Change-Id: Id924f79c8b2c720297ebc49bf9c5d4ddd6d52547
Signed-off-by: Han-Wen Nienhuys <hanwen@google.com>

Change-Id: I1e85a951488177993521ce177fdea212615e6164
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>

Change-Id: I585242507815aad4aa0103fd55a6c369e342ab33

ObjectId is serializable, and so are its subtypes. Ensure that
serialization does not follow the hash collision chain internal to the
ObjectIdOwnerMap, otherwise completely unrelated objects may get
serialized when a RevObject is serialized.

Note that serializing a RevCommit or RevTag may serialize quite a few
objects due to the parent/object links they contain. A user has no real
control over how many objects will be written when a RevCommit is
serialized. C.f [1]. This change does not resolve that, but in any case
this internal hash collision chain link should not participate in
serialization.

[1] https://github.com/gitblit/gitblit/pull/1141

Change-Id: Ice331a9dc80a59ca360fcc04adaff8b5e750d847
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>

Hoist ObjectIdSet up to lib as part of the public API and add
the interface to some common types like PackIndex and JGit custom
ObjectId map types.  This cleans up wrapper code in a number of
places by allowing direct use of the types as an ObjectIdSet.

Future commits can now rely on ObjectIdSet as a simple read-only
type to check a set of objects from a number of storage options.

Change-Id: Ib62b062421d475bd52abd6c84a73916ef36e084b

Java compiler must generate synthetic access methods for private methods
and fields of the enclosing class if they are accessed from inner
classes and vice versa.

While invisible in the code, those synthetic access methods exist in the
bytecode and seem to produce some extra execution overhead at runtime
(compared with the direct access to this fields or methods), see
https://git.eclipse.org/r/58948/.

By removing the "private" access modifier from affected methods and
fields we help compiler to avoid generation of synthetic access methods
and hope to improve execution performance.

To validate changes, one can either use javap or use Bytecode Outline
plugin in Eclipse. In both cases one should look for "synthetic
access$<number>" methods at the end of the class and inner class files
in question - there should be none.

NB: don't mix this "synthetic access$" methods up with "public synthetic
bridge" methods generated to allow generic method override return types.

Change-Id: Ie7b65f251ec4452d5a5ed48aa0f272cf49a9aecd
Signed-off-by: Andrey Loskutov <loskutov@gmx.de>

OwnerMap is about 200 ms faster than SubclassMap, more friendly to the
GC, and uses less storage: testing the "Counting objects" part of
PackWriter on 1886362 objects:

  ObjectIdSubclassMap:
    load factor 50%
    table: 4194304 (wasted 2307942)
    ms spent 36998 36009 34795 34703 34941 35070 34284 34511 34638 34256
    ms avg 34800 (last 9 runs)

  ObjectIdOwnerMap:
    load factor 100%
    table: 2097152 (wasted 210790)
    directory: 1024
    ms spent 36842 35112 34922 34703 34580 34782 34165 34662 34314 34140
    ms avg 34597 (last 9 runs)

The major difference with OwnerMap is entries must extend from
ObjectIdOwnerMap.Entry, where the OwnerMap has injected its own
private "next" field into each object. This allows the OwnerMap to use
a singly linked list for chaining collisions within a bucket. By
putting collisions in a linked list, we gain the entire table back for
the SHA-1 bits to index their own "private" slot.

Unfortunately this means that each object can appear in at most ONE
OwnerMap, as there is only one "next" field within the object instance
to thread into the map. For types that are very object map heavy like
RevWalk (entity RevObject) and PackWriter (entity ObjectToPack) this
is sufficient, these entity types are only put into one map by their
container.  By introducing a new map type, we don't break existing
applications that might be trying to use ObjectIdSubclassMap to track
RevCommits they obtained from a RevWalk.

The OwnerMap uses less memory. Each object uses 1 reference more (so
we're up 1,886,362 references), but the table is 1/2 the size (2^20
rather than 2^21). The table itself wastes only 210,790 slots, rather
than 2,307,942. So OwnerMap is wasting 200k fewer references.

OwnerMap is more friendly to the GC, because it hardly ever generates
garbage. As the map reaches its 100% load factor target, it doubles in
size by allocating additional segment arrays of 2048 entries. (So the
first grow allocates 1 segment, second 2 segments, third 4 segments,
etc.)  These segments are hooked into the pre-allocated directory of
1024 spaces. This permits the map to grow to 2 million objects before
the directory itself has to grow. By using segments of 2048 entries,
we are asking the GC to acquire 8,204 bytes in a 32 bit JVM. This is
easier to satisfy then 2,307,942 bytes (for the 512k table that is
just an intermediate step in the SubclassMap). By reusing the
previously allocated segments (they are re-hashed in-place) we don't
release any memory during a table grow.

When the directory grows, it does so by discarding the old one and
using one that is 4x larger (so the directory goes to 4096 entries on
its first grow). A directory of size 4096 can handle up to 8 millon
objects. The second directory grow (16384) goes to 33 million objects.
At that point we're starting to really push the limits of the JVM
heap, but at least its many small arrays. Previously SubclassMap would
need a table of 67108864 entries to handle that object count, which
needs a single contiguous allocation of 256 MiB. That's hard to come
by in a 32 bit JVM. Instead OwnerMap uses 8192 arrays of about 8 KiB
each. This is much easier to fit into a fragmented heap.

Change-Id: Ia4acf5cfbf7e9b71bc7faa0db9060f6a969c0c50
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>