This reverts commit cd2c3f9055.
Reason for revert: Looks to be causing angle failures on initial clear test
Original change's description:
> Require mips to be allocated at texture creation time and disable late allocations.
>
> If we get a non-mipped texture for a draw that wants to be use mip map filter, we
> will copy the texture into a new mipped texture.
>
> Clean up of unused code in the GPU backends for reallocating for mips will be done
> in a follow up CL.
>
> Bug: skia:
> Change-Id: Idab588c1abf4bbbf7eeceb3727d500e5df274188
> Reviewed-on: https://skia-review.googlesource.com/132830
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Greg Daniel <egdaniel@google.com>
TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com
Change-Id: I49f0ace52f2586d61b451630b2e6aae84b420b81
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:
Reviewed-on: https://skia-review.googlesource.com/133041
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
If we get a non-mipped texture for a draw that wants to be use mip map filter, we
will copy the texture into a new mipped texture.
Clean up of unused code in the GPU backends for reallocating for mips will be done
in a follow up CL.
Bug: skia:
Change-Id: Idab588c1abf4bbbf7eeceb3727d500e5df274188
Reviewed-on: https://skia-review.googlesource.com/132830
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Use GrContext::FallbackTextHelper in SkTextBlobCacheDiffCanvas to
replicate glyph generation logic for fallback text during analysis. This
ensures that we correctly handle these fallback cases when using
distance field or paths for text rendering.
R=herb@google.com, jvanverth@google.com
Bug: skia:7913
Change-Id: I3067c4f1bd09231a564ac7c4cd89efcb876d2abd
Reviewed-on: https://skia-review.googlesource.com/132285
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Khusal Sagar <khushalsagar@chromium.org>
I'm looking at making SkArenaAlloc independent of Skia.
Here's a bit of low-hanging fruit.
Nothing is using makeSkSp(). Good... it seems real dangerous.
Change-Id: Ib7154e7948a3c6d828376ef37935636b3b4695ff
Reviewed-on: https://skia-review.googlesource.com/132824
Commit-Queue: Mike Klein <mtklein@chromium.org>
Commit-Queue: Herb Derby <herb@google.com>
Auto-Submit: Mike Klein <mtklein@chromium.org>
Reviewed-by: Herb Derby <herb@google.com>
This is the mechanism we'd want to add back if we add an sRGBA
SkColorType. This will break the DM "srgb" config hard.
Landing this first should help flush out GMs that are incidentally using
sRGB offscreens.
Updated tests/AnimatedImageTest.cpp to avoid linear sRGB 8888 surfaces.
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel
Change-Id: Idb5035cf4d60fcd1dc24c303d43a406fc4a603fa
Reviewed-on: https://skia-review.googlesource.com/132261
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Brian Osman <brianosman@google.com>
Pathops sorts line intersections to determine which edges to keep.
If the intersections are very short, they may get discarded and the
adjacent edge is used instead.
If a pair of edges are 180 degrees apart, and an adjacent edge is
part of the sort, it is ambiguous whether it is inside or outside
the span. Add logic to look for this and evaluate the original data
rather than the adjacent edge.
In a separate CL, I'll add a specialization for rect/rect ops.
R=halcanary@google.com
Bug: skia:8049
Change-Id: I8d88d5520051d41303ea683e7d6b844f2afa9937
Reviewed-on: https://skia-review.googlesource.com/132661
Commit-Queue: Hal Canary <halcanary@google.com>
Reviewed-by: Hal Canary <halcanary@google.com>
Auto-Submit: Cary Clark <caryclark@skia.org>
This reverts commit 74b390d6b1.
Reason for revert: reverting to add patch for valgrind
Original change's description:
> Revert "Have draw(Text|PosText|PosTextH) use a single entry on the device"
>
> This reverts commit 4225b3220e.
>
> Reason for revert: made valgrind unhappy.
>
> Original change's description:
> > Have draw(Text|PosText|PosTextH) use a single entry on the device
> >
> > Handle the positioning of drawText at the canvas layer. Simplify
> > the code by removing similar implementations.
> >
> > Change-Id: I8b711783435072f560e29fca1dd934fa2e345ed2
> > Reviewed-on: https://skia-review.googlesource.com/127131
> > Reviewed-by: Ben Wagner <bungeman@google.com>
> > Commit-Queue: Herb Derby <herb@google.com>
>
> TBR=jvanverth@google.com,bungeman@google.com,herb@google.com
>
> Change-Id: I65c9d30ae6ecb1f87e8660e56d8f8ce5daab7551
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Reviewed-on: https://skia-review.googlesource.com/132403
> Reviewed-by: Hal Canary <halcanary@google.com>
> Commit-Queue: Hal Canary <halcanary@google.com>
TBR=jvanverth@google.com,halcanary@google.com,bungeman@google.com,herb@google.com
Change-Id: I9bbb73aac447b51eb8215ac42331759fa4c9fa45
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/132580
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Herb Derby <herb@google.com>
No test tests, ok? Gotta draw the line somewhere.
Change-Id: I52284272f74a72a858910a143ecd4ad68fe3f1de
Reviewed-on: https://skia-review.googlesource.com/132410
Commit-Queue: Mike Klein <mtklein@chromium.org>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: Mike Klein <mtklein@chromium.org>
Reviewed-by: Brian Osman <brianosman@google.com>
It's actually possible for an intersection to be out-of-range on both
the intersected edges (e.g., below both bottom points), because
floating point. So we need to clamp against both edges.
Bug: 846014
Change-Id: I9fe25a1fcd3b5242af7b1ee36b17f1e968aeb836
Reviewed-on: https://skia-review.googlesource.com/132323
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Stephen White <senorblanco@chromium.org>
This reverts commit 4225b3220e.
Reason for revert: made valgrind unhappy.
Original change's description:
> Have draw(Text|PosText|PosTextH) use a single entry on the device
>
> Handle the positioning of drawText at the canvas layer. Simplify
> the code by removing similar implementations.
>
> Change-Id: I8b711783435072f560e29fca1dd934fa2e345ed2
> Reviewed-on: https://skia-review.googlesource.com/127131
> Reviewed-by: Ben Wagner <bungeman@google.com>
> Commit-Queue: Herb Derby <herb@google.com>
TBR=jvanverth@google.com,bungeman@google.com,herb@google.com
Change-Id: I65c9d30ae6ecb1f87e8660e56d8f8ce5daab7551
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/132403
Reviewed-by: Hal Canary <halcanary@google.com>
Commit-Queue: Hal Canary <halcanary@google.com>
Handle the positioning of drawText at the canvas layer. Simplify
the code by removing similar implementations.
Change-Id: I8b711783435072f560e29fca1dd934fa2e345ed2
Reviewed-on: https://skia-review.googlesource.com/127131
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Herb Derby <herb@google.com>
What makes an info valid (or invalid)? Nothing to do with
color space.
Bug: skia:
Change-Id: I6795efa9aa74ab0d65935c5ddccc1058f8e0b112
Reviewed-on: https://skia-review.googlesource.com/131780
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Change-Id: I40d046c66240ab40794aa008861a1974f7f9182c
Reviewed-on: https://skia-review.googlesource.com/131620
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
PS5: Removes SkDestinationSurfaceColorMode, tracking of mipmap
mode on GrTexture, sRGB decode state per-texture. Because we
were often choosing sRGB configs for RGB color types, legacy
rendering would then be incorrect (too dark). So...
PS7: Stops ever using sRGB pixel configs when translating
image info or color type. Also removes a bunch of GrCaps bits
and a GrContextOption that are no longer relevant.
PS9: Adjusts surface creation unit test expectations, and
changes the raster rules accordingly.
At this point, sRGB configs are (obviously) going to be broken.
Locally, I ran 8888, gl, and the gbr- versions of both. Across
all GMs x configs, there are 13 diffs. 12 are GMs that create
surfaces with a color-space attached (and thus, the offscreen
is no longer getting sRGB pixel config). The only remainder
constructs an SkPictureImageGenerator, (with an attached color
space) and renders it to the gbr-gl canvas, which triggers a
a tagged surface inside the generator.
Bug: skia:
Change-Id: Ie5edfa157dd799f3121e8173fc4f97f6c8ed6789
Reviewed-on: https://skia-review.googlesource.com/131282
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Most of this is (obviously) not necessary to do, but once
I started, I figured I'd just get it all. Tools (nanobench,
DM, skiaserve), all GMs, benches, and unit tests, plus support
code (command line parsing and config stuff).
This is almost entirely mechanical.
Bug: skia:
Change-Id: I209500f8df8c5bd43f8298ff26440d1c4d7425fb
Reviewed-on: https://skia-review.googlesource.com/131153
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This is a reland of 331c266ed7
Original change's description:
> Use GrVkMemoryAllocator for vulkan memory allocations in ganesh.
>
> Besides using the new allocator, the big logical change is that map
> and unmap calls form GrVkMemory are specc'd to map the entire GrVkAlloc
> instead of a specific offset and size as they did before. As a
> consequence of this, we move the handling of non-coherent alignment
> for flush/invalidate calls to GrVkMemory instead of the callers.
>
> Bug: skia:
> Change-Id: I794d713106602f27aa7e808c306bbb69fd2b67be
> Reviewed-on: https://skia-review.googlesource.com/130021
> Commit-Queue: Greg Daniel <egdaniel@google.com>
> Reviewed-by: Jim Van Verth <jvanverth@google.com>
Bug: skia:
Change-Id: Ia9a4192d344449fb444d2adaa1d62ff1ede4b21d
Reviewed-on: https://skia-review.googlesource.com/131083
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
feels so good.
Change-Id: I24d223957945ee81f34e5815a8e9afe32ef5004e
Reviewed-on: https://skia-review.googlesource.com/131110
Commit-Queue: Mike Klein <mtklein@chromium.org>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Auto-Submit: Mike Klein <mtklein@chromium.org>
Reviewed-by: Brian Salomon <bsalomon@google.com>
This reverts commit 32a4910e57.
Reason for revert: SkMatrix::toString use has been removed from flutter and has been picked up in fuchsia
Original change's description:
> Revert "remove toString"
>
> This reverts commit 5191880cbf.
>
> Reason for revert: broke flutter
>
> Original change's description:
> > remove toString
> >
> > toString may have been used by obsolete debugger only
> > find out if that is so
> >
> > R=​brianosman@google.com,bsalomon@google.com
> >
> > Docs-Preview: https://skia.org/?cl=119894
> > Bug:830651
> > Change-Id: I737f19b7d3fbc869bea2f443fa3b5ed7c1393ffd
> > Reviewed-on: https://skia-review.googlesource.com/119894
> > Commit-Queue: Cary Clark <caryclark@google.com>
> > Reviewed-by: Brian Salomon <bsalomon@google.com>
>
> TBR=bsalomon@google.com,brianosman@google.com,caryclark@google.com,caryclark@skia.org
>
> Change-Id: I9f81de6c3615ee0608bcea9081b77239b4b8816c
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: 830651
> Reviewed-on: https://skia-review.googlesource.com/129340
> Reviewed-by: Cary Clark <caryclark@google.com>
> Commit-Queue: Cary Clark <caryclark@google.com>
TBR=bsalomon@google.com,brianosman@google.com,caryclark@google.com,caryclark@skia.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: 830651
Change-Id: Ida8725b6051132d8c46faf99358a8fcc1bcabf34
Reviewed-on: https://skia-review.googlesource.com/129623
Reviewed-by: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@google.com>
Commit-Queue: Cary Clark <caryclark@google.com>
When support for out-of-range intersections was added in
3b5a3fa8b1, it was intended to support
splitting edges that are almost flat, where merging with the top or
bottom vertex would cause visual artifacts. However, it triggers too
often for other, non-nearly-flat cases, causing simplify() to loop
infinitely.
The fix is to support out-of-range intersections only if they differ by
1/2 machine epsilon. This also generalizes the
out_of_range_and_collinear() check, so it was removed.
Bug: 838978
Change-Id: I238f2b90e4b7ad647ecf072427ee38726e549581
Reviewed-on: https://skia-review.googlesource.com/130458
Reviewed-by: Stephen White <senorblanco@chromium.org>
This reverts commit 331c266ed7.
Reason for revert: breaking an intel vulkan bot
Original change's description:
> Use GrVkMemoryAllocator for vulkan memory allocations in ganesh.
>
> Besides using the new allocator, the big logical change is that map
> and unmap calls form GrVkMemory are specc'd to map the entire GrVkAlloc
> instead of a specific offset and size as they did before. As a
> consequence of this, we move the handling of non-coherent alignment
> for flush/invalidate calls to GrVkMemory instead of the callers.
>
> Bug: skia:
> Change-Id: I794d713106602f27aa7e808c306bbb69fd2b67be
> Reviewed-on: https://skia-review.googlesource.com/130021
> Commit-Queue: Greg Daniel <egdaniel@google.com>
> Reviewed-by: Jim Van Verth <jvanverth@google.com>
TBR=egdaniel@google.com,jvanverth@google.com,bsalomon@google.com
Change-Id: I5237c00625dc95d3d9b36c1e5591762988d85562
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:
Reviewed-on: https://skia-review.googlesource.com/131081
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Remove wide and narrow, adjust remaining configs to match the new CPU
configs, and fix several formatting problems in the help output.
Bug: skia:
Change-Id: I0946ac407d88a922e7f62220e7f4b39ae8c2e469
Reviewed-on: https://skia-review.googlesource.com/131000
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Besides using the new allocator, the big logical change is that map
and unmap calls form GrVkMemory are specc'd to map the entire GrVkAlloc
instead of a specific offset and size as they did before. As a
consequence of this, we move the handling of non-coherent alignment
for flush/invalidate calls to GrVkMemory instead of the callers.
Bug: skia:
Change-Id: I794d713106602f27aa7e808c306bbb69fd2b67be
Reviewed-on: https://skia-review.googlesource.com/130021
Commit-Queue: Greg Daniel <egdaniel@google.com>
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Change-Id: I6b08cd586d313e3bc41c0da90698fc26ae1a8bb8
Reviewed-on: https://skia-review.googlesource.com/130822
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Change-Id: Ia122050cabf68671c27e8aef3c69ec038c790899
Reviewed-on: https://skia-review.googlesource.com/130461
Commit-Queue: Ben Wagner <bungeman@google.com>
Commit-Queue: Bruce Wang <brucewang@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
This reverts commit 4f078f7cfa.
Reason for revert: Seems to have caused these breakages:
* https://chromium-swarm.appspot.com/task?id=3dc6788dc7b77010&refresh=10
* https://chromium-swarm.appspot.com/task?id=3dc67452db797c10&refresh=10
Original change's description:
> fonts: Use correct SurfaceProps in analysis canvas for remoting.
>
> SaveLayer may not preserve the behaviour for lcd text, in which case
> the surfaceProps used for desc generation are inconsistent.
>
> R=herb@google.com
>
> Bug: 829622
> Change-Id: I3adfc6780f26e4eb333a8aff76eaa4b5a9f0a0a7
> Reviewed-on: https://skia-review.googlesource.com/129557
> Reviewed-by: Herb Derby <herb@google.com>
> Commit-Queue: Khusal Sagar <khushalsagar@chromium.org>
TBR=herb@google.com,khushalsagar@chromium.org
Change-Id: I0372a3d834f8c4c929feb71c2b6b13739443eaaa
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 829622
Reviewed-on: https://skia-review.googlesource.com/130680
Reviewed-by: Ravi Mistry <rmistry@google.com>
Commit-Queue: Ravi Mistry <rmistry@google.com>
SaveLayer may not preserve the behaviour for lcd text, in which case
the surfaceProps used for desc generation are inconsistent.
R=herb@google.com
Bug: 829622
Change-Id: I3adfc6780f26e4eb333a8aff76eaa4b5a9f0a0a7
Reviewed-on: https://skia-review.googlesource.com/129557
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Khusal Sagar <khushalsagar@chromium.org>
We're not going to need the bit.
I've rewritten "esrgb" and "srgbnl" to express themselves the way I'd
like them to work. Their images are supressed in Gold already.
Change-Id: I6da58cc75dcb998cbfcf9a8f65de31c030adb494
Reviewed-on: https://skia-review.googlesource.com/130506
Reviewed-by: Brian Osman <brianosman@google.com>
This reverts commit bfb2a05af1.
Reason for revert: Seems to cause failures in chromeos and chromecast test bots. eg:
https://chromium-swarm.appspot.com/task?id=3dc5be8269e3b410&refresh=10https://chromium-swarm.appspot.com/task?id=3dc5d62dfdc99010&refresh=10
Original change's description:
> GrTessellator: fix for ping-pong split fuzzer hang.
>
> Change 3b5a3fa8b1 introduced support for
> splitting on out-of-range intersections. However, this is only necessary
> for correctness when the edge is nearly-flat (the top and bottom
> points only differ by 1/2 machine epsilon in the primary sort criterion).
> In other cases, it can cause repeated splitting and re-merging of edges,
> as the intersection code (being approximate and not exact) may produce a
> ping-pong set of intersections.
>
> The fix is to support out-of-range intersections only if they differ by
> 1/2 machine epsilon. This also generalizes the
> out_of_range_and_collinear() check, so it was removed.
>
> Bug: 838978
> Change-Id: I134f7eff3f15707e0d68de11c55f7fadce4ff8e7
> Reviewed-on: https://skia-review.googlesource.com/130448
> Reviewed-by: Robert Phillips <robertphillips@google.com>
> Commit-Queue: Stephen White <senorblanco@chromium.org>
TBR=robertphillips@google.com,senorblanco@chromium.org
Change-Id: I3fa62423d3875665397adcecb94b467b9b6611cc
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 838978
Reviewed-on: https://skia-review.googlesource.com/130522
Reviewed-by: Ravi Mistry <rmistry@google.com>
Commit-Queue: Ravi Mistry <rmistry@google.com>
Change 3b5a3fa8b1 introduced support for
splitting on out-of-range intersections. However, this is only necessary
for correctness when the edge is nearly-flat (the top and bottom
points only differ by 1/2 machine epsilon in the primary sort criterion).
In other cases, it can cause repeated splitting and re-merging of edges,
as the intersection code (being approximate and not exact) may produce a
ping-pong set of intersections.
The fix is to support out-of-range intersections only if they differ by
1/2 machine epsilon. This also generalizes the
out_of_range_and_collinear() check, so it was removed.
Bug: 838978
Change-Id: I134f7eff3f15707e0d68de11c55f7fadce4ff8e7
Reviewed-on: https://skia-review.googlesource.com/130448
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Stephen White <senorblanco@chromium.org>
Here's our drawing pipeline if we always blend as encoded.
I think it's very intuitive.
Change-Id: I0e531c5da1f6279d0da1f19b84d6317d99942da3
Reviewed-on: https://skia-review.googlesource.com/130131
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
* relocate all SkSG-related files under modules/sksg/
* fix various tidbits to make non-sksg builds possible
* drop obsolete SampleSGInval.cpp
Change-Id: I54e6c5bb1a09f45030fa8d607b3eb3f7cba78957
Reviewed-on: https://skia-review.googlesource.com/130025
Commit-Queue: Florin Malita <fmalita@chromium.org>
Reviewed-by: Mike Klein <mtklein@google.com>
Change-Id: I309b39425afc9b45095241eeb299096bc426afed
Reviewed-on: https://skia-review.googlesource.com/130029
Reviewed-by: Herb Derby <herb@google.com>
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Herb Derby <herb@google.com>
Change-Id: I26ed83bb613a83a0fd73f2f8b868b775be1ce968
Reviewed-on: https://skia-review.googlesource.com/130121
Commit-Queue: Mike Klein <mtklein@chromium.org>
Auto-Submit: Mike Klein <mtklein@chromium.org>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: I72e1084e6a14b3c1fa3f0a5c2100f13c6bcb24c7
Reviewed-on: https://skia-review.googlesource.com/130062
Commit-Queue: Mike Klein <mtklein@chromium.org>
Auto-Submit: Mike Klein <mtklein@chromium.org>
Reviewed-by: Brian Osman <brianosman@google.com>
- opaque and unpremul are never the same...
there's no reason to ever premul opaque sources.
- under optimization, they won't always be a deterministic
tweak the the premul steps (though right now they still are).
Change-Id: I5669b3dba83774326c07d5a8f16b1d2ce2b22aae
Reviewed-on: https://skia-review.googlesource.com/130061
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: Mike Klein <mtklein@chromium.org>
Change-Id: I3bb3ec19ee856ce4070a58a57b3bb8e8a170a5b9
Reviewed-on: https://skia-review.googlesource.com/130024
Commit-Queue: Mike Klein <mtklein@chromium.org>
Auto-Submit: Mike Klein <mtklein@chromium.org>
Reviewed-by: Brian Osman <brianosman@google.com>
Lots TODO.
Change-Id: I95edb764b85a5140d432adb506c3b537869e6df4
Reviewed-on: https://skia-review.googlesource.com/129933
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Brian Osman <brianosman@google.com>
Auto-Submit: Mike Klein <mtklein@chromium.org>
When erasing the prior frame rect, we need to update the frame rect to
take scaling into account. We already do if the prior frame was
provided; also do so if it was not provided.
Note that this only affects an image with a restore previous frame that
is being scaled. webp does not support restore previous, so this will
not affect AnimatedImageDrawable in Android.
Change-Id: I2d9a4ad45262a0e7afd1134958aff5c6e2ca6687
Reviewed-on: https://skia-review.googlesource.com/115646
Reviewed-by: Derek Sollenberger <djsollen@google.com>
Commit-Queue: Leon Scroggins <scroggo@google.com>
Bug: oss-fuzz:8509
Change-Id: I13b1a77e1549070827a7cc534b062ec85aad255e
Reviewed-on: https://skia-review.googlesource.com/129930
Commit-Queue: Mike Reed <reed@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Bug: 78866720
The client in Android calls newPictureSnapshot, which results in copying
the mutable SkBitmap into a newly allocated one in each frame. Avoid
this by calling SkMakeImageFromRasterBitmap with
kNever_SkCopyPixelsMode. Make SkAnimatedImage copy on write, by copying
before decoding if the bitmap's pixel ref is not unique.
Android's AnimatedImageDrawable's current architecture only decodes one
frame in advance, so it will never need to perform the copy on write.
This will save one bitmap allocation per GIF frame.
Add a test to verify that copy on write works as expected.
Change-Id: I87eb6e84089096cd2d618b91fb627fc58677e66a
Reviewed-on: https://skia-review.googlesource.com/129841
Reviewed-by: Leon Scroggins <scroggo@google.com>
Commit-Queue: Leon Scroggins <scroggo@google.com>
Auto-Submit: Leon Scroggins <scroggo@google.com>
Change-Id: I00ca97ce7f584bdcc042f7e76fb0f8116978b799
Reviewed-on: https://skia-review.googlesource.com/129761
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
