Commit Graph

849 Commits

Author SHA1 Message Date
joshualitt
08da4f22d7 Revert of removing GrDrawEffect (patchset #4 id:60001 of https://codereview.chromium.org/571163002/)
Reason for revert:
reverting to unblock another revert

Original issue's description:
> removing GrDrawEffect
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/8ddbe8b9366c8c59c4fb55f01f253de8a0b37d6e

R=bsalomon@google.com
TBR=bsalomon@google.com
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Author: joshualitt@chromium.org

Review URL: https://codereview.chromium.org/577593003
2014-09-16 07:17:28 -07:00
joshualitt
8ddbe8b936 removing GrDrawEffect
BUG=skia:
R=bsalomon@google.com

Author: joshualitt@chromium.org

Review URL: https://codereview.chromium.org/571163002
2014-09-16 06:46:30 -07:00
joshualitt
249af15fb8 BUG=skia:
R=bsalomon@google.com, egdaniel@google.com, jvanverth@google.com, robertphillips@google.com

Author: joshualitt@chromium.org

Review URL: https://codereview.chromium.org/543623004
2014-09-15 11:41:14 -07:00
reed
841010f4bd disable conservative rasterclip on gpu, until we understand perspective
BUG=skia:
R=bsalomon@google.com

Author: reed@google.com

Review URL: https://codereview.chromium.org/561763002
2014-09-10 07:23:32 -07:00
reed
d954498c01 Revert of Revert of allow canvas to force conservative clips (for speed) (patchset #1 id:1 of https://codereview.chromium.org/554033003/)
Reason for revert:
May just rebaseline, plus want to see the results of the chrome tests, so re-trying this CL.

Original issue's description:
> Revert of allow canvas to force conservative clips (for speed) (patchset #7 id:120001 of https://codereview.chromium.org/541593005/)
>
> Reason for revert:
> multipicturedraw failed on nvprmsaa -- don't know why yet
>
> Original issue's description:
> > Allow SkCanvas to be initialized to force conservative rasterclips. This has the following effects:
> >
> > 1. Queries to the current clip will be conservatively large. This can mean the quickReject may return false more often.
> >
> > 2. The conservative clips mean less work is done.
> >
> > 3. Enabled by default for Gpu, Record, and NoSaveLayer canvases.
> >
> > 4. API is private for now.
> >
> > Committed: https://skia.googlesource.com/skia/+/27a5e656c3d6ef22f9cb34de18e1b960da3aa241
>
> TBR=robertphillips@google.com,bsalomon@google.com,mtklein@google.com,junov@google.com
> NOTREECHECKS=true
> NOTRY=true
>
> Committed: https://skia.googlesource.com/skia/+/6f09709519b79a1159f3826645f1c5fbc101ee11

R=robertphillips@google.com, bsalomon@google.com, mtklein@google.com, junov@google.com, reed@google.com
TBR=bsalomon@google.com, junov@google.com, mtklein@google.com, reed@google.com, robertphillips@google.com
NOTREECHECKS=true
NOTRY=true

Author: reed@chromium.org

Review URL: https://codereview.chromium.org/560713002
2014-09-09 18:46:22 -07:00
reed
6f09709519 Revert of allow canvas to force conservative clips (for speed) (patchset #7 id:120001 of https://codereview.chromium.org/541593005/)
Reason for revert:
multipicturedraw failed on nvprmsaa -- don't know why yet

Original issue's description:
> Allow SkCanvas to be initialized to force conservative rasterclips. This has the following effects:
>
> 1. Queries to the current clip will be conservatively large. This can mean the quickReject may return false more often.
>
> 2. The conservative clips mean less work is done.
>
> 3. Enabled by default for Gpu, Record, and NoSaveLayer canvases.
>
> 4. API is private for now.
>
> Committed: https://skia.googlesource.com/skia/+/27a5e656c3d6ef22f9cb34de18e1b960da3aa241

R=robertphillips@google.com, bsalomon@google.com, mtklein@google.com, junov@google.com
TBR=bsalomon@google.com, junov@google.com, mtklein@google.com, robertphillips@google.com
NOTREECHECKS=true
NOTRY=true

Author: reed@google.com

Review URL: https://codereview.chromium.org/554033003
2014-09-09 12:51:10 -07:00
reed
27a5e656c3 Allow SkCanvas to be initialized to force conservative rasterclips. This has the following effects:
1. Queries to the current clip will be conservatively large. This can mean the quickReject may return false more often.

2. The conservative clips mean less work is done.

3. Enabled by default for Gpu, Record, and NoSaveLayer canvases.

4. API is private for now.

R=robertphillips@google.com, bsalomon@google.com, mtklein@google.com, junov@google.com

Author: reed@google.com

Review URL: https://codereview.chromium.org/541593005
2014-09-09 12:19:30 -07:00
bsalomon
655ad128d0 Make GrEffectStage copy constructor work correctly with pending execution refs.
R=egdaniel@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/551083002
2014-09-08 12:06:20 -07:00
bsalomon
18444b1649 Remove debug-only ref of GrEffectStage
BUG=skia:2889
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/553973002
2014-09-08 10:45:58 -07:00
bsalomon
2a9ca782ba Make GrDrawState and GrEffectStage use the pending io/exec ref mechanisms.
BUG=skia:2889
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/542723004
2014-09-05 14:27:43 -07:00
bsalomon
49f085dddf "NULL !=" = NULL
R=reed@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/544233002
2014-09-05 13:34:00 -07:00
bsalomon
52e9d63f71 Don't take a ref on GP in AutoEffectRestore.
BUG=skia:2889
R=joshualitt@chromium.org, robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/546043002
2014-09-05 12:23:12 -07:00
bsalomon
ac8d6193ea Add method to GrProgramResource to record that pending a execution was completed.
BUG=skia:2889
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/543833002
2014-09-04 14:13:44 -07:00
bsalomon
95740981c3 Add GrProgramElement base class for GrEffect with deferred exec ref.
BUG=skia:2889
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/537773004
2014-09-04 13:12:37 -07:00
bsalomon
00b76bd750 Add reference base class to GrGpuResourcewith pending IO references.
BUG=skia:2889
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/533343002
2014-09-03 14:05:49 -07:00
robertphillips
98d709bc8d Reorganize Layer Hoisting code
With the new MultiPictureDraw API the GrContext will be performing the layer hoisting (instead of the SkGpuDevice). This CL being moving the layer hoisting functionality to GrLayerHoister rather then dumping it straight into GrContext.

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/531733003
2014-09-02 10:20:50 -07:00
bsalomon
6267f81f3d Move MakeDivByTextureWHMatrix to GrCoordTransform
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/522873005
2014-08-29 15:05:53 -07:00
bsalomon
32d0b3b546 Simplify SkGpuDevice construction
R=robertphillips@google.com, reed@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/506413004
2014-08-29 07:50:23 -07:00
bsalomon
23b406cc04 Make SkGpuDevice constructors private.
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/517033002
2014-08-28 14:06:42 -07:00
bsalomon
744998e666 Make textures register with GrResourceCache2 as scratch.
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/510053003
2014-08-28 09:54:34 -07:00
mtklein
533eb782ed Convert BBH APIs to use SkRect.
Still TODO: convert internals of SkTileGrid.cpp and SkRTree.cpp to work in floats too.

NOTREECHECKS=true

BUG=skia:1021
R=robertphillips@google.com, reed@google.com, mtklein@google.com

Author: mtklein@chromium.org

Review URL: https://codereview.chromium.org/511613002
2014-08-27 10:39:42 -07:00
bsalomon
169612621f Make GrGpuResources register with GrResourceCache2 after fully constructed.
Committed: https://skia.googlesource.com/skia/+/d68b3e491bb765beb7ca5e4ac8e0c80dedf5a83b

R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/504313002
2014-08-26 14:01:07 -07:00
bsalomon
a2b0af82f0 Revert of Make GrGpuResources register with GrResourceCache2 after fully constructed. (patchset #3 of https://codereview.chromium.org/504313002/)
Reason for revert:
Breaking the bots

Original issue's description:
> Make GrGpuResources register with GrResourceCache2 after fully constructed.
>
> Committed: https://skia.googlesource.com/skia/+/d68b3e491bb765beb7ca5e4ac8e0c80dedf5a83b

R=robertphillips@google.com
TBR=robertphillips@google.com
NOTREECHECKS=true
NOTRY=true

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/488463004
2014-08-26 13:11:11 -07:00
bsalomon
d68b3e491b Make GrGpuResources register with GrResourceCache2 after fully constructed.
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/504313002
2014-08-26 12:51:20 -07:00
kkinnunen
ec56e45454 Implement NV_path_rendering on OpenGL ES
Implement support for NV_path_rendering on OpenGL ES. Use
glProgramPathFragmentInputGenNV function call instead of glPathTexGenNV to
communicate transforms to fragment shader.

The intention is that the NVPR paths will be drawn with the same shader program
as non-NVPR geometry. For NVPR calls, the GPU will skip the vertex shader and
just run the fragment shader.

After program is linked, query the locations of the fragment shader inputs with
glGetResourceLocation. The location will be used to set the transforms with
glProgramPathFragmentInputGenNV.

The functions and their workings are documented in:

glProgramPathFragmentInputGenNV
https://www.opengl.org/registry/specs/NV/path_rendering.txt
(note: addition as of API version 1.3)

glGetResourceLocation
https://www.opengl.org/registry/specs/ARB/program_interface_query.txt
http://www.opengl.org/registry/doc/glspec44.core.pdf
(function is in core Open GL 4.4)

Note: glProgramPathFragmentInputGenNV could be used also for OpenGL. However,
using seems to trigger a bug in the driver. Disable this feature on OpenGL at
least until the driver is fixed and released. The bug manifests in shadertext
test, where the lower-left text pair is missing. Valgrind catches a bad read
for the test and causes the context to OOM reproducibly.

R=bsalomon@google.com, cdalton@nvidia.com, joshualitt@google.com, joshualitt@chromium.org

Author: kkinnunen@nvidia.com

Review URL: https://codereview.chromium.org/367643004
2014-08-25 22:21:16 -07:00
joshualitt
30ba436f04 Initial refactor of shaderbuilder to prepare for geometry shaders
gitignore for eclipse

BUG=skia:
R=bsalomon@google.com, bsalomon@chromium.org

Author: joshualitt@chromium.org

Review URL: https://codereview.chromium.org/491673002
2014-08-21 20:18:45 -07:00
robertphillips
64bf76722e Split up monolithic EXPERIMENTAL_drawPicture method
This is prep for the new multi-picture API.

This is split out of (Add new API to allow layer hoisting/atlasing across picture piles - https://codereview.chromium.org/474623002/)

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/489103002
2014-08-21 13:07:35 -07:00
bsalomon
c8dc1f74b6 Add GrResourceCache2.
Currently it just replaces GrGpu as the owner of the linked list of resources.

Committed: https://skia.googlesource.com/skia/+/94ce9ac8624dbb45656b8f5c992fad9c9ff3ee5f

R=mtklein@google.com, robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/481443002
2014-08-21 13:02:13 -07:00
bsalomon
1d89ddc4a3 Revert of Add GrResourceCache2. (patchset #4 of https://codereview.chromium.org/481443002/)
Reason for revert:
Likely caused a leak detected in Chromium after last Skia roll.

Original issue's description:
> Add GrResourceCache2.
>
> Currently it just replaces GrGpu as the owner of the linked list of resources.
>
> Committed: https://skia.googlesource.com/skia/+/94ce9ac8624dbb45656b8f5c992fad9c9ff3ee5f

R=mtklein@google.com, robertphillips@google.com
TBR=mtklein@google.com, robertphillips@google.com
NOTREECHECKS=true
NOTRY=true

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/477323006
2014-08-19 14:20:58 -07:00
bsalomon
94ce9ac862 Add GrResourceCache2.
Currently it just replaces GrGpu as the owner of the linked list of resources.

R=robertphillips@google.com, mtklein@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/481443002
2014-08-19 08:21:25 -07:00
reed
bbe7a2ad32 eliminate code related to SkBitmap::Config
BUG=skia:
R=reed@google.com, mike@reedtribe.org

Author: reed@chromium.org

Review URL: https://codereview.chromium.org/483593002
2014-08-18 08:13:03 -07:00
krajcevski
b1aded8edd Add flag to bench/render pictures
R=robertphillips@google.com, bsalomon@google.com

Author: krajcevski@google.com

Review URL: https://codereview.chromium.org/464423003
2014-08-18 07:52:17 -07:00
krajcevski
69a5560ea5 Add entry point for passing options to the GrContextFactory
R=bsalomon@google.com, mtklein@google.com, robertphillips@google.com

Author: krajcevski@google.com

Review URL: https://codereview.chromium.org/465073002
2014-08-13 10:46:31 -07:00
george
b62508be81 Add a parameter GR_ALWAYS_ALLOCATE_ON_HEAP to allow for only ever creating temporary buffers on the heap instead of allowing stack allocation
Unfortunately this can't be a GR_GL_* config like we'd discussed because there are SkAutoSMalloc buffers that are built outside of gl/

R=bsalomon@google.com, bsalomon
BUG=skia:

Author: george@mozilla.com

Review URL: https://codereview.chromium.org/459263003
2014-08-12 18:00:47 -07:00
krajcevski
9c6d4d744a Change GR_COMPRESS_ALPHA_MASK from compile-time flag to run-time. We do this by introducing an Options struct to be passed to a GrContext on creation.
R=robertphillips@google.com, bsalomon@google.com

Author: krajcevski@google.com

Review URL: https://codereview.chromium.org/459033002
2014-08-12 07:26:26 -07:00
bsalomon
d4cb922ea1 Consider index8 a compressed config.
R=krajcevski@google.com, robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/457333004
2014-08-11 14:19:09 -07:00
cdalton
c7103a104f Wrap NV_path_rendering API with GrGLPathRendering
Adds a GrGLPathRendering class that wraps the NV_path_rendering
extension and manages its various API versions. It also provides
backup implementations when certain NVpr methods from later API
versions are not present on the current system.

BUG=skia:
R=bungeman@google.com, bsalomon@google.com, kkinnunen@nvidia.com

Author: cdalton@nvidia.com

Review URL: https://codereview.chromium.org/444223002
2014-08-11 14:05:05 -07:00
bsalomon
9c0822a415 Always use both a color and coverage attribute in GrAARectRenderer.
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/435743002
2014-08-11 11:07:48 -07:00
reed
d5fa1a455a add drawPicture variant that takes a matrix and paint
will need some staging strategy, since chrome and blink have overrides of onDrawPicture

R=robertphillips@google.com, fmalita@google.com, bsalomon@google.com, mtklein@google.com

Author: reed@google.com

Review URL: https://codereview.chromium.org/448793004
2014-08-09 11:08:05 -07:00
joshualitt
bab82ed05b Pretty print of shaders
BUG=skia:
R=bsalomon@chromium.org, senorblanco@chromium.org, bsalomon@google.com

Author: joshualitt@chromium.org

Review URL: https://codereview.chromium.org/437593004
2014-08-08 09:41:42 -07:00
senorblanco
be129b26f1 Remove external SkImageFilter cache, and rename UniqueIDCache -> Cache.
There Can Only Be One.... Cache for SkImageFilter.

R=bsalomon@google.com

BUG=skia:

Author: senorblanco@chromium.org

Review URL: https://codereview.chromium.org/452923002
2014-08-08 07:14:35 -07:00
joshualitt
c863ab07c2 Cleanup of context initialization
Cleanup context initialization on linux

BUG=skia:
R=bsalomon@chromium.org, bsalomon@google.com

Author: joshualitt@chromium.org

Review URL: https://codereview.chromium.org/422323003
2014-08-07 13:48:50 -07:00
bungeman
760ba8d392 Revert "Adds a GrGLPathRendering class that wraps the NV_path_rendering"
This reverts commit 5672da0fa5.

This appears to be blocking the Skia roll by causing failures in the blink layout tests on the canvas-lost-gpu-context.html test.

The bisect for this can be seen at:

https://codereview.chromium.org/449473002/

Original issue's description:
> Adds a GrGLPathRendering class that wraps the NV_path_rendering
> extension and manages its various API versions. It also provides
> backup implementations when certain NVpr methods from later API
> versions are not present on the current system.
>
> Committed: https://skia.googlesource.com/skia/+/5672da0fa54f31c9727568e9dd5fe82c6e1585bc
2014-08-06 11:15:50 -04:00
egdaniel
6d7f5faeb3 Enable gpu Debug markers if tracing category skia.gpu is enabled
BUG=skia:
R=bsalomon@google.com

Author: egdaniel@google.com

Review URL: https://codereview.chromium.org/438023004
2014-08-04 12:54:39 -07:00
cdalton
5672da0fa5 Adds a GrGLPathRendering class that wraps the NV_path_rendering
extension and manages its various API versions. It also provides
backup implementations when certain NVpr methods from later API
versions are not present on the current system.

R=bsalomon@google.com, kkinnunen@nvidia.com, markkilgard@gmail.com

Author: cdalton@nvidia.com

Review URL: https://codereview.chromium.org/437473002
2014-08-04 11:19:11 -07:00
bsalomon
0673676854 Don't compare coord change matrices to determine effect compatibility when using explicit local coords.
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/438053002
2014-08-04 10:56:40 -07:00
bsalomon
01c8da1eef Remove unused matrix param from GrContext/GrDrawTarget rect drawing functions.
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/441623005
2014-08-04 09:21:30 -07:00
senorblanco
55b6d8be99 Implement a persistent uniqueID-based cache for SkImageFilter.
Add a unique ID to SkImageFilter, and use it as part
of a persistent cache of image-filtered results. This is used for
caching frame-to-frame coherent filters.

We also keep track of which filter subtrees do not reference the
src input, and use a GenID of zero for the src input in that case.
That way, subtrees which are not dependent on the filter input can be
cached independently of it.

This gives approximately a 4X speedup on
letmespellitoutforyou.com/samples/svg/filter_terrain.svg on Z620
and Nexus10. The cache key consists of the uniqueID of the filter, the
clip bounds, the CTM and the genID of the input bitmap.

Since this does not yet handle the case where the input primitives
(and part of the resulting filter tree) are unchanged, we have
to keep around the external cache for that painting case.
When the work to cache unchanging input primitives is done, the
old cache can be removed, and the new UniqueIDCache will be renamed
to Cache.

R=bsalomon@google.com, mtklein@google.com

Author: senorblanco@chromium.org

Review URL: https://codereview.chromium.org/414483003
2014-07-30 11:26:46 -07:00
cdalton
c8f5204667 Incorporate glStencilThenCover* nvpr methods
Adds the glStencilThenCover* nvpr methods to GrGLInterface and starts
using them. When drawing multible paths, this will make it so we only
have to send the index/transform data once. It will also allow the
driver to save time internally.

The glStencilThenCover* methods are a newer addition, so they aren't
available on every driver. In the event that they are not present, we
emulate them using the existing glStencil*/glCover* methods.

BUG=skia:
R=markkilgard@gmail.com, bsalomon@google.com

Author: cdalton@nvidia.com

Review URL: https://codereview.chromium.org/423173004
2014-07-29 15:25:51 -07:00
egdaniel
b414f25cf9 Remove support in the gpu to use hardware AA lines. Current code paths could not reach this code and thus was never used.
The blend opt flag kDisableBlend should now be able to be removed, however it is left in this CL and will removed in its own.

BUG=skia:
R=bsalomon@google.com

Author: egdaniel@google.com

Review URL: https://codereview.chromium.org/427823002
2014-07-29 13:15:47 -07:00