[AutoUpgrade] Don't lose attributes when upgrading mem intrinsics

[arichardson]

Summary

The original AutoUpgrade code from 1e68724d24ba38de7c7cdb2e1939d78c8b37cc0d
did not retain existing attributes. I noticed this in some downstream test
cases but it turns out there is also one affected testcase upstream.

[arichardson]

Fix another test that benefits from this change

[nikic]

llvm/lib/IR/AutoUpgrade.cpp

4080

Don't you need to remove the attributes the the 4th argument (that is being dropped) from the attribute list?

[arichardson]

llvm/lib/IR/AutoUpgrade.cpp

4080

Oh yes that's a good point. I guess it hasn't mattered so far since I've not seen an attribute on that argument, let me see what happens if there is one.

[arichardson]

Address review comment by @nikic

llvm/lib/IR/AutoUpgrade.cpp

4080

Not sure if there is an better way of generating the new AttributeList, creating a new one seemed like the easiest way.

[nikic]

LGTM

llvm/lib/IR/AutoUpgrade.cpp

if (CI->arg_size() != 5) {
       DefaultCase();
       return;
     }
     // Remove alignment argument (3), and add alignment attributes to the
     // dest/src pointers.
     Value *Args[4] = {CI->getArgOperand(0), CI->getArgOperand(1),
                       CI->getArgOperand(2), CI->getArgOperand(4)};
     NewCall = Builder.CreateCall(NewFn, Args);
+    AttributeList OldAttrs = CI->getAttributes();
+    AttributeList NewAttrs = AttributeList::get(
+        C, OldAttrs.getFnAttrs(), OldAttrs.getRetAttrs(),
+        {OldAttrs.getParamAttrs(0), OldAttrs.getParamAttrs(1),
+         OldAttrs.getParamAttrs(2), OldAttrs.getParamAttrs(4)});
+    NewCall->setAttributes(NewAttrs);
     auto *MemCI = cast<MemIntrinsic>(NewCall);
     // All mem intrinsics support dest alignment.
     const ConstantInt *Align = cast<ConstantInt>(CI->getArgOperand(3));
     MemCI->setDestAlignment(Align->getMaybeAlignValue());
     // Memcpy/Memmove also support source alignment.
     if (auto *MTI = dyn_cast<MemTransferInst>(MemCI))
       MTI->setSourceAlignment(Align->getMaybeAlignValue());
     break;

llvm/test/Bitcode/upgrade-memory-intrinsics.ll

 ; CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %p1, i8* %p2, i64 50, i1 false)
 ; CHECK: call void @llvm.memmove.p0i8.p0i8.i64(i8* %p2, i8* %p3, i64 1000, i1 false)
   call void @llvm.memset.p0i8.i64(i8* %p1, i8 55, i64 100, i32 0, i1 false)
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %p1, i8* %p2, i64 50, i32 0, i1 false)
   call void @llvm.memmove.p0i8.p0i8.i64(i8* %p2, i8* %p3, i64 1000, i32 0, i1 false)
   ret void
 }
 
+; Make sure that attributes are not dropped
+define void @test3(i8* %p1, i8* %p2, i8* %p3) {
+; CHECK-LABEL: @test
+; CHECK: call void @llvm.memset.p0i8.i64(i8* nonnull align 4 %p1, i8 signext 55, i64 zeroext 100, i1 immarg false)
+; CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull align 4 %p1, i8* readonly align 4 %p2, i64 zeroext 50, i1 immarg false)
+; CHECK: call void @llvm.memmove.p0i8.p0i8.i64(i8* nonnull align 4 %p2, i8* readonly align 4 %p3, i64 zeroext 1000, i1 immarg false)
+  call void @llvm.memset.p0i8.i64(i8* nonnull %p1, i8 signext 55, i64 zeroext 100, i32 signext 4, i1 immarg false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull %p1, i8* readonly %p2, i64 zeroext 50, i32 signext 4, i1 immarg false)
+  call void @llvm.memmove.p0i8.p0i8.i64(i8* nonnull %p2, i8* readonly %p3, i64 zeroext 1000, i32 signext 4, i1 immarg false)
+  ret void
+}
+
 ; CHECK: declare void @llvm.memset.p0i8.i64(i8* nocapture writeonly, i8, i64, i1 immarg)
 ; CHECK: declare void @llvm.memcpy.p0i8.p0i8.i64(i8* noalias nocapture writeonly, i8* noalias nocapture readonly, i64, i1 immarg)
 ; CHECK: declare void @llvm.memmove.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i1 immarg)
 declare void @llvm.memset.p0i8.i64(i8* nocapture writeonly, i8, i64, i32, i1)
 declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly , i8* nocapture readonly, i64, i32, i1)
 declare void @llvm.memmove.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i32, i1)
 

llvm/test/Transforms/IROutliner/illegal-memset.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt -S -verify -iroutliner -ir-outlining-no-cost -no-ir-sim-intrinsics < %s | FileCheck %s
 
 ; This test checks that we do not outline memset intrinsics since it requires
 ; extra address space checks.
 
 declare void @llvm.memset.p0i8.i64(i8* nocapture writeonly, i8, i64, i32, i1)
 
 define i64 @function1(i64 %x, i64 %z, i64 %n) {
 ; CHECK-LABEL: @function1(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[POOL:%.*]] = alloca [59 x i64], align 4
 ; CHECK-NEXT:    [[TMP:%.*]] = bitcast [59 x i64]* [[POOL]] to i8*
-; CHECK-NEXT:    call void @llvm.memset.p0i8.i64(i8* align 4 [[TMP]], i8 0, i64 236, i1 false)
+; CHECK-NEXT:    call void @llvm.memset.p0i8.i64(i8* nonnull align 4 [[TMP]], i8 0, i64 236, i1 false)
 ; CHECK-NEXT:    call void @outlined_ir_func_0(i64 [[N:%.*]], i64 [[X:%.*]], i64 [[Z:%.*]])
 ; CHECK-NEXT:    ret i64 0
 ;
 entry:
   %pool = alloca [59 x i64], align 4
   %tmp = bitcast [59 x i64]* %pool to i8*
   call void @llvm.memset.p0i8.i64(i8* nonnull %tmp, i8 0, i64 236, i32 4, i1 false)
   %cmp3 = icmp eq i64 %n, 0
   %a = add i64 %x, %z
   %c = add i64 %x, %z
   ret i64 0
 }
 
 define i64 @function2(i64 %x, i64 %z, i64 %n) {
 ; CHECK-LABEL: @function2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[POOL:%.*]] = alloca [59 x i64], align 4
 ; CHECK-NEXT:    [[TMP:%.*]] = bitcast [59 x i64]* [[POOL]] to i8*
-; CHECK-NEXT:    call void @llvm.memset.p0i8.i64(i8* align 4 [[TMP]], i8 0, i64 236, i1 false)
+; CHECK-NEXT:    call void @llvm.memset.p0i8.i64(i8* nonnull align 4 [[TMP]], i8 0, i64 236, i1 false)
 ; CHECK-NEXT:    call void @outlined_ir_func_0(i64 [[N:%.*]], i64 [[X:%.*]], i64 [[Z:%.*]])
 ; CHECK-NEXT:    ret i64 0
 ;
 entry:
   %pool = alloca [59 x i64], align 4
   %tmp = bitcast [59 x i64]* %pool to i8*
   call void @llvm.memset.p0i8.i64(i8* nonnull %tmp, i8 0, i64 236, i32 4, i1 false)
   %cmp3 = icmp eq i64 %n, 0
   %a = add i64 %x, %z
   %c = add i64 %x, %z
   ret i64 0
 }

llvm/test/Transforms/IROutliner/outline-memset.ll

 ; CHECK-NEXT:    ret i64 0
 ;
 ;
 ; CHECK: define internal void @outlined_ir_func_0(
 ; CHECK-NEXT:  newFuncRoot:
 ; CHECK-NEXT:    br label [[ENTRY_TO_OUTLINE:%.*]]
 ; CHECK:       entry_to_outline:
 ; CHECK-NEXT:    [[TMP:%.*]] = bitcast [59 x i64]* [[TMP0:%.*]] to i8*
-; CHECK-NEXT:    call void @llvm.memset.p0i8.i64(i8* align 4 [[TMP]], i8 0, i64 236, i1 false)
+; CHECK-NEXT:    call void @llvm.memset.p0i8.i64(i8* nonnull align 4 [[TMP]], i8 0, i64 236, i1 false)
 ; CHECK-NEXT:    [[CMP3:%.*]] = icmp eq i64 [[TMP1:%.*]], 0
 ; CHECK-NEXT:    [[A:%.*]] = add i64 [[TMP2:%.*]], [[TMP3:%.*]]
 ; CHECK-NEXT:    [[C:%.*]] = add i64 [[TMP2]], [[TMP3]]
 ; CHECK-NEXT:    br label [[ENTRY_AFTER_OUTLINE_EXITSTUB:%.*]]
 ; CHECK:       entry_after_outline.exitStub:
 ; CHECK-NEXT:    ret void
 ;