e15d72adac
Currently the naming scheme is a bit funky; the specializations are named after the original function followed by an arbitrary decimal number. This makes it hard to debug inlined specializations of recursive functions. With this patch I am adding ".specialized." in between of the original name and the suffix, which is now a single increment counter.
28 lines
829 B
LLVM
28 lines
829 B
LLVM
; RUN: opt -S --passes="ipsccp<func-spec>" -force-specialization < %s | FileCheck %s
|
|
|
|
; Tests that `bar` has been specialized and that the compiler did not crash
|
|
; while attempting to promote the alloca in `entry`.
|
|
; CHECK: bar.specialized.1
|
|
|
|
@block = internal constant [8 x i8] zeroinitializer, align 1
|
|
|
|
define dso_local void @entry() {
|
|
%1 = alloca i32, align 4
|
|
call void @foo(ptr nonnull %1)
|
|
ret void
|
|
}
|
|
|
|
define internal void @foo(ptr nocapture readnone %0) {
|
|
%2 = alloca i32, align 4
|
|
call void @bar(ptr nonnull %2, ptr nonnull @block)
|
|
call void @bar(ptr nonnull %2, ptr nonnull getelementptr inbounds ([8 x i8], ptr @block, i64 0, i64 4))
|
|
ret void
|
|
}
|
|
|
|
define internal void @bar(ptr nocapture readonly %0, ptr nocapture readonly %1) {
|
|
%3 = load i32, ptr %0, align 4
|
|
%4 = load i32, ptr %1, align 4
|
|
ret void
|
|
}
|
|
|