Teach valuetracking that byval arguments with a specified alignment are

aligned.

Teach memcpyopt to not give up all hope when confonted with an underaligned
memcpy feeding an overaligned byval.  If the *source* of the memcpy can be
determined to be adequeately aligned, or if it can be forced to be, we can
eliminate the memcpy.

This addresses PR9794.  We now compile the example into:

define i32 @f(%struct.p* nocapture byval align 8 %q) nounwind ssp {
entry:
  %call = call i32 @g(%struct.p* byval align 8 %q) nounwind
  ret i32 %call
}

in both x86-64 and x86-32 mode.  We still don't get a tailcall though,
because tailcalls apparently can't handle byval.

llvm-svn: 131884

1 files changed, 20 insertions, 0 deletions

diff --git a/llvm/test/Transforms/MemCpyOpt/memcpy.ll b/llvm/test/Transforms/MemCpyOpt/memcpy.ll
index b387d32a7d5..5c6a94ce5c9 100644
--- a/llvm/test/Transforms/MemCpyOpt/memcpy.ll
+++ b/llvm/test/Transforms/MemCpyOpt/memcpy.ll
@@ -109,3 +109,23 @@ define void @test6(i8 *%P) {
 ; CHECK-NEXT: ret void
 }
 
+
+; PR9794 - Should forward memcpy into byval argument even though the memcpy
+; isn't itself 8 byte aligned.
+%struct.p = type { i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32 }
+
+define i32 @test7(%struct.p* nocapture byval align 8 %q) nounwind ssp {
+entry:
+  %agg.tmp = alloca %struct.p, align 4
+  %tmp = bitcast %struct.p* %agg.tmp to i8*
+  %tmp1 = bitcast %struct.p* %q to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp, i8* %tmp1, i64 48, i32 4, i1 false)
+  %call = call i32 @g(%struct.p* byval align 8 %agg.tmp) nounwind
+  ret i32 %call
+; CHECK: @test7
+; CHECK: call i32 @g(%struct.p* byval align 8 %q) nounwind
+}
+
+declare i32 @g(%struct.p* byval align 8)
+
+