llvm-for-llvmta/test/CodeGen/Mips/tail16.ll

21 lines
525 B
LLVM

; RUN: llc -mtriple=mipsel-linux-gnu -march=mipsel -mattr=mips16 -mattr=+soft-float -mips16-hard-float -relocation-model=pic < %s | FileCheck %s
; Function Attrs: nounwind optsize
define float @h() {
entry:
%call = tail call float bitcast (float (...)* @g to float ()*)()
ret float %call
; CHECK: .ent h
; CHECK: save $16, $ra, $18, 32
; CHECK: lw ${{[0-9]+}}, %got(__mips16_call_stub_sf_0)(${{[0-9]+}})
; CHECK: restore $16, $ra, $18, 32
; CHECK: .end h
}
; Function Attrs: optsize
declare float @g(...)