wip RawFilePath 2x git-annex find speedup

Finally builds (oh the agoncy of making it build), but still very
unmergable, only Command.Find is included and lots of stuff is badly
hacked to make it compile.

Benchmarking vs master, this git-annex find is significantly faster!
Specifically:

	num files	old	new	speedup
	48500		4.77	3.73	28%
	12500		1.36	1.02	66%
	20		0.075	0.074	0% (so startup time is unchanged)

That's without really finishing the optimization. Things still to do:

* Eliminate all the fromRawFilePath, toRawFilePath, encodeBS,
  decodeBS conversions.
* Use versions of IO actions like getFileStatus that take a RawFilePath.
* Eliminate some Data.ByteString.Lazy.toStrict, which is a slow copy.
* Use ByteString for parsing git config to speed up startup.

It's likely several of those will speed up git-annex find further.
And other commands will certianly benefit even more.
This commit is contained in:
Joey Hess 2019-11-26 15:27:22 -04:00
parent 6a97ff6b3a
commit 067aabdd48
No known key found for this signature in database
GPG key ID: DB12DB0FF05F8F38
61 changed files with 380 additions and 296 deletions

View file

@ -145,7 +145,7 @@ updateFromLog db (oldtree, currtree) = do
recordAnnexBranchTree db currtree
flushDbQueue db
where
go ti = case extLogFileKey remoteContentIdentifierExt (getTopFilePath (DiffTree.file ti)) of
go ti = case extLogFileKey remoteContentIdentifierExt (toRawFilePath (getTopFilePath (DiffTree.file ti))) of
Nothing -> return ()
Just k -> do
l <- Log.getContentIdentifiers k

View file

@ -128,28 +128,28 @@ addExportedLocation :: ExportHandle -> Key -> ExportLocation -> IO ()
addExportedLocation h k el = queueDb h $ do
void $ insertUnique $ Exported ik ef
let edirs = map
(\ed -> ExportedDirectory (toSFilePath (fromExportDirectory ed)) ef)
(\ed -> ExportedDirectory (toSFilePath (fromRawFilePath (fromExportDirectory ed))) ef)
(exportDirectories el)
putMany edirs
where
ik = toIKey k
ef = toSFilePath (fromExportLocation el)
ef = toSFilePath $ fromRawFilePath $ fromExportLocation el
removeExportedLocation :: ExportHandle -> Key -> ExportLocation -> IO ()
removeExportedLocation h k el = queueDb h $ do
deleteWhere [ExportedKey ==. ik, ExportedFile ==. ef]
let subdirs = map (toSFilePath . fromExportDirectory)
let subdirs = map (toSFilePath . fromRawFilePath . fromExportDirectory)
(exportDirectories el)
deleteWhere [ExportedDirectoryFile ==. ef, ExportedDirectorySubdir <-. subdirs]
where
ik = toIKey k
ef = toSFilePath (fromExportLocation el)
ef = toSFilePath $ fromRawFilePath $ fromExportLocation el
{- Note that this does not see recently queued changes. -}
getExportedLocation :: ExportHandle -> Key -> IO [ExportLocation]
getExportedLocation (ExportHandle h _) k = H.queryDbQueue h $ do
l <- selectList [ExportedKey ==. ik] []
return $ map (mkExportLocation . fromSFilePath . exportedFile . entityVal) l
return $ map (mkExportLocation . toRawFilePath . fromSFilePath . exportedFile . entityVal) l
where
ik = toIKey k
@ -159,13 +159,13 @@ isExportDirectoryEmpty (ExportHandle h _) d = H.queryDbQueue h $ do
l <- selectList [ExportedDirectorySubdir ==. ed] []
return $ null l
where
ed = toSFilePath $ fromExportDirectory d
ed = toSFilePath $ fromRawFilePath $ fromExportDirectory d
{- Get locations in the export that might contain a key. -}
getExportTree :: ExportHandle -> Key -> IO [ExportLocation]
getExportTree (ExportHandle h _) k = H.queryDbQueue h $ do
l <- selectList [ExportTreeKey ==. ik] []
return $ map (mkExportLocation . fromSFilePath . exportTreeFile . entityVal) l
return $ map (mkExportLocation . toRawFilePath . fromSFilePath . exportTreeFile . entityVal) l
where
ik = toIKey k
@ -181,21 +181,21 @@ getExportTreeKey (ExportHandle h _) el = H.queryDbQueue h $ do
map (fromIKey . exportTreeKey . entityVal)
<$> selectList [ExportTreeFile ==. ef] []
where
ef = toSFilePath (fromExportLocation el)
ef = toSFilePath (fromRawFilePath $ fromExportLocation el)
addExportTree :: ExportHandle -> Key -> ExportLocation -> IO ()
addExportTree h k loc = queueDb h $
void $ insertUnique $ ExportTree ik ef
where
ik = toIKey k
ef = toSFilePath (fromExportLocation loc)
ef = toSFilePath (fromRawFilePath $ fromExportLocation loc)
removeExportTree :: ExportHandle -> Key -> ExportLocation -> IO ()
removeExportTree h k loc = queueDb h $
deleteWhere [ExportTreeKey ==. ik, ExportTreeFile ==. ef]
where
ik = toIKey k
ef = toSFilePath (fromExportLocation loc)
ef = toSFilePath (fromRawFilePath $ fromExportLocation loc)
-- An action that is passed the old and new values that were exported,
-- and updates state.
@ -220,7 +220,7 @@ mkExportDiffUpdater removeold addnew h srcek dstek i = do
Nothing -> return ()
Just k -> liftIO $ addnew h (asKey k) loc
where
loc = mkExportLocation $ getTopFilePath $ Git.DiffTree.file i
loc = mkExportLocation $ toRawFilePath $ getTopFilePath $ Git.DiffTree.file i
runExportDiffUpdater :: ExportDiffUpdater -> ExportHandle -> Sha -> Sha -> Annex ()
runExportDiffUpdater updater h old new = do

View file

@ -235,7 +235,7 @@ reconcileStaged qh = do
where
go cur indexcache = do
(l, cleanup) <- inRepo $ pipeNullSplit diff
changed <- procdiff l False
changed <- procdiff (map decodeBL' l) False
void $ liftIO cleanup
-- Flush database changes immediately
-- so other processes can see them.
@ -262,7 +262,8 @@ reconcileStaged qh = do
-- perfect. A file could start with this and not be a
-- pointer file. And a pointer file that is replaced with
-- a non-pointer file will match this.
, Param $ "-G^" ++ toInternalGitPath (pathSeparator:objectDir)
, Param $ "-G^" ++ fromRawFilePath (toInternalGitPath $
toRawFilePath (pathSeparator:objectDir))
-- Don't include files that were deleted, because this only
-- wants to update information for files that are present
-- in the index.
@ -277,7 +278,7 @@ reconcileStaged qh = do
procdiff (info:file:rest) changed = case words info of
((':':_srcmode):dstmode:_srcsha:dstsha:_change:[])
-- Only want files, not symlinks
| dstmode /= fmtTreeItemType TreeSymlink -> do
| dstmode /= decodeBS' (fmtTreeItemType TreeSymlink) -> do
maybe noop (reconcile (asTopFilePath file))
=<< catKey (Ref dstsha)
procdiff rest True
@ -293,11 +294,11 @@ reconcileStaged qh = do
caches <- liftIO $ SQL.getInodeCaches ikey (SQL.ReadHandle qh)
keyloc <- calcRepo (gitAnnexLocation key)
keypopulated <- sameInodeCache keyloc caches
p <- fromRepo $ fromTopFilePath file
filepopulated <- sameInodeCache p caches
p <- fromRepo $ toRawFilePath . fromTopFilePath file
filepopulated <- sameInodeCache (fromRawFilePath p) caches
case (keypopulated, filepopulated) of
(True, False) ->
populatePointerFile (Restage True) key keyloc p >>= \case
populatePointerFile (Restage True) key (toRawFilePath keyloc) p >>= \case
Nothing -> return ()
Just ic -> liftIO $
SQL.addInodeCaches ikey [ic] (SQL.WriteHandle qh)