Evan Cheng 400073d546 On recent Intel u-arch's, folding loads into some unary SSE instructions can ... be non-optimal. To be precise, we should avoid folding loads if the instructions only update part of the destination register, and the non-updated part is not needed. e.g. cvtss2sd, sqrtss. Unfolding the load from these instructions breaks the partial register dependency and it can improve performance. e.g. movss (%rdi), %xmm0 cvtss2sd %xmm0, %xmm0 instead of cvtss2sd (%rdi), %xmm0 An alternative method to break dependency is to clear the register first. e.g. xorps %xmm0, %xmm0 cvtss2sd (%rdi), %xmm0 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@91672 91177308-0d34-0410-b5e6-96231b3b80d8		2009-12-18 07:40:29 +00:00
..
Alpha	Eliminate more uses of llvm-as and llvm-dis.	2009-09-09 00:09:15 +00:00
ARM	Handle ARM inline asm "w" constraints with 64-bit ("d") registers.	2009-12-18 01:03:29 +00:00
Blackfin	Move Blackfin intrinsics into the Target/Blackfin directory.	2009-10-15 18:50:52 +00:00
CBackend	Eliminate more uses of llvm-as and llvm-dis.	2009-09-09 00:09:15 +00:00
CellSPU	Revert this dag combine change:	2009-12-17 00:40:05 +00:00
CPP	fix PR5295 where the .ll parser didn't reject a function after a global	2009-10-25 23:22:50 +00:00
Generic	While this test is testing a problem in the generic part of codegen,	2009-11-27 16:04:14 +00:00
Mips	Support PIC loading of constant pool entries	2009-11-25 12:17:58 +00:00
MSP430	Lower setcc branchless, if this is profitable.	2009-12-11 23:01:29 +00:00
PIC16	While this test is testing a problem in the generic part of codegen,	2009-11-27 16:04:14 +00:00
PowerPC	Do better with physical reg operands (typically, from inline asm)	2009-12-16 00:29:41 +00:00
SPARC	Eliminate some redundant llvm-as calls.	2009-10-12 09:31:55 +00:00
SystemZ	Eliminate more uses of llvm-as and llvm-dis.	2009-09-09 00:09:15 +00:00
Thumb	Add test case for the phi reuse patch.	2009-12-18 00:11:44 +00:00
Thumb2	Make this test pass on Linux.	2009-12-16 07:35:25 +00:00
X86	On recent Intel u-arch's, folding loads into some unary SSE instructions can	2009-12-18 07:40:29 +00:00
XCore	Add XCore support for indirectbr / blockaddress.	2009-11-18 23:20:42 +00:00