llvm/test/CodeGen/X86/combine-lds.ll

; RUN: llc < %s -march=x86 -mattr=+sse2 | grep fldl | count 1

define double @doload64(i64 %x) nounwind  {
	%tmp717 = bitcast i64 %x to double
	ret double %tmp717
}
Eliminate more uses of llvm-as and llvm-dis. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@81290 91177308-0d34-0410-b5e6-96231b3b80d8 2009-09-08 23:54:48 +00:00			`; RUN: llc < %s -march=x86 -mattr=+sse2 \| grep fldl \| count 1`
Xform bitconvert(build_pair(load a, load b)) to a single load if the load locations are at the right offset from each other. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@51008 91177308-0d34-0410-b5e6-96231b3b80d8 2008-05-12 23:04:07 +00:00
For whatever the reason, x86 CallingConv::Fast (i.e. fastcc) was not passing scalar arguments in registers. This patch defines a new fastcc CC which is slightly different from the FastCall CC. In addition to passing integer arguments in ECX and EDX, it also specify doubles are passed in 8-byte slots which are 8-byte aligned (instead of 4-byte aligned). This avoids a potential performance hazard where doubles span cacheline boundaries. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55807 91177308-0d34-0410-b5e6-96231b3b80d8 2008-09-04 22:59:58 +00:00			`define double @doload64(i64 %x) nounwind {`
Xform bitconvert(build_pair(load a, load b)) to a single load if the load locations are at the right offset from each other. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@51008 91177308-0d34-0410-b5e6-96231b3b80d8 2008-05-12 23:04:07 +00:00			`%tmp717 = bitcast i64 %x to double`
			`ret double %tmp717`
			`}`