[llvm-commits] [llvm] r117154 - in /llvm/trunk: lib/Transforms/InstCombine/InstCombineCalls.cpp test/Transforms/InstCombine/neon-intrinsics.ll

[Bob Wilson]

Author: bwilson
Date: Fri Oct 22 16:41:48 2010
New Revision: 117154

URL: http://llvm.org/viewvc/llvm-project?rev=117154&view=rev
Log:
Teach instcombine to set the alignment arguments for NEON load/store intrinsics.

Added:
    llvm/trunk/test/Transforms/InstCombine/neon-intrinsics.ll
Modified:
    llvm/trunk/lib/Transforms/InstCombine/InstCombineCalls.cpp

Modified: llvm/trunk/lib/Transforms/InstCombine/InstCombineCalls.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/InstCombine/InstCombineCalls.cpp?rev=117154&r1=117153&r2=117154&view=diff
==============================================================================
--- llvm/trunk/lib/Transforms/InstCombine/InstCombineCalls.cpp (original)
+++ llvm/trunk/lib/Transforms/InstCombine/InstCombineCalls.cpp Fri Oct 22 16:41:48 2010
@@ -702,6 +702,32 @@
     }
     break;
 
+  case Intrinsic::arm_neon_vld1:
+  case Intrinsic::arm_neon_vld2:
+  case Intrinsic::arm_neon_vld3:
+  case Intrinsic::arm_neon_vld4:
+  case Intrinsic::arm_neon_vld2lane:
+  case Intrinsic::arm_neon_vld3lane:
+  case Intrinsic::arm_neon_vld4lane:
+  case Intrinsic::arm_neon_vst1:
+  case Intrinsic::arm_neon_vst2:
+  case Intrinsic::arm_neon_vst3:
+  case Intrinsic::arm_neon_vst4:
+  case Intrinsic::arm_neon_vst2lane:
+  case Intrinsic::arm_neon_vst3lane:
+  case Intrinsic::arm_neon_vst4lane: {
+    unsigned MemAlign = GetOrEnforceKnownAlignment(II->getArgOperand(0));
+    unsigned AlignArg = II->getNumArgOperands() - 1;
+    ConstantInt *IntrAlign = dyn_cast<ConstantInt>(II->getArgOperand(AlignArg));
+    if (IntrAlign && IntrAlign->getZExtValue() < MemAlign) {
+      II->setArgOperand(AlignArg,
+                        ConstantInt::get(Type::getInt32Ty(II->getContext()),
+                                         MemAlign, false));
+      return II;
+    }
+    break;
+  }
+
   case Intrinsic::stackrestore: {
     // If the save is right next to the restore, remove the restore.  This can
     // happen when variable allocas are DCE'd.

Added: llvm/trunk/test/Transforms/InstCombine/neon-intrinsics.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/InstCombine/neon-intrinsics.ll?rev=117154&view=auto
==============================================================================
--- llvm/trunk/test/Transforms/InstCombine/neon-intrinsics.ll (added)
+++ llvm/trunk/test/Transforms/InstCombine/neon-intrinsics.ll Fri Oct 22 16:41:48 2010
@@ -0,0 +1,25 @@
+; RUN: opt < %s -instcombine -S | FileCheck %s
+
+; The alignment arguments for NEON load/store intrinsics can be increased
+; by instcombine.  Check for this.
+
+; CHECK: vld4.v2i32({{.*}}, i32 32)
+; CHECK: vst4.v2i32({{.*}}, i32 16)
+
+ at x = common global [8 x i32] zeroinitializer, align 32
+ at y = common global [8 x i32] zeroinitializer, align 16
+
+%struct.__neon_int32x2x4_t = type { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> }
+
+define void @test() nounwind ssp {
+  %tmp1 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8* bitcast ([8 x i32]* @x to i8*), i32 1)
+  %tmp2 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 0
+  %tmp3 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 1
+  %tmp4 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 2
+  %tmp5 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 3
+  call void @llvm.arm.neon.vst4.v2i32(i8* bitcast ([8 x i32]* @y to i8*), <2 x i32> %tmp2, <2 x i32> %tmp3, <2 x i32> %tmp4, <2 x i32> %tmp5, i32 1)
+  ret void
+}
+
+declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8*, i32) nounwind readonly
+declare void @llvm.arm.neon.vst4.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind