Benjamin Kramer 59957500f9 CodeGenPrepare: Add a transform to turn selects into branches in some cases. ... This came up when a change in block placement formed a cmov and slowed down a hot loop by 50%: ucomisd (%rdi), %xmm0 cmovbel %edx, %esi cmov is a really bad choice in this context because it doesn't get branch prediction. If we emit it as a branch, an out-of-order CPU can do a better job (if the branch is predicted right) and avoid waiting for the slow load+compare instruction to finish. Of course it won't help if the branch is unpredictable, but those are really rare in practice. This patch uses a dumb conservative heuristic, it turns all cmovs that have one use and a direct memory operand into branches. cmovs usually save some code size, so we disable the transform in -Os mode. In-Order architectures are unlikely to benefit as well, those are included in the "predictableSelectIsExpensive" flag. It would be better to reuse branch probability info here, but BPI doesn't support select instructions currently. It would make sense to use the same heuristics as the if-converter pass, which does the opposite direction of this transform. Test suite shows a small improvement here and there on corei7-level machines, but the actual results depend a lot on the used microarchitecture. The transformation is currently disabled by default and available by passing the -enable-cgp-select2branch flag to the code generator. Thanks to Chandler for the initial test case to him and Evan Cheng for providing me with comments and test-suite numbers that were more stable than mine :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@156234 91177308-0d34-0410-b5e6-96231b3b80d8		2012-05-05 12:49:22 +00:00
..
Analysis	Rename the Region::block_iterator to Region::block_node_iterator, and	2012-05-04 20:55:23 +00:00
Archive	Include cstdio in a few place that depended on getting it transitively through StringExtras.h	2012-03-23 11:35:30 +00:00
AsmParser
Bitcode	llvm::SwitchInst	2012-03-11 06:09:17 +00:00
CodeGen	Add a new target hook "predictableSelectIsExpensive".	2012-05-05 12:49:14 +00:00
DebugInfo	Fix a C++11 UDL conflict.	2012-04-04 20:33:56 +00:00
ExecutionEngine	RuntimeDyld cleanup:	2012-05-01 10:41:12 +00:00
Linker	Add a flag to the struct type finder to collect only those types which have	2012-04-21 23:59:16 +00:00
MC	MC: Unknown assembler directives are now hard errors.	2012-05-01 18:38:27 +00:00
Object	Fix the implementation of MachOObjectFile::isSectionZeroInit so it follows the MachO spec.	2012-05-02 02:31:28 +00:00
Support	This patch adds a new NVPTX back-end to LLVM which supports code generation for NVIDIA PTX 3.0. This back-end will (eventually) replace the current PTX back-end, while maintaining compatibility with it.	2012-05-04 20:18:50 +00:00
TableGen	Move llvm-tblgen's StringMatcher into the TableGen library so it can	2012-05-02 17:32:48 +00:00
Target	Add a new target hook "predictableSelectIsExpensive".	2012-05-05 12:49:14 +00:00
Transforms	CodeGenPrepare: Add a transform to turn selects into branches in some cases.	2012-05-05 12:49:22 +00:00
VMCore	Reapply r155682, making constant folding more consistent, with a fix to work	2012-04-27 17:50:22 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile