1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
|
; RUN: llc -verify-machineinstrs -mcpu=pwr7 -O2 -relocation-model=pic < %s | FileCheck %s
target datalayout = "e-m:e-i64:64-n32:64"
target triple = "powerpc64le-unknown-linux-gnu"
; Test back-to-back stores of TLS variables to ensure call sequences no
; longer overlap.
@__once_callable = external thread_local global ptr
@__once_call = external thread_local global ptr
define i64 @call_once(i64 %flag, ptr %ptr) {
entry:
%var = alloca ptr, align 8
store ptr %ptr, ptr %var, align 8
store ptr %var, ptr @__once_callable, align 8
store ptr @__once_call_impl, ptr @__once_call, align 8
ret i64 %flag
}
; CHECK-LABEL: call_once:
; CHECK: addi 3, {{[0-9]+}}, __once_callable@got@tlsgd@l
; CHECK: bl __tls_get_addr(__once_callable@tlsgd)
; CHECK-NEXT: nop
; FIXME: We could check here for 'std {{[0-9]+}}, 0(3)', but that no longer
; works because, with new scheduling freedom, we create a copy of R3 based on the
; initial scheduling, but don't coalesce it again after we move the instructions
; so that the copy is no longer necessary.
; CHECK: addi 3, {{[0-9]+}}, __once_call@got@tlsgd@l
; CHECK: bl __tls_get_addr(__once_call@tlsgd)
; CHECK-NEXT: nop
; FIXME: We don't really need the copy here either, we could move the store up.
; CHECK: mr [[REG1:[0-9]+]], 3
; CHECK: std {{[0-9]+}}, 0([[REG1]])
declare void @__once_call_impl()
|
