Chris Lattner 474ed0aa9b Improve an integer select optimization in two ways: ... 1. generalize (select (x == 0), -1, 0) -> (sign_bit (x - 1)) to: (select (x == 0), -1, y) -> (sign_bit (x - 1)) | y 2. Handle the identical pattern that happens with !=: (select (x != 0), y, -1) -> (sign_bit (x - 1)) | y cmov is often high latency and can't fold immediates or memory operands. For example for (x == 0) ? -1 : 1, before we got: < testb %sil, %sil < movl $-1, %ecx < movl $1, %eax < cmovel %ecx, %eax now we get: > cmpb $1, %sil > sbbl %eax, %eax > orl $1, %eax llvm-svn: 120929		2010-12-05 01:23:24 +00:00
..
Analysis	Also inore '()' while creating mdnode name from ObjC symbol name.	2010-12-03 23:40:45 +00:00
Archive	Merge System into Support.	2010-11-29 18:16:10 +00:00
AsmParser	Add a new 'hotpatch' attribute. This attribute will insert a two-byte no-op	2010-10-25 15:37:09 +00:00
Bitcode	Generalize the darwin wrapper hack to work with generic macho triples as well as darwin ones.	2010-11-29 23:29:54 +00:00
CodeGen	Remove PHIElimination's private copy of SkipPHIsAndLabels.	2010-12-04 20:40:15 +00:00
CompilerDriver	Now to chant the magical incantation that will exorcise the System library	2010-11-29 19:44:50 +00:00
ExecutionEngine	Remove unneeded zero arrays.	2010-12-04 15:28:22 +00:00
Linker	Merge System into Support.	2010-11-29 18:16:10 +00:00
MC	Once the layout is done we don't need to keep updating which fragments are	2010-12-04 22:47:22 +00:00
Object	Merge System into Support.	2010-11-29 18:16:10 +00:00
Support	Silence 'may be used uninitialized in this function' warnings. Static analysis	2010-12-04 20:20:34 +00:00
Target	Improve an integer select optimization in two ways:	2010-12-05 01:23:24 +00:00
Transforms	PR5207: Rename overloaded APInt methods set(), clear(), flip() to	2010-12-01 08:53:58 +00:00
VMCore	Remove unneeded zero arrays.	2010-12-04 15:28:22 +00:00
Makefile	Add LLVMObject Library.	2010-11-15 03:21:41 +00:00