commit d986ede313227a5f473f83c8ba52130255955514
author Sjoerd Meijer <sjoerd.meijer@arm.com> Mon Sep 24 12:02:50 2018 +0000
committer Sjoerd Meijer <sjoerd.meijer@arm.com> Mon Sep 24 12:02:50 2018 +0000
tree 80aca1adfaa4dbbbeba0d7494484d9bbfb33a951
parent 5555c009026eba9aef07544731ef6cd421cc7ccc

[ARM] Do not fuse VADD and VMUL on the Cortex-M4 and Cortex-M33

A sequence of VMUL and VADD instructions always give the same or better
performance than a fused VMLA instruction on the Cortex-M4 and Cortex-M33.
Executing the VMUL and VADD back-to-back requires the same cycles, but
having separate instructions allows scheduling to avoid the hazard between
these 2 instructions.

Differential Revision: https://reviews.llvm.org/D52289

llvm-svn: 342874

llvm/test/CodeGen/Thumb2/float-intrinsics-float.ll

diff --git a/llvm/test/CodeGen/Thumb2/float-intrinsics-float.ll b/llvm/test/CodeGen/Thumb2/float-intrinsics-float.ll
index 847aeac..8ee2af0 100644
--- a/llvm/test/CodeGen/Thumb2/float-intrinsics-float.ll
+++ b/llvm/test/CodeGen/Thumb2/float-intrinsics-float.ll
@@ -1,5 +1,6 @@
 ; RUN: llc < %s -mtriple=thumbv7-none-eabi   -mcpu=cortex-m3                    | FileCheck %s -check-prefix=CHECK -check-prefix=SOFT -check-prefix=NONE
-; RUN: llc < %s -mtriple=thumbv7-none-eabihf -mcpu=cortex-m4                    | FileCheck %s -check-prefix=CHECK -check-prefix=HARD -check-prefix=SP -check-prefix=VMLA
+; RUN: llc < %s -mtriple=thumbv7-none-eabihf -mcpu=cortex-m4                    | FileCheck %s -check-prefix=CHECK -check-prefix=HARD -check-prefix=SP -check-prefix=NO-VMLA
+; RUN: llc < %s -mtriple=thumbv7-none-eabihf -mcpu=cortex-m33                   | FileCheck %s -check-prefix=CHECK -check-prefix=HARD -check-prefix=SP -check-prefix=NO-VMLA
 ; RUN: llc < %s -mtriple=thumbv7-none-eabihf -mcpu=cortex-m7                    | FileCheck %s -check-prefix=CHECK -check-prefix=HARD -check-prefix=DP -check-prefix=VFP  -check-prefix=FP-ARMv8  -check-prefix=VMLA
 ; RUN: llc < %s -mtriple=thumbv7-none-eabihf -mcpu=cortex-m7 -mattr=+fp-only-sp | FileCheck %s -check-prefix=CHECK -check-prefix=HARD -check-prefix=SP -check-prefix=FP-ARMv8 -check-prefix=VMLA
 ; RUN: llc < %s -mtriple=thumbv7-none-eabihf -mcpu=cortex-a7                    | FileCheck %s -check-prefix=CHECK -check-prefix=HARD -check-prefix=DP -check-prefix=NEON -check-prefix=VFP4 -check-prefix=NO-VMLA
@@ -188,8 +189,6 @@
   ret float %1
 }
 
-; FIXME: why does cortex-m4 use vmla, while cortex-a7 uses vmul+vadd?
-; (these should be equivalent, even the rounding is the same)
 declare float     @llvm.fmuladd.f32(float %a, float %b, float %c)
 define float @fmuladd_f(float %a, float %b, float %c) {
 ; CHECK-LABEL: fmuladd_f: