Benjamin Kramer 44c76d239a Emit a more efficient magic number multiplication for exact sdivs. ... We have to do this in DAGBuilder instead of DAGCombiner, because the exact bit is lost after building. struct foo { char x[24]; }; long bar(struct foo *a, struct foo *b) { return a-b; } is now compiled into movl 4(%esp), %eax subl 8(%esp), %eax sarl $3, %eax imull $-1431655765, %eax, %eax instead of movl 4(%esp), %eax subl 8(%esp), %eax movl $715827883, %ecx imull %ecx movl %edx, %eax shrl $31, %eax sarl $2, %edx addl %eax, %edx movl %edx, %eax llvm-svn: 134695		2011-07-08 10:31:30 +00:00
..
Analysis	Preserve debug loc.	2011-07-05 21:48:22 +00:00
Archive
AsmParser	Replace the existing forms of ConstantArray::get() with a single form	2011-06-22 09:24:39 +00:00
Bitcode	type can be null	2011-07-07 05:29:18 +00:00
CodeGen	Emit a more efficient magic number multiplication for exact sdivs.	2011-07-08 10:31:30 +00:00
CompilerDriver
ExecutionEngine	Fix the ridiculous SubtargetFeatures API where it implicitly expects CPU name to	2011-06-30 01:53:36 +00:00
Linker
MC	Eliminate asm parser's dependency on TargetMachine:	2011-07-08 01:53:10 +00:00
Object	Use memcmp.	2011-07-05 20:28:00 +00:00
Support	Recognize mipseb as alias for mips for symmetry with mipsel.	2011-07-07 16:53:52 +00:00
Target	Eliminate asm parser's dependency on TargetMachine:	2011-07-08 01:53:10 +00:00
Transforms	Make GVN look through extractvalues for recognised intrinsics. GVN can then CSE ops that match values produced by the intrinsics.	2011-07-08 01:50:54 +00:00
VMCore	Introduce "expect" intrinsic instructions.	2011-07-06 18:22:43 +00:00
CMakeLists.txt
Makefile