[llvm] [AMDGPU][test]added unique and sort options for update_mc_test_check script (PR #111769)

[Joe Nash via llvm-commits]

================
@@ -0,0 +1,8 @@
+# NOTE: Assertions have been autogenerated by utils/update_mc_test_checks.py UTC_ARGS: --sort
+# RUN: llvm-mc -triple=amdgcn -mcpu=gfx1100 -disassemble -show-encoding %s 2>&1 | FileCheck -check-prefixes=CHECK %s
+
+0x00,0x00,0x00,0x7e
+# CHECK: v_nop                                   ; encoding: [0x00,0x00,0x00,0x7e]
----------------
Sisyph wrote:

Perhaps you could exclude disasm file sorting from this patch? It seems like the other cases are working well, and we don't need to do it all at once. I don't see how sorting on the hex is going to be useful. The bits that encode the opcode would make sense to sort on, but that varies so much from instruction to instruction it will be horrible to maintain. With sorting on the output, can you redirect errors to null? Or maybe we separate those test files with both passing and error tests?

https://github.com/llvm/llvm-project/pull/111769