This was a bit disappointing, I was hoping for a 2x speedup. But, I think
the metadata lookup is wasting a lot of time and also needs to be made to
stream.
The changes to catObjectStreamLsTree were benchmarked to not also speed
up --all around 3% more. Seems I managed to make it polymorphic after all.
planned to use for an optimisation
most things using stagedDetails were not expecting to get dup files in a
conflicted merge and deal with them, so converted them to use
inRepoDetails.
Turns out the %(rest) trick was not needed. Instead, just maintain a
list of files we've asked for, and each cat-file response is for the
next file in the list.
This actually benchmarks 25% faster than before! Very surprising, but it
must be due to needing to shove less data through the pipe, and parse
less.
This assumes that no location log files will have a newline or carriage
return in their name. catObjectStream skips any such files due to
cat-file not supporting them.
Keys have been prevented from containing newlines since 2011,
commit 480495beb4. If some old repo
had a key with a newline in it, --all will just skip processing that key.
Other things, like .git/annex/unused files certianly assume no newlines in
keys too, and AFAICR, such keys never actually worked.
Carriage return is escaped by preSanitizeKeyName since 2013. WORM keys
generated before that point could perhaps contain a CR. (URL probably not,
http probably doesn't support an URL with a raw CR in it.) So, added
a warning in fsck about such keys. Although, fsck --all will naturally
skip them, so won't be able to warn about them. Not entirely
satisfactory, but I'll bet there are not really any such keys in
existence.
Thanks to Lukey for finding this optimisation.
Only supported by some special remotes: directory
I need to check the rest and they're currently missing methods until I do.
git-annex sync --no-content does not yet use this to do imports
Looked into this, and dropKey from web actually removes the url,
so git-annex won't try to get content from it.
So, if lockContent were implemented for web, and the web was left as the
only thing containing an object, another repo could at the same time
drop from web and remove its url, leaving no way to get the object.
Add to that, of course, the web is typically set untrusted, and so
implementing lockContent would not then be useful.
Similar reasoning applies to the bittorrent special remote, as well
as the fact that it does not even implement checkKey.
Audited for openFile and openFd, and this fixes all the ones I found
where an async exception could prevent the file getting closed.
Except for the lock pool, which is a whole other can of worms.
Masking ensures that EndStderrHandler gets written, so the helper
threads shut down.
However, nothing currently guarantees that calls to closeP2PSshConnection
are async exception safe, so made a note about it.
At this point, I've audited all calls to async, and made them all async
exception safe, except for ones in the assistant, and a few in leaf
commands (remotedaemon, enable-tor, multicast, p2p) which don't need to
be.
This handles all createProcessSuccess callers, and aside from process
pools, the complete conversion of all process running to async exception
safety should be complete now.
Also, was able to remove from Utility.Process the old API that I now
know was not a good idea. And proof it was bad: The code size went *down*,
despite there being a fair bit of boilerplate for some future API to
reduce.
Added annex.skipunknown git config, that can be set to false to change the
behavior of commands like `git annex get foo*`, to not skip over files/dirs
that are not checked into git and are explicitly listed in the command
line.
Significant complexity was needed to handle git-annex add, which uses some
git ls-files calls, but needs to not use --error-unmatch because of course
the files are not known to git.
annex.skipunknown is planned to change to default to false in a
git-annex release in early 2022. There's a todo for that.
Todo item is done at last.
Might later want to think about testing some other types of remotes that
can be tested locally. The git remote itself is probably already well
enough tested by the test suite that testremote is not needed. Could
test things like bup, or rsync to a local directory. Or even external,
although that would require embedding an external special remote program
into the test suite..
The only price paid is one additional MVar read per write to the journal.
Presumably writing a journal file dominiates over a MVar read time by
several orders of magnitude.
--batch does not get the speedup because then it needs to notice when
another process has made a change. Also made the assistant and other damon
modes bypass the optimisation, which would not help them anyway.
* Added sync --only-annex, which syncs the git-annex branch and annexed
content but leaves managing the other git branches up to you.
* Added annex.synconlyannex git config setting, which can also be set with
git-annex config to configure sync in all clones of the repo.
Use case is then the user has their own git workflow, and wants to use
git-annex without disrupting that, so they sync --only-annex to get the
git-annex stuff in sync in addition to their usual git workflow.
When annex.synconlyannex is set, --not-only-annex can be used to override
it.
It's not entirely clear what --only-annex --commit or --only-annex
--push should do, and I left that combination not documented because I
don't know if I might want to change the current behavior, which is that
such options do not override the --only-annex. My gut feeling is that
there is no good reasons to use such combinations; if you want to use
your own git workflow, you'll be doing your own committing and pulling
and pushing.
A subtle question is, how should import/export special remotes be handled?
Importing updates their remote tracking branch and merges it into master.
If --only-annex prevented that git branch stuff, then it would prevent
exporting to the special remote, in the case where it has changes that
were not imported yet, because there would be a unresolved conflict.
I decided that it's best to treat the fact that there's a remote tracking
branch for import/export as an implementation detail in this case. The more
important thing is that an import/export special remote is entirely annexed
content, and so it makes a lot of sense that --only-annex will still sync
with it.
The use case is basically the user having forgotten, so --help would be
best, but it would be quite hard to include this in --help, since it may
even have to spin up an external special remote program.
I also considered --umm but typoed it the first time I tried it as
--uum, and while memorable, it's too cutesy. --whatelse is good because
it explicitly asks, what other params, besides the ones I've given?
This is a first step toward that goal, using the ProposedAccepted type
in RemoteConfig lets initremote/enableremote reject bad parameters that
were passed in a remote's configuration, while avoiding enableremote
rejecting bad parameters that have already been stored in remote.log
This does not eliminate every place where a remote config is parsed and a
default value is used if the parse false. But, I did fix several
things that expected foo=yes/no and so confusingly accepted foo=true but
treated it like foo=no. There are still some fields that are parsed with
yesNo but not not checked when initializing a remote, and there are other
fields that are parsed in other ways and not checked when initializing a
remote.
This also lays groundwork for rejecting unknown/typoed config keys.
ifAnnexed in a bare repo passes to git cat-file :./filename , which it
refuses to do since the repo is bare.
Note that, reinject somefile someannexedfile in a bare repo silently does
nothing, because someannexedfile is never actually an annexed worktree
file, because the repo is bare.
options make it easier to override annex.largefiles configuration
(and potentially safer as it avoids bugs like the smudge bug fixed
in the last release)
Deleted some old comments that were posted to the man page discussing such
options.
Updated docs that used -c annex.largefiles to use the options.
Note that addSmallOverridden was needed to avoid the clean filter running
on the file. It would be possible to make addFile also update the index
directly, rather than going via git add. However, it was not necessary,
and I want to avoid breaking on some edge case, particularly if the code in
addSmallOverridden has some oversight.
Also, when annex.addunlocked is set and annex.largefiles does not match a file,
git annex add --force-large works, but git status will then show the file
as added, with a unstaged modification. The unstaged modification adds the
file to git. This is identical behavior to using -c annex.largefiles=nothing
when annex.addunlocked is set. This does not prevent committing what was
intended to be added. I have not gotten to the bottom of why git thinks
the file is modified and runs it through the clean filter in this case.
smudge: When annex.largefiles=anything, files that were already stored in
git, and have not been modified could sometimes be converted to being
stored in the annex. Changes in 7.20191024 made this more of a problem.
This case is now detected and prevented.
The git add behavior changes could be avoided if it turns out to be
really annoying, but then it would need to behave the old way when
annex.dotfiles=false and the new way when annex.dotfiles=true. I'd
rather not have the config option result in such divergent behavior as
`git annex add .` skipping a dotfile (old) vs adding to annex (new).
Note that the assistant always adds dotfiles to the annex.
This is surprising, but not new behavior. Might be worth making it also
honor annex.dotfiles, but I wonder if perhaps some user somewhere uses
it and keeps large files in a directory that happens to begin with a
dot. Since dotfiles and dotdirs are a unix culture thing, and the
assistant users may not be part of that culture, it seems best to keep
its current behavior for now.
* annex.addunlocked can be set to an expression with the same format used by
annex.largefiles, in case you want to default to unlocking some files but
not others.
* annex.addunlocked can be configured by git-annex config.
Added a git-annex-matching-expression man page, broken out from
tips/largefiles.
A tricky consequence of this is that git-annex add --relaxed
honors annex.addunlocked, but an expression might want to know the size
or content of an url, which it's not going to download. I decided it was
better not to fail, and just dummy up some plausible data in that case.
Performance impact should be negligible. The global config is already
loaded for annex.largefiles. The expression only has to be parsed once,
and in the simple true/false case, it should not do any additional work
matching it.
Remove dup definitions and just use the RawFilePath one. </> etc are
enough faster that it's probably faster than building a String directly,
although I have not benchmarked.
My ByteString rewrite oversimplified it, resulting in any _ in a journal
file turning into a / in the git-annex branch, which was often the wrong
filename, or sometimes (//) an invalid filename that git
refused to add.
git-annex find is now RawFilePath end to end, no string conversions.
So is git-annex get when it does not need to get anything.
So this is a major milestone on optimisation.
Benchmarks indicate around 30% speedup in both commands.
Probably many other performance improvements. All or nearly all places
where a file is statted use RawFilePath now.
Since the sqlite branch uses blobs extensively, there are some
performance benefits, ByteStrings now get stored and retrieved w/o
conversion in some cases like in Database.Export.
The parser and looking up config keys in the map should both be faster
due to using ByteString.
I had hoped this would speed up startup time, but any improvement to
that was too small to measure. Seems worth keeping though.
Note that the parser breaks up the ByteString, but a config map ends up
pointing to the config as read, which is retained in memory until every
value from it is no longer used. This can change memory usage
patterns marginally, but won't affect git-annex.
While L.toStrict copies, profiling showed it was only around 0.3% of
git-annex find runtime. Does not seem worth optimising that, which would
probably involve either a major refactoring, or a use of
UnsafeInterleaveIO.
Also, it seems to me that the latter would need to read chunks, and
preappend the leftover part to the next chunk. But a strict ByteString
append itself is a copy, so I'm not convinced that would be faster than
L.toStrict.
* git-lfs: The url provided to initremote/enableremote will now be
stored in the git-annex branch, allowing enableremote to be used without
an url. initremote --sameas can be used to add additional urls.
* git-lfs: When there's a git remote with an url that's known to be
used for git-lfs, automatically enable the special remote.
Delete the old export dbs on upgrade.
Testing this an exporting to a directory with both exporttree=yes and
importtree=yes, it refused to let an interrupted export proceed after
upgrade, with "unsafe to overwrite file". An import resolved the
problem.
It will be populated automatically by the next command that needs data
from it, the same way it gets populated in a fresh clone. That may be a
little expensive, but it's a one time cost, and no slower than in a
fresh clone.
The old db is cleaned up when a new incremental fsck is started.
The incremental fsck won't pick up where the old one left off, but I
consider this a minor enough thing that it can just be documented and
won't be a problem.
The test suite found a bug; select_ can fail now because a uniqueness
constrain has been added.
Now the test suite passes.
Also, I'm satisfied the changed PersistField instances work.
Looking over what changed, and what I've already tested, Key, FilePath,
and InodeCache are known working; ContentIdentifier is trivial
ByteString to blob; and SSha is trivial String to varchar. Both are
tested by the test suite. I've also tested the new FileSize and
EpochTime instances already, and they work.
Bearing in mind that these indexes are really uniqueness constraints
that just happen to also make sqlite generate indexes.
In Database.ContentIndentifier, the ContentIndentifiersKeyRemoteCidIndex
is fine as a uniqueness constraint because it contains all rows from the
table. The ContentIndentifiersCidRemoteIndex is also ok because there
can only be one key for a given (cid, uuid) combination.
In Database.Export, the new ExportTreeFileKeyIndex is the same pair as
the old ExportTreeKeyFileIndex (previously ExportTreeIndex). And
in Database.Keys.SQL, the new InodeCacheKeyIndex is the same pair as the
old KeyInodeCacheIndex.
Added annex.gitaddtoannex configuration. Setting it to false prevents
git add from usually adding files to the annex.
(Unless the file was annexed before, or a renamed annexed file is detected.)
Currently left at true; some users are encouraging it be set to false.
The only good thing about it is it does not require a major version bump
to improve the database. That will need to happen at some point though.
Potentially very very slow in a large repository.
Ugly use of raw sql.
This is used by a special remote with sameas-uuid=
The remote's uuid is the sameas-uuid, but it needs to get
its RemoteConfig from the annex-config-uuid.
Straightforward, except for the issue of how to reverse LockAdjustment.
With --unlock, a commit that modifies/adds unlocked files gets reverse
adjusted to use locked files. That's fairly reasonable, I think.
But reversing --lock by unlocking all modified files feels wrong. Maybe
that's just because repositories typically seem to still have mostly
locked files in them (unless one is in an adjusted unlocked branch of
course!)
It may be that eventually how to reverse both will need to be configurable,
I don't know.
* Automatically convert direct mode repositories to v7 with adjusted
unlocked branches and set annex.thin.
* init: When run on a crippled filesystem with --version=5,
will error out, since version 7 is needed for adjusted unlocked branch.
* direct: This command always errors out as direct mode is no longer
supported.
* indirect: This command has become a deprecated noop.
* proxy: This command is deprecated because it was only needed in direct
mode. (But it continues to work.)
Also removed mentions of direct mode throughough the documentation.
I have not removed all the direct mode code yet.
Rather than direct mode, which this is a small step on the path to
removing.
Init on a crippled filesystem already used v7 adjusted branches,
and like that, this doesn't pose any interoperability issues with old
versions of git-annex that clone the same repo, because files are only
unlocked on the adjusted branch.
This means that Command.Move and Command.Get don't need to
manually set the stage, and is a lot cleaner conceptually.
Also, this makes Command.Sync.syncFile use the worker pool better.
In the scenario where it first downloads content and then uploads it to
some other remotes, it will start in TransferStage, then enter VerifyStage
and then go back to TransferStage for each transfer to the remotes.
Before, it entered CleanupStage after the download, and stayed in it for
the upload, so too many transfer jobs could run at the same time.
Note that, in Remote.Git, it uses runTransfer and also verifyKeyContent
inside onLocal. That has a Annex state for the remote, with no worker pool.
So the resulting calls to enteringStage won't block in there.
While Remote.Git.copyToRemote does do checksum verification, I
realized that should not use a verification slot in the WorkerPool
to do it. Because, it's reading back from eg, a removable disk to checksum.
That will contend with other writes to that disk. It's best to treat
that checksum verification as just part of the transer. So, removed the todo
item about that, as there's nothing needing to be done.
get, move, copy, sync: When -J or annex.jobs has enabled concurrency,
checksum verification uses a separate job pool than is used for
downloads, to keep bandwidth saturated.
Not yet done for upload checksum verification, but that only affects
remotes on local disks.
The hoped for optimisation of CommandStart with -J did not materialize.
In fact, not runnign CommandStart in parallel is slower than -J3.
So, CommandStart are still run in parallel.
(The actual bad performance I've been seeing with -J in my big repo
has to do with building the remoteList.)
But, this is still progress toward making -J faster, because it gets rid
of the onlyActionOn roadblock in the way of making CommandCleanup jobs
run separate from CommandPerform jobs.
Added OnlyActionOn constructor for ActionItem which fixes the
onlyActionOn breakage in the last commit.
Made CustomOutput include an ActionItem, so even things using it can
specify OnlyActionOn.
In Command.Move and Command.Sync, there were CommandStarts that used
includeCommandAction, so output messages, which is no longer allowed.
Fixed by using startingCustomOutput, but that's still not quite right,
since it prevents message display for the includeCommandAction run
inside it too.
When running multiple concurrent actions, the cleanup phase is run in a
separate queue than the main action queue. This can make some commands
faster, because less time is spent on bookkeeping in between each file
transfer.
But as far as I can see, nothing will be sped up much by this yet, because
all the existing cleanup actions are very light-weight. This is just groundwork
for deferring checksum verification to cleanup time.
This change does mean that if the user expects -J2 will mean that they see no
more than 2 jobs running at a time, they may be surprised to see 4 in some
cases (if the cleanup actions are slow enough to notice).
It might also make sense to enable background cleanup without the -J,
for at least one cleanup action. Indeed, that's the behavior that -J1
has now. At some point in the future, it make make sense to make the
behavior with no -J the same as -J1. The only reason it's not currently
is that git-annex can build w/o concurrent-output, and also any bugs
in concurrent-output (such as perhaps misbehaving on non-VT100 compatible
terminals) are avoided by default by only using it when -J is used.
Importing from a special remote honors its preferred content too; unwanted
files are not imported. But, some preferred content expressions can't be
checked before files are imported, and trying to import with such an
expression will fail.
Tested this with scenarios including changing the preferred content
expression and making sure merging the import didn't delete files that were
no longer wanted.
There was one minor inefficiency mentioned in the todo that I punted on.
Make the import have the previous import as a parent, so eg `git log --stat`
displays a useful diff.
Also a minor optimisation, only calculate the depth of the imported history
once.
Prevents merging the import from deleting the non-preferred files from
the branch it's merged into.
adjustTree previously appended the new list of items to the old, which
could result in it generating a tree with multiple files with the same
name. That is not good and confuses some parts of git. Gave it a
function to resolve such conflicts.
That allowed dealing with the problem of what happens when the import
contains some files (or subtrees) with the same name as files that were
filtered out of the export. The files from the import win.
This includes a note about how include= and exclude= match when exporting
a subtree. I don't know if the note is prominent enough, but the
behavior seems unsurprising enough.
This will let import try to match preferred content expressions before
downloading the content and generating its key.
If an expression needs a key, it preferredContentParser with
preferredContentKeylessTokens will fail to parse it.
standard and groupwanted are not in preferredContentKeylessTokens
because they may refer to an expression that refers to a key.
That needs further work to support them.
Added the ability to run one job per CPU (core), by setting annex.jobs=cpus,
or using option --jobs=cpus or -Jcpus.
Built with future expansion in mind, including not defaulting matching on
Concurrency so more constructors can later be added, and using "cpu"
instead of "0".
