Similar to 806761a762
-mtriple= specifies the full target triple while -march= merely sets the
architecture part of the default target triple (e.g. Windows, macOS),
leaving a target triple which may not make sense.
Therefore, -march= is error-prone and not recommended for tests without a target
triple. The issue has been benign as we recognize xcore-apple-darwin as ELF instead
of rejecting it outrightly.
21 lines
386 B
LLVM
21 lines
386 B
LLVM
; RUN: llc < %s -mtriple=xcore -mcpu=xs1b-generic | FileCheck %s
|
|
|
|
define ptr @addr_G() {
|
|
entry:
|
|
; CHECK-LABEL: addr_G:
|
|
; CHECK: get r11, id
|
|
ret ptr @G
|
|
}
|
|
|
|
@G = thread_local global i32 15
|
|
; CHECK: .section .dp.data,"awd",@progbits
|
|
; CHECK: G:
|
|
; CHECK: .long 15
|
|
; CHECK: .long 15
|
|
; CHECK: .long 15
|
|
; CHECK: .long 15
|
|
; CHECK: .long 15
|
|
; CHECK: .long 15
|
|
; CHECK: .long 15
|
|
; CHECK: .long 15
|