Commit by dkazakov :: r1153471 krita/trunk/koffice/krita/image/ (12 files in 4 dirs):

Added a KisSwappedDataStore

This is a highlevel interface for a couple KisChunkAllocator and
KisMemoryWindow. This class is supposed to be used for swapping a
single tile out and restoring it back.

Commit by dkazakov :: r1152107 krita/trunk/koffice/krita/image/ (7 files in 3 dirs):

Added a sliding memory window for an upcoming swapper

It just provides random access to the file through memory mapping.

Commit by dkazakov :: r1151254 krita/trunk/koffice/krita/image/ (8 files in 3 dirs):

Added a chunk allocator for an upcoming swapper

Commit by dkazakov :: r1151067 krita/trunk/koffice/krita/image/ (4 files):

Clone layers work at last!

Now the updates of cloned layers are manageged specially, so they work
perfectly.

Btw, this was the last item of my todo list before i could call the
layer stack "usable". So now it is really so!

There are still some bugs present, but they do not depend on my work
(i hope):
1) The move tool sometimes refuses to move clone layers.
2) I cant add a Filter Mask to a clone or group layer.

Commit by dkazakov :: r1151035 krita/trunk/koffice/krita/image/ (3 files):

Fix a transparency mask.

It didn't work properly on a layers which have non-transparent default
pixel.

Commit by dkazakov :: r1151021 krita/trunk/koffice/krita/image/kis_layer.cc:

Fixed a race condition during creation of a projection for a layer.

Commit by dkazakov :: r1150938 krita/trunk/koffice/krita/image/ (6 files in 2 dirs):

Fixed a bug in New Adjustment Layer dialog

When the filter of the node changes too often (like the user is
constantly clicking the buttons of New Adjustment Layer dialog) some
walkers can be outdated, so now we use a checksum to catch this case.

Commit by dkazakov :: r1150512 krita/trunk/koffice/krita/image/ (3 files in 2 dirs):

Fix a bug of fastBitBlt

Source and destination points must coinside for fastBitBlt.
BUG:244750

Commit by dkazakov :: r1150492 krita/trunk/koffice/krita/image/ (8 files):

Added configurations for the update scheduler

• useUpdateScheduler - you can disable the scheduler by setting this option to false. KisProjection backend will be used instead.
• updatePatchWidth/updatePatchHeight the size of the patches produced after splitting huge update areas
• maxCollectAlpha/maxMergeAlpha/maxMergeCollectAlpha the parameters of optimization process

Commit by dkazakov :: r1150241 krita/trunk/koffice/krita/image/ (2 files):

Add documentation to KisIndirectPaintingSupport

Commit by dkazakov :: r1150240 krita/trunk/koffice/krita/image/ (4 files):

Fix a race condition between indirect painting and merging threads.

Commit by dkazakov :: r1150239 krita/trunk/koffice/krita/ (10 files in 2 dirs):

Major clean-up for KisToolFreehand

Now merging of a temporary target is a responsibility of
KisIndirectPaintingSupport

Commit by dkazakov :: r1150156 krita/trunk/koffice/krita/image/kis_simple_update_queue.cpp:

Return back queue'ing of updates (instead of stacking)

Commit by dkazakov :: r1150153 krita/trunk/koffice/krita/image/kis_updater_context.h:

Quite a silly flickering bug :)

I forgot about the key aim of the scheduler =)

Commit by dkazakov :: r1149823 krita/trunk/koffice/krita/image/ (6 files in 2 dirs):

Some optimizations for KisSimpleUpdateQueue

1) The update areas are now split into 512x512px patches
2) The update "queue" now works like a "stack" =) It means that it
   will first update area, where the user is painting *right now*. If
   the updater have missed some areas, they will be postponed until
   better days, because more recent areas have higher priority.

Commit by dkazakov :: r1149785 krita/trunk/koffice/krita/image/tiles3/kis_tile_hash_table_p.h:

The shared pointers are set to NULL automatically, there is no need to do it one more time

Commit by dkazakov :: r1149784 krita/trunk/koffice/krita/image/tiles3/ (kis_tile_data.cc kis_tile_data.h):

Make KisMemoryPool do a real job :)

He-he, i connected it wrongly the previous time, so it cached objects
of about 50 bytes =) Now it works with actual tile data. Though
benchmarks don't show anything =(

Commit by dkazakov :: r1149761 krita/trunk/koffice/krita/image/ (2 files in 2 dirs):

Fix fastBitBlt and activate it back

Commit by dkazakov :: r1148653 krita/trunk/koffice/krita/image/kis_painter.cc:

Disable fastBitBlt temporarily

Commit by dkazakov :: r1148574 krita/trunk/koffice/krita/image/kis_painter.cc:

Fix a KisPainter::bitBlt to add dirty rect when doing fast bitBlt

It should fix a bug reported by Lukas.