The earlier CL doesn't change the flag definition so it's not
turned on yet.
Bug: skia:
Change-Id: Id278ae5fc27d703ab7f6628bed95093d32cd7d0b
TBR: caryclark@google.com, fmalita@chromium.org
Reviewed-on: https://skia-review.googlesource.com/28161
Commit-Queue: Yuqian Li <liyuqian@google.com>
Reviewed-by: Yuqian Li <liyuqian@google.com>
ARMv7 can pass 16 floats as function arguments. We've been slicing that
as 8 2-float vectors. This CL switches to 4 4-float vectors.
We'll now operate on 4 pixels at a time instead of 2, at the expense of
keeping the d-vectors (mostly used for blending) on the stack. It'll
be interesting to see how this plays out performance-wise.
One nice side effect is now both ARMv7 and ARMv8 use 4-float NEON
vectors. Most of the code is now shared, with just a couple checks
to use new instructions added in ARMv8.
It looks like we do see a ~15% win:
$ bin/droid out/monobench SkRasterPipeline_srgb 200
Before: 644.029ns
After: 547.301ns
ARMv8: 453.838ns (just for reference)
Change-Id: I184ff29a36499e3cdb4c284809d40880b02c2236
Reviewed-on: https://skia-review.googlesource.com/27701
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
change underscore to space if needed
remove bmh_ prefix
TBR=caryclark@google.com
Bug: skia:
Change-Id: I9d4d29c7ff91d9d29bf8740d163724f371e5e211
Reviewed-on: https://skia-review.googlesource.com/28044
Reviewed-by: Cary Clark <caryclark@skia.org>
Commit-Queue: Cary Clark <caryclark@skia.org>
The documents were generated by running
tools/bookmaker against docs/*.bmh
No-Try: true
Docs-Preview: https://skia.org/?cl=28002
Change-Id: I7d7dd73cdea4a234c6175646d5b8938e1af3117a
Reviewed-on: https://skia-review.googlesource.com/28002
Reviewed-by: Cary Clark <caryclark@skia.org>
Commit-Queue: Cary Clark <caryclark@skia.org>
Bug: skia:6896
Change-Id: If2ed083703e6300b9d2a86cfb298011bbf54e018
Reviewed-on: https://skia-review.googlesource.com/28180
Reviewed-by: Eric Boren <borenet@google.com>
Also adds testing of copying ImageStorageAccess and ref counts of proxies held by cloned FPs.
Change-Id: Ia23220bf65b4df83d1c874b46d8525cc3540f716
Reviewed-on: https://skia-review.googlesource.com/28004
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
This is needed since there are currently no files in the include dir, and
some of our tests that check all the public headers complain if it doesn't
exists.
Bug: skia:6896
Change-Id: I3f70293b64b5096a55d6ba0ea5f4e6ebbfd7f62b
Reviewed-on: https://skia-review.googlesource.com/28003
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Eric Boren <borenet@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
bookmaker is a tool that generates documentation
backends from a canonical markup. Documentation for
bookmaker itself is evolving at docs/usingBookmaker.bmh,
which is visible online at skia.org/user/api/bmh_usingBookmaker
Change-Id: Ic76ddf29134895b5c2ebfbc84603e40ff08caf09
Reviewed-on: https://skia-review.googlesource.com/28000
Commit-Queue: Cary Clark <caryclark@google.com>
Reviewed-by: Cary Clark <caryclark@google.com>
We will then rebase the tests in Google3 before completely removing
SK_NO_ANALYTIC_AA.
Bug: skia:
Change-Id: I47ec0cde78374bd53f7cdf7c0f2f6607ca6c18c2
Reviewed-on: https://skia-review.googlesource.com/27801
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Yuqian Li <liyuqian@google.com>
Also use sk_sp for ref management of owned light object.
Change-Id: I2fa8427a9374351996c09f992f74cc83008605f1
Reviewed-on: https://skia-review.googlesource.com/28040
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
The bot was broken in https://skia-review.googlesource.com/c/27800/
Depot Tools' version is just a wrapper which looks for clang-format
under buildtools, which doesn't exist on the bots.
Bug: skia:6893
Change-Id: I32ae6c358735c9971e533064aa50a0e0520dd05a
Reviewed-on: https://skia-review.googlesource.com/28020
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Eric Boren <borenet@google.com>
In https://skia-review.googlesource.com/c/26363/ (Remove origin field from GrSurface) I just passed
an extra GrSurfaceOrigin parameter to all these methods. Besides being verbose this also led to
the "discard" method having an origin (to support the GrVkGpuCommandBuffer). It think this approach
is better and is plausible if the GrGpuCommandBuffer is viewed as an intermediary between the
GrProxy-based Ops and the Gpu.
In isolation this CL doesn't really show why we want to percolate the Proxy down. Once GrSurface
no longer has an origin a lot of the GrGpu methods need it passed in explicitly. By having the
GrGpuCommandBuffer get the proxy it can then pass the origin to GrGpu and removes a layer of
functions with an extra origin parameter.
Change-Id: Ie223fdee930171a32a5923155a0322e9a9c2aaa9
Reviewed-on: https://skia-review.googlesource.com/27980
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Bug: skia:6893
Change-Id: I54480ddd1ad1989661393500044f9164a99ce7b0
Reviewed-on: https://skia-review.googlesource.com/27700
Reviewed-by: Ravi Mistry <rmistry@google.com>
Commit-Queue: Eric Boren <borenet@google.com>
The FCLocker lock cannot be held while calling fTFCache::add. The
~SkTypeface_fontconfig method must take the FCLocker lock to destroy its
FcPattern since a typeface may be destroyed by any last owner. However,
fTFCache may be the last owner of some of its cached typefaces, and so a
purge when calling ::add to make space for the new entry may cause a
typeface to be destroyed. As a result, createTypefaceFromFcPattern must
not hold the FCLocker lock when calling fTFCache::add.
Fortunately, the FCLocker lock is only needed to serialize calls into
FontConfig. If acquire and release were free then they would be used
around each individual call to FontConfig. As a result it is fine to
give up the lock at any point in Skia code so long as no FontConfig
calls are made while the FCLocker lock is not held.
Change-Id: I37224d4b38bf88ace482496ce7530c84158d2d2e
Reviewed-on: https://skia-review.googlesource.com/27663
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
https://skia-review.googlesource.com/c/26363 (Remove origin field from GrSurface) is
already too large. This pulls some of the cosmetic changes out for separate review.
Change-Id: I1d8b95522144b2f4cbd916ef38faa3dde6f78087
Reviewed-on: https://skia-review.googlesource.com/27840
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Change-Id: I71cf04b12be95a54b7fb47d048ba1f8672ed9a8f
Reviewed-on: https://skia-review.googlesource.com/27760
Commit-Queue: Ben Wagner <bungeman@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Change-Id: I3e993e271cb5e26816d37c70d9ad62acce3ed84c
Reviewed-on: https://skia-review.googlesource.com/27800
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Also removes the hard stop gradient conditional compilation macro.
Change-Id: Idd56b5f3061db063d650a0090406ec192acc0b53
Reviewed-on: https://skia-review.googlesource.com/27661
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
This reverts commit dbce07d803.
Reason for revert: need rebaseline in Google3
Original change's description:
> Remove the temporary flag SK_NO_ANALYTIC_AA
>
> It seems that this temp flag lives for too long...
>
> Bug: skia:
> Change-Id: I48b49af8b7f372abac88a8cd5440def3791e635a
> Reviewed-on: https://skia-review.googlesource.com/27662
> Reviewed-by: Cary Clark <caryclark@google.com>
> Reviewed-by: Florin Malita <fmalita@chromium.org>
> Commit-Queue: Yuqian Li <liyuqian@google.com>
TBR=caryclark@google.com,fmalita@chromium.org,liyuqian@google.com,reed@google.com
Change-Id: I35a91923f16f4e1f530fae20599240cde73438b6
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:
Reviewed-on: https://skia-review.googlesource.com/27722
Reviewed-by: Yuqian Li <liyuqian@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
This has been disabled for almost two years.
Bug: skia:
Change-Id: Idc1dbf2220514947d2ccd91968e6e173d44d1b86
Reviewed-on: https://skia-review.googlesource.com/27740
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
It seems that this temp flag lives for too long...
Bug: skia:
Change-Id: I48b49af8b7f372abac88a8cd5440def3791e635a
Reviewed-on: https://skia-review.googlesource.com/27662
Reviewed-by: Cary Clark <caryclark@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Yuqian Li <liyuqian@google.com>
There are not many registers on 32-bit x86, and we're using most to pass
Stage function arguments. This means few are available as temporaries,
and we're forced to hit the stack all the time. xmm registers are the
most egregious example: we use all 8 registers pass data, leaving none
free as temporaries.
This CL cuts things down pretty dramatically, from passing 5 general
purpose and 8 xmm registers to 2 general purpose and 4 xmm registers.
One of the two general purpose registers is a pointer to space on the
stack where we store all those other values.
Every stage function needs to use the program pointer, so that stays in
a general purpose register. Almost every stage uses the r,g,b,a
vectors, so they stay in xmm registers. The rest (destination x,y, the
tail mask, a pointer to tricky constants, and the dr,dg,db,da vectors)
now live on the stack.
The generated code is about 20K smaller and runs about 20% faster.
$ out/monobench SkRasterPipeline_srgb 200
Before: 358.784ns
After: 282.563ns
Change-Id: Icc117af95c1a81c41109984b32e0841022f0d1a6
Reviewed-on: https://skia-review.googlesource.com/27620
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Again, this will be useful when GrSurface does not have a origin field.
Change-Id: I0c4ef5b441eaf38f3489e8631aced11b11fdb6ac
Reviewed-on: https://skia-review.googlesource.com/27540
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Change-Id: Icae3c6ce80a0bef097ea1010a4d065cc9d5a4c88
Reviewed-on: https://skia-review.googlesource.com/27560
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
This will allow Flutter to workaround the lack of this extension on a
small number of devices.
Bug: skia:
Change-Id: I46e54aaffc0551a7a17758ebc2016b39bda19abf
Reviewed-on: https://skia-review.googlesource.com/27440
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This reverts commit a4ce4b1f6b.
Fix SkPathRef deserialization malloc crash
If the path says it has more points/verbs/etc than the buffer could
be holding, then resetToSize could try to allocate something huge
and crash.
Bug: skia:
Change-Id: I23b8870e9f74386aca89fb8f9a60d3b452044094
Reviewed-on: https://skia-review.googlesource.com/26805
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Bug: skia:
Change-Id: I59185b8acc00b6df854c613ba29d1002111fcb42
Reviewed-on: https://skia-review.googlesource.com/27240
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
This makes Engines (task execution strategies: serial, thread, fork)
pluggable just like most of the rest of ok. It removes the thread and
process limits, as I find myself rarely caring about what they are
exactly. Instead of limiting to num-cores, we just allow any number of
concurrent threads, and any number of concurrent child processes subject
to OS limitations.
Change-Id: Icef49d86818fe9a4b7380efb60e73e40bc2e6b73
Reviewed-on: https://skia-review.googlesource.com/27140
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
In a future world where GrSurface no longer has an origin it will be
useful for the GrPipeline to be holding the GrRenderTargetProxy (which will
still have an origin).
Change-Id: I743a8cc07b6b92f8116227fb77b7c37da43cde8a
Reviewed-on: https://skia-review.googlesource.com/26804
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Bug: skia:
Change-Id: I1449f0f4d7d9ab6225d98c601eafa7461a2a7dde
Reviewed-on: https://skia-review.googlesource.com/27120
Commit-Queue: Brian Osman <brianosman@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Remove color clamp to allow for negative and greater than 1.0
values, when converting to scRGB.
Bug: b/62347704
Change-Id: I5b811d5dd464cf3bb126ad435a7d27c4563fdacc
Reviewed-on: https://skia-review.googlesource.com/26862
Commit-Queue: Stan Iliev <stani@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This reverts commit df6660f64e.
Reason for revert: crashing dm on android and windows
Original change's description:
> Fix SkPathRef deserialization malloc crash
>
> If the path says it has more points/verbs/etc than the buffer could
> be holding, then resetToSize could try to allocate something huge
> and crash.
>
> Change-Id: I40e8db87e4f61abb23217281ab0365c6af222fa3
> Reviewed-on: https://skia-review.googlesource.com/24802
> Reviewed-by: Mike Reed <reed@google.com>
> Commit-Queue: Mike Reed <reed@google.com>
TBR=bungeman@google.com,reed@google.com,enne@chromium.org
Change-Id: I06fa89c05b785652b097ae04e7ebc001a7c176b2
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/26944
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>
