##// END OF EJS Templates
exchange: improve computation of relevant markers for large repos...
exchange: improve computation of relevant markers for large repos Compute the candidate nodes with relevant markers directly from keys of the predecessors/successors/children dictionaries of obsstore. This is faster than iterating over all nodes directly. This test could be further improved for repositories with relative few markers compared to the repository size, but this is no longer hot already. With the current loop structure, the obshashrange use works as well as before as it passes lists with a single node. Adjust the interface by allowing revision lists as well as node lists. This helps cases that computes ancestors as it reduces the materialisation cost. Use this in _pushdiscoveryobsmarker and _getbundleobsmarkerpart. Improve the latter further by directly using ancestors(). Performance benchmarks show notable and welcome improvement to no-op push and pull (that would also apply to other push/pull). This apply to push and pull done without evolve. ### push/pull Benchmark parameter # bin-env-vars.hg.flavor = default # benchmark.variants.explicit-rev = none # benchmark.variants.protocol = ssh # benchmark.variants.revs = none ## benchmark.name = hg.command.pull # data-env-vars.name = mercurial-devel-2024-03-22-zstd-sparse-revlog before: 5.968537 seconds after: 5.668507 seconds (-5.03%, -0.30) # data-env-vars.name = tryton-devel-2024-03-22-zstd-sparse-revlog before: 1.446232 seconds after: 0.835553 seconds (-42.23%, -0.61) # data-env-vars.name = netbsd-src-draft-2024-09-19-zstd-sparse-revlog before: 5.777412 seconds after: 2.523454 seconds (-56.32%, -3.25) ## benchmark.name = hg.command.push # data-env-vars.name = mercurial-devel-2024-03-22-zstd-sparse-revlog before: 6.155501 seconds after: 5.885072 seconds (-4.39%, -0.27) # data-env-vars.name = tryton-devel-2024-03-22-zstd-sparse-revlog before: 1.491054 seconds after: 0.934882 seconds (-37.30%, -0.56) # data-env-vars.name = netbsd-src-draft-2024-09-19-zstd-sparse-revlog before: 5.902494 seconds after: 2.957644 seconds (-49.89%, -2.94) There is not notable different in these result using the "rust" flavor instead of the "default". The performance impact on the same operation when using evolve were also tested and no impact was noted.

File last commit:

r52756:f4733654 default
r52789:8583d138 default
Show More
timestamp.py
128 lines | 3.9 KiB | text/x-python | PythonLexer
# Copyright Mercurial Contributors
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
from __future__ import annotations
import functools
import os
import stat
from .. import error
rangemask = 0x7FFFFFFF
@functools.total_ordering
class timestamp(tuple):
"""
A Unix timestamp with optional nanoseconds precision,
modulo 2**31 seconds.
A 3-tuple containing:
`truncated_seconds`: seconds since the Unix epoch,
truncated to its lower 31 bits
`subsecond_nanoseconds`: number of nanoseconds since `truncated_seconds`.
When this is zero, the sub-second precision is considered unknown.
`second_ambiguous`: whether this timestamp is still "reliable"
(see `reliable_mtime_of`) if we drop its sub-second component.
"""
def __new__(cls, value):
truncated_seconds, subsec_nanos, second_ambiguous = value
value = (truncated_seconds & rangemask, subsec_nanos, second_ambiguous)
return super(timestamp, cls).__new__(cls, value)
def __eq__(self, other):
raise error.ProgrammingError(
'timestamp should never be compared directly'
)
def __gt__(self, other):
raise error.ProgrammingError(
'timestamp should never be compared directly'
)
def get_fs_now(vfs):
"""return a timestamp for "now" in the current vfs
This will raise an exception if no temporary files could be created.
"""
tmpfd, tmpname = vfs.mkstemp()
try:
return mtime_of(os.fstat(tmpfd))
finally:
os.close(tmpfd)
vfs.unlink(tmpname)
def zero():
"""
Returns the `timestamp` at the Unix epoch.
"""
return tuple.__new__(timestamp, (0, 0))
def mtime_of(stat_result):
"""
Takes an `os.stat_result`-like object and returns a `timestamp` object
for its modification time.
"""
try:
# TODO: add this attribute to `osutil.stat` objects,
# see `mercurial/cext/osutil.c`.
#
# This attribute is also not available on Python 2.
nanos = stat_result.st_mtime_ns
except AttributeError:
# https://docs.python.org/2/library/os.html#os.stat_float_times
# "For compatibility with older Python versions,
# accessing stat_result as a tuple always returns integers."
secs = stat_result[stat.ST_MTIME]
subsec_nanos = 0
else:
billion = int(1e9)
secs = nanos // billion
subsec_nanos = nanos % billion
return timestamp((secs, subsec_nanos, False))
def reliable_mtime_of(stat_result, present_mtime):
"""Same as `mtime_of`, but return `None` or a `Timestamp` with
`second_ambiguous` set if the date might be ambiguous.
A modification time is reliable if it is older than "present_time" (or
sufficiently in the future).
Otherwise a concurrent modification might happens with the same mtime.
"""
file_mtime = mtime_of(stat_result)
file_second = file_mtime[0]
file_ns = file_mtime[1]
boundary_second = present_mtime[0]
boundary_ns = present_mtime[1]
# If the mtime of the ambiguous file is younger (or equal) to the starting
# point of the `status` walk, we cannot garantee that another, racy, write
# will not happen right after with the same mtime and we cannot cache the
# information.
#
# However if the mtime is far away in the future, this is likely some
# mismatch between the current clock and previous file system operation. So
# mtime more than one days in the future are considered fine.
if boundary_second == file_second:
if file_ns and boundary_ns:
if file_ns < boundary_ns:
return timestamp((file_second, file_ns, True))
return None
elif boundary_second < file_second < (3600 * 24 + boundary_second):
return None
else:
return file_mtime