f1987c74ee
Similar to 806761a7629df268c8aed49657aeccffa6bca449 -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.
43 lines
973 B
LLVM
43 lines
973 B
LLVM
; RUN: llc < %s -mtriple=xcore | FileCheck %s
|
|
|
|
@a = external dso_local constant [0 x i32], section ".cp.rodata"
|
|
@b = external dso_local global [0 x i32]
|
|
|
|
define ptr @f1() nounwind {
|
|
entry:
|
|
; CHECK-LABEL: f1:
|
|
; CHECK: ldaw r11, cp[a+4]
|
|
; CHECK: mov r0, r11
|
|
%0 = getelementptr [0 x i32], ptr @a, i32 0, i32 1
|
|
ret ptr %0
|
|
}
|
|
|
|
define ptr @f2() nounwind {
|
|
entry:
|
|
; CHECK-LABEL: f2:
|
|
; CHECK: ldaw r0, dp[b+4]
|
|
%0 = getelementptr [0 x i32], ptr @b, i32 0, i32 1
|
|
ret ptr %0
|
|
}
|
|
|
|
; Don't fold negative offsets into cp / dp accesses to avoid a relocation
|
|
; error if the address + addend is less than the start of the cp / dp.
|
|
|
|
define ptr @f3() nounwind {
|
|
entry:
|
|
; CHECK-LABEL: f3:
|
|
; CHECK: ldaw r11, cp[a]
|
|
; CHECK: sub r0, r11, 4
|
|
%0 = getelementptr [0 x i32], ptr @a, i32 0, i32 -1
|
|
ret ptr %0
|
|
}
|
|
|
|
define ptr @f4() nounwind {
|
|
entry:
|
|
; CHECK-LABEL: f4:
|
|
; CHECK: ldaw [[REG:r[0-9]+]], dp[b]
|
|
; CHECK: sub r0, [[REG]], 4
|
|
%0 = getelementptr [0 x i32], ptr @b, i32 0, i32 -1
|
|
ret ptr %0
|
|
}
|