[PowerPC] Emit VMX loads/stores for aligned ops to avoid adding swaps on LE

This patch corresponds to review:
https://reviews.llvm.org/D26861

It also fixes PR30730.

Committing on behalf of Lei Huang.

llvm-svn: 287679

5 files changed, 83 insertions, 59 deletions

diff --git a/llvm/test/CodeGen/PowerPC/ppc64-i128-abi.ll b/llvm/test/CodeGen/PowerPC/ppc64-i128-abi.ll
index 924e04a34a9..4a8fd90db3e 100644
--- a/llvm/test/CodeGen/PowerPC/ppc64-i128-abi.ll
+++ b/llvm/test/CodeGen/PowerPC/ppc64-i128-abi.ll
@@ -1,5 +1,6 @@
 ; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu \
-; RUN:   -mcpu=pwr8 < %s | FileCheck %s -check-prefix=CHECK-LE
+; RUN:   -mcpu=pwr8 < %s | FileCheck %s -check-prefix=CHECK-LE \
+; RUN:   --implicit-check-not xxswapd
 
 ; RUN: llc -verify-machineinstrs -mtriple=powerpc64-unknown-linux-gnu \
 ; RUN:   -mcpu=pwr8 < %s | FileCheck %s -check-prefix=CHECK-BE
@@ -8,13 +9,15 @@
 ; RUN:   -mcpu=pwr8 -mattr=-vsx < %s | FileCheck %s -check-prefix=CHECK-NOVSX
 
 ; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu \
-; RUN:   -mcpu=pwr8 -mattr=-vsx < %s | FileCheck %s -check-prefix=CHECK-NOVSX
+; RUN:   -mcpu=pwr8 -mattr=-vsx < %s | FileCheck %s -check-prefix=CHECK-NOVSX \
+; RUN:   --implicit-check-not xxswapd
 
 ; RUN: llc -verify-machineinstrs -mtriple=powerpc64-unknown-linux-gnu \
 ; RUN:   -mcpu=pwr8 -mattr=-vsx < %s | FileCheck %s -check-prefix=CHECK-BE-NOVSX
 
 ; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu \
-; RUN:   -mcpu=pwr8 -mattr=-vsx < %s | FileCheck %s -check-prefix=CHECK-LE-NOVSX
+; RUN:   -mcpu=pwr8 -mattr=-vsx < %s | \
+; RUN:   FileCheck %s -check-prefix=CHECK-LE-NOVSX --implicit-check-not xxswapd
 
 ; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu \
 ; RUN:   -mcpu=pwr9 -ppc-vsr-nums-as-vr < %s | FileCheck %s \
@@ -26,7 +29,7 @@
 
 ; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu \
 ; RUN:   -mcpu=pwr9 -mattr=-power9-vector -mattr=-direct-move < %s | \
-; RUN:   FileCheck %s -check-prefix=CHECK-LE
+; RUN:   FileCheck %s -check-prefix=CHECK-LE --implicit-check-not xxswapd
 
 @x = common global <1 x i128> zeroinitializer, align 16
 @y = common global <1 x i128> zeroinitializer, align 16
@@ -199,8 +202,7 @@ define <1 x i128> @call_v1i128_increment_by_one() nounwind {
        ret <1 x i128> %ret
 
 ; CHECK-LE-LABEL: @call_v1i128_increment_by_one
-; CHECK-LE: lxvd2x [[PARAM:[0-9]+]], {{[0-9]+}}, {{[0-9]+}}
-; CHECK-LE: xxswapd 34, [[PARAM]]
+; CHECK-LE: lvx 2, {{[0-9]+}}, {{[0-9]+}}
 ; CHECK-LE: bl v1i128_increment_by_one
 ; CHECK-LE: blr
 
@@ -229,10 +231,8 @@ define <1 x i128> @call_v1i128_increment_by_val() nounwind {
        ret <1 x i128> %ret
 
 ; CHECK-LE-LABEL: @call_v1i128_increment_by_val
-; CHECK-LE: lxvd2x [[PARAM1:[0-9]+]], {{[0-9]+}}, {{[0-9]+}}
-; CHECK-LE: lxvd2x [[PARAM2:[0-9]+]], {{[0-9]+}}, {{[0-9]+}}
-; CHECK-LE-DAG: xxswapd 34, [[PARAM1]]
-; CHECK-LE-DAG: xxswapd 35, [[PARAM2]]
+; CHECK-LE: lvx 2, {{[0-9]+}}, {{[0-9]+}}
+; CHECK-LE: lvx 3, {{[0-9]+}}, {{[0-9]+}}
 ; CHECK-LE: bl v1i128_increment_by_val
 ; CHECK-LE: blr
 
diff --git a/llvm/test/CodeGen/PowerPC/swaps-le-1.ll b/llvm/test/CodeGen/PowerPC/swaps-le-1.ll
index cb83bf262b4..29c94f89543 100644
--- a/llvm/test/CodeGen/PowerPC/swaps-le-1.ll
+++ b/llvm/test/CodeGen/PowerPC/swaps-le-1.ll
@@ -13,6 +13,12 @@
 ; RUN:  -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu < %s \
 ; RUN:  | FileCheck -check-prefix=NOOPTSWAP %s
 
+; LH: 2016-11-17
+;   Updated align attritue from 16 to 8 to keep swap instructions tests.
+;   Changes have been made on little-endian to use lvx and stvx
+;   instructions instead of lxvd2x/xxswapd and xxswapd/stxvd2x for
+;   aligned vectors with elements up to 4 bytes
+
 ; This test was generated from the following source:
 ;
 ; #define N 4096
@@ -29,10 +35,10 @@
 ;   }
 ; }
 
-@cb = common global [4096 x i32] zeroinitializer, align 16
-@cc = common global [4096 x i32] zeroinitializer, align 16
-@cd = common global [4096 x i32] zeroinitializer, align 16
-@ca = common global [4096 x i32] zeroinitializer, align 16
+@cb = common global [4096 x i32] zeroinitializer, align 8
+@cc = common global [4096 x i32] zeroinitializer, align 8
+@cd = common global [4096 x i32] zeroinitializer, align 8
+@ca = common global [4096 x i32] zeroinitializer, align 8
 
 define void @foo() {
 entry:
@@ -42,63 +48,63 @@ vector.body:
   %index = phi i64 [ 0, %entry ], [ %index.next.3, %vector.body ]
   %0 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cb, i64 0, i64 %index
   %1 = bitcast i32* %0 to <4 x i32>*
-  %wide.load = load <4 x i32>, <4 x i32>* %1, align 16
+  %wide.load = load <4 x i32>, <4 x i32>* %1, align 8
   %2 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cc, i64 0, i64 %index
   %3 = bitcast i32* %2 to <4 x i32>*
-  %wide.load13 = load <4 x i32>, <4 x i32>* %3, align 16
+  %wide.load13 = load <4 x i32>, <4 x i32>* %3, align 8
   %4 = add nsw <4 x i32> %wide.load13, %wide.load
   %5 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cd, i64 0, i64 %index
   %6 = bitcast i32* %5 to <4 x i32>*
-  %wide.load14 = load <4 x i32>, <4 x i32>* %6, align 16
+  %wide.load14 = load <4 x i32>, <4 x i32>* %6, align 8
   %7 = mul nsw <4 x i32> %4, %wide.load14
   %8 = getelementptr inbounds [4096 x i32], [4096 x i32]* @ca, i64 0, i64 %index
   %9 = bitcast i32* %8 to <4 x i32>*
-  store <4 x i32> %7, <4 x i32>* %9, align 16
+  store <4 x i32> %7, <4 x i32>* %9, align 8
   %index.next = add nuw nsw i64 %index, 4
   %10 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cb, i64 0, i64 %index.next
   %11 = bitcast i32* %10 to <4 x i32>*
-  %wide.load.1 = load <4 x i32>, <4 x i32>* %11, align 16
+  %wide.load.1 = load <4 x i32>, <4 x i32>* %11, align 8
   %12 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cc, i64 0, i64 %index.next
   %13 = bitcast i32* %12 to <4 x i32>*
-  %wide.load13.1 = load <4 x i32>, <4 x i32>* %13, align 16
+  %wide.load13.1 = load <4 x i32>, <4 x i32>* %13, align 8
   %14 = add nsw <4 x i32> %wide.load13.1, %wide.load.1
   %15 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cd, i64 0, i64 %index.next
   %16 = bitcast i32* %15 to <4 x i32>*
-  %wide.load14.1 = load <4 x i32>, <4 x i32>* %16, align 16
+  %wide.load14.1 = load <4 x i32>, <4 x i32>* %16, align 8
   %17 = mul nsw <4 x i32> %14, %wide.load14.1
   %18 = getelementptr inbounds [4096 x i32], [4096 x i32]* @ca, i64 0, i64 %index.next
   %19 = bitcast i32* %18 to <4 x i32>*
-  store <4 x i32> %17, <4 x i32>* %19, align 16
+  store <4 x i32> %17, <4 x i32>* %19, align 8
   %index.next.1 = add nuw nsw i64 %index.next, 4
   %20 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cb, i64 0, i64 %index.next.1
   %21 = bitcast i32* %20 to <4 x i32>*
-  %wide.load.2 = load <4 x i32>, <4 x i32>* %21, align 16
+  %wide.load.2 = load <4 x i32>, <4 x i32>* %21, align 8
   %22 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cc, i64 0, i64 %index.next.1
   %23 = bitcast i32* %22 to <4 x i32>*
-  %wide.load13.2 = load <4 x i32>, <4 x i32>* %23, align 16
+  %wide.load13.2 = load <4 x i32>, <4 x i32>* %23, align 8
   %24 = add nsw <4 x i32> %wide.load13.2, %wide.load.2
   %25 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cd, i64 0, i64 %index.next.1
   %26 = bitcast i32* %25 to <4 x i32>*
-  %wide.load14.2 = load <4 x i32>, <4 x i32>* %26, align 16
+  %wide.load14.2 = load <4 x i32>, <4 x i32>* %26, align 8
   %27 = mul nsw <4 x i32> %24, %wide.load14.2
   %28 = getelementptr inbounds [4096 x i32], [4096 x i32]* @ca, i64 0, i64 %index.next.1
   %29 = bitcast i32* %28 to <4 x i32>*
-  store <4 x i32> %27, <4 x i32>* %29, align 16
+  store <4 x i32> %27, <4 x i32>* %29, align 8
   %index.next.2 = add nuw nsw i64 %index.next.1, 4
   %30 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cb, i64 0, i64 %index.next.2
   %31 = bitcast i32* %30 to <4 x i32>*
-  %wide.load.3 = load <4 x i32>, <4 x i32>* %31, align 16
+  %wide.load.3 = load <4 x i32>, <4 x i32>* %31, align 8
   %32 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cc, i64 0, i64 %index.next.2
   %33 = bitcast i32* %32 to <4 x i32>*
-  %wide.load13.3 = load <4 x i32>, <4 x i32>* %33, align 16
+  %wide.load13.3 = load <4 x i32>, <4 x i32>* %33, align 8
   %34 = add nsw <4 x i32> %wide.load13.3, %wide.load.3
   %35 = getelementptr inbounds [4096 x i32], [4096 x i32]* @cd, i64 0, i64 %index.next.2
   %36 = bitcast i32* %35 to <4 x i32>*
-  %wide.load14.3 = load <4 x i32>, <4 x i32>* %36, align 16
+  %wide.load14.3 = load <4 x i32>, <4 x i32>* %36, align 8
   %37 = mul nsw <4 x i32> %34, %wide.load14.3
   %38 = getelementptr inbounds [4096 x i32], [4096 x i32]* @ca, i64 0, i64 %index.next.2
   %39 = bitcast i32* %38 to <4 x i32>*
-  store <4 x i32> %37, <4 x i32>* %39, align 16
+  store <4 x i32> %37, <4 x i32>* %39, align 8
   %index.next.3 = add nuw nsw i64 %index.next.2, 4
   %40 = icmp eq i64 %index.next.3, 4096
   br i1 %40, label %for.end, label %vector.body
diff --git a/llvm/test/CodeGen/PowerPC/swaps-le-2.ll b/llvm/test/CodeGen/PowerPC/swaps-le-2.ll
index 0963b92609f..e7751a194f7 100644
--- a/llvm/test/CodeGen/PowerPC/swaps-le-2.ll
+++ b/llvm/test/CodeGen/PowerPC/swaps-le-2.ll
@@ -2,6 +2,13 @@
 
 ; Test swap removal when a vector splat must be adjusted to make it legal.
 ;
+
+; LH: 2016-11-17
+;   Updated align attritue from 16 to 8 to keep swap instructions tests.
+;   Changes have been made on little-endian to use lvx and stvx
+;   instructions instead of lxvd2x/xxswapd and xxswapd/stxvd2x for
+;   aligned vectors with elements up to 4 bytes
+
 ; Test generated from following C code:
 ;
 ; vector char vc = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15};
@@ -28,37 +35,37 @@
 ;   vir = (vector int){vi[1], vi[1], vi[1], vi[1]};
 ; }
 
-@vc = global <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, align 16
-@vs = global <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, align 16
-@vi = global <4 x i32> <i32 0, i32 1, i32 2, i32 3>, align 16
-@vcr = common global <16 x i8> zeroinitializer, align 16
-@vsr = common global <8 x i16> zeroinitializer, align 16
-@vir = common global <4 x i32> zeroinitializer, align 16
+@vc = global <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, align 8
+@vs = global <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, align 8
+@vi = global <4 x i32> <i32 0, i32 1, i32 2, i32 3>, align 8
+@vcr = common global <16 x i8> zeroinitializer, align 8
+@vsr = common global <8 x i16> zeroinitializer, align 8
+@vir = common global <4 x i32> zeroinitializer, align 8
 
 ; Function Attrs: nounwind
 define void @cfoo() {
 entry:
-  %0 = load <16 x i8>, <16 x i8>* @vc, align 16
+  %0 = load <16 x i8>, <16 x i8>* @vc, align 8
   %vecinit30 = shufflevector <16 x i8> %0, <16 x i8> undef, <16 x i32> <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
-  store <16 x i8> %vecinit30, <16 x i8>* @vcr, align 16
+  store <16 x i8> %vecinit30, <16 x i8>* @vcr, align 8
   ret void
 }
 
 ; Function Attrs: nounwind
 define void @sfoo() {
 entry:
-  %0 = load <8 x i16>, <8 x i16>* @vs, align 16
+  %0 = load <8 x i16>, <8 x i16>* @vs, align 8
   %vecinit14 = shufflevector <8 x i16> %0, <8 x i16> undef, <8 x i32> <i32 6, i32 6, i32 6, i32 6, i32 6, i32 6, i32 6, i32 6>
-  store <8 x i16> %vecinit14, <8 x i16>* @vsr, align 16
+  store <8 x i16> %vecinit14, <8 x i16>* @vsr, align 8
   ret void
 }
 
 ; Function Attrs: nounwind
 define void @ifoo() {
 entry:
-  %0 = load <4 x i32>, <4 x i32>* @vi, align 16
+  %0 = load <4 x i32>, <4 x i32>* @vi, align 8
   %vecinit6 = shufflevector <4 x i32> %0, <4 x i32> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
-  store <4 x i32> %vecinit6, <4 x i32>* @vir, align 16
+  store <4 x i32> %vecinit6, <4 x i32>* @vir, align 8
   ret void
 }
 
diff --git a/llvm/test/CodeGen/PowerPC/vsx-ldst.ll b/llvm/test/CodeGen/PowerPC/vsx-ldst.ll
index a146182de99..d8dd635aab5 100644
--- a/llvm/test/CodeGen/PowerPC/vsx-ldst.ll
+++ b/llvm/test/CodeGen/PowerPC/vsx-ldst.ll
@@ -14,8 +14,10 @@
 
 ; RUN: llc -verify-machineinstrs -mcpu=pwr8 -mattr=+vsx -O2 \
 ; RUN:   -mtriple=powerpc64le-unknown-linux-gnu < %s > %t
-; RUN: grep lxvd2x < %t | count 6
-; RUN: grep stxvd2x < %t | count 6
+; RUN: grep lxvd2x < %t | count 3
+; RUN: grep lvx < %t | count 3
+; RUN: grep stxvd2x < %t | count 3
+; RUN: grep stvx < %t | count 3
 
 ; RUN: llc -verify-machineinstrs -mcpu=pwr9 -O2 \
 ; RUN:   -mtriple=powerpc64le-unknown-linux-gnu < %s > %t
diff --git a/llvm/test/CodeGen/PowerPC/vsx.ll b/llvm/test/CodeGen/PowerPC/vsx.ll
index 92b0eaff9fb..b15dc4bede1 100644
--- a/llvm/test/CodeGen/PowerPC/vsx.ll
+++ b/llvm/test/CodeGen/PowerPC/vsx.ll
@@ -1,8 +1,17 @@
-; RUN: llc -verify-machineinstrs -mcpu=pwr7 -mtriple=powerpc64-unknown-linux-gnu -mattr=+vsx < %s | FileCheck %s
-; RUN: llc -verify-machineinstrs -mcpu=pwr7 -mtriple=powerpc64-unknown-linux-gnu -mattr=+vsx < %s | FileCheck -check-prefix=CHECK-REG %s
-; RUN: llc -verify-machineinstrs -mcpu=pwr7 -mtriple=powerpc64-unknown-linux-gnu -mattr=+vsx -fast-isel -O0 < %s | FileCheck %s
-; RUN: llc -verify-machineinstrs -mcpu=pwr7 -mtriple=powerpc64-unknown-linux-gnu -mattr=+vsx -fast-isel -O0 < %s | FileCheck -check-prefix=CHECK-FISL %s
-; RUN: llc -verify-machineinstrs -mcpu=pwr8 -mtriple=powerpc64le-unknown-linux-gnu -mattr=+vsx < %s | FileCheck -check-prefix=CHECK-LE %s
+; RUN: llc -verify-machineinstrs -mcpu=pwr7 \
+; RUN:   -mtriple=powerpc64-unknown-linux-gnu -mattr=+vsx < %s | FileCheck %s
+; RUN: llc -verify-machineinstrs -mcpu=pwr7 \
+; RUN:   -mtriple=powerpc64-unknown-linux-gnu -mattr=+vsx < %s | \
+; RUN:   FileCheck -check-prefix=CHECK-REG %s
+; RUN: llc -verify-machineinstrs -mcpu=pwr7 \
+; RUN:   -mtriple=powerpc64-unknown-linux-gnu -mattr=+vsx -fast-isel -O0 < %s |\
+; RUN:   FileCheck %s
+; RUN: llc -verify-machineinstrs -mcpu=pwr7 \
+; RUN:   -mtriple=powerpc64-unknown-linux-gnu -mattr=+vsx -fast-isel -O0 < %s |\
+; RUN:   FileCheck -check-prefix=CHECK-FISL %s
+; RUN: llc -verify-machineinstrs -mcpu=pwr8 \
+; RUN:   -mtriple=powerpc64le-unknown-linux-gnu -mattr=+vsx < %s | \
+; RUN:   FileCheck -check-prefix=CHECK-LE %s
 
 define double @test1(double %a, double %b) {
 entry:
@@ -645,8 +654,8 @@ define <4 x float> @test32(<4 x float>* %a) {
 ; CHECK-FISL: blr
 
 ; CHECK-LE-LABEL: @test32
-; CHECK-LE: lxvd2x [[V1:[0-9]+]], 0, 3
-; CHECK-LE: xxswapd 34, [[V1]]
+; CHECK-LE: lvx 2, 0, 3
+; CHECK-LE-NOT: xxswapd
 ; CHECK-LE: blr
 }
 
@@ -663,8 +672,8 @@ define void @test33(<4 x float>* %a, <4 x float> %b) {
 ; CHECK-FISL: blr
 
 ; CHECK-LE-LABEL: @test33
-; CHECK-LE: xxswapd [[V1:[0-9]+]], 34
-; CHECK-LE: stxvd2x [[V1]], 0, 3
+; CHECK-LE-NOT: xxswapd
+; CHECK-LE: stvx 2, 0, 3
 ; CHECK-LE: blr
 }
 
@@ -716,8 +725,8 @@ define <4 x i32> @test34(<4 x i32>* %a) {
 ; CHECK-FISL: blr
 
 ; CHECK-LE-LABEL: @test34
-; CHECK-LE: lxvd2x [[V1:[0-9]+]], 0, 3
-; CHECK-LE: xxswapd 34, [[V1]]
+; CHECK-LE: lvx 2, 0, 3
+; CHECK-LE-NOT: xxswapd
 ; CHECK-LE: blr
 }
 
@@ -734,8 +743,8 @@ define void @test35(<4 x i32>* %a, <4 x i32> %b) {
 ; CHECK-FISL: blr
 
 ; CHECK-LE-LABEL: @test35
-; CHECK-LE: xxswapd [[V1:[0-9]+]], 34
-; CHECK-LE: stxvd2x [[V1]], 0, 3
+; CHECK-LE-NOT: xxswapd
+; CHECK-LE: stvx 2, 0, 3
 ; CHECK-LE: blr
 }
 
@@ -1154,9 +1163,9 @@ define <2 x i32> @test80(i32 %v) {
 ; CHECK-LE-DAG: mtvsrd [[R1:[0-9]+]], 3
 ; CHECK-LE-DAG: xxswapd  [[V1:[0-9]+]], [[R1]]
 ; CHECK-LE-DAG: addi [[R2:[0-9]+]], {{[0-9]+}}, .LCPI
-; CHECK-LE-DAG: lxvd2x [[V2:[0-9]+]], 0, [[R2]]
+; CHECK-LE-DAG: lvx 3, 0, [[R2]]
 ; CHECK-LE-DAG: xxspltw 34, [[V1]]
-; CHECK-LE-DAG: xxswapd 35, [[V2]]
+; CHECK-LE-NOT: xxswapd 35, [[V2]]
 ; CHECK-LE: vadduwm 2, 2, 3
 ; CHECK-LE: blr
 }