Commit Graph

27 Commits

Author SHA1 Message Date
robertphillips
478dd72336 Fix layer hoisting image filter corner cases
This CL fixes 5 bugs related to hoisting image filters:

For image filters the src layer (the one prior to filtering) often needs to be
smaller then the final layer. This requires the saveLayer's optional bounds
to be stored (in SkLayerInfo.h and SkRecordDraw.cpp) and then used in
compute_source_rect and carried around in GrCachedLayer.

The image filters can add an extra offset to the final draw operation.
This is now computed in GrLayerHoister::FilterLayer and  carried around in
GrCachedLayer.

Filtered layers must use exact matches. This is now done in GrLayerCache::lock.

The filter cache requires a valid matrix so it can compute the correct offset.
This is now done in GrLayerHoister::FilterLayer.

Filtered layers need to be drawn with drawSprite while unfiltered (and therefore
hopefully atlased) layers can be drawn with drawBitmap. This is now done in
draw_replacement_bitmap.

Committed: https://skia.googlesource.com/skia/+/702eb9622102599d94ab6798e6227cf29f48c2d3

Review URL: https://codereview.chromium.org/803183003
2014-12-16 08:25:55 -08:00
robertphillips
86f032832f Revert of Fix layer hoisting image filter corner cases (patchset #3 id:40001 of https://codereview.chromium.org/803183003/)
Reason for revert:
Breaking DM

Original issue's description:
> Fix layer hoisting image filter corner cases
>
> This CL fixes 5 bugs related to hoisting image filters:
>
> For image filters the src layer (the one prior to filtering) often needs to be
> smaller then the final layer. This requires the saveLayer's optional bounds
> to be stored (in SkLayerInfo.h and SkRecordDraw.cpp) and then used in
> compute_source_rect and carried around in GrCachedLayer.
>
> The image filters can add an extra offset to the final draw operation.
> This is now computed in GrLayerHoister::FilterLayer and  carried around in
> GrCachedLayer.
>
> Filtered layers must use exact matches. This is now done in GrLayerCache::lock.
>
> The filter cache requires a valid matrix so it can compute the correct offset.
> This is now done in GrLayerHoister::FilterLayer.
>
> Filtered layers need to be drawn with drawSprite while unfiltered (and therefore
> hopefully atlased) layers can be drawn with drawBitmap. This is now done in
> draw_replacement_bitmap.
>
> Committed: https://skia.googlesource.com/skia/+/702eb9622102599d94ab6798e6227cf29f48c2d3

TBR=bsalomon@google.com
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/790643009
2014-12-16 07:41:37 -08:00
robertphillips
702eb96221 Fix layer hoisting image filter corner cases
This CL fixes 5 bugs related to hoisting image filters:

For image filters the src layer (the one prior to filtering) often needs to be
smaller then the final layer. This requires the saveLayer's optional bounds
to be stored (in SkLayerInfo.h and SkRecordDraw.cpp) and then used in
compute_source_rect and carried around in GrCachedLayer.

The image filters can add an extra offset to the final draw operation.
This is now computed in GrLayerHoister::FilterLayer and  carried around in
GrCachedLayer.

Filtered layers must use exact matches. This is now done in GrLayerCache::lock.

The filter cache requires a valid matrix so it can compute the correct offset.
This is now done in GrLayerHoister::FilterLayer.

Filtered layers need to be drawn with drawSprite while unfiltered (and therefore
hopefully atlased) layers can be drawn with drawBitmap. This is now done in
draw_replacement_bitmap.

Review URL: https://codereview.chromium.org/803183003
2014-12-16 07:27:30 -08:00
kkinnunen
b5f4961b30 Fix GpuLayerCacheTest compile error
Fix GpuLayerCacheTest compile error:
GpuLayerCacheTest.cpp:120:34: error: comparison between signed and
unsigned integer expressions [-Werror=sign-compare]

gcc --version
gcc (Ubuntu/Linaro 4.6.4-1ubuntu1~12.04) 4.6.4

Review URL: https://codereview.chromium.org/794003003
2014-12-12 05:29:12 -08:00
robertphillips
e99d499caa Fuse GrReplacements and GrLayerCache
The conversion step from GrCachedLayer to ReplacementInfo isn't necessary.

Review URL: https://codereview.chromium.org/769533004
2014-12-03 07:33:57 -08:00
robertphillips
01d6e5f462 Use variable length key (rather than accumulated matrix) as save layer hoisting key
Adding the rendering canvas' CTM to the layer hoisting key (i.e., Add support for hoisting layers in pictures drawn with a matrix - https://codereview.chromium.org/748853002/) has increased the cache miss rate due to accumulated floating point error. This CL fixes part of the issue by using the chain of operation indices leading to each saveLayer as the key. The canvas' CTM must still form part of the key but should be less subject to accumulated error.

BUG=skia:2315

Review URL: https://codereview.chromium.org/753253002
2014-12-01 09:09:27 -08:00
mtklein
04c9695055 Remove Picture deletion listeners.
Looks like we can just have ~SkPicture put the message on the bus directly.

BUG=skia:3144

Review URL: https://codereview.chromium.org/751663002
2014-11-24 08:20:58 -08:00
robertphillips
4ab5a9048c Discard atlas after every MultiPictureDraw::draw
This is intended to prevent ghosting on tiled architectures.

This CL also defers creation of the atlas (and its texture) until it is actually needed.

Committed: https://skia.googlesource.com/skia/+/6d5b5455743414ddb11d2b8c1fe9d7959f2b853d

Review URL: https://codereview.chromium.org/678403002
2014-10-29 13:56:03 -07:00
bsalomon
f2703d83da rename GrTextureDesc->GrSurfaceDesc, GrTextureFlags->GrSurfaceFlags
Review URL: https://codereview.chromium.org/682223002
2014-10-28 14:33:06 -07:00
robertphillips
fd61ed0d79 Alter layer hoisting to only hoist layers for one canvas at a time
This CL alters layer hoisting to defer creation of the free floating layers until they are actually needed (rather than creating _all_ the hoisted layers at the start).

It also fixes a pre vs. post Concat bug with how matrices were being accumulated.

BUG=skia:2315

Review URL: https://codereview.chromium.org/657383004
2014-10-28 07:21:44 -07:00
robertphillips
3aac6e0848 Add clip to layer cache
This CL adds the clip region to the GPU layer hoisting image cache. It also switches back to the old caching behavior of using the entire CTM in the cache key rather then just the upper 2x2. This latter change is to focus more on hoisting rather then caching.

It also includes 2 smaller fixes:
a) layer's that have an image filter are no longer atlased (b.c. doing so complicates applying the image filter)

b) the result of clipping the layer's bounds to the current clip is used as the hoisted layer's size. This reduces the amount of pixels drawn to match a normal (non-hoisted) draw pass.

Review URL: https://codereview.chromium.org/640773004
2014-10-20 08:52:40 -07:00
robertphillips
7bb9ed756e Fix bug in GrCachedLayer reuse
In the new MultiPictureDraw tests a single hoisted layer is reused multiple times. The previous plot locking scheme allowed GrCachedLayer objects to be aggressively deleted prematurely leaving the reusing GrHoistedLayer objects with dangling pointers.

This CL changes adds a new pseudo-ref to GrCachedLayer. (It can't be a real ref since the cached layers aren't deleted when it goes to 0).

NOTRY=true

Committed: https://skia.googlesource.com/skia/+/5c481666c9678f43e039ad605457be3854cf8de3

Review URL: https://codereview.chromium.org/640323002
2014-10-10 11:38:29 -07:00
robertphillips
ed42059d2f Move offset and CTM from LayerCache Key to per-hoisted-layer info
This CL reduces the amount of information used in the layer cache key:
- the stop value isn't needed since the start value uniquely identifies the layer in the picture.
- only the upper-left 2x2 portion of the CTM should be used as a key for looking up cached layers.
- individual layers can be redraw in different locations so the final offset cannot be a part of the key.

Since this data is no longer stored in the cached layer, but is still required to draw the cached layer, it is now stored in the per-layer information (i.e., HoistedLayer).

This is split out of (Fix sub-picture layer rendering bugs - https://codereview.chromium.org/597293002/).

BUG=skia:2315
R=egdaniel@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/609403003
2014-09-29 11:39:38 -07:00
robertphillips
4aa6dfc0b7 Separate replacement creation from layer discovery
This is a simple refactoring that sets the stage for eliminating GrReplacements::ReplacementInfo and splitting up EXPERIMENTAL_drawPicture to support multi picture draw.

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/559603004
2014-09-17 07:50:47 -07:00
robertphillips
4815fe5a0a Fix bug in layer hoisting transition to SkRecord backend
Care must be taken when setting up the initial CTM matrix for partial SkRecord playbacks b.c. all the setMatrix calls will concatenate with the initial matrix (which may be different then the CTM that is required to draw correctly).

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

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/549143003
2014-09-16 10:32:43 -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
robertphillips
6f294af43b Refactor GrLayerCache for new API
The only substantive change in this CL is skipping atlasing
for any layers that are involved in nesting. Prior versions
have allowed nesting layers to be atlased.

--------------------------------------------------------

All times are on Windows with a repeat count of 200.
Format is: <time in ms> (<# of glBindframebuffer calls>)

How to interpret this:

For the boogie page:

 both columns should be the same (since boogie has no
        nested layers)
 without the new API the tiled case doesn't show any real
        benefit from hoisting

For the carsvg page:

 The nesting change does increase the number of FBO
      switches but doesn't kill performance
 Because of the location & size of the layers the
      tile case does show some improvement (even
      without the new API)

boogie
-------

		   w/o nested change   w/ nested change

simple                5.62 (811)           N/A

tiled                 7.72 (811)           N/A

simple w/ hoisting    5.23 (409)        5.77 (409)
(but no caching)

tiled w/ hoisting     7.57 (809)        7.49 (809)
(but no caching)

carsvg
------

		   w/o nested change   w/ nested change

simple 		      60.37 (141990)        N/A

tiled                115.13 (256929)        N/A

simple w/ hoisting    41.57 (64570)      42.82 (72279)
(but no caching)

tiled w/ hoisting     84.24 (154352)     84.71 (165630)
(but no caching)

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/476833004
2014-08-18 08:50:03 -07:00
robertphillips
0c42332859 Add CTM to the cached layers' key and reduce render target pingponging in layer pre-rendering
The CTM is required on the key so we will re-render when necessary.
The SkGpuDevice change ensures big layers in the midst of atlas-able layers don't cause a render target switch.

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/433553002
2014-07-31 11:02:38 -07:00
robertphillips
09dde64628 Now always use Atlas for layer cache
This is setup for running cluster telemetry tests

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/429593002
2014-07-29 07:14:49 -07:00
robertphillips
320c92380f Add plot-based purging to GrLayerCache
This CL allows a GrPlot full of atlased layer to be purged from the atlas to make room for new layers.

R=jvanverth@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/411703003
2014-07-29 06:07:19 -07:00
robertphillips
d771f6bc27 Add auto purging for SkPicture-related Ganesh resources (esp. layers)
This is intended to lower the bookkeeping burden for the Layer Caching feature. Cached layers are now automatically purged when a picture is deleted.

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/408923002
2014-07-22 10:18:06 -07:00
bsalomon
e904c09a3a Fix alpha textures in NV ES3 contexts on Windows.
Make unit tests iterate over all the rendering GL context types rather than using kNative.

Fix the extension printing when gStartupSpew is set.

R=jvanverth@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/398183002
2014-07-17 10:50:59 -07:00
robertphillips
21048b588f Always have GrLayer's rect be valid
This CL just makes atlasing easier/clearer since there is a separate atlased query method. Not using the rect as a signal also simplifies the rendering of the layer in SkGpuDevice.cpp.

This is calved off from (Add atlased layer purging - https://codereview.chromium.org/367073002/)

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/384233002
2014-07-15 19:46:35 -07:00
Robert Phillips
cfaeec446d Remove Skia's use of the default SkPicture constructor and multi-clone
This cannot be landed until (Chrome: Switch to one-at-a-time SkPicture::clone interface - https://codereview.chromium.org/380323002/) has landed.

R=mtklein@google.com
TBR=reed@google.com

Review URL: https://codereview.chromium.org/388833003
2014-07-13 12:00:50 -04:00
robertphillips
952841bf41 Begin atlasing
This CL makes it possible for pulled-forward-layers to be atlased. It currently has a couple glaring limitations (which is why it is disabled):

1) the atlased layers cannot be purged nor aged out
2) the texture backing the atlas is not pulled from (or returned to) the resource cache

#1 is on hold until we have a recycling rectanizer

A separate major limitation (the non-atlased layers aren't cached) is blocked until we can transmute entries in the resource cache from scratch to non-scratch while potentially preserving their contents.

Committed: https://skia.googlesource.com/skia/+/55e61f0ef4e5c8c34ac107deaadc9b4ffef3111b

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/354533004
2014-06-30 08:26:50 -07:00
robertphillips
e462f2bed3 Revert of Begin atlasing (https://codereview.chromium.org/354533004/)
Reason for revert:
Sigh

Original issue's description:
> Begin atlasing
>
> This CL makes it possible for pulled-forward-layers to be atlased. It currently has a couple glaring limitations (which is why it is disabled):
>
> 1) the atlased layers cannot be purged nor aged out
> 2) the texture backing the atlas is not pulled from (or returned to) the resource cache
>
> #1 is on hold until we have a recycling rectanizer
>
> A separate major limitation (the non-atlased layers aren't cached) is blocked until we can transmute entries in the resource cache from scratch to non-scratch while potentially preserving their contents.
>
> Committed: https://skia.googlesource.com/skia/+/55e61f0ef4e5c8c34ac107deaadc9b4ffef3111b

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

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/359953002
2014-06-29 17:16:27 -07:00
robertphillips
55e61f0ef4 Begin atlasing
This CL makes it possible for pulled-forward-layers to be atlased. It currently has a couple glaring limitations (which is why it is disabled):

1) the atlased layers cannot be purged nor aged out
2) the texture backing the atlas is not pulled from (or returned to) the resource cache

#1 is on hold until we have a recycling rectanizer

A separate major limitation (the non-atlased layers aren't cached) is blocked until we can transmute entries in the resource cache from scratch to non-scratch while potentially preserving their contents.

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/354533004
2014-06-29 15:08:31 -07:00