[utils] fixing update_mir_test_checks.py's language (mir/IR/etc) detection

[rtereshin]

Summary

Adding a little more robust language detection for utils/update_mir_test_checks.py parsing it from -x <LANG> arg of llc if it's not obvious from the test file name.

This patch is dependent upon https://reviews.llvm.org/D43603

[bogner]

Looks good, pending what we decide to do about the testing.

test/CodeGen/AArch64/test-update_mir_test_checks.py-itself.mir

 # UNSUPPORTED: system-windows
 #
 # RUN: sed -e 's/^# RUN-NESTED:/# RUN:/' %s | sed -e 's/^# SOURCE.*$//' > %t \
 # RUN:   && %S/../../../utils/update_mir_test_checks.py --add-vreg-checks \
 # RUN:     %t 2>&1 | FileCheck %s --check-prefix=STDERR \
 # RUN:   && cat %t | FileCheck %s --check-prefix=SOURCE
 #
 # RUN-NESTED: llc -mtriple aarch64-apple-ios -run-pass instruction-select %s \
 # RUN-NESTED:     -x mir -verify-machineinstrs -simplify-mir -o - | FileCheck %s
 
 # STDERR-NOT:  WARNING
 # STDERR:      WARNING: [[TMP_FILE_NAME:/.*\.tmp]]: Skipping non-llc RUN line:
 # STDERR-SAME:     sed -e {{.*}} RUN-NESTED{{.*}}SOURCE{{.*}}STDERR{{.*}}SOURCE
 # STDERR-NOT:  WARNING
 
+# SOURCE-LABEL: # NOTE: Assertions have been autogenerated by
+# SOURCE-SAME:  utils/update_mir_test_checks.py
+#
 # SOURCE:      WARNING
 # SOURCE-NEXT: WARNING
 # SOURCE-NEXT:     sed
 # SOURCE-NEXT: WARNING
 #
 # SOURCE-LABEL: --- |
 # SOURCE-NEXT:   target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
 # SOURCE-NEXT:   target triple = "aarch64-apple-ios"

utils/update_mir_test_checks.py

 import os
 import re
 import subprocess
 import sys
 
 RUN_LINE_RE = re.compile('^\s*[;#]\s*RUN:\s*(.*)$')
 TRIPLE_ARG_RE = re.compile(r'-mtriple[= ]([^ ]+)')
 MARCH_ARG_RE = re.compile(r'-march[= ]([^ ]+)')
+LANG_ARG_RE = re.compile(r'-x[= ]([^ ]+)')
 TRIPLE_IR_RE = re.compile(r'^\s*target\s+triple\s*=\s*"([^"]+)"$')
 CHECK_PREFIX_RE = re.compile('--?check-prefix(?:es)?[= ](\S+)')
 CHECK_RE = re.compile(r'^\s*[;#]\s*([^:]+?)(?:-NEXT|-NOT|-DAG|-LABEL)?:')
 
 MIR_FUNC_NAME_RE = re.compile(r' *name: *(?P<func>[A-Za-z0-9_.-]+)')
 MIR_BODY_BEGIN_RE = re.compile(r' *body: *\|')
 MIR_BASIC_BLOCK_RE = re.compile(r' *bb\.[0-9]+.*:$')
 VREG_RE = re.compile(r'(%[0-9]+)(?::[a-z0-9_]+)?(?:\([<>a-z0-9 ]+\))?')
def __call__(self, args, ir):
             stdout = subprocess.check_output('{} {}'.format(self.bin, args),
                                              shell=True, stdin=ir_file)
             # Fix line endings to unix CR style.
             stdout = stdout.replace('\r\n', '\n')
         return stdout
 
 
 class Run:
-    def __init__(self, prefixes, cmd_args, triple):
+    def __init__(self, prefixes, cmd_args, triple, lang):
         self.prefixes = prefixes
         self.cmd_args = cmd_args
         self.triple = triple
+        self.lang = lang
 
     def __getitem__(self, index):
         return [self.prefixes, self.cmd_args, self.triple][index]
 
 
 def log(msg, verbose=True):
     if verbose:
         print(msg, file=sys.stderr)
for l in run_lines:
         m = TRIPLE_ARG_RE.search(llc_cmd)
         if m:
             triple = m.group(1)
         # If we find -march but not -mtriple, use that.
         m = MARCH_ARG_RE.search(llc_cmd)
         if m and not triple:
             triple = '{}--'.format(m.group(1))
 
+        m = LANG_ARG_RE.search(llc_cmd)
+        lang = (m.group(1) if m else test.split('.')[-1]).lower()
+
         cmd_args = llc_cmd[len('llc'):].strip()
         cmd_args = cmd_args.replace('< %s', '').replace('%s', '').strip()
 
         check_prefixes = [item for m in CHECK_PREFIX_RE.finditer(filecheck_cmd)
                           for item in m.group(1).split(',')]
         if not check_prefixes:
             check_prefixes = ['CHECK']
         all_prefixes += check_prefixes
 
-        run_list.append(Run(check_prefixes, cmd_args, triple))
+        run_list.append(Run(check_prefixes, cmd_args, triple, lang))
 
     # Remove any common prefixes. We'll just leave those entirely alone.
     common_prefixes = set([prefix for prefix in all_prefixes
                            if all_prefixes.count(prefix) > 1])
     for run in run_list:
         run.prefixes = [p for p in run.prefixes if p not in common_prefixes]
 
     return run_list, common_prefixes
def update_test_file(llc, test, remove_common_prefixes=False,
     log('Rewriting FileCheck prefixes: {}'.format(prefix_set), verbose)
 
     if remove_common_prefixes:
         prefix_set.update(common_prefixes)
     elif common_prefixes:
         warn('Ignoring common prefixes: {}'.format(common_prefixes),
              test_file=test)
 
-    comment_char = '#' if test.endswith('.mir') else ';'
+    comment_char = '#' if any(run.lang == 'mir' for run in run_list) else ';'
     autogenerated_note = ('{} NOTE: Assertions have been autogenerated by '
                           'utils/{}'.format(comment_char,
                                             os.path.basename(__file__)))
     output_lines = []
     output_lines.append(autogenerated_note)
 
     for input_line in input_lines:
         if input_line == autogenerated_note: