mirror of
https://github.com/RPCS3/llvm.git
synced 2024-11-25 21:00:00 +00:00
db6cbe1ff1
add <rd>, sp, #<imm8> ldr <rd>, [sp, #<imm8>] When the offset from sp is multiple of 4 and in range of 0-1020. This saves code size by utilizing 16-bit instructions. rdar://9321541 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129971 91177308-0d34-0410-b5e6-96231b3b80d8
24 lines
601 B
LLVM
24 lines
601 B
LLVM
; RUN: llc < %s -mtriple=thumbv7-apple-darwin | FileCheck %s
|
|
|
|
; Use sp, #imm to lower frame indices when the offset is multiple of 4
|
|
; and in the range of 0-1020. This saves code size by utilizing
|
|
; 16-bit instructions.
|
|
; rdar://9321541
|
|
|
|
define i32 @t() nounwind {
|
|
entry:
|
|
; CHECK: t:
|
|
; CHECK: sub sp, #12
|
|
; CHECK-NOT: sub
|
|
; CHECK: add r0, sp, #4
|
|
; CHECK: add r1, sp, #8
|
|
; CHECK: mov r2, sp
|
|
%size = alloca i32, align 4
|
|
%count = alloca i32, align 4
|
|
%index = alloca i32, align 4
|
|
%0 = call i32 @foo(i32* %count, i32* %size, i32* %index) nounwind
|
|
ret i32 %0
|
|
}
|
|
|
|
declare i32 @foo(i32*, i32*, i32*)
|