wireproto: add streams to frame-based protocol
Previously, the frame-based protocol was just a series of frames,
with each frame associated with a request ID.
In order to scale the protocol, we'll want to enable the use of
compression. While it is possible to enable compression at the
socket/pipe level, this has its disadvantages. The big one is it
undermines the point of frames being standalone, atomic units that
can be read and written: if you add compression above the framing
protocol, you are back to having a stream-based protocol as opposed
to something frame-based.
So in order to preserve frames, compression needs to occur at
the frame payload level.
Compressing each frame's payload individually will limit compression
ratios because the window size of the compressor will be limited
by the max frame size, which is 32-64kb as currently defined. It
will also add CPU overhead, as it is more efficient for compressors
to operate on fewer, larger blocks of data than more, smaller blocks.
So compressing each frame independently is out.
This means we need to compress each frame's payload as if it is part
of a larger stream.
The simplest approach is to have 1 stream per connection. This
could certainly work. However, it has disadvantages (documented below).
We could also have 1 stream per RPC/command invocation. (This is the
model HTTP/2 goes with.) This also has disadvantages.
The main disadvantage to one global stream is that it has the very
real potential to create CPU bottlenecks doing compression. Networks
are only getting faster and the performance of single CPU cores has
been relatively flat. Newer compression formats like zstandard offer
better CPU cycle efficiency than predecessors like zlib. But it still
all too common to saturate your CPU with compression overhead long
before you saturate the network pipe.
The main disadvantage with streams per request is that you can't
reap the benefits of the compression context for multiple requests.
For example, if you send 1000 RPC requests (or HTTP/2 requests for
that matter), the response to each would have its own compression
context. The overall size of the raw responses would be larger because
compression contexts wouldn't be able to reference data from another
request or response.
The approach for streams as implemented in this commit is to support
N streams per connection and for streams to potentially span requests
and responses. As explained by the added internals docs, this
facilitates servers and clients delegating independent streams and
compression to independent threads / CPU cores. This helps alleviate
the CPU bottleneck of compression. This design also allows compression
contexts to be reused across requests/responses. This can result in
improved compression ratios and less overhead for compressors and
decompressors having to build new contexts.
Another feature that was defined was the ability for individual frames
within a stream to declare whether that individual frame's payload
uses the content encoding (read: compression) defined by the stream.
The idea here is that some servers may serve data from a combination
of caches and dynamic resolution. Data coming from caches may be
pre-compressed. We want to facilitate servers being able to essentially
stream bytes from caches to the wire with minimal overhead. Being
able to mix and match with frames are compressed within a stream
enables these types of advanced server functionality.
This commit defines the new streams mechanism. Basic code for
supporting streams in frames has been added. But that code is
seriously lacking and doesn't fully conform to the defined protocol.
For example, we don't close any streams. And support for content
encoding within streams is not yet implemented. The change was
rather invasive and I didn't think it would be reasonable to implement
the entire feature in a single commit.
For the record, I would have loved to reuse an existing multiplexing
protocol to build the new wire protocol on top of. However, I couldn't
find a protocol that offers the performance and scaling characteristics
that I desired. Namely, it should support multiple compression
contexts to facilitate scaling out to multiple CPU cores and
compression contexts should be able to live longer than single RPC
requests. HTTP/2 *almost* fits the bill. But the semantics of HTTP
message exchange state that streams can only live for a single
request-response. We /could/ tunnel on top of HTTP/2 streams and
frames with HEADER and DATA frames. But there's no guarantee that
HTTP/2 libraries and proxies would allow us to use HTTP/2 streams
and frames without the HTTP message exchange semantics defined in
RFC 7540 Section 8. Other RPC protocols like gRPC tunnel are built
on top of HTTP/2 and thus preserve its semantics of stream per
RPC invocation. Even QUIC does this. We could attempt to invent a
higher-level stream that spans HTTP/2 streams. But this would be
violating HTTP/2 because there is no guarantee that HTTP/2 streams
are routed to the same server. The best we can do - which is what
this protocol does - is shoehorn all request and response data into
a single HTTP message and create streams within. At that point, we've
defined a Content-Type in HTTP parlance. It just so happens our
media type can also work as a standalone, stream-based protocol,
without leaning on HTTP or similar protocol.
Differential Revision: https://phab.mercurial-scm.org/D2907
# statichttprepo.py - simple http repository class for mercurial
#
# This provides read-only repo access to repositories exported via static http
#
# Copyright 2005-2007 Matt Mackall <mpm@selenic.com>
#
# 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 absolute_import
import errno
from .i18n import _
from . import (
changelog,
error,
localrepo,
manifest,
namespaces,
pathutil,
scmutil,
store,
url,
util,
vfs as vfsmod,
)
urlerr = util.urlerr
urlreq = util.urlreq
class httprangereader(object):
def __init__(self, url, opener):
# we assume opener has HTTPRangeHandler
self.url = url
self.pos = 0
self.opener = opener
self.name = url
def __enter__(self):
return self
def __exit__(self, exc_type, exc_value, traceback):
self.close()
def seek(self, pos):
self.pos = pos
def read(self, bytes=None):
req = urlreq.request(self.url)
end = ''
if bytes:
end = self.pos + bytes - 1
if self.pos or end:
req.add_header('Range', 'bytes=%d-%s' % (self.pos, end))
try:
f = self.opener.open(req)
data = f.read()
code = f.code
except urlerr.httperror as inst:
num = inst.code == 404 and errno.ENOENT or None
raise IOError(num, inst)
except urlerr.urlerror as inst:
raise IOError(None, inst.reason[1])
if code == 200:
# HTTPRangeHandler does nothing if remote does not support
# Range headers and returns the full entity. Let's slice it.
if bytes:
data = data[self.pos:self.pos + bytes]
else:
data = data[self.pos:]
elif bytes:
data = data[:bytes]
self.pos += len(data)
return data
def readlines(self):
return self.read().splitlines(True)
def __iter__(self):
return iter(self.readlines())
def close(self):
pass
# _RangeError and _HTTPRangeHandler were originally in byterange.py,
# which was itself extracted from urlgrabber. See the last version of
# byterange.py from history if you need more information.
class _RangeError(IOError):
"""Error raised when an unsatisfiable range is requested."""
class _HTTPRangeHandler(urlreq.basehandler):
"""Handler that enables HTTP Range headers.
This was extremely simple. The Range header is a HTTP feature to
begin with so all this class does is tell urllib2 that the
"206 Partial Content" response from the HTTP server is what we
expected.
"""
def http_error_206(self, req, fp, code, msg, hdrs):
# 206 Partial Content Response
r = urlreq.addinfourl(fp, hdrs, req.get_full_url())
r.code = code
r.msg = msg
return r
def http_error_416(self, req, fp, code, msg, hdrs):
# HTTP's Range Not Satisfiable error
raise _RangeError('Requested Range Not Satisfiable')
def build_opener(ui, authinfo):
# urllib cannot handle URLs with embedded user or passwd
urlopener = url.opener(ui, authinfo)
urlopener.add_handler(_HTTPRangeHandler())
class statichttpvfs(vfsmod.abstractvfs):
def __init__(self, base):
self.base = base
def __call__(self, path, mode='r', *args, **kw):
if mode not in ('r', 'rb'):
raise IOError('Permission denied')
f = "/".join((self.base, urlreq.quote(path)))
return httprangereader(f, urlopener)
def join(self, path):
if path:
return pathutil.join(self.base, path)
else:
return self.base
return statichttpvfs
class statichttppeer(localrepo.localpeer):
def local(self):
return None
def canpush(self):
return False
class statichttprepository(localrepo.localrepository):
supported = localrepo.localrepository._basesupported
def __init__(self, ui, path):
self._url = path
self.ui = ui
self.root = path
u = util.url(path.rstrip('/') + "/.hg")
self.path, authinfo = u.authinfo()
vfsclass = build_opener(ui, authinfo)
self.vfs = vfsclass(self.path)
self.cachevfs = vfsclass(self.vfs.join('cache'))
self._phasedefaults = []
self.names = namespaces.namespaces()
self.filtername = None
try:
requirements = scmutil.readrequires(self.vfs, self.supported)
except IOError as inst:
if inst.errno != errno.ENOENT:
raise
requirements = set()
# check if it is a non-empty old-style repository
try:
fp = self.vfs("00changelog.i")
fp.read(1)
fp.close()
except IOError as inst:
if inst.errno != errno.ENOENT:
raise
# we do not care about empty old-style repositories here
msg = _("'%s' does not appear to be an hg repository") % path
raise error.RepoError(msg)
# setup store
self.store = store.store(requirements, self.path, vfsclass)
self.spath = self.store.path
self.svfs = self.store.opener
self.sjoin = self.store.join
self._filecache = {}
self.requirements = requirements
self.manifestlog = manifest.manifestlog(self.svfs, self)
self.changelog = changelog.changelog(self.svfs)
self._tags = None
self.nodetagscache = None
self._branchcaches = {}
self._revbranchcache = None
self.encodepats = None
self.decodepats = None
self._transref = None
def _restrictcapabilities(self, caps):
caps = super(statichttprepository, self)._restrictcapabilities(caps)
return caps.difference(["pushkey"])
def url(self):
return self._url
def local(self):
return False
def peer(self):
return statichttppeer(self)
def wlock(self, wait=True):
raise error.LockUnavailable(0, _('lock not available'), 'lock',
_('cannot lock static-http repository'))
def lock(self, wait=True):
raise error.Abort(_('cannot lock static-http repository'))
def _writecaches(self):
pass # statichttprepository are read only
def instance(ui, path, create):
if create:
raise error.Abort(_('cannot create new static-http repository'))
return statichttprepository(ui, path[7:])