llvm-mirror/test/CodeGen/Mips/2008-07-06-fadd64.ll

; RUN: llc -march=mips -mattr=single-float  < %s | FileCheck %s

define double @dofloat(double %a, double %b) nounwind {
entry:
; CHECK: __adddf3
	fadd double %a, %b		; <double>:0 [#uses=1]
	ret double %0
}
Re-enable some of the disabled tests. Use FileCheck instead of grep to check output. llvm-svn: 140263 2011-09-21 19:36:30 +02:00			`; RUN: llc -march=mips -mattr=single-float < %s \| FileCheck %s`
Batch 2 of Mips CodeGen tests llvm-svn: 54507 2008-08-08 06:03:25 +02:00
			`define double @dofloat(double %a, double %b) nounwind {`
			`entry:`
Re-enable some of the disabled tests. Use FileCheck instead of grep to check output. llvm-svn: 140263 2011-09-21 19:36:30 +02:00			`; CHECK: __adddf3`
Split the Add, Sub, and Mul instruction opcodes into separate integer and floating-point opcodes, introducing FAdd, FSub, and FMul. For now, the AsmParser, BitcodeReader, and IRBuilder all preserve backwards compatability, and the Core LLVM APIs preserve backwards compatibility for IR producers. Most front-ends won't need to change immediately. This implements the first step of the plan outlined here: http://nondot.org/sabre/LLVMNotes/IntegerOverflow.txt llvm-svn: 72897 2009-06-05 00:49:04 +02:00			`fadd double %a, %b ; <double>:0 [#uses=1]`
Batch 2 of Mips CodeGen tests llvm-svn: 54507 2008-08-08 06:03:25 +02:00			`ret double %0`
			`}`