| e56e84046fb601882235512491f4fa528c88523a |
|
14-May-2014 |
commit-bot@chromium.org <commit-bot@chromium.org@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Reland the 4x -> 1x allocation from "4x allocation in PipeController is probably overkill."
BUG=372671
R=mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/283933002
git-svn-id: http://skia.googlecode.com/svn/trunk@14734 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| f700fb2f139cc38f6ba3571db6cfe0a2e1bbc8c0 |
|
14-May-2014 |
commit-bot@chromium.org <commit-bot@chromium.org@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Manually revert "4x allocation in PipeController is probably overkill."
This reverts commit 2d91efffdb57646a495de5bf859ff281ef86dd12.
Conflicts:
src/pipe/SkGPipeWrite.cpp
BUG=372671
R=mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/277653004
git-svn-id: http://skia.googlecode.com/svn/trunk@14725 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| 2d91efffdb57646a495de5bf859ff281ef86dd12 |
|
07-May-2014 |
commit-bot@chromium.org <commit-bot@chromium.org@2bbb7eff-a529-9590-31e7-b0007b416f81> |
4x allocation in PipeController is probably overkill.
When verylargebitmap GM runs in cross-process pipe mode, we're
requestBlock()ing ~200M to carry the bitmaps. The current
implementation ends up allocating ~800M, which is a bit wasteful.
SkGPipeWrite already rounds up to 16K, so just rely on that.
This change exposed several bugs in pipe:
- we don't reserve enough space in drawVertices
- we don't reserve enough space for factory names in cross-process mode
- we don't quite have the right check in needOpBytes to see if we needed to send off the current block and allocate a new one
SETUP_NOTIFY and generally calling doNotify() more often than necessary made things hard to debug and understand. Now the pipe always waits to send off its current block until it needs more space than that block can provide, or it's the final block. We can put these back if we need the proactive flushing, but it seems not necessary?
Removed an assert in DeferredCanvasTest, which is somtimes 2 (Debug), sometimes 3 (Release). It seemed like the other asserts were more essential, and this one was more of a white-box assertion. Still sound if we remove it?
BUG=skia:2478
R=scroggo@google.com, mtklein@google.com, reed@google.com, junov@chromium.org
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/267863002
git-svn-id: http://skia.googlecode.com/svn/trunk@14613 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| 1f2f338e23789f3eef168dcbd8171a28820ba6c1 |
|
29-Aug-2013 |
robertphillips@google.com <robertphillips@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Split SkDevice into SkBaseDevice and SkBitmapDevice
https://codereview.chromium.org/22978012/
git-svn-id: http://skia.googlecode.com/svn/trunk@10995 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| 9b051a375ba6d6b61cea98f35834cd032aaa5347 |
|
20-Aug-2013 |
robertphillips@google.com <robertphillips@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Revert r10830 (Split SkDevice out of SkRasterDevice) until we can get Chromium ready.
git-svn-id: http://skia.googlecode.com/svn/trunk@10835 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| 3055b700189afdd02486ed8f2279cea1d8897243 |
|
20-Aug-2013 |
robertphillips@google.com <robertphillips@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Split SkDevice out of SkBitmapDevice
https://codereview.chromium.org/22978012/
git-svn-id: http://skia.googlecode.com/svn/trunk@10830 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| 74b7ffda687c66d46ac3cfa4f2baedd4c62e3fbe |
|
30-Apr-2013 |
scroggo@google.com <scroggo@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Fixes for piping bitmaps with encoded data.
Similar goals as https://codereview.chromium.org/14437012.
Builds on patch set 1 from that issue
(https://codereview.chromium.org/14437012/#ps1).
Instead of the changes in patch set 2 from that issue, this
changes SkOrderedWriteBuffer::writeBitmap to store whether an
SkBitmapHeap was used when to store the index of the SkBitmap.
SkOrderedReadBuffer::readBitmap now uses that information to
distinguish between using the heap and unflattening.
In addition, writeBitmap now records the width/height first in
all cases. If now SkBitmapHeapReader is attached, but an
SkBitmapHeap was used to record the bitmap, reading will fail
and provide the same red SkBitmap as in the case where the
SkBitmap was encoded but could not be decoded.
Updates the PICTURE_VERSION as well.
The key differences in this CL to look at are in:
SkOrderedWriteBuffer,
SkOrderedReadBuffer,
and SkPicture.
BUG=
R=djsollen@google.com
Review URL: https://codereview.chromium.org/14230022
git-svn-id: http://skia.googlecode.com/svn/trunk@8917 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| e5f48243bdbed2662be7a31be0888abc273b09e8 |
|
25-Feb-2013 |
scroggo@google.com <scroggo@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Fix a bunch of warnings, mainly around rowBytes.
My recent change changed the way SkBitmap::fRowBytes is stored,
and parameter/return values referring to rowBytes were changed
to type size_t. Change the storage back, and eliminate warnings
resulting from returning a size_t.
Review URL: https://codereview.appspot.com/7396059
git-svn-id: http://skia.googlecode.com/svn/trunk@7855 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| 0e51577a14f903ffeafa117a75954baeb173ffb9 |
|
07-Jan-2013 |
humper@google.com <humper@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
fix warning from scalar --> int32 conversion
BUG=
Review URL: https://codereview.appspot.com/7065050
git-svn-id: http://skia.googlecode.com/svn/trunk@7061 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| bcdf2ec50dfd170959cc2db67c49f6dac084be03 |
|
20-Sep-2012 |
scroggo@google.com <scroggo@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
In bench_pictures, use a pool of tiles for multicore drawing.
Also includes some code cleanup and code sharing.
Allow setting the number of threads on the command line.
Rename ThreadSafePipeController::playback to ::draw, to be the same
as SkPicture so DrawTileToCanvas can take a template parameter.
Disallow multithreading with GPU turned on.
Display help information with improper tiled arguments.
BUG=https://code.google.com/p/skia/issues/detail?id=871
Review URL: https://codereview.appspot.com/6536050
git-svn-id: http://skia.googlecode.com/svn/trunk@5602 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| 8e073ba2d63f4d2aab7532f1421df00bb1f88003 |
|
31-Aug-2012 |
scroggo@google.com <scroggo@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Rename my DeferredPipeController to not conflict with the other.
Fixes build.
Review URL: https://codereview.appspot.com/6506055
git-svn-id: http://skia.googlecode.com/svn/trunk@5372 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| 58b4ead36c62d8c0256ee4da554f3df2744d904c |
|
31-Aug-2012 |
scroggo@google.com <scroggo@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Perform multi core rendering in bench_pictures.
Add a flag in SkGPipeWriter for threadsafe drawing.
Add a deferred pipe controller to SamplePipeControllers, which can
be called to play back in multiple threads.
Depends on http://codereview.appspot.com/6459105/
Review URL: https://codereview.appspot.com/6482068
git-svn-id: http://skia.googlecode.com/svn/trunk@5371 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| b073d920c90bc95f6c216327acd321a4fc809bfa |
|
08-Jun-2012 |
scroggo@google.com <scroggo@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Add a tiled pipe mode to SampleApp.
Using the "Pipe" checkbox in SamleApp, you can now
toggle amongst three states: no pipe, a normal pipe
(or others depending on #defines), and a tiled pipe.
You can also use shift + p to toggle.
Review URL: https://codereview.appspot.com/6303055
git-svn-id: http://skia.googlecode.com/svn/trunk@4213 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|
| 72c9672ce274a3b6cb40800d66374edf25b157a3 |
|
06-Jun-2012 |
scroggo@google.com <scroggo@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> |
Add tiled rendering as an option to GM.
Use an SkGPipe to play back drawing into tiles.
This will help us to debug differences in drawing
while tiled.
Pass --tiledPipe to gm to use the tiled pipe.
Review URL: https://codereview.appspot.com/6295050
git-svn-id: http://skia.googlecode.com/svn/trunk@4199 2bbb7eff-a529-9590-31e7-b0007b416f81
/external/skia/src/pipe/utils/SamplePipeControllers.cpp
|