diff options
| author | Sanjay Patel <spatel@rotateright.com> | 2016-03-23 21:40:53 +0000 |
|---|---|---|
| committer | Sanjay Patel <spatel@rotateright.com> | 2016-03-23 21:40:53 +0000 |
| commit | c062937b5ea1101f5a408359763903d9f2df90dd (patch) | |
| tree | 638f547c9754595e1c6b6fc2431505103613475c | |
| parent | 825efc59cb586f9698d9b0d9e763f80aa2f1cdd5 (diff) | |
reorganize llc checks script to allow more flexibility; NFCI
The goal is to enhance this script to be used with opt and clang:
Group all of the regexes together, so it's easier to see what's going on.
This will make it easier to break main() up into pieces too.
Also, note that some of the regexes are for x86-specific asm.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@264197 91177308-0d34-0410-b5e6-96231b3b80d8
| -rwxr-xr-x | utils/update_llc_test_checks.py | 59 |
1 files changed, 31 insertions, 28 deletions
diff --git a/utils/update_llc_test_checks.py b/utils/update_llc_test_checks.py index 4ce034df308..781fab54f69 100755 --- a/utils/update_llc_test_checks.py +++ b/utils/update_llc_test_checks.py @@ -15,7 +15,7 @@ import sys import tempfile import re - +# Invoke the tool that is being tested. def llc(args, cmd_args, ir): with open(ir) as ir_file: stdout = subprocess.check_output(args.llc_binary + ' ' + cmd_args, @@ -25,33 +25,45 @@ def llc(args, cmd_args, ir): return stdout -ASM_SCRUB_WHITESPACE_RE = re.compile(r'(?!^(| \w))[ \t]+', flags=re.M) -ASM_SCRUB_TRAILING_WHITESPACE_RE = re.compile(r'[ \t]+$', flags=re.M) -ASM_SCRUB_SHUFFLES_RE = ( +# RegEx: this is where the magic happens. + +SCRUB_WHITESPACE_RE = re.compile(r'(?!^(| \w))[ \t]+', flags=re.M) +SCRUB_TRAILING_WHITESPACE_RE = re.compile(r'[ \t]+$', flags=re.M) +SCRUB_X86_SHUFFLES_RE = ( re.compile( r'^(\s*\w+) [^#\n]+#+ ((?:[xyz]mm\d+|mem) = .*)$', flags=re.M)) -ASM_SCRUB_SP_RE = re.compile(r'\d+\(%(esp|rsp)\)') -ASM_SCRUB_RIP_RE = re.compile(r'[.\w]+\(%rip\)') -ASM_SCRUB_KILL_COMMENT_RE = re.compile(r'^ *#+ +kill:.*\n') +SCRUB_X86_SP_RE = re.compile(r'\d+\(%(esp|rsp)\)') +SCRUB_X86_RIP_RE = re.compile(r'[.\w]+\(%rip\)') +SCRUB_KILL_COMMENT_RE = re.compile(r'^ *#+ +kill:.*\n') + +RUN_LINE_RE = re.compile('^\s*;\s*RUN:\s*(.*)$') +IR_FUNCTION_RE = re.compile('^\s*define\s+(?:internal\s+)?[^@]*@(\w+)\s*\(') +ASM_FUNCTION_RE = re.compile( + r'^_?(?P<f>[^:]+):[ \t]*#+[ \t]*@(?P=f)\n[^:]*?' + r'(?P<body>^##?[ \t]+[^:]+:.*?)\s*' + r'^\s*(?:[^:\n]+?:\s*\n\s*\.size|\.cfi_endproc|\.globl|\.comm|\.(?:sub)?section)', + flags=(re.M | re.S)) +CHECK_PREFIX_RE = re.compile('--check-prefix=(\S+)') +CHECK_RE = re.compile(r'^\s*;\s*([^:]+?)(?:-NEXT|-NOT|-DAG|-LABEL)?:') def scrub_asm(asm): # Scrub runs of whitespace out of the assembly, but leave the leading # whitespace in place. - asm = ASM_SCRUB_WHITESPACE_RE.sub(r' ', asm) + asm = SCRUB_WHITESPACE_RE.sub(r' ', asm) # Expand the tabs used for indentation. asm = string.expandtabs(asm, 2) # Detect shuffle asm comments and hide the operands in favor of the comments. - asm = ASM_SCRUB_SHUFFLES_RE.sub(r'\1 {{.*#+}} \2', asm) + asm = SCRUB_X86_SHUFFLES_RE.sub(r'\1 {{.*#+}} \2', asm) # Generically match the stack offset of a memory operand. - asm = ASM_SCRUB_SP_RE.sub(r'{{[0-9]+}}(%\1)', asm) + asm = SCRUB_X86_SP_RE.sub(r'{{[0-9]+}}(%\1)', asm) # Generically match a RIP-relative memory operand. - asm = ASM_SCRUB_RIP_RE.sub(r'{{.*}}(%rip)', asm) + asm = SCRUB_X86_RIP_RE.sub(r'{{.*}}(%rip)', asm) # Strip kill operands inserted into the asm. - asm = ASM_SCRUB_KILL_COMMENT_RE.sub('', asm) + asm = SCRUB_KILL_COMMENT_RE.sub('', asm) # Strip trailing whitespace. - asm = ASM_SCRUB_TRAILING_WHITESPACE_RE.sub(r'', asm) + asm = SCRUB_TRAILING_WHITESPACE_RE.sub(r'', asm) return asm @@ -66,15 +78,6 @@ def main(): parser.add_argument('tests', nargs='+') args = parser.parse_args() - run_line_re = re.compile('^\s*;\s*RUN:\s*(.*)$') - ir_function_re = re.compile('^\s*define\s+(?:internal\s+)?[^@]*@(\w+)\s*\(') - asm_function_re = re.compile( - r'^_?(?P<f>[^:]+):[ \t]*#+[ \t]*@(?P=f)\n[^:]*?' - r'(?P<body>^##?[ \t]+[^:]+:.*?)\s*' - r'^\s*(?:[^:\n]+?:\s*\n\s*\.size|\.cfi_endproc|\.globl|\.comm|\.(?:sub)?section)', - flags=(re.M | re.S)) - check_prefix_re = re.compile('--check-prefix=(\S+)') - check_re = re.compile(r'^\s*;\s*([^:]+?)(?:-NEXT|-NOT|-DAG|-LABEL)?:') autogenerated_note = ('; NOTE: Assertions have been autogenerated by ' 'utils/update_llc_test_checks.py') @@ -85,7 +88,7 @@ def main(): test_lines = [l.rstrip() for l in f] run_lines = [m.group(1) - for m in [run_line_re.match(l) for l in test_lines] if m] + for m in [RUN_LINE_RE.match(l) for l in test_lines] if m] if args.verbose: print >>sys.stderr, 'Found %d RUN lines:' % (len(run_lines),) for l in run_lines: @@ -106,7 +109,7 @@ def main(): llc_cmd_args = llc_cmd_args.replace('< %s', '').replace('%s', '').strip() check_prefixes = [m.group(1) - for m in check_prefix_re.finditer(filecheck_cmd)] + for m in CHECK_PREFIX_RE.finditer(filecheck_cmd)] if not check_prefixes: check_prefixes = ['CHECK'] @@ -124,7 +127,7 @@ def main(): print >>sys.stderr, 'Extracted FileCheck prefixes: ' + str(prefixes) raw_asm = llc(args, llc_args, test) # Build up a dictionary of all the function bodies. - for m in asm_function_re.finditer(raw_asm): + for m in ASM_FUNCTION_RE.finditer(raw_asm): if not m: continue f = m.group('f') @@ -158,7 +161,7 @@ def main(): for l in test_lines: if is_in_function_start: if l.lstrip().startswith(';'): - m = check_re.match(l) + m = CHECK_RE.match(l) if not m or m.group(1) not in prefix_set: fixed_lines.append(l) continue @@ -187,7 +190,7 @@ def main(): if l.strip() == ';': continue # And skip any CHECK lines. We'll build our own. - m = check_re.match(l) + m = CHECK_RE.match(l) if m and m.group(1) in prefix_set: continue # Collect the remaining lines in the function body and look for the end @@ -201,7 +204,7 @@ def main(): continue fixed_lines.append(l) - m = ir_function_re.match(l) + m = IR_FUNCTION_RE.match(l) if not m: continue name = m.group(1) |