace 0-X with X*-1.

PR34078
------------------------------------------------------------------------
Change TypePath in RepositoryPath in Workspace
The file was modified/llvm/branches/release_50llvm.src
The file was modified/llvm/branches/release_50/lib/Transforms/Scalar/Reassociate.cppllvm.src/lib/Transforms/Scalar/Reassociate.cpp
The file was modified/llvm/branches/release_50/test/Transforms/Reassociate/canonicalize-neg-const.llllvm.src/test/Transforms/Reassociate/canonicalize-neg-const.ll
Revision 311602 by hans:
Merging r311565:
------------------------------------------------------------------------
r311565 | hans | 2017-08-23 08:43:28 -0700 (Wed, 23 Aug 2017) | 8 lines

LowerAtomic: Don't skip optnone functions; atomic still need lowering (PR34020)

The lowering isn't really an optimization, so optnone shouldn't make a
difference. ARM relies on the pass running when using "-mthread-model
single", because in that mode, it doesn't run AtomicExpand. See bug for
more details.

Differential Revision: https://reviews.llvm.org/D37040
------------------------------------------------------------------------
Change TypePath in RepositoryPath in Workspace
The file was modified/llvm/branches/release_50llvm.src
The file was modified/llvm/branches/release_50/lib/Transforms/Scalar/LowerAtomic.cppllvm.src/lib/Transforms/Scalar/LowerAtomic.cpp
The file was modified/llvm/branches/release_50/test/Feature/optnone-opt.llllvm.src/test/Feature/optnone-opt.ll
The file was modified/llvm/branches/release_50/test/Transforms/LowerAtomic/atomic-swap.llllvm.src/test/Transforms/LowerAtomic/atomic-swap.ll