Chris Lattner 72dd9bdcc5 Lower v8i16 multiply into this code: ... li r5, lo16(LCPI1_0) lis r6, ha16(LCPI1_0) lvx v4, r6, r5 vmulouh v5, v3, v2 vmuleuh v2, v3, v2 vperm v2, v2, v5, v4 where v4 is: LCPI1_0: ; <16 x ubyte> .byte 2 .byte 3 .byte 18 .byte 19 .byte 6 .byte 7 .byte 22 .byte 23 .byte 10 .byte 11 .byte 26 .byte 27 .byte 14 .byte 15 .byte 30 .byte 31 This is 5.07x faster on the G5 (measured) than lowering to scalar code + loads/stores. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27789 91177308-0d34-0410-b5e6-96231b3b80d8		2006-04-18 03:43:48 +00:00
..
Analysis	Implement value #'ing for vector operations, implementing	2006-04-14 05:10:20 +00:00
Archive	more C++ daintiness	2005-12-26 14:31:26 +00:00
AsmParser	Make sure CVS versions of yacc and lex files get distributed.	2006-04-12 20:57:05 +00:00
Bytecode	use isValidOperands instead of duplicating checks	2006-04-08 04:09:19 +00:00
CodeGen	Fix handling of calls in functions that use vectors. This fixes a crash on	2006-04-17 22:10:08 +00:00
Debugger	Add the README files to the distribution.	2006-04-13 06:39:24 +00:00
ExecutionEngine	Get JIT/Interpreter working on Windows again.	2006-03-24 02:53:49 +00:00
Linker	Add shufflevector support	2006-04-08 01:19:47 +00:00
Support	Qualify dwarf namespace inside llvm namespace.	2006-02-27 22:37:23 +00:00
System	Add checks for __OpenBSD__.	2006-04-17 17:55:41 +00:00
Target	Lower v8i16 multiply into this code:	2006-04-18 03:43:48 +00:00
Transforms	Turn x86 unaligned load/store intrinsics into aligned load/store instructions	2006-04-17 22:26:56 +00:00
VMCore	Move these ctors out of line	2006-04-14 22:20:32 +00:00
Makefile