summaryrefslogtreecommitdiffstats
path: root/llvm/test/CodeGen/PowerPC/ctr-minmaxnum.ll
diff options
context:
space:
mode:
authorLei Huang <lei@ca.ibm.com>2017-10-12 16:43:33 +0000
committerLei Huang <lei@ca.ibm.com>2017-10-12 16:43:33 +0000
commit0724fea2da637883f1461e12ff46d596a816f758 (patch)
tree167aa66c20867839db6bc0c1556d29ee9845e528 /llvm/test/CodeGen/PowerPC/ctr-minmaxnum.ll
parentc8ffffe4625bc0b0f790824eaa2a591943dff862 (diff)
downloadbcm5719-llvm-0724fea2da637883f1461e12ff46d596a816f758.tar.gz
bcm5719-llvm-0724fea2da637883f1461e12ff46d596a816f758.zip
[PowerPC] Add profitablilty check for conversion to mtctr loops
Add profitability checks for modifying counted loops to use the mtctr instruction. The latency of mtctr is only justified if there are more than 4 comparisons that will be removed as a result. Usually counted loops are formed relatively early and before unrolling, so most low trip count loops often don't survive. However we want to ensure that if they do, we do not mistakenly update them to mtctr loops. Use CodeMetrics to ensure we are only doing this for small loops with small trip counts. Differential Revision: https://reviews.llvm.org/D38212 llvm-svn: 315592
Diffstat (limited to 'llvm/test/CodeGen/PowerPC/ctr-minmaxnum.ll')
-rw-r--r--llvm/test/CodeGen/PowerPC/ctr-minmaxnum.ll47
1 files changed, 42 insertions, 5 deletions
diff --git a/llvm/test/CodeGen/PowerPC/ctr-minmaxnum.ll b/llvm/test/CodeGen/PowerPC/ctr-minmaxnum.ll
index 2b0a7cceb68..e38f851620b 100644
--- a/llvm/test/CodeGen/PowerPC/ctr-minmaxnum.ll
+++ b/llvm/test/CodeGen/PowerPC/ctr-minmaxnum.ll
@@ -1,6 +1,5 @@
-; RUN: llc -verify-machineinstrs -mcpu=pwr7 < %s | FileCheck %s
-; RUN: llc -verify-machineinstrs -mcpu=a2q < %s | FileCheck %s --check-prefix=QPX
-target triple = "powerpc64-unknown-linux-gnu"
+; RUN: llc -mtriple=powerpc64-unknown-linux-gnu -verify-machineinstrs -mcpu=pwr7 < %s | FileCheck %s
+; RUN: llc -mtriple=powerpc64-unknown-linux-gnu -verify-machineinstrs -mcpu=a2q < %s | FileCheck %s --check-prefix=QPX
declare float @fabsf(float)
@@ -38,6 +37,9 @@ loop_exit:
; CHECK-LABEL: test1:
; CHECK-NOT: mtctr
; CHECK: bl fminf
+; CHECK-NOT: bl fminf
+; CHECK-NOT: mtctr
+; CHECK: blr
define void @test1v(<4 x float> %f, <4 x float>* %fp) {
entry:
@@ -48,7 +50,7 @@ loop_body:
%0 = call <4 x float> @llvm.minnum.v4f32(<4 x float> %f, <4 x float> <float 1.0, float 1.0, float 1.0, float 1.0>)
store <4 x float> %0, <4 x float>* %fp, align 16
%1 = add i64 %invar_address.dim.0.01, 1
- %2 = icmp eq i64 %1, 2
+ %2 = icmp eq i64 %1, 4
br i1 %2, label %loop_exit, label %loop_body
loop_exit:
@@ -56,8 +58,15 @@ loop_exit:
}
; CHECK-LABEL: test1v:
+; CHECK: bl fminf
+; CHECK-NOT: mtctr
+; CHECK: bl fminf
; CHECK-NOT: mtctr
; CHECK: bl fminf
+; CHECK-NOT: mtctr
+; CHECK: bl fminf
+; CHECK-NOT: bl fminf
+; CHECK: blr
; QPX-LABEL: test1v:
; QPX: mtctr
@@ -83,6 +92,9 @@ loop_exit:
; CHECK-LABEL: test1a:
; CHECK-NOT: mtctr
; CHECK: bl fminf
+; CHECK-NOT: bl fminf
+; CHECK-NOT: mtctr
+; CHECK: blr
define void @test2(float %f, float* %fp) {
entry:
@@ -103,6 +115,9 @@ loop_exit:
; CHECK-LABEL: test2:
; CHECK-NOT: mtctr
; CHECK: bl fmaxf
+; CHECK-NOT: bl fmaxf
+; CHECK-NOT: mtctr
+; CHECK: blr
define void @test2v(<4 x double> %f, <4 x double>* %fp) {
entry:
@@ -113,7 +128,7 @@ loop_body:
%0 = call <4 x double> @llvm.maxnum.v4f64(<4 x double> %f, <4 x double> <double 1.0, double 1.0, double 1.0, double 1.0>)
store <4 x double> %0, <4 x double>* %fp, align 16
%1 = add i64 %invar_address.dim.0.01, 1
- %2 = icmp eq i64 %1, 2
+ %2 = icmp eq i64 %1, 4
br i1 %2, label %loop_exit, label %loop_body
loop_exit:
@@ -121,8 +136,15 @@ loop_exit:
}
; CHECK-LABEL: test2v:
+; CHECK: bl fmax
+; CHECK-NOT: mtctr
+; CHECK: bl fmax
+; CHECK-NOT: mtctr
+; CHECK: bl fmax
; CHECK-NOT: mtctr
; CHECK: bl fmax
+; CHECK-NOT: bl fmax
+; CHECK: blr
; QPX-LABEL: test2v:
; QPX: mtctr
@@ -148,6 +170,9 @@ loop_exit:
; CHECK-LABEL: test2a:
; CHECK-NOT: mtctr
; CHECK: bl fmaxf
+; CHECK-NOT: bl fmaxf
+; CHECK-NOT: mtctr
+; CHECK: blr
define void @test3(double %f, double* %fp) {
entry:
@@ -168,6 +193,9 @@ loop_exit:
; CHECK-LABEL: test3:
; CHECK-NOT: mtctr
; CHECK: bl fmin
+; CHECK-NOT: bl fmin
+; CHECK-NOT: mtctr
+; CHECK: blr
define void @test3a(double %f, double* %fp) {
entry:
@@ -188,6 +216,9 @@ loop_exit:
; CHECK-LABEL: test3a:
; CHECK-NOT: mtctr
; CHECK: bl fmin
+; CHECK-NOT: bl fmin
+; CHECK-NOT: mtctr
+; CHECK: blr
define void @test4(double %f, double* %fp) {
entry:
@@ -208,6 +239,9 @@ loop_exit:
; CHECK-LABEL: test4:
; CHECK-NOT: mtctr
; CHECK: bl fmax
+; CHECK-NOT: bl fmax
+; CHECK-NOT: mtctr
+; CHECK: blr
define void @test4a(double %f, double* %fp) {
entry:
@@ -228,4 +262,7 @@ loop_exit:
; CHECK-LABEL: test4a:
; CHECK-NOT: mtctr
; CHECK: bl fmax
+; CHECK-NOT: bl fmax
+; CHECK-NOT: mtctr
+; CHECK: blr
OpenPOWER on IntegriCloud