282a74c7f0
This is a reland of 6612943010
Fixed: Unaligned reads, unspecified evaluation order.
Original change's description:
> [regexp] Bytecode peephole optimization
>
> Bytecodes used by the regular expression interpreter often occur in
> specific sequences. The number of dispatches in the interpreter can be
> reduced if those sequences are combined into a single bytecode.
>
> This CL adds a peephole optimization pass for regexp bytecodes.
> This pass checks the generated bytecode for pre-defined sequences that
> can be merged into a single bytecode.
>
> With the currently implemented bytecode sequences a speedup of 1.12x on
> regex-dna and octane-regexp is achieved.
>
> Bug: v8:9330
> Change-Id: I827f93273a5848e5963c7e3329daeb898995d151
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1813743
> Commit-Queue: Patrick Thier <pthier@google.com>
> Reviewed-by: Peter Marshall <petermarshall@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63992}
Cq-Include-Trybots: luci.v8.try:v8_linux64_ubsan_rel_ng
Cq-Include-Trybots: luci.v8.try:v8_linux_gcc_rel
Bug: v8:9330,chromium:1008502,chromium:1008631
Change-Id: Ib9fc395b6809aa1debdb54d9fba5b7f09a235e5b
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1828917
Reviewed-by: Peter Marshall <petermarshall@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#64064}
68 lines
1.8 KiB
Python
Executable File
68 lines
1.8 KiB
Python
Executable File
#!/usr/bin/env python
|
|
# Copyright 2019 the V8 project authors. All rights reserved.
|
|
# Use of this source code is governed by a BSD-style license that can be
|
|
# found in the LICENSE file.
|
|
|
|
"""
|
|
python %prog trace-file
|
|
|
|
Parses output generated by v8 with flag --trace-regexp-bytecodes and generates
|
|
a list of the most common sequences.
|
|
"""
|
|
|
|
from __future__ import print_function
|
|
|
|
import sys
|
|
import re
|
|
import collections
|
|
|
|
def parse(file, seqlen):
|
|
# example:
|
|
# pc = 00, sp = 0, curpos = 0, curchar = 0000000a ..., bc = PUSH_BT, 02, 00, 00, 00, e8, 00, 00, 00 .......
|
|
rx = re.compile(r'pc = (?P<pc>[0-9a-f]+), sp = (?P<sp>\d+), '
|
|
r'curpos = (?P<curpos>\d+), curchar = (?P<char_hex>[0-9a-f]+) '
|
|
r'(:?\.|\()(?P<char>\.|\w)(:?\.|\)), bc = (?P<bc>\w+), .*')
|
|
total = 0
|
|
bc_cnt = [None] * seqlen
|
|
for i in xrange(seqlen):
|
|
bc_cnt[i] = {}
|
|
last = [None] * seqlen
|
|
with open(file) as f:
|
|
l = f.readline()
|
|
while l:
|
|
l = l.strip()
|
|
if l.startswith("Start bytecode interpreter"):
|
|
for i in xrange(seqlen):
|
|
last[i] = collections.deque(maxlen=i+1)
|
|
|
|
match = rx.search(l)
|
|
if match:
|
|
total += 1
|
|
bc = match.group('bc')
|
|
for i in xrange(seqlen):
|
|
last[i].append(bc)
|
|
key = ' --> '.join(last[i])
|
|
bc_cnt[i][key] = bc_cnt[i].get(key,0) + 1
|
|
|
|
l = f.readline()
|
|
return bc_cnt, total
|
|
|
|
def print_most_common(d, seqlen, total):
|
|
sorted_d = sorted(d.items(), key=lambda kv: kv[1], reverse=True)
|
|
for (k,v) in sorted_d:
|
|
if v*100/total < 1.0:
|
|
return
|
|
print("{}: {} ({} %)".format(k,v,(v*100/total)))
|
|
|
|
def main(argv):
|
|
max_seq = 7
|
|
bc_cnt, total = parse(argv[1],max_seq)
|
|
for i in xrange(max_seq):
|
|
print()
|
|
print("Most common of length {}".format(i+1))
|
|
print()
|
|
print_most_common(bc_cnt[i], i, total)
|
|
|
|
if __name__ == '__main__':
|
|
main(sys.argv)
|