changeset 39638:d292328e0143

exchangev2: fetch manifest revisions Now that the server has support for retrieving manifest data, we can implement the client bits to call it. We teach the changeset fetching code to capture the manifest revisions that are encountered on incoming changesets. We then feed this into a new function which filters out known manifests and then batches up manifest data requests to the server. This is different from the previous wire protocol in a few notable ways. First, the client fetches manifest data separately and explicitly. Before, we'd ask the server for data pertaining to some changesets (via a "getbundle" command) and manifests (and files) would be sent automatically. Providing an API for looking up just manifest data separately gives clients much more flexibility for manifest management. For example, a client may choose to only fetch manifest data on demand instead of prefetching it (i.e. partial clone). Second, we send N commands to the server for manifest retrieval instead of 1. This property has a few nice side-effects. One is that the deterministic nature of the requests lends itself to server-side caching. For example, say the remote has 50,000 manifests. If the server is configured to cache responses, each time a new commit arrives, you will have a cache miss and need to regenerate all outgoing data. But if you makes N requests requesting 10,000 manifests each, a new commit will still yield cache hits on the initial, unchanged manifest batches/requests. A derived benefit from these properties is that resumable clone is conceptually simpler to implement. When making a monolithic request for all of the repository data, recovering from an interrupted clone is hard because the server was in the driver's seat and was maintaining state about all the data that needed transferred. With the client driving fetching, the client can persist the set of unfetched entities and retry/resume a fetch if something goes wrong. Or we can fetch all data N changesets at a time and slowly build up a repository. This approach is drastically easier to implement when we have server APIs exposing low-level repository primitives (such as manifests and files). We don't yet support tree manifests. But it should be possible to implement that with the existing wire protocol command. Differential Revision: https://phab.mercurial-scm.org/D4489
author Gregory Szorc <gregory.szorc@gmail.com>
date Wed, 05 Sep 2018 09:09:57 -0700
parents c7a7c7e844e5
children 0e03e6a44dee
files mercurial/exchangev2.py tests/test-wireproto-exchangev2.t
diffstat 2 files changed, 203 insertions(+), 0 deletions(-) [+]
line wrap: on
line diff
--- a/mercurial/exchangev2.py	Wed Sep 05 09:09:52 2018 -0700
+++ b/mercurial/exchangev2.py	Wed Sep 05 09:09:57 2018 -0700
@@ -16,6 +16,7 @@
 )
 from . import (
     bookmarks,
+    error,
     mdiff,
     phases,
     pycompat,
@@ -57,6 +58,8 @@
                                remote.url(), pullop.gettransaction,
                                explicit=pullop.explicitbookmarks)
 
+    _fetchmanifests(repo, tr, remote, csetres['manifestnodes'])
+
 def _pullchangesetdiscovery(repo, remote, heads, abortwhenunrelated=True):
     """Determine which changesets need to be pulled."""
 
@@ -121,6 +124,8 @@
                                     unit=_('chunks'),
                                     total=meta.get(b'totalitems'))
 
+    manifestnodes = {}
+
     def linkrev(node):
         repo.ui.debug('add changeset %s\n' % short(node))
         # Linkrev for changelog is always self.
@@ -129,6 +134,12 @@
     def onchangeset(cl, node):
         progress.increment()
 
+        revision = cl.changelogrevision(node)
+
+        # We need to preserve the mapping of changelog revision to node
+        # so we can set the linkrev accordingly when manifests are added.
+        manifestnodes[cl.rev(node)] = revision.manifest
+
     nodesbyphase = {phase: set() for phase in phases.phasenames}
     remotebookmarks = {}
 
@@ -178,4 +189,106 @@
         'added': added,
         'nodesbyphase': nodesbyphase,
         'bookmarks': remotebookmarks,
+        'manifestnodes': manifestnodes,
     }
+
+def _fetchmanifests(repo, tr, remote, manifestnodes):
+    rootmanifest = repo.manifestlog.getstorage(b'')
+
+    # Some manifests can be shared between changesets. Filter out revisions
+    # we already know about.
+    fetchnodes = []
+    linkrevs = {}
+    seen = set()
+
+    for clrev, node in sorted(manifestnodes.iteritems()):
+        if node in seen:
+            continue
+
+        try:
+            rootmanifest.rev(node)
+        except error.LookupError:
+            fetchnodes.append(node)
+            linkrevs[node] = clrev
+
+        seen.add(node)
+
+    # TODO handle tree manifests
+
+    # addgroup() expects 7-tuple describing revisions. This normalizes
+    # the wire data to that format.
+    def iterrevisions(objs, progress):
+        for manifest in objs:
+            node = manifest[b'node']
+
+            if b'deltasize' in manifest:
+                basenode = manifest[b'deltabasenode']
+                delta = next(objs)
+            elif b'revisionsize' in manifest:
+                basenode = nullid
+                revision = next(objs)
+                delta = mdiff.trivialdiffheader(len(revision)) + revision
+            else:
+                continue
+
+            yield (
+                node,
+                manifest[b'parents'][0],
+                manifest[b'parents'][1],
+                # The value passed in is passed to the lookup function passed
+                # to addgroup(). We already have a map of manifest node to
+                # changelog revision number. So we just pass in the
+                # manifest node here and use linkrevs.__getitem__ as the
+                # resolution function.
+                node,
+                basenode,
+                delta,
+                # Flags not yet supported.
+                0
+            )
+
+            progress.increment()
+
+    progress = repo.ui.makeprogress(_('manifests'), unit=_('chunks'),
+                                    total=len(fetchnodes))
+
+    # Fetch manifests 10,000 per command.
+    # TODO have server advertise preferences?
+    # TODO make size configurable on client?
+    batchsize = 10000
+
+    # We send commands 1 at a time to the remote. This is not the most
+    # efficient because we incur a round trip at the end of each batch.
+    # However, the existing frame-based reactor keeps consuming server
+    # data in the background. And this results in response data buffering
+    # in memory. This can consume gigabytes of memory.
+    # TODO send multiple commands in a request once background buffering
+    # issues are resolved.
+
+    added = []
+
+    for i in pycompat.xrange(0, len(fetchnodes), batchsize):
+        batch = [node for node in fetchnodes[i:i + batchsize]]
+        if not batch:
+            continue
+
+        with remote.commandexecutor() as e:
+            objs = e.callcommand(b'manifestdata', {
+                b'tree': b'',
+                b'nodes': batch,
+                b'fields': {b'parents', b'revision'},
+            }).result()
+
+            # Chomp off header object.
+            next(objs)
+
+            added.extend(rootmanifest.addgroup(
+                iterrevisions(objs, progress),
+                linkrevs.__getitem__,
+                weakref.proxy(tr)))
+
+    progress.complete()
+
+    return {
+        'added': added,
+    }
--- a/tests/test-wireproto-exchangev2.t	Wed Sep 05 09:09:52 2018 -0700
+++ b/tests/test-wireproto-exchangev2.t	Wed Sep 05 09:09:57 2018 -0700
@@ -76,6 +76,24 @@
   add changeset e96ae20f4188
   add changeset caa2a465451d
   checking for updated bookmarks
+  sending 1 commands
+  sending command manifestdata: {
+    'fields': set([
+      'parents',
+      'revision'
+    ]),
+    'nodes': [
+      '\x99/Gy\x02\x9a=\xf8\xd0fm\x00\xbb\x92OicN&A',
+      '\xa9\x88\xfbCX>\x87\x1d\x1e\xd5u\x0e\xe0t\xc6\xd8@\xbb\xbf\xc8',
+      '\xec\x80NH\x8c \x88\xc25\t\x9a\x10 u\x13\xbe\xcd\xc3\xdd\xa5',
+      '\x04\\\x7f9\'\xda\x13\xe7Z\xf8\xf0\xe4\xf0HI\xe4a\xa9x\x0f',
+      '7\x9c\xb0\xc2\xe6d\\y\xdd\xc5\x9a\x1dG\'\xa9\xfb\x83\n\xeb&'
+    ],
+    'tree': ''
+  }
+  received frame(size=11; request=1; stream=2; streamflags=stream-begin; type=command-response; flags=continuation)
+  received frame(size=922; request=1; stream=2; streamflags=; type=command-response; flags=continuation)
+  received frame(size=0; request=1; stream=2; streamflags=; type=command-response; flags=eos)
   updating the branch cache
   new changesets 3390ef850073:caa2a465451d (3 drafts)
 
@@ -101,6 +119,16 @@
   o  0 3390ef850073fbc2f0dfff2244342c8e9229013a public
   
 
+All manifests should have been transferred
+
+  $ hg -R client-simple debugindex -m
+     rev linkrev nodeid       p1           p2
+       0       0 992f4779029a 000000000000 000000000000
+       1       1 a988fb43583e 992f4779029a 000000000000
+       2       2 ec804e488c20 a988fb43583e 000000000000
+       3       3 045c7f3927da 992f4779029a 000000000000
+       4       4 379cb0c2e664 045c7f3927da 000000000000
+
 Cloning only a specific revision works
 
   $ hg --debug clone -U -r 4432d83626e8 http://localhost:$HGPORT client-singlehead
@@ -146,6 +174,21 @@
   add changeset 3390ef850073
   add changeset 4432d83626e8
   checking for updated bookmarks
+  sending 1 commands
+  sending command manifestdata: {
+    'fields': set([
+      'parents',
+      'revision'
+    ]),
+    'nodes': [
+      '\x99/Gy\x02\x9a=\xf8\xd0fm\x00\xbb\x92OicN&A',
+      '\xa9\x88\xfbCX>\x87\x1d\x1e\xd5u\x0e\xe0t\xc6\xd8@\xbb\xbf\xc8'
+    ],
+    'tree': ''
+  }
+  received frame(size=11; request=1; stream=2; streamflags=stream-begin; type=command-response; flags=continuation)
+  received frame(size=376; request=1; stream=2; streamflags=; type=command-response; flags=continuation)
+  received frame(size=0; request=1; stream=2; streamflags=; type=command-response; flags=eos)
   updating the branch cache
   new changesets 3390ef850073:4432d83626e8
 
@@ -157,6 +200,11 @@
   o  0 3390ef850073fbc2f0dfff2244342c8e9229013a public
   
 
+  $ hg debugindex -m
+     rev linkrev nodeid       p1           p2
+       0       0 992f4779029a 000000000000 000000000000
+       1       1 a988fb43583e 992f4779029a 000000000000
+
 Incremental pull works
 
   $ hg --debug pull
@@ -204,6 +252,22 @@
   add changeset e96ae20f4188
   add changeset caa2a465451d
   checking for updated bookmarks
+  sending 1 commands
+  sending command manifestdata: {
+    'fields': set([
+      'parents',
+      'revision'
+    ]),
+    'nodes': [
+      '\xec\x80NH\x8c \x88\xc25\t\x9a\x10 u\x13\xbe\xcd\xc3\xdd\xa5',
+      '\x04\\\x7f9\'\xda\x13\xe7Z\xf8\xf0\xe4\xf0HI\xe4a\xa9x\x0f',
+      '7\x9c\xb0\xc2\xe6d\\y\xdd\xc5\x9a\x1dG\'\xa9\xfb\x83\n\xeb&'
+    ],
+    'tree': ''
+  }
+  received frame(size=11; request=1; stream=2; streamflags=stream-begin; type=command-response; flags=continuation)
+  received frame(size=559; request=1; stream=2; streamflags=; type=command-response; flags=continuation)
+  received frame(size=0; request=1; stream=2; streamflags=; type=command-response; flags=eos)
   updating the branch cache
   new changesets cd2534766bec:caa2a465451d (3 drafts)
   (run 'hg update' to get a working copy)
@@ -220,6 +284,14 @@
   o  0 3390ef850073fbc2f0dfff2244342c8e9229013a public
   
 
+  $ hg debugindex -m
+     rev linkrev nodeid       p1           p2
+       0       0 992f4779029a 000000000000 000000000000
+       1       1 a988fb43583e 992f4779029a 000000000000
+       2       2 ec804e488c20 a988fb43583e 000000000000
+       3       3 045c7f3927da 992f4779029a 000000000000
+       4       4 379cb0c2e664 045c7f3927da 000000000000
+
 Phase-only update works
 
   $ hg -R ../server-simple phase --public -r caa2a465451dd
@@ -331,6 +403,24 @@
   checking for updated bookmarks
   adding remote bookmark book-1
   adding remote bookmark book-2
+  sending 1 commands
+  sending command manifestdata: {
+    'fields': set([
+      'parents',
+      'revision'
+    ]),
+    'nodes': [
+      '\x99/Gy\x02\x9a=\xf8\xd0fm\x00\xbb\x92OicN&A',
+      '\xa9\x88\xfbCX>\x87\x1d\x1e\xd5u\x0e\xe0t\xc6\xd8@\xbb\xbf\xc8',
+      '\xec\x80NH\x8c \x88\xc25\t\x9a\x10 u\x13\xbe\xcd\xc3\xdd\xa5',
+      '\x04\\\x7f9\'\xda\x13\xe7Z\xf8\xf0\xe4\xf0HI\xe4a\xa9x\x0f',
+      '7\x9c\xb0\xc2\xe6d\\y\xdd\xc5\x9a\x1dG\'\xa9\xfb\x83\n\xeb&'
+    ],
+    'tree': ''
+  }
+  received frame(size=11; request=1; stream=2; streamflags=stream-begin; type=command-response; flags=continuation)
+  received frame(size=922; request=1; stream=2; streamflags=; type=command-response; flags=continuation)
+  received frame(size=0; request=1; stream=2; streamflags=; type=command-response; flags=eos)
   updating the branch cache
   new changesets 3390ef850073:caa2a465451d (1 drafts)