view tests/hghave @ 34824:e2ad93bcc084

revlog: introduce an experimental flag to slice chunks reads when too sparse

Delta chains can become quite sparse if there is a lot of unrelated data between
relevant pieces. Right now, revlog always reads all the necessary data for the
delta chain in one single read. This can lead to a lot of unrelated data to be
read (see issue5482 for more details).
One can use the `experimental.maxdeltachainspan` option with a large value
(or -1) to easily produce a very sparse delta chain.

This change introduces the ability to slice the chunks retrieval into  multiple
reads, skipping large sections of unrelated data. Preliminary testing shows
interesting results. For example the peak memory consumption to read a manifest
on a large repository is reduced from 600MB to 250MB (200MB without
maxdeltachainspan). However, the slicing itself and the multiple reads can have
an negative impact on performance. This is why the new feature is hidden behind
an experimental flag.

Future changesets will add various parameters to control the slicing heuristics.
We hope to experiment a wide variety of repositories during 4.4 and hopefully
turn the feature on by default in 4.5.
As a first try, the algorithm itself is prone to deep changes. However, we wish
to define APIs and have a baseline to work on.

#!/usr/bin/env python
"""Test the running system for features availability. Exit with zero
if all features are there, non-zero otherwise. If a feature name is
prefixed with "no-", the absence of feature is tested.
"""

from __future__ import absolute_import, print_function

import hghave
import optparse
import os
import sys

checks = hghave.checks

def list_features():
    for name, feature in sorted(checks.items()):
        desc = feature[1]
        print(name + ':', desc)

def test_features():
    failed = 0
    for name, feature in checks.items():
        check, _ = feature
        try:
            check()
        except Exception as e:
            print("feature %s failed:  %s" % (name, e))
            failed += 1
    return failed

parser = optparse.OptionParser("%prog [options] [features]")
parser.add_option("--test-features", action="store_true",
                  help="test available features")
parser.add_option("--list-features", action="store_true",
                  help="list available features")

def _loadaddon():
    if 'TESTDIR' in os.environ:
        # loading from '.' isn't needed, because `hghave` should be
        # running at TESTTMP in this case
        path = os.environ['TESTDIR']
    else:
        path = '.'

    if not os.path.exists(os.path.join(path, 'hghaveaddon.py')):
        return

    sys.path.insert(0, path)
    try:
        import hghaveaddon
        assert hghaveaddon  # silence pyflakes
    except BaseException as inst:
        sys.stderr.write('failed to import hghaveaddon.py from %r: %s\n'
                         % (path, inst))
        sys.exit(2)
    sys.path.pop(0)

if __name__ == '__main__':
    options, args = parser.parse_args()
    _loadaddon()
    if options.list_features:
        list_features()
        sys.exit(0)

    if options.test_features:
        sys.exit(test_features())

    hghave.require(args)