Added scripts/calls.py for viewing the callgraph directly

This commit is contained in:
Christopher Haster
2022-02-19 14:12:17 -06:00
parent 2cdabe810d
commit f5286abe7a
5 changed files with 211 additions and 22 deletions

View File

@@ -28,6 +28,7 @@ SRC ?= $(wildcard *.c)
OBJ := $(SRC:%.c=$(BUILDDIR)%.o)
DEP := $(SRC:%.c=$(BUILDDIR)%.d)
ASM := $(SRC:%.c=$(BUILDDIR)%.s)
CGI := $(SRC:%.c=$(BUILDDIR)%.ci)
ifdef DEBUG
override CFLAGS += -O0 -g3
@@ -54,6 +55,7 @@ ifdef BUILDDIR
override TESTFLAGS += --build-dir="$(BUILDDIR:/=)"
override CODEFLAGS += --build-dir="$(BUILDDIR:/=)"
override DATAFLAGS += --build-dir="$(BUILDDIR:/=)"
override CALLSFLAGS += --build-dir="$(BUILDDIR:/=)"
endif
ifneq ($(NM),nm)
override CODEFLAGS += --nm-tool="$(NM)"
@@ -84,6 +86,10 @@ code: $(OBJ)
data: $(OBJ)
./scripts/data.py -S $^ $(DATAFLAGS)
.PHONY: calls
calls: $(CGI)
./scripts/calls.py $^ $(CALLSFLAGS)
.PHONY: test
test:
./scripts/test.py $(TESTFLAGS)
@@ -111,11 +117,15 @@ $(BUILDDIR)%.o: %.c
$(BUILDDIR)%.s: %.c
$(CC) -S $(CFLAGS) $< -o $@
$(BUILDDIR)%.ci: %.c
$(CC) -c -MMD -fcallgraph-info=su $(CFLAGS) $< -o $(@:.ci=.o)
# clean everything
.PHONY: clean
clean:
rm -f $(TARGET)
rm -f $(OBJ)
rm -f $(CGI)
rm -f $(DEP)
rm -f $(ASM)
rm -f $(BUILDDIR)tests/*.toml.*

170
scripts/calls.py Executable file
View File

@@ -0,0 +1,170 @@
#!/usr/bin/env python3
#
# Script to show the callgraph in a human readable manner. Basically just a
# wrapper aroung GCC's -fcallgraph-info flag.
#
import os
import glob
import itertools as it
import re
import csv
import collections as co
CI_PATHS = ['*.ci']
def collect(paths, **args):
# parse the vcg format
k_pattern = re.compile('([a-z]+)\s*:', re.DOTALL)
v_pattern = re.compile('(?:"(.*?)"|([a-z]+))', re.DOTALL)
def parse_vcg(rest):
def parse_vcg(rest):
node = []
while True:
rest = rest.lstrip()
m = k_pattern.match(rest)
if not m:
return (node, rest)
k, rest = m.group(1), rest[m.end(0):]
rest = rest.lstrip()
if rest.startswith('{'):
v, rest = parse_vcg(rest[1:])
assert rest[0] == '}', "unexpected %r" % rest[0:1]
rest = rest[1:]
node.append((k, v))
else:
m = v_pattern.match(rest)
assert m, "unexpected %r" % rest[0:1]
v, rest = m.group(1) or m.group(2), rest[m.end(0):]
node.append((k, v))
node, rest = parse_vcg(rest)
assert rest == '', "unexpected %r" % rest[0:1]
return node
# collect into functions
results = co.defaultdict(lambda: (None, None, set()))
f_pattern = re.compile(r'([^\\]*)\\n([^:]*)')
for path in paths:
with open(path) as f:
vcg = parse_vcg(f.read())
for k, graph in vcg:
if k != 'graph':
continue
for k, info in graph:
if k == 'node':
info = dict(info)
m = f_pattern.match(info['label'])
if m:
function, file = m.groups()
_, _, targets = results[info['title']]
results[info['title']] = (file, function, targets)
elif k == 'edge':
info = dict(info)
_, _, targets = results[info['sourcename']]
targets.add(info['targetname'])
else:
continue
if not args.get('everything'):
for source, (s_file, s_function, _) in list(results.items()):
# discard internal functions
if s_file.startswith('<') or s_file.startswith('/usr/include'):
del results[source]
# flatten into a list
flat_results = []
for _, (s_file, s_function, targets) in results.items():
for target in targets:
if target not in results:
continue
t_file, t_function, _ = results[target]
flat_results.append((s_file, s_function, t_file, t_function))
return flat_results
def main(**args):
# find sizes
if not args.get('use', None):
# find .ci files
paths = []
for path in args['ci_paths']:
if os.path.isdir(path):
path = path + '/*.ci'
for path in glob.glob(path):
paths.append(path)
if not paths:
print('no .ci files found in %r?' % args['ci_paths'])
sys.exit(-1)
results = collect(paths, **args)
else:
with open(args['use']) as f:
r = csv.DictReader(f)
results = [
( result['file'],
result['function'],
result['callee_file'],
result['callee_function'])
for result in r]
# write results to CSV
if args.get('output'):
with open(args['output'], 'w') as f:
w = csv.writer(f)
w.writerow(['file', 'function', 'callee_file', 'callee_function'])
for file, func, c_file, c_func in sorted(results):
w.writerow((file, func, c_file, c_func))
# print results
def dedup_entries(results, by='function'):
entries = co.defaultdict(lambda: set())
for file, func, c_file, c_func in results:
entry = (file if by == 'file' else func)
entries[entry].add(c_file if by == 'file' else c_func)
return entries
def print_entries(by='function'):
entries = dedup_entries(results, by=by)
for name, callees in sorted(entries.items()):
print(name)
for i, c_name in enumerate(sorted(callees)):
print(" -> %s" % c_name)
if args.get('quiet'):
pass
elif args.get('files'):
print_entries(by='file')
else:
print_entries(by='function')
if __name__ == "__main__":
import argparse
import sys
parser = argparse.ArgumentParser(
description="Find code size at the function level.")
parser.add_argument('ci_paths', nargs='*', default=CI_PATHS,
help="Description of where to find *.ci files. May be a directory \
or a list of paths. Defaults to %r." % CI_PATHS)
parser.add_argument('-v', '--verbose', action='store_true',
help="Output commands that run behind the scenes.")
parser.add_argument('-o', '--output',
help="Specify CSV file to store results.")
parser.add_argument('-u', '--use',
help="Don't parse callgraph files, instead use this CSV file.")
parser.add_argument('-A', '--everything', action='store_true',
help="Include builtin and libc specific symbols.")
parser.add_argument('--files', action='store_true',
help="Show file-level calls.")
parser.add_argument('-q', '--quiet', action='store_true',
help="Don't show anything, useful with -o.")
parser.add_argument('--build-dir',
help="Specify the relative build directory. Used to map object files \
to the correct source files.")
sys.exit(main(**vars(parser.parse_args())))

View File

@@ -49,8 +49,9 @@ def collect(paths, **args):
if args.get('build_dir'):
file = re.sub('%s/*' % re.escape(args['build_dir']), '', file)
# discard internal functions
if func.startswith('__'):
continue
if not args.get('everything'):
if func.startswith('__'):
continue
# discard .8449 suffixes created by optimizer
func = re.sub('\.[0-9]+', '', func)
flat_results.append((file, func, size))
@@ -128,19 +129,19 @@ def main(**args):
def sorted_entries(entries):
if args.get('size_sort'):
return sorted(entries.items(), key=lambda x: (-x[1], x))
return sorted(entries, key=lambda x: (-x[1], x))
elif args.get('reverse_size_sort'):
return sorted(entries.items(), key=lambda x: (+x[1], x))
return sorted(entries, key=lambda x: (+x[1], x))
else:
return sorted(entries.items())
return sorted(entries)
def sorted_diff_entries(entries):
if args.get('size_sort'):
return sorted(entries.items(), key=lambda x: (-x[1][1], x))
return sorted(entries, key=lambda x: (-x[1][1], x))
elif args.get('reverse_size_sort'):
return sorted(entries.items(), key=lambda x: (+x[1][1], x))
return sorted(entries, key=lambda x: (+x[1][1], x))
else:
return sorted(entries.items(), key=lambda x: (-x[1][3], x))
return sorted(entries, key=lambda x: (-x[1][3], x))
def print_header(by=''):
if not args.get('diff'):
@@ -153,7 +154,7 @@ def main(**args):
if not args.get('diff'):
print_header(by=by)
for name, size in sorted_entries(entries):
for name, size in sorted_entries(entries.items()):
print("%-36s %7d" % (name, size))
else:
prev_entries = dedup_entries(prev_results, by=by)
@@ -161,7 +162,7 @@ def main(**args):
print_header(by='%s (%d added, %d removed)' % (by,
sum(1 for old, _, _, _ in diff.values() if not old),
sum(1 for _, new, _, _ in diff.values() if not new)))
for name, (old, new, diff, ratio) in sorted_diff_entries(diff):
for name, (old, new, diff, ratio) in sorted_diff_entries(diff.items()):
if ratio or args.get('all'):
print("%-36s %7s %7s %+7d%s" % (name,
old or "-",
@@ -211,6 +212,8 @@ if __name__ == "__main__":
help="Specify CSV file to diff code size against.")
parser.add_argument('-a', '--all', action='store_true',
help="Show all functions, not just the ones that changed.")
parser.add_argument('-A', '--everything', action='store_true',
help="Include builtin and libc specific symbols.")
parser.add_argument('-s', '--size-sort', action='store_true',
help="Sort by size.")
parser.add_argument('-S', '--reverse-size-sort', action='store_true',

View File

@@ -55,8 +55,9 @@ def collect(paths, **args):
for (file, func), (hits, count) in reduced_funcs.items():
# discard internal/testing functions (test_* injected with
# internal testing)
if func.startswith('__') or func.startswith('test_'):
continue
if not args.get('everything'):
if func.startswith('__') or func.startswith('test_'):
continue
# discard .8449 suffixes created by optimizer
func = re.sub('\.[0-9]+', '', func)
results.append((file, func, hits, count))
@@ -245,6 +246,8 @@ if __name__ == "__main__":
help="Specify CSV file to diff code size against.")
parser.add_argument('-a', '--all', action='store_true',
help="Show all functions, not just the ones that changed.")
parser.add_argument('-A', '--everything', action='store_true',
help="Include builtin and libc specific symbols.")
parser.add_argument('--files', action='store_true',
help="Show file-level coverage.")
parser.add_argument('--summary', action='store_true',

View File

@@ -49,8 +49,9 @@ def collect(paths, **args):
if args.get('build_dir'):
file = re.sub('%s/*' % re.escape(args['build_dir']), '', file)
# discard internal functions
if func.startswith('__'):
continue
if not args.get('everything'):
if func.startswith('__'):
continue
# discard .8449 suffixes created by optimizer
func = re.sub('\.[0-9]+', '', func)
flat_results.append((file, func, size))
@@ -128,19 +129,19 @@ def main(**args):
def sorted_entries(entries):
if args.get('size_sort'):
return sorted(entries.items(), key=lambda x: (-x[1], x))
return sorted(entries, key=lambda x: (-x[1], x))
elif args.get('reverse_size_sort'):
return sorted(entries.items(), key=lambda x: (+x[1], x))
return sorted(entries, key=lambda x: (+x[1], x))
else:
return sorted(entries.items())
return sorted(entries)
def sorted_diff_entries(entries):
if args.get('size_sort'):
return sorted(entries.items(), key=lambda x: (-x[1][1], x))
return sorted(entries, key=lambda x: (-x[1][1], x))
elif args.get('reverse_size_sort'):
return sorted(entries.items(), key=lambda x: (+x[1][1], x))
return sorted(entries, key=lambda x: (+x[1][1], x))
else:
return sorted(entries.items(), key=lambda x: (-x[1][3], x))
return sorted(entries, key=lambda x: (-x[1][3], x))
def print_header(by=''):
if not args.get('diff'):
@@ -153,7 +154,7 @@ def main(**args):
if not args.get('diff'):
print_header(by=by)
for name, size in sorted_entries(entries):
for name, size in sorted_entries(entries.items()):
print("%-36s %7d" % (name, size))
else:
prev_entries = dedup_entries(prev_results, by=by)
@@ -161,7 +162,7 @@ def main(**args):
print_header(by='%s (%d added, %d removed)' % (by,
sum(1 for old, _, _, _ in diff.values() if not old),
sum(1 for _, new, _, _ in diff.values() if not new)))
for name, (old, new, diff, ratio) in sorted_diff_entries(diff):
for name, (old, new, diff, ratio) in sorted_diff_entries(diff.items()):
if ratio or args.get('all'):
print("%-36s %7s %7s %+7d%s" % (name,
old or "-",
@@ -211,6 +212,8 @@ if __name__ == "__main__":
help="Specify CSV file to diff data size against.")
parser.add_argument('-a', '--all', action='store_true',
help="Show all functions, not just the ones that changed.")
parser.add_argument('-A', '--everything', action='store_true',
help="Include builtin and libc specific symbols.")
parser.add_argument('-s', '--size-sort', action='store_true',
help="Sort by size.")
parser.add_argument('-S', '--reverse-size-sort', action='store_true',