From: Chris Lattner <sabre@nondot.org>
Date: Fri, 14 Jan 2011 00:01:01 +0000 (+0000)
Subject: fix PR8961 - a fast isel miscompilation where we'd insert a new instruction
X-Git-Url: http://demsky.eecs.uci.edu/git/?a=commitdiff_plain;h=d7540414935fb30405ed07a9b193d3f8748f5274;p=oota-llvm.git

fix PR8961 - a fast isel miscompilation where we'd insert a new instruction
after sext's generated for addressing that got folded.  Previously we compiled
test5 into:

_test5:                                 ## @test5
## BB#0:
        movq    -8(%rsp), %rax          ## 8-byte Reload
        movq    (%rdi,%rax), %rdi
        addq    %rdx, %rdi
        movslq  %esi, %rax
        movq    %rax, -8(%rsp)          ## 8-byte Spill
        movq    %rdi, %rax
        ret

which is insane and wrong.  Now we produce:

_test5:                                 ## @test5
## BB#0:
	movslq	%esi, %rax
	movq	(%rdi,%rax), %rax
	addq	%rdx, %rax
	ret



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123414 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/lib/Target/X86/X86FastISel.cpp b/lib/Target/X86/X86FastISel.cpp
index f29d127c85d..9d42ac2e470 100644
--- a/lib/Target/X86/X86FastISel.cpp
+++ b/lib/Target/X86/X86FastISel.cpp
@@ -1933,7 +1933,7 @@ bool X86FastISel::TryToFoldLoad(MachineInstr *MI, unsigned OpNo,
     XII.foldMemoryOperandImpl(*FuncInfo.MF, MI, OpNo, AddrOps, Size, Alignment);
   if (Result == 0) return false;
 
-  MI->getParent()->insert(MI, Result);
+  FuncInfo.MBB->insert(FuncInfo.InsertPt, Result);
   MI->eraseFromParent();
   return true;
 }
diff --git a/test/CodeGen/X86/fast-isel-gep.ll b/test/CodeGen/X86/fast-isel-gep.ll
index 577dd7223a4..622a1ff831d 100644
--- a/test/CodeGen/X86/fast-isel-gep.ll
+++ b/test/CodeGen/X86/fast-isel-gep.ll
@@ -70,3 +70,20 @@ entry:
 ; X64: test4:
 ; X64: 128(%r{{.*}},%r{{.*}},8)
 }
+
+; PR8961 - Make sure the sext for the GEP addressing comes before the load that
+; is folded.
+define i64 @test5(i8* %A, i32 %I, i64 %B) nounwind {
+  %v8 = getelementptr i8* %A, i32 %I
+  %v9 = bitcast i8* %v8 to i64*
+  %v10 = load i64* %v9
+  %v11 = add i64 %B, %v10
+  ret i64 %v11
+; X64: test5:
+; X64: movslq	%esi, %rax
+; X64-NEXT: movq	(%rdi,%rax), %rax
+; X64-NEXT: addq	%rdx, %rax
+; X64-NEXT: ret
+}
+
+