[llvm] [AMDGPU] Skip -mattr=dumpcode test on big-endian hosts (PR #127064)

[Jay Foad via llvm-commits]

https://github.com/jayfoad created https://github.com/llvm/llvm-project/pull/127064

I did not know that the output is different depending on the host
endianness. This should be fixed properly but in the mean time this
patch fixes the buildbot failures.


>From d6e41646c7446415b1bd9cfaa63c3d2edb0c9bee Mon Sep 17 00:00:00 2001
From: Jay Foad <jay.foad at amd.com>
Date: Thu, 13 Feb 2025 14:17:19 +0000
Subject: [PATCH] [AMDGPU] Skip -mattr=dumpcode test on big-endian hosts

I did not know that the output is different depending on the host
endianness. This should be fixed properly but in the mean time this
patch fixes the buildbot failures.
---
 llvm/test/CodeGen/AMDGPU/dumpcode.ll | 1 +
 1 file changed, 1 insertion(+)

diff --git a/llvm/test/CodeGen/AMDGPU/dumpcode.ll b/llvm/test/CodeGen/AMDGPU/dumpcode.ll
index 1acec2997aa8f..27205cb644436 100644
--- a/llvm/test/CodeGen/AMDGPU/dumpcode.ll
+++ b/llvm/test/CodeGen/AMDGPU/dumpcode.ll
@@ -1,4 +1,5 @@
 ; RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -mattr=dumpcode -filetype=obj < %s | llvm-objcopy --dump-section .AMDGPU.disasm=- - /dev/null | FileCheck %s -check-prefix=GFX10
+; REQUIRES: host-byteorder-little-endian
 
 ; GFX10: f:
 ; GFX10-NEXT: BB0_0: