T0b1/wasmtime
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Cranelift: Simplify leaf functions that do not use the stack (#2960)

Browse Source 
* Cranelift AArch64: Simplify leaf functions that do not use the stack

Leaf functions that do not use the stack (e.g. do not clobber any
callee-saved registers) do not need a frame record.

Copyright (c) 2021, Arm Limited.
This commit is contained in:
Anton Kirilov
2021-08-27 11:12:37 +01:00
committed by GitHub
parent 12515e6646
commit 7b98be1bee
34 changed files with 650 additions and 1385 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
cranelift/filetests/filetests/isa/aarch64/simd-pairwise-add.clif
View File

@@ -11,10 +11,7 @@ block0(v0: i8x16):
return v3
}
; check: stp fp
; nextln: mov fp, sp
; nextln: saddlp v0.8h, v0.16b
; nextln: ldp fp, lr, [sp], #16
; check: saddlp v0.8h, v0.16b
; nextln: ret
function %fn2(i8x16) -> i16x8 {
@@ -25,10 +22,7 @@ block0(v0: i8x16):
return v3
}
; check: stp fp
; nextln: mov fp, sp
; nextln: uaddlp v0.8h, v0.16b
; nextln: ldp fp, lr, [sp], #16
; check: uaddlp v0.8h, v0.16b
; nextln: ret
function %fn3(i16x8) -> i32x4 {
@@ -39,10 +33,7 @@ block0(v0: i16x8):
return v3
}
; check: stp fp
; nextln: mov fp, sp
; nextln: saddlp v0.4s, v0.8h
; nextln: ldp fp, lr, [sp], #16
; check: saddlp v0.4s, v0.8h
; nextln: ret
function %fn4(i16x8) -> i32x4 {
@@ -53,10 +44,7 @@ block0(v0: i16x8):
return v3
}
; check: stp fp
; nextln: mov fp, sp
; nextln: uaddlp v0.4s, v0.8h
; nextln: ldp fp, lr, [sp], #16
; check: uaddlp v0.4s, v0.8h
; nextln: ret
function %fn5(i8x16, i8x16) -> i16x8 {
@@ -67,12 +55,9 @@ block0(v0: i8x16, v1: i8x16):
return v4
}
; check: stp fp
; nextln: mov fp, sp
; nextln: sxtl v0.8h, v0.8b
; check: sxtl v0.8h, v0.8b
; nextln: sxtl2 v1.8h, v1.16b
; nextln: addp v0.8h, v0.8h, v1.8h
; nextln: ldp fp, lr, [sp], #16
; nextln: ret
function %fn6(i8x16, i8x16) -> i16x8 {
@@ -83,12 +68,9 @@ block0(v0: i8x16, v1: i8x16):
return v4
}
; check: stp fp
; nextln: mov fp, sp
; nextln: uxtl v0.8h, v0.8b
; check: uxtl v0.8h, v0.8b
; nextln: uxtl2 v1.8h, v1.16b
; nextln: addp v0.8h, v0.8h, v1.8h
; nextln: ldp fp, lr, [sp], #16
; nextln: ret
function %fn7(i8x16) -> i16x8 {
@@ -99,12 +81,9 @@ block0(v0: i8x16):
return v3
}
; check: stp fp
; nextln: mov fp, sp
; nextln: uxtl v1.8h, v0.8b
; check: uxtl v1.8h, v0.8b
; nextln: sxtl2 v0.8h, v0.16b
; nextln: addp v0.8h, v1.8h, v0.8h
; nextln: ldp fp, lr, [sp], #16
; nextln: ret
function %fn8(i8x16) -> i16x8 {
@@ -115,10 +94,7 @@ block0(v0: i8x16):
return v3
}
; check: stp fp
; nextln: mov fp, sp
; nextln: sxtl v1.8h, v0.8b
; check: sxtl v1.8h, v0.8b
; nextln: uxtl2 v0.8h, v0.16b
; nextln: addp v0.8h, v1.8h, v0.8h
; nextln: ldp fp, lr, [sp], #16
; nextln: ret