2 files changed, 21 insertions, 3 deletions
diff --git a/llvm/test/CodeGen/ARM/fmacs.ll b/llvm/test/CodeGen/ARM/fmacs.ll
index aa492708c0b..027991ef2c9 100644
--- a/llvm/test/CodeGen/ARM/fmacs.ll
+++ b/llvm/test/CodeGen/ARM/fmacs.ll
@@ -3,6 +3,8 @@
 ; RUN: llc -mtriple=arm-eabi -mcpu=cortex-a8 %s -o - | FileCheck %s -check-prefix=A8
 ; RUN: llc -mtriple=arm-eabi -mcpu=cortex-a9 %s -o - | FileCheck %s -check-prefix=A9
 ; RUN: llc -mtriple=arm-linux-gnueabi -mcpu=cortex-a9 -float-abi=hard %s -o - | FileCheck %s -check-prefix=HARD
+; RUN: llc -mtriple=arm-linux-gnueabi -mcpu=cortex-m4 -float-abi=hard %s -o - | FileCheck %s -check-prefix=VMLA
+; RUN: llc -mtriple=arm-linux-gnueabi -mcpu=cortex-m33 -float-abi=hard %s -o - | FileCheck %s -check-prefix=VMLA
 
 define float @t1(float %acc, float %a, float %b) {
 entry:
@@ -15,6 +17,21 @@ entry:
 ; A8-LABEL: t1:
 ; A8: vmul.f32
 ; A8: vadd.f32
+
+; VMLA-LABEL: t1:
+; VMLA:       vmul.f32
+; VMLA-NEXT:  vadd.f32
+
+  %0 = fmul float %a, %b
+  %1 = fadd float %acc, %0
+	ret float %1
+}
+
+define float @vlma_minsize(float %acc, float %a, float %b) #0 {
+entry:
+; VMLA-LABEL: vlma_minsize:
+; VLMA:       vmla.f32  s0, s1, s2
+
   %0 = fmul float %a, %b
   %1 = fadd float %acc, %0
 	ret float %1
@@ -102,3 +119,5 @@ entry:
   %3 = fadd float %1, %2
   ret float %3
 }
+
+attributes #0 = { minsize nounwind optsize }
diff --git a/llvm/test/CodeGen/Thumb2/float-intrinsics-float.ll b/llvm/test/CodeGen/Thumb2/float-intrinsics-float.ll
index 847aeacd2f9..8ee2af03eca 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 @@ define float @round_f(float %a) {
   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: