AMDGPU: Change insertion point of si_mask_branch
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Wed, 10 Aug 2016 19:11:42 +0000 (19:11 +0000)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Wed, 10 Aug 2016 19:11:42 +0000 (19:11 +0000)
commit57431c9680c8cde0f6cf0e85bed8507b307b080f
tree5886987b418c5df95368608f2a1f8ddc03240e07
parentb920e9987df2fa99dffa973c66c364b3675e2412
AMDGPU: Change insertion point of si_mask_branch

Insert before the skip branch if one is created.
This is a somewhat more natural placement relative
to the skip branches, and makes it possible to implement
analyzeBranch for skip blocks.

The test changes are mostly due to a quirk where
the block label is not emitted if there is a terminator
that is not also a branch.

llvm-svn: 278273
llvm/lib/Target/AMDGPU/SIInstructions.td
llvm/lib/Target/AMDGPU/SILowerControlFlow.cpp
llvm/test/CodeGen/AMDGPU/convergent-inlineasm.ll
llvm/test/CodeGen/AMDGPU/skip-if-dead.ll
llvm/test/CodeGen/AMDGPU/uniform-loop-inside-nonuniform.ll
llvm/test/CodeGen/AMDGPU/valu-i1.ll
llvm/test/CodeGen/AMDGPU/wqm.ll