Replace func name with regex for update test scripts

[ggeorgakoudis]

Summary

The patch adds an argument to update test scripts, such as update_cc_test_checks, for replacing a function name matching a regex. This functionality is needed to match generated function signatures that include file hashes. Example:

The function signature for the following function:

__omp_offloading_50_b84c41e__Z9ftemplateIiET_i_l30_worker

with --replace-function-regex "__omp_offloading_[0-9]+_[a-z0-9]+_(.*)" will become:

CHECK-LABEL: @{{__omp_offloading_[0-9]+_[a-z0-9]+__Z9ftemplateIiET_i_l30_worker}}(

[ggeorgakoudis]

Fix to check regex input exists

[ggeorgakoudis]

Move implementation in common, add tests

[ggeorgakoudis]

Remove testing print

[jdoerfert]

Cool, almost able to generate offloading tests :) LG

[ggeorgakoudis]

Fix for multiple capture groups

[ggeorgakoudis]

Fix (another) for multiple capture groups

[ggeorgakoudis]

Change regex to avoid failing windows test

[ggeorgakoudis]

Add triple in test to avoid spurious failures

[thakis]

This breaks check-llvm on linux and mac:
http://45.33.8.238/linux/41398/step_12.txt
http://45.33.8.238/macm1/5291/step_10.txt
http://lab.llvm.org:8011/#/builders/109
(and on my local build too)

Please take a look, and please revert for now if it takes a while to fix.

[ggeorgakoudis]

In D97107#2617897, @thakis wrote:

This breaks check-llvm on linux and mac:
http://45.33.8.238/linux/41398/step_12.txt
http://45.33.8.238/macm1/5291/step_10.txt
http://lab.llvm.org:8011/#/builders/109
(and on my local build too)

Please take a look, and please revert for now if it takes a while to fix.

Reverted, I'm going to test better, fix and re-land.

[ggeorgakoudis]

NFC change in update_llc_test_checks for compatibility

[ggeorgakoudis]

Fix for upstreaming

[nikic]

Looks like this has broken update_analyze_test_checks.py.

clang/test/utils/update_cc_test_checks/Inputs/generated-funcs-regex.c

+// RUN: %clang_cc1 -fopenmp %s -emit-llvm -o - | FileCheck %s
+
+void use(int);
+
+void foo(int a)
+{
+    #pragma omp target
+        use(a);
+}

clang/test/utils/update_cc_test_checks/Inputs/generated-funcs-regex.c.expected

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --include-generated-funcs --replace-function-regex "__omp_offloading_[0-9]+_[a-z0-9]+_(.*)_l[0-9]+"
+// RUN: %clang_cc1 -fopenmp %s -emit-llvm -o - | FileCheck %s
+
+void use(int);
+
+void foo(int a)
+{
+    #pragma omp target
+        use(a);
+}
+// CHECK-LABEL: @foo(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[A_CASTED:%.*]] = alloca i64, align 8
+// CHECK-NEXT:    store i32 [[A:%.*]], i32* [[A_ADDR]], align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[A_ADDR]], align 4
+// CHECK-NEXT:    [[CONV:%.*]] = bitcast i64* [[A_CASTED]] to i32*
+// CHECK-NEXT:    store i32 [[TMP0]], i32* [[CONV]], align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i64, i64* [[A_CASTED]], align 8
+// CHECK-NEXT:    call void @{{__omp_offloading_[0-9]+_[a-z0-9]+_foo_l[0-9]+}}(i64 [[TMP1]]) [[ATTR3:#.*]]
+// CHECK-NEXT:    ret void
+//
+//
+// CHECK-LABEL: @{{__omp_offloading_[0-9]+_[a-z0-9]+_foo_l[0-9]+}}(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[A_ADDR:%.*]] = alloca i64, align 8
+// CHECK-NEXT:    store i64 [[A:%.*]], i64* [[A_ADDR]], align 8
+// CHECK-NEXT:    [[CONV:%.*]] = bitcast i64* [[A_ADDR]] to i32*
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[CONV]], align 8
+// CHECK-NEXT:    call void @use(i32 signext [[TMP0]])
+// CHECK-NEXT:    ret void
+//

clang/test/utils/update_cc_test_checks/generated-funcs-regex.test

+## Test that CHECK lines are generated for clang-generated functions replaced
+## by regex
+
+## RUN: cp %S/Inputs/generated-funcs-regex.c %t-generated-funcs-regex.c && %update_cc_test_checks --include-generated-funcs --replace-function-regex "__omp_offloading_[0-9]+_[a-z0-9]+_(.*)_l[0-9]+" -- %t-generated-funcs-regex.c
+# RUN: diff -u %S/Inputs/generated-funcs-regex.c.expected %t-generated-funcs-regex.c
+
+## Check that re-running update_cc_test_checks doesn't change the output
+# RUN: %update_cc_test_checks %t-generated-funcs-regex.c
+# RUN: diff -u %S/Inputs/generated-funcs-regex.c.expected %t-generated-funcs-regex.c

llvm/utils/UpdateTestChecks/common.py

def parse_commandline_args(parser):
   parser.add_argument('-u', '--update-only', action='store_true',
                       help='Only update test if it was already autogened')
   parser.add_argument('--force-update', action='store_true',
                       help='Update test even if it was autogened by a different script')
   parser.add_argument('--enable', action='store_true', dest='enabled', default=True,
                        help='Activate CHECK line generation from this point forward')
   parser.add_argument('--disable', action='store_false', dest='enabled',
                       help='Deactivate CHECK line generation from this point forward')
+  parser.add_argument('--replace-function-regex', nargs='+', default=[],
+                      help='List of regular expressions to replace matching function names')
   args = parser.parse_args()
   global _verbose
   _verbose = args.verbose
   return args
 
 
 class InputLineInfo(object):
   def __init__(self, line, line_number, args, argv):
def __str__(self):
     return self.scrub
 
 class FunctionTestBuilder:
   def __init__(self, run_list, flags, scrubber_args):
     self._verbose = flags.verbose
     self._record_args = flags.function_signature
     self._check_attributes = flags.check_attributes
     self._scrubber_args = scrubber_args
+    # Strip double-quotes if input was read by UTC_ARGS
+    self._replace_function_regex = list(map(lambda x: x.strip('"'), flags.replace_function_regex))
     self._func_dict = {}
     self._func_order = {}
     for tuple in run_list:
       for prefix in tuple[0]:
         self._func_dict.update({prefix:dict()})
         self._func_order.update({prefix: []})
 
   def finish_and_get_func_dict(self):
for m in function_re.finditer(raw_tool_output):
             else:
               # This means a previous RUN line produced a body for this function
               # that is different from the one produced by this current RUN line,
               # so the body can't be common accross RUN lines. We use None to
               # indicate that.
               self._func_dict[prefix][func] = None
               continue
 
+        # Replace function names matching the regex.
+        for regex in self._replace_function_regex:
+          # Pattern that matches capture groups in the regex in leftmost order.
+          group_regex = re.compile('\(.*?\)')
+          # Replace function name with regex.
+          match = re.match(regex, func)
+          if match:
+            func_repl = regex
+            # Replace any capture groups with their matched strings.
+            for g in match.groups():
+              func_repl = group_regex.sub(g, func_repl, count=1)
+            func = '{{' + func_repl + '}}'
+
+          # Replace all calls to regex matching functions.
+          matches = re.findall(regex, scrubbed_body)
+          for found in matches:
+            print('found', found)
+            func_repl = regex
+            # Replace any capture groups with their matched strings.
+            func_repl = group_regex.sub(found, func_repl, count=1)
+            # Substitute function call names that match the regex with the same
+            # capture groups set.
+            scrubbed_body = re.sub(func_repl, '{{' + func_repl + '}}', scrubbed_body)
+
         self._func_dict[prefix][func] = function_body(
             scrubbed_body, scrubbed_extra, args_and_sig, attrs)
         self._func_order[prefix].append(func)
 
   def _get_failed_prefixes(self):
     # This returns the list of those prefixes that failed to match any function,
     # because there were conflicting bodies produced by different RUN lines, in
     # all instances of the prefix. Effectively, this prefix is unused and should
if action.const is not None:  # action stores a constant (usually True/False)
       # Skip actions with different constant values (this happens with boolean
       # --foo/--no-foo options)
       if value != action.const:
         continue
     if parser.get_default(action.dest) == value:
       continue  # Don't add default values
     autogenerated_note_args += action.option_strings[0] + ' '
     if action.const is None:  # action takes a parameter
+      if action.nargs == '+':
+        value = ' '.join(map(lambda v: '"' + v.strip('"') + '"', value))
       autogenerated_note_args += '%s ' % value
   if autogenerated_note_args:
     autogenerated_note_args = ' %s %s' % (UTC_ARGS_KEY, autogenerated_note_args[:-1])
   return autogenerated_note_args
 
 
 def check_for_command(line, parser, args, argv, argparse_callback):
     cmd_m = UTC_ARGS_CMD.match(line)