Summary:
The goal of the whole series is to have logging around linknode fixup. It's a slow operation, and we have two heuristics to make it faster. Unfortunately we have no idea about how well these heuristics actually help. This series of diffs aims to fix this problem. After this series is landed and logging is enabled, I hope that we can find out if we really these heuristics at all.
Let's log more data to make it easier to analyze where and how often does
linknode adjusing happen.
Test Plan: Run test-remotefilelog-linknodes.t
Reviewers: #fbhgext, quark
Reviewed By: #fbhgext, quark
Subscribers: quark
Differential Revision: https://phab.mercurial-scm.org/D838
Summary:
The goal of the whole series is to have logging around linknode fixup. It's a slow operation, and we have two heuristics to make it faster. Unfortunately we have no idea about how well these heuristics actually help. This series of diffs aims to fix this problem. After this series is landed and logging is enabled, I hope that we can find out if we really these heuristics at all.
Test Plan: Run test-remotefilelog-linknodes.t
Reviewers: #fbhgext, quark
Reviewed By: #fbhgext, quark
Differential Revision: https://phab.mercurial-scm.org/D837
Summary:
The goal of the whole series is to have logging around linknode fixup. It's a slow operation, and we have two heuristics to make it faster. Unfortunately we have no idea about how well these heuristics actually help. This series of diffs aims to fix this problem. After this series is landed and logging is enabled, I hope that we can find out if we really these heuristics at all.
Let's store all the fastlog logging inside fastlog function. This refactoring helps in the next diffs
Test Plan: Run test-remotefilelog-linknodes.t
Reviewers: #fbhgext, quark
Reviewed By: #fbhgext, quark
Differential Revision: https://phab.mercurial-scm.org/D836
Summary:
The goal of the whole series is to have logging around linknode fixup. It's a slow operation, and we have two heuristics to make it faster. Unfortunately we have no idea about how well these heuristics actually help. This series of diffs aims to fix this problem. After this series is landed and logging is enabled, I hope that we can find out if we really these heuristics at all.
Let's change the name of the logging category. In the next diffs I'll add more
logging, not only related to fastlog, so current name may be misleading.
Test Plan: Run test-remotefilelog-linknodes.t
Reviewers: #fbhgext, quark
Reviewed By: #fbhgext, quark
Differential Revision: https://phab.mercurial-scm.org/D835
This is a simple `rt -i` fix for changes caused by upstream 83b647ce1cec and
976f5a9c1fd2. We probably want to change our template implementation to
de-dup hashes in the future.
This diff adds an option --cwd-list to hg sparse. This will return the
contents of the current directory. The files that are in the sparse profile are
annotated with a '-' indicator.
Test Plan:
Tested by running the command 'hg sparse --cwd-list' in various folders.
Differential Revision: https://phab.mercurial-scm.org/D731
Summary: Regression that wasn't tested by the tests before
Test Plan: This is now tested with a new test
Reviewers: #fbhgext, mjpieters
Reviewed By: mjpieters
Differential Revision: https://phab.mercurial-scm.org/D823
Summary:
I think this is a better way of surfacing this warning. It also allows
us to tweak the message so we can point at new documentation I have already
written up.
Test Plan: Updated test
Reviewers: #fbhgext, mbthomas, durham, singhsrb
Reviewed By: #fbhgext, mbthomas, durham, singhsrb
Subscribers: singhsrb, quark
Differential Revision: https://phab.mercurial-scm.org/D812
D802 added two test cases in the test but the test cases are only used
for a very little portion of the test and therefore, result in needless double
execution for most of the test. Therefore, this commit removes the test cases
and instead, handles the different cases by passing the corresponding config
option.
Test Plan:
ran test-tweakdefaults.t
Differential Revision: https://phab.mercurial-scm.org/D814
Resolving obsoleted revisions (`repo.revs('obsolete()')`) is slow. Let's
move that filtering logic from read code path to write code path. So it does
not get executed for every command.
Test Plan:
`hg bookmark --hidden --profile`. Make sure obsstore loading logic
disappears from the profiling output.
Differential Revision: https://phab.mercurial-scm.org/D806
p4fastimport's reposetup will break other lfs commands or verify. Let's
limit that side effect to p4fastimport related commands. So they won't
affect users with p4fastimport enabled.
Differential Revision: https://phab.mercurial-scm.org/D801
Tested by trying to connect to a non-phabricator URL that responds with HTML to
a POST request and by trying to run hg ssl when not on the corp network.
Depends on D799
Differential Revision: https://phab.mercurial-scm.org/D800
Add a --sparse option to `hg diff` which filters the diff to include only files
that are part of the sparse profile.
When files are moved or copied across the sparse profile boundary, they will
show as added or removed (for moves) in the diff. This is the same behaviour
as `hg diff --root`.
Test Plan:
Expand `tests/test-sparse-diff.t`
Differential Revision: https://phab.mercurial-scm.org/D797
When commands like 'diff' request the contents of files from a workingfilectx,
normally this is satisfied by reading from disk. For files outside the sparse
checkout this doesn't work, so instead we must return the content from the
working copy parent.
Test Plan:
`tests/test-sparse-diff.t`
Differential Revision: https://phab.mercurial-scm.org/D788
We must correctly return the value from the function we are decorating.
Test Plan:
run tests
Differential Revision: https://phab.mercurial-scm.org/D767
Use naming consistent with Mercurial conventions and call the result of
bin(id) a node.
Test Plan:
None
Differential Revision: https://phab.mercurial-scm.org/D766
Summary:
Adding a config to control (only enable/disable for now) developer
warnings. This allows disabling the warnings on platforms like servers.
Test Plan: Updated and ran test-tweakdefaults.t
Reviewers: #fbhgext, durham
Reviewed By: #fbhgext, durham
Differential Revision: https://phab.mercurial-scm.org/D802
Summary:
The _adjustlinknode function in remotefilectx currently has a slow case where
it must prefetch history from the server to lookup the correct linkcode.
This change makes use of fastlog (if enabled for the repo) to speed up the
linknode lookup as fastlog will already have the correct linkrev. If fastlog
is not enabled or the fastlog call fails for some reason then it will fall
back to the old method.
Test Plan: Added new test cases in test-remotefilelog-linknodes.t
Reviewers: ikostia, stash, #fbhgext
Differential Revision: https://phab.mercurial-scm.org/D620
This makes the client side pushes also include tree packs in the push.
Also adds a test to see that the tree parts round trip between two tree-only
clients.
Differential Revision: https://phab.mercurial-scm.org/D721
When serving data from a stored bundle, if we need to rebundle the data first,
we need to make sure that the tree data stored within the bundle gets packaged
up as well.
If we're not rebundling, then things work fine as is since it just forwards the
parts along.
Differential Revision: https://phab.mercurial-scm.org/D720
Previously, when serving the contents of a stored bundle, we would explicitly
pick out the changegroup and serve only that. Let's also send along all other
parts that are stored in the bundle.
At the moment the only other parts that are stored in the bundle are the ones in
the partforwardingwhitelist in processparts, which is only treemanifest packs
right now. Later we'll serve bookmarks using this same pattern.
Differential Revision: https://phab.mercurial-scm.org/D719
Makes the infinitepush server store the received treepack parts in the bundle.
Future patches will enable serving the trees from the bundle, and sending the
trees from the client.
Differential Revision: https://phab.mercurial-scm.org/D718
Upstream Mercurial now has a function that handles the iteration over bundle
parts. As part of making infinitepush handle more part types, let's override
processparts and completely take over part processing.
Initially the part processing will just mimic the existing handlers, but there's
a config flag that causes it to simply redirect the received parts into the new
bundle. This will be useful later for storing all bundle parts (like tree packs
and obsmarkers).
Differential Revision: https://phab.mercurial-scm.org/D717
Previously the logic that created a bundle from a changegroup part and put it in
the store was all in the same function. Let's split it up.
This will be useful in a later part where we stop using the default bundle2 part
handling logic and instead handle each part manually (allowing us to store other
parts in the bundle).
Differential Revision: https://phab.mercurial-scm.org/D716
Summary: Just as with the previous patch, this will reduce errors and cognitive overhead.
Test Plan: `rt`
Reviewers: #fbhgext, stash
Reviewed By: #fbhgext, stash
Differential Revision: https://phab.mercurial-scm.org/D793
Summary:
This has been a big source of confusion and errors when sporadically hacking on
this script that usually doubles the time it takes to add a new feature. This
is the more sane way to do it anyway.
Test Plan: `rt`
Reviewers: #fbhgext, stash
Reviewed By: #fbhgext, stash
Differential Revision: https://phab.mercurial-scm.org/D792
Summary:
If you're working n multiple terminals on the same repository, it is
often useful to know if a command you're about to run might block due to an
already-held lock on the repository. This upgrade to scm-prompt gives the
user this information.
Test Plan: updated the test
Reviewers: #fbhgext, stash
Reviewed By: #fbhgext, stash
Subscribers: stash
Differential Revision: https://phab.mercurial-scm.org/D790
The remotefilelog annotate logic was not keeping track of which commits it had
seen when doing the bfs ancestor walk. For very mergy histories this resulted in
a ton of duplicate work.
Differential Revision: https://phab.mercurial-scm.org/D771
In hybrid tree/flat mode, if a local commit accidentally got marked as public,
then we would attempt to access the server, which of course didn't have the
commit and failed. This caused an error on the client, when in reality we
should've just fallen back to the flat manifest. Let's eat the error and the
message in this case.
Differential Revision: https://phab.mercurial-scm.org/D772
Previously, if during the middle a transaction something needed to access a tree
for a commit that was created during the transaction, the manifestlog could find
that tree in a side store it kept for current transaction trees. Unfortunatley
we were only storing the data, and not the parents, so if logic needed the
parents it failed with a key error. The fix is to just store the parents as
well.
This was caught by the sparse extension, which looks at the current tree during
a commit when a sparse profile changes.
With rules of nested paths, the order of rules matters.
Note: this does not make dirsync handle recursive changes. i.e. a mirror
from path1 to path2 won't trigger a rule matching path2.
Differential Revision: https://phab.mercurial-scm.org/D770
Previously, undolog runs for every command, which sometimes adds unwanted
noticeable overhead like `hg status` with a large obsstore. This patch makes
undo smarter to only log commands that actually write things.
This is done by making undolog only run for transactions, updates (legacy
code that does not require a transaction). Tests are changed to reflect
`hg status` does not trigger undolog, and the new code seems to work better
with chg.
Differential Revision: https://phab.mercurial-scm.org/D739
