1; When an SCC got split due to inlining, we have two mechanisms for reprocessing the updated SCC, first is UR.UpdatedC
2; that repeatedly rerun the new, current SCC; second is a worklist for all newly split SCCs. We need to avoid rerun of
3; the same SCC when the SCC is set to be processed by both mechanisms back to back. In pathological cases, such extra,
4; redundant rerun could cause exponential size growth due to inlining along cycles.
5;
6; The test cases here illustrates potential redundant rerun and how it's prevented, however there's no extra inlining
7; even if we allow the redundant rerun. In real code, when inliner makes different decisions for different call sites
8; of the same caller-callee edge, we could end up getting more recursive inlining without SCC mutation.
9;
10; REQUIRES: asserts
11; RUN: opt < %s -passes='cgscc(inline)' -inline-threshold=500 -debug-only=cgscc -S 2>&1 | FileCheck %s
12
13; CHECK: Running an SCC pass across the RefSCC: [(test1_a, test1_b, test1_c)]
14; CHECK: Enqueuing the existing SCC in the worklist:(test1_b)
15; CHECK: Enqueuing a newly formed SCC:(test1_c)
16; CHECK: Enqueuing a new RefSCC in the update worklist: [(test1_b)]
17; CHECK: Switch an internal ref edge to a call edge from 'test1_a' to 'test1_c'
18; CHECK: Switch an internal ref edge to a call edge from 'test1_a' to 'test1_a'
19; CHECK: Re-running SCC passes after a refinement of the current SCC: (test1_c, test1_a)
20; CHECK: Skipping redundant run on SCC: (test1_c, test1_a)
21
22declare void @external(i32 %seed)
23
24define void @test1_a(i32 %num) {
25entry:
26  call void @test1_b(i32 %num)
27  call void @external(i32 %num)
28  ret void
29}
30
31define void @test1_b(i32 %num) {
32entry:
33  call void @test1_c(i32 %num)
34  call void @test1_a(i32 %num)
35  call void @external(i32 %num)
36  ret void
37}
38
39define void @test1_c(i32 %num) #0 {
40  call void @test1_a(i32 %num)
41  ret void
42}
43
44attributes #0 = { noinline nounwind optnone }
45