]> granicus.if.org Git - llvm/commit
projects / llvm / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 0a25d0e) | patch
[AMDGPU] Run always inliner early in opt
authorStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Thu, 16 Mar 2017 16:11:46 +0000 (16:11 +0000)
committerStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Thu, 16 Mar 2017 16:11:46 +0000 (16:11 +0000)
commit1537930ea1c4e172fea6fa0767f1337860f82bd5
treefb265896acba5ad3d2fef65c6f3cf402106296dbtree | snapshot
parent0a25d0e2466f1f414b6c6cfed9aad603313fdf99commit | diff
[AMDGPU] Run always inliner early in opt

We can mark functions to always inline early in the opt. Since we do not have
call support this early inlining creates opportunities for inter-procedural
optimizations which would not occur otherwise.

Differential Revision: https://reviews.llvm.org/D31016

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@297958 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/AMDGPU/AMDGPUTargetMachine.cpp diff | blob | history
test/CodeGen/AMDGPU/early-inline.ll [new file with mode: 0644] blob
Unnamed repository; edit this file 'description' to name the repository.