projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 93ed1b3) | patch
[NVPTX] allow register copy between float and int
authorJingyue Wu <jingyue@google.com>
Sat, 1 Aug 2015 18:02:12 +0000 (18:02 +0000)
committerJingyue Wu <jingyue@google.com>
Sat, 1 Aug 2015 18:02:12 +0000 (18:02 +0000)
commit6ebf387fddffb87ce18c40fcca02568e8c6f645d
treed67047253bc22e45572d4ca962b140d71f7c8872tree | snapshot
parent93ed1b33882a110a5fe8dcd4988bc14249013287commit | diff
[NVPTX] allow register copy between float and int

Summary:
Fixes PR24303. With Bruno's WIP (D11197) on PeepholeOptimizer, across-class
register copying (e.g. i32 to f32) becomes possible. Enhance
NVPTXInstrInfo::copyPhysReg to handle these cases.

Reviewers: jholewinski

Subscribers: eliben, jholewinski, llvm-commits, bruno

Differential Revision: http://reviews.llvm.org/D11622

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@243839 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/NVPTX/NVPTXInstrInfo.cpp diff | blob | history
test/CodeGen/NVPTX/reg-copy.ll [new file with mode: 0644] blob
C/C++ LLVM-based compilers that forbids OOTA behaviors