CodeGen peephole: fold redundant phys reg copies

[oota-llvm.git] / test / CodeGen / X86 / phys_subreg_coalesce-3.ll

diff --git a/test/CodeGen/X86/phys_subreg_coalesce-3.ll b/test/CodeGen/X86/phys_subreg_coalesce-3.ll
index f23669ed9a4abeae70a9d4c03077b9faa3eda555..74e3d1291c05aa21e691c684aad38ef8c4a84418 100644 (file)
--- a/test/CodeGen/X86/phys_subreg_coalesce-3.ll
+++ b/test/CodeGen/X86/phys_subreg_coalesce-3.ll
@@ -1,8 +1,13 @@
-; RUN: llc < %s -mtriple=i386-apple-darwin | FileCheck %s
+; RUN: llc < %s -verify-machineinstrs -mtriple=i386-apple-darwin -mcpu=corei7 | FileCheck %s
 ; rdar://5571034
 
+; This requires physreg joining, %vreg13 is live everywhere:
+; 304L         %CL<def> = COPY %vreg13:sub_8bit; GR32_ABCD:%vreg13
+; 320L         %vreg15<def> = COPY %vreg19; GR32:%vreg15 GR32_NOSP:%vreg19
+; 336L         %vreg15<def> = SAR32rCL %vreg15, %EFLAGS<imp-def,dead>, %CL<imp-use,kill>; GR32:%vreg15
+
 define void @foo(i32* nocapture %quadrant, i32* nocapture %ptr, i32 %bbSize, i32 %bbStart, i32 %shifts) nounwind ssp {
-; CHECK: foo:
+; CHECK-LABEL: foo:
 entry:
   %j.03 = add i32 %bbSize, -1                     ; <i32> [#uses=2]
   %0 = icmp sgt i32 %j.03, -1                     ; <i1> [#uses=1]
@@ -20,11 +25,11 @@ bb:                                               ; preds = %bb, %bb.nph
   %indvar = phi i32 [ 0, %bb.nph ], [ %indvar.next, %bb ] ; <i32> [#uses=3]
   %j.06 = sub i32 %j.03, %indvar                  ; <i32> [#uses=1]
   %tmp11 = sub i32 %tmp10, %indvar                ; <i32> [#uses=1]
-  %scevgep = getelementptr i32* %ptr, i32 %tmp11  ; <i32*> [#uses=1]
-  %1 = load i32* %scevgep, align 4                ; <i32> [#uses=1]
+  %scevgep = getelementptr i32, i32* %ptr, i32 %tmp11  ; <i32*> [#uses=1]
+  %1 = load i32, i32* %scevgep, align 4                ; <i32> [#uses=1]
   %2 = ashr i32 %j.06, %shifts                    ; <i32> [#uses=1]
   %3 = and i32 %2, 65535                          ; <i32> [#uses=1]
-  %4 = getelementptr inbounds i32* %quadrant, i32 %1 ; <i32*> [#uses=1]
+  %4 = getelementptr inbounds i32, i32* %quadrant, i32 %1 ; <i32*> [#uses=1]
   store i32 %3, i32* %4, align 4
   %indvar.next = add i32 %indvar, 1               ; <i32> [#uses=2]
   %exitcond = icmp eq i32 %indvar.next, %bbSize   ; <i1> [#uses=1]