[Utils] Add prefix parameter in update test checks to avoid FileCheck conflicts

[ggeorgakoudis]

Summary

IR values convert to check prefix FileCheck variables for IR checks. For example, nameless values, e.g., %0, convert to check prefix TMP FileCheck variables, e.g., [[TMP0:%.*]]. This check prefix may clash with named values that have the same name and that causes auto-generated tests to fail. Currently a warning is emitted to change the names of the IR values but this is not always possible, if for example they are generated by clang. Manual intervention to fix the FileCheck variable names is too tedious. This patch add a parameter to prefix conflicting FileCheck variable names with a user-provided string to automate the process.

[jdoerfert]

Cool, thanks. LGTM, two nits.

llvm/utils/UpdateTestChecks/common.py

532

Move the re.match into a helper (also used below), maybe: may_clash_with_default_check_prefix_name(check_prefix, var)

563

I would not add the and not _prefix_filecheck... stuff here because if the other condition holds a warning seems appropriate, maybe the prefix was chosen pooreley, etc.

[ggeorgakoudis]

Update for comments

clang/test/utils/update_cc_test_checks/Inputs/resolve-tmp-conflict.cpp

+// RUN: %clang_cc1 -std=c++11 -triple x86_64-unknown-linux-gnu -emit-llvm %s -o - | FileCheck %s
+
+void foo(int a) {
+  int &tmp0 = a;
+  int &&tmp1 = 1;
+  tmp1 = a;
+  return;
+}

clang/test/utils/update_cc_test_checks/Inputs/resolve-tmp-conflict.cpp.expected

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --function-signature --prefix-filecheck-ir-name _
+// RUN: %clang_cc1 -std=c++11 -triple x86_64-unknown-linux-gnu -emit-llvm %s -o - | FileCheck %s
+
+// CHECK-LABEL: define {{[^@]+}}@_Z3fooi
+// CHECK-SAME: (i32 [[A:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[_TMP0:%.*]] = alloca i32*, align 8
+// CHECK-NEXT:    [[_TMP1:%.*]] = alloca i32*, align 8
+// CHECK-NEXT:    [[REF_TMP:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    store i32 [[A]], i32* [[A_ADDR]], align 4
+// CHECK-NEXT:    store i32* [[A_ADDR]], i32** [[_TMP0]], align 8
+// CHECK-NEXT:    store i32 1, i32* [[REF_TMP]], align 4
+// CHECK-NEXT:    store i32* [[REF_TMP]], i32** [[_TMP1]], align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[A_ADDR]], align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[_TMP1]], align 8
+// CHECK-NEXT:    store i32 [[TMP0]], i32* [[TMP1]], align 4
+// CHECK-NEXT:    ret void
+//
+void foo(int a) {
+  int &tmp0 = a;
+  int &&tmp1 = 1;
+  tmp1 = a;
+  return;
+}

clang/test/utils/update_cc_test_checks/resolve-tmp-conflict.test

+## Test that CHECK lines generated avoid naming conflicts with FileCheck IR variables
+
+# RUN: cp %S/Inputs/resolve-tmp-conflict.cpp %t.cpp && %update_cc_test_checks --function-signature --prefix-filecheck-ir-name _ %t.cpp
+# RUN: diff -u %S/Inputs/resolve-tmp-conflict.cpp.expected %t.cpp
+
+## Check that re-running update_cc_test_checks doesn't change the output
+# RUN: %update_cc_test_checks %t.cpp
+# RUN: diff -u %S/Inputs/resolve-tmp-conflict.cpp.expected %t.cpp

llvm/utils/UpdateTestChecks/common.py

class string:
     expandtabs = str.expandtabs
 else:
   import string
 
 ##### Common utilities for update_*test_checks.py
 
 
 _verbose = False
+_prefix_filecheck_ir_name = ''
 
 def parse_commandline_args(parser):
   parser.add_argument('--include-generated-funcs', action='store_true',
                       help='Output checks for functions not in source')
   parser.add_argument('-v', '--verbose', action='store_true',
                       help='Show verbose output')
   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')
+  parser.add_argument('--prefix-filecheck-ir-name', default='',
+                      help='Add a prefix to FileCheck IR value names to avoid conflicts with scripted names')
   args = parser.parse_args()
   global _verbose
   _verbose = args.verbose
   return args
 
 
 class InputLineInfo(object):
   def __init__(self, line, line_number, args, argv):
     self.line = line
     self.line_number = line_number
     self.args = args
     self.argv = argv
 
 
 class TestInfo(object):
   def __init__(self, test, parser, script_name, input_lines, args, argv,
                comment_prefix, argparse_callback):
     self.parser = parser
     self.argparse_callback = argparse_callback
     self.path = test
     self.args = args
+    if args.prefix_filecheck_ir_name:
+      global _prefix_filecheck_ir_name
+      _prefix_filecheck_ir_name = args.prefix_filecheck_ir_name
     self.argv = argv
     self.input_lines = input_lines
     self.run_lines = find_run_lines(test, self.input_lines)
     self.comment_prefix = comment_prefix
     if self.comment_prefix is None:
       if self.path.endswith('.mir'):
         self.comment_prefix = '#'
       else:

 
 # Return true if this kind of IR value is "global", basically if it matches '#{{.*}}'.
 def is_global_scope_ir_value_match(match):
     return nameless_values[get_idx_from_ir_value_match(match)].global_ir_prefix is not None
 
 # Create a FileCheck variable name based on an IR name.
 def get_value_name(var, check_prefix):
   var = var.replace('!', '')
+  # This is a nameless value, prepend check_prefix.
   if var.isdigit():
     var = check_prefix + var
+  else:
+    # This is a named value that clashes with the check_prefix, prepend with _prefix_filecheck_ir_name,
+    # if it has been defined.
+    if check_prefix and re.match(r'^' + check_prefix + r'[0-9]+?$', var, re.IGNORECASE) \
+      and _prefix_filecheck_ir_name:
+      var = _prefix_filecheck_ir_name + var
   var = var.replace('.', '_')
   var = var.replace('-', '_')
   return var.upper()
 
 # Create a FileCheck variable from regex.
 def get_value_definition(var, match):
   # for backwards compatibility we check locals with '.*'
   if is_local_def_ir_value_match(match):

 def generalize_check_lines(lines, is_analyze, vars_seen, global_vars_seen):
   # This gets called for each match that occurs in
   # a line. We transform variables we haven't seen
   # into defs, and variables we have seen into uses.
   def transform_line_vars(match):
     pre, check = get_ir_prefix_from_ir_value_match(match)
     var = get_name_from_ir_value_match(match)
     for nameless_value in nameless_values:
-        if nameless_value.check_prefix and re.match(r'^' + nameless_value.check_prefix + r'[0-9]+?$', var, re.IGNORECASE):
-            warn("Change IR value name '%s' to prevent possible conflict with scripted FileCheck name." % (var,))
+        if nameless_value.check_prefix and re.match(r'^' + nameless_value.check_prefix + r'[0-9]+?$', var, re.IGNORECASE) \
+          and not _prefix_filecheck_ir_name:
+          warn("Change IR value name '%s' or use -prefix-ir-filecheck-name to prevent possible conflict"
+            " with scripted FileCheck name." % (var,))
     key = (var, get_check_key_from_ir_value_match(match))
     is_local_def = is_local_def_ir_value_match(match)
     if is_local_def and key in vars_seen:
       rv = get_value_use(var, match, get_nameless_check_prefix_from_ir_value_match(match))
     elif not is_local_def and key in global_vars_seen:
       rv = get_value_use(var, match, global_vars_seen[key])
     else:
       if is_local_def: