When reading large indexes from disk, a portion of the time is
dominated in malloc() calls. This can be mitigated by allocating a
large block of memory and manage it ourselves via memory pools.
This change moves the cache entry allocation to be on top of memory
pools.
Design:
The index_state struct will gain a notion of an associated memory_pool
from which cache_entries will be allocated from. When reading in the
index from disk, we have information on the number of entries and
their size, which can guide us in deciding how large our initial
memory allocation should be. When an index is discarded, the
associated memory_pool will be discarded as well - so the lifetime of
a cache_entry is tied to the lifetime of the index_state that it was
allocated for.
In the case of a Split Index, the following rules are followed. 1st,
some terminology is defined:
Terminology:
- 'the_index': represents the logical view of the index
- 'split_index': represents the "base" cache entries. Read from the
split index file.
'the_index' can reference a single split_index, as well as
cache_entries from the split_index. `the_index` will be discarded
before the `split_index` is. This means that when we are allocating
cache_entries in the presence of a split index, we need to allocate
the entries from the `split_index`'s memory pool. This allows us to
follow the pattern that `the_index` can reference cache_entries from
the `split_index`, and that the cache_entries will not be freed while
they are still being referenced.
Managing transient cache_entry structs:
Cache entries are usually allocated for an index, but this is not always
the case. Cache entries are sometimes allocated because this is the
type that the existing checkout_entry function works with. Because of
this, the existing code needs to handle cache entries associated with an
index / memory pool, and those that only exist transiently. Several
strategies were contemplated around how to handle this:
Chosen approach:
An extra field was added to the cache_entry type to track whether the
cache_entry was allocated from a memory pool or not. This is currently
an int field, as there are no more available bits in the existing
ce_flags bit field. If / when more bits are needed, this new field can
be turned into a proper bit field.
Alternatives:
1) Do not include any information about how the cache_entry was
allocated. Calling code would be responsible for tracking whether the
cache_entry needed to be freed or not.
Pro: No extra memory overhead to track this state
Con: Extra complexity in callers to handle this correctly.
The extra complexity and burden to not regress this behavior in the
future was more than we wanted.
2) cache_entry would gain knowledge about which mem_pool allocated it
Pro: Could (potentially) do extra logic to know when a mem_pool no
longer had references to any cache_entry
Con: cache_entry would grow heavier by a pointer, instead of int
We didn't see a tangible benefit to this approach
3) Do not add any extra information to a cache_entry, but when freeing a
cache entry, check if the memory exists in a region managed by existing
mem_pools.
Pro: No extra memory overhead to track state
Con: Extra computation is performed when freeing cache entries
We decided tracking and iterating over known memory pool regions was
less desirable than adding an extra field to track this stae.
Signed-off-by: Jameson Miller <jamill@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
It has been observed that the time spent loading an index with a large
number of entries is partly dominated by malloc() calls. This change
is in preparation for using memory pools to reduce the number of
malloc() calls made to allocate cahce entries when loading an index.
Add an API to allocate and discard cache entries, abstracting the
details of managing the memory backing the cache entries. This commit
does actually change how memory is managed - this will be done in a
later commit in the series.
This change makes the distinction between cache entries that are
associated with an index and cache entries that are not associated with
an index. A main use of cache entries is with an index, and we can
optimize the memory management around this. We still have other cases
where a cache entry is not persisted with an index, and so we need to
handle the "transient" use case as well.
To keep the congnitive overhead of managing the cache entries, there
will only be a single discard function. This means there must be enough
information kept with the cache entry so that we know how to discard
them.
A summary of the main functions in the API is:
make_cache_entry: create cache entry for use in an index. Uses specified
parameters to populate cache_entry fields.
make_empty_cache_entry: Create an empty cache entry for use in an index.
Returns cache entry with empty fields.
make_transient_cache_entry: create cache entry that is not used in an
index. Uses specified parameters to populate
cache_entry fields.
make_empty_transient_cache_entry: create cache entry that is not used in
an index. Returns cache entry with
empty fields.
discard_cache_entry: A single function that knows how to discard a cache
entry regardless of how it was allocated.
Signed-off-by: Jameson Miller <jamill@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert the base_sha1 member of struct split_index to use struct
object_id and rename it base_oid. Include cache.h to make the structure
visible.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Rename C++ keyword in order to bring the codebase closer to being able
to be compiled with a C++ compiler.
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In a96d3cc3f6 ("cache-tree: reject entries with null sha1", 2017-04-21)
we made sure that broken cache entries do not get propagated to new
trees. Part of that was making sure not to re-use an existing cache
tree that includes a null oid.
It did so by dropping the cache tree in 'do_write_index()' if one of
the entries contains a null oid. In split index mode however, there
are two invocations to 'do_write_index()', one for the shared index
and one for the split index. The cache tree is only written once, to
the split index.
As we only loop through the elements that are effectively being
written by the current invocation, that may not include the entry with
a null oid in the split index (when it is already written to the
shared index), where we write the cache tree. Therefore in split
index mode we may still end up writing the cache tree, even though
there is an entry with a null oid in the index.
Fix this by checking for null oids in prepare_to_write_split_index,
where we loop the entries of the shared index as well as the entries for
the split index.
This fixes t7009 with GIT_TEST_SPLIT_INDEX. Also add a new test that's
more specifically showing the problem.
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A common pattern to free a piece of memory and assign NULL to the
pointer that used to point at it has been replaced with a new
FREE_AND_NULL() macro.
* ab/free-and-null:
*.[ch] refactoring: make use of the FREE_AND_NULL() macro
coccinelle: make use of the "expression" FREE_AND_NULL() rule
coccinelle: add a rule to make "expression" code use FREE_AND_NULL()
coccinelle: make use of the "type" FREE_AND_NULL() rule
coccinelle: add a rule to make "type" code use FREE_AND_NULL()
git-compat-util: add a FREE_AND_NULL() wrapper around free(ptr); ptr = NULL
Replace occurrences of `free(ptr); ptr = NULL` which weren't caught by
the coccinelle rule. These fall into two categories:
- free/NULL assignments one after the other which coccinelle all put
on one line, which is functionally equivalent code, but very ugly.
- manually spotted occurrences where the NULL assignment isn't right
after the free() call.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When split-index is being used, we have two cache_entry arrays in
index_state->cache[] and index_state->split_index->base->cache[].
index_state->cache[] may share the same entries with base->cache[] so
we can quickly determine what entries are shared. This makes memory
management tricky, we can't free base->cache[] until we know
index_state->cache[] does not point to any of those entries.
unshare_split_index() is added for this purpose, to find shared
entries and either duplicate them in index_state->cache[], or discard
them. Either way it should be safe to free base->cache[] after
unshare_split_index().
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Also use the functions in cmd_update_index() in
builtin/update-index.c.
These functions will be used in a following commit to tweak
our use of the split-index feature depending on the setting
of a configuration variable.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a semantic patch for converting certain calls of memcpy(3) to
COPY_ARRAY() and apply that transformation to the code base. The result
is
shorter and safer code. For now only consider calls where source and
destination have the same type, or in other words: easy cases.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Many instances of duplicate words (e.g. "the the path") and
a few typoes are fixed, originally in multiple patches.
wildmatch: fix duplicate words of "the"
t: fix duplicate words of "output"
transport-helper: fix duplicate words of "read"
Git.pm: fix duplicate words of "return"
path: fix duplicate words of "look"
pack-protocol.txt: fix duplicate words of "the"
precompose-utf8: fix typo of "sequences"
split-index: fix typo
worktree.c: fix typo
remote-ext: fix typo
utf8: fix duplicate words of "the"
git-cvsserver: fix duplicate words
Signed-off-by: Li Peng <lip@dtdream.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If you have a large work tree but only make changes in a subset, then
$GIT_DIR/index's size should be stable after a while. If you change
branches that touch something else, $GIT_DIR/index's size may grow
large that it becomes as slow as the unified index. Do --split-index
again occasionally to force all changes back to the shared index and
keep $GIT_DIR/index small.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We know the positions of replaced entries via the replace bitmap in
"link" extension, so the "name" path does not have to be stored (it's
still in the shared index). With this, we also have a way to
distinguish additions vs replacements at load time and can catch
broken "link" extensions.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We are sure that after merge_base_index() is done. cache-tree can
still be used with the final index. So don't destroy cache tree.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
CE_REMOVE'd entries are removed here because only parts of the code
base (unpack_trees in fact) test this bit when they look for the
presence of an entry. Leaving them may confuse the code ignores this
bit and expects to see a real entry.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
prepare_to_write_split_index() does the major work, classifying
deleted, updated and added entries. write_link_extension() then just
writes it down.
An observation is, deleting an entry, then adding it back is recorded
as "entry X is deleted, entry X is added", not "entry X is replaced".
This is simpler, with small overhead: a replaced entry is stored
without its path, a new entry is store with its path.
A note about unpack_trees() and the deduplication code inside
prepare_to_write_split_index(). Usually tracking updated/removed
entries via read-cache API is enough. unpack_trees() manipulates the
index in a different way: it throws the entire source index out,
builds up a new one, copying/duplicating entries (using dup_entry)
from the source index over if necessary, then returns the new index.
A naive solution would be marking the entire source index "deleted"
and add their duplicates as new. That could bring $GIT_DIR/index back
to the original size. So we try harder and memcmp() between the
original and the duplicate to see if it needs updating.
We could avoid memcmp() too, by avoiding duplicating the original
entry in dup_entry(). The performance gain this way is within noise
level and it complicates unpack-trees.c. So memcmp() is the preferred
way to deal with deduplication.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The large part of this patch just follows CE_ENTRY_CHANGED
marks. replace_index_entry() is updated to update
split_index->base->cache[] as well so base->cache[] does not reference
to a freed entry.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Entries that belong to the base index should not be freed. Mark
CE_REMOVE to track them.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This split-index mode is designed to keep write cost proportional to
the number of changes the user has made, not the size of the work
tree. (Read cost is another matter, to be dealt separately.)
This mode stores index info in a pair of $GIT_DIR/index and
$GIT_DIR/sharedindex.<SHA-1>. sharedindex is large and unchanged over
time while "index" is smaller and updated often. Format details are in
index-format.txt, although not everything is implemented in this
patch.
Shared indexes are not automatically removed, because it's unclear if
the shared index is needed by any (even temporary) indexes by just
looking at it. After a while you'll collect stale shared indexes. The
good news is one shared index is useable for long, until
$GIT_DIR/index becomes too big and sluggish that the new shared index
must be created.
The safest way to clean shared indexes is to turn off split index
mode, so shared files are all garbage, delete them all, then turn on
split index mode again.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
