Preston Gurd 4b0d66f924 Pad Short Functions for Intel Atom ... The current Intel Atom microarchitecture has a feature whereby when a function returns early then it is slightly faster to execute a sequence of NOP instructions to wait until the return address is ready, as opposed to simply stalling on the ret instruction until the return address is ready. When compiling for X86 Atom only, this patch will run a pass, called "X86PadShortFunction" which will add NOP instructions where less than four cycles elapse between function entry and return. It includes tests. This patch has been updated to address Nadav's review comments - Optimize only at >= O1 and don't do optimization if -Os is set - Stores MachineBasicBlock* instead of BBNum - Uses DenseMap instead of std::map - Fixes placement of braces Patch by Andy Zhang. llvm-svn: 171879		2013-01-08 18:27:24 +00:00
..
Analysis	Move CallGraphSCCPass.h into the Analysis tree; that's where the	2013-01-07 15:26:48 +00:00
Archive	Move all of the header files which are involved in modelling the LLVM IR	2013-01-02 11:36:10 +00:00
AsmParser	There was a switch fall-through in the parser for textual LLVM that caused	2013-01-07 13:32:38 +00:00
Bitcode	Move all of the header files which are involved in modelling the LLVM IR	2013-01-02 11:36:10 +00:00
CodeGen	Add fp128 rtlib function names to LLVM	2013-01-08 17:09:59 +00:00
DebugInfo	Whitespace and 80-col.	2013-01-07 22:40:48 +00:00
ExecutionEngine	[Object][ELF] Add a maximum alignment. This is used by createELFObjectFile to create a properly aligned reader.	2013-01-04 20:36:28 +00:00
IR	Remove the llvm-local DW_TAG_vector_type tag and add a test to	2013-01-08 01:53:52 +00:00
Linker	Move TypeFinder.h into the IR tree, it clearly belongs with the IR library.	2013-01-07 15:43:51 +00:00
MC	Rename statistic for instruction fragments -> relaxable	2013-01-08 17:41:59 +00:00
Object	[Object][ELF] Add a maximum alignment. This is used by createELFObjectFile to create a properly aligned reader.	2013-01-04 20:36:28 +00:00
Option	Resort the #include lines in include/... and lib/... with the	2013-01-02 10:22:59 +00:00
Support	Remove the llvm-local DW_TAG_vector_type tag and add a test to	2013-01-08 01:53:52 +00:00
TableGen	Revert r171140. We don't actually need to support #NAME. Because NAME by itself is interpreted just fine.	2013-01-07 05:09:33 +00:00
Target	Pad Short Functions for Intel Atom	2013-01-08 18:27:24 +00:00
Transforms	Code cleanup: refactor the switch statements in the generation of reduction variables into an IR builder call.	2013-01-08 17:37:45 +00:00
CMakeLists.txt	Rename VMCore directory to IR.	2013-01-02 09:10:48 +00:00
LLVMBuild.txt	Rename VMCore directory to IR.	2013-01-02 09:10:48 +00:00
Makefile	Rename VMCore directory to IR.	2013-01-02 09:10:48 +00:00