This reverts commit e02d3caab8.
Reason for revert: New logic triggers out-of-date assert. Reverting until I can fix that, too.
Original change's description:
> Add ImageToColorSpace helper in SkImageFilter
>
> Share this logic among a couple filters that need it. This also fixes a
> bug that showed up in the morhpology GM for GPU color space configs.
>
> BUG=skia:
>
> Change-Id: Ic686b07aff80e58e14a86108703bfbb3cf524979
> Reviewed-on: https://skia-review.googlesource.com/6475
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Brian Osman <brianosman@google.com>
>
TBR=bsalomon@google.com,robertphillips@google.com,brianosman@google.com,reviews@skia.org
BUG=skia:
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Change-Id: I00b444dfaaa9b5981f7b33b34419cf9795b52ddb
Reviewed-on: https://skia-review.googlesource.com/6480
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Share this logic among a couple filters that need it. This also fixes a
bug that showed up in the morhpology GM for GPU color space configs.
BUG=skia:
Change-Id: Ic686b07aff80e58e14a86108703bfbb3cf524979
Reviewed-on: https://skia-review.googlesource.com/6475
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This forces all of our testing tools to run with the discrete GPU in
laptop systems that have that option.
BUG=skia:
Change-Id: Ibd7629d6de5f063cdf219b3c7469210af5085d90
Reviewed-on: https://skia-review.googlesource.com/6474
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Brian Osman <brianosman@google.com>
GrMatrixConvolutionFilter is used in two places right now.
The GaussianBlur code ensures that we blur in the space of
the source image, so no conversion is needed. This changes
SkMatrixConvolutionImageFilter to give the same guarantee.
BUG=skia:
Change-Id: I1a6199d4893f2e579dc6fa809c59cd195bd62f90
Reviewed-on: https://skia-review.googlesource.com/6467
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Tag bitmap as sRGB, and adjust gain/bias in color-managed configs to
produce results that are less blown out.
Also added a colorized version of the GM, to validate that gamut
conversion is working.
BUG=skia:
Change-Id: I333988dcdaa1272121e8aa731b4188c942fe19d8
Reviewed-on: https://skia-review.googlesource.com/6466
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
GrRenderTargetOpList now stores the IDs along side each op.
This should put us closer to using proxy IDs and not forcing early render target instantiation as many comments point towards.
Change-Id: I1ee82b01a0818a80d2bcac39fdf3a4ee7dccecc9
Reviewed-on: https://skia-review.googlesource.com/6403
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Correct handling of kGray, k565, k4444 etc. is still a TODO.
SkImage_Generator and SkImage_Gpu are still TODOs.
BUG=skia:6021
Change-Id: Ib53d97d3a866b2b4934fd85c10100855743a8fab
Reviewed-on: https://skia-review.googlesource.com/6396
Commit-Queue: Matt Sarett <msarett@google.com>
Reviewed-by: Mike Reed <reed@google.com>
GrContext still doesn't convert, but it has the source and destination
color spaces, and call sites are supplying appropriate values where it
makes sense.
BUG=skia:
Change-Id: Ia88733125b8090776cfc9b0dc8030cce365b0b8b
Reviewed-on: https://skia-review.googlesource.com/6400
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Upgrading most Windows bots in the Skolo to version 1607 (build number 14393).
Adding IntelIris540 and GTX1070 bot.
BUG=skia:
Change-Id: Ide7a09a4061d1e387e1d2a22be31b14b13f2fa96
Reviewed-on: https://skia-review.googlesource.com/6357
Reviewed-by: Ravi Mistry <rmistry@google.com>
Reviewed-by: Eric Boren <borenet@google.com>
Commit-Queue: Ben Wagner <benjaminwagner@google.com>
Since the in/out parameter is a const SkPixmap without the proper
color table, there is no way to tell the client about it without
modifying the const SkPixmap. Rather than cheating, just return
false.
Change-Id: I63fdf57febc59e1ee9af13aa6eb9b253d19bcb17
Reviewed-on: https://skia-review.googlesource.com/6414
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Leon Scroggins <scroggo@google.com>
This reverts commit 9bcda719e8.
Reason for revert: No idea why the bots failed with the original CL. Putting it back in.
Original change's description:
> Revert "Fix issue in SkDebugCanvas where filter canvas prevents GrOp bounds from drawing"
>
> This reverts commit 4bf98e7e80.
>
> Reason for revert: Seems to have caused some compile bots to break-
> https://chromium-swarm.appspot.com/task?id=333c47cadb764f10&refresh=10
>
> Original change's description:
> > Fix issue in SkDebugCanvas where filter canvas prevents GrOp bounds from drawing
> >
> > Change-Id: I3446bfc42c4cf521916a03aa0f367cd38b4fd370
> > Reviewed-on: https://skia-review.googlesource.com/6401
> > Reviewed-by: Ben Wagner <bungeman@google.com>
> > Commit-Queue: Brian Salomon <bsalomon@google.com>
> >
>
> TBR=bsalomon@google.com,bungeman@google.com,reviews@skia.org
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
>
> Change-Id: I0a73fe8c13967233dec950c317693d13d738722a
> Reviewed-on: https://skia-review.googlesource.com/6408
> Commit-Queue: Ravi Mistry <rmistry@google.com>
> Reviewed-by: Ravi Mistry <rmistry@google.com>
>
TBR=bsalomon@google.com,bungeman@google.com,rmistry@google.com,reviews@skia.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Change-Id: Iad544f883804d13f04640498b4b63f735d33840e
Reviewed-on: https://skia-review.googlesource.com/6409
Commit-Queue: Ravi Mistry <rmistry@google.com>
Reviewed-by: Ravi Mistry <rmistry@google.com>
This reverts commit 4bf98e7e80.
Reason for revert: Seems to have caused some compile bots to break-
https://chromium-swarm.appspot.com/task?id=333c47cadb764f10&refresh=10
Original change's description:
> Fix issue in SkDebugCanvas where filter canvas prevents GrOp bounds from drawing
>
> Change-Id: I3446bfc42c4cf521916a03aa0f367cd38b4fd370
> Reviewed-on: https://skia-review.googlesource.com/6401
> Reviewed-by: Ben Wagner <bungeman@google.com>
> Commit-Queue: Brian Salomon <bsalomon@google.com>
>
TBR=bsalomon@google.com,bungeman@google.com,reviews@skia.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Change-Id: I0a73fe8c13967233dec950c317693d13d738722a
Reviewed-on: https://skia-review.googlesource.com/6408
Commit-Queue: Ravi Mistry <rmistry@google.com>
Reviewed-by: Ravi Mistry <rmistry@google.com>
- The base class already stores a color space xform, so just forward our
factory argument and use that one.
- We had factories that took arbitrary coefficients, but they were only
used in the unit test. Make the assumption of Mitchell's coefficients
really hard-coded, which lets us skip transposing, storing, comparing,
etc...
BUG=skia:
Change-Id: Ia1d4f72d32c201cba1da24d61b90250f383336d1
Reviewed-on: https://skia-review.googlesource.com/6394
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
99% of the time (or maybe 100%?) morphology will trigger pad_image,
so the input texture will already be in the destimation color space.
If that doesn't happen, then just force the source to be converted,
which keeps the morphology effect and driver code simple.
BUG=skia:
Change-Id: I98876af4f9e9a5da031973213ed76349752ce68f
Reviewed-on: https://skia-review.googlesource.com/6388
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Plumb calls down to SkCodec.
Add a gm
Change-Id: I16da24eb739295ab72f487df02f19968151443f3
Reviewed-on: https://skia-review.googlesource.com/6287
Reviewed-by: Matt Sarett <msarett@google.com>
Commit-Queue: Leon Scroggins <scroggo@google.com>
This clarifies ownership rules throughout SkFontHost_Mac.cpp, as well as
tidying up various bits of code. Clarifying ownership means replacing
AutoCFRelease with UniqueCFRef which is based on std::unique_ptr. Most of
the cleanup is removing now dead code and properly indenting.
Change-Id: I6d3a225d62b5d0f2f48a9e70c1a24317faa06747
Reviewed-on: https://skia-review.googlesource.com/6297
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
Change-Id: I36b8532c57d0b6004a5fd283e30a506df89a4fa6
Reviewed-on: https://skia-review.googlesource.com/6387
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: Iedfe5bd019ca1171ab09de569f74c57975aa54c9
Reviewed-on: https://skia-review.googlesource.com/6384
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: I80f951976558a284e55386e0a368f08bd835d8ca
Reviewed-on: https://skia-review.googlesource.com/6359
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
std::vector needs to be able to assign objects contained inside it. With
const member variables, this isn't possible. Remove the consts so
SkGIFLZWBlock can be assigned.
BUG=skia:6072
Change-Id: I990dc80fb1c49fbd584712c6d0c1154c2da36e85
Reviewed-on: https://skia-review.googlesource.com/6362
Reviewed-by: Leon Scroggins <scroggo@google.com>
Commit-Queue: Leon Scroggins <scroggo@google.com>
This function will always be useful. Don't make it look/feel like a
temporary hack.
BUG=skia:
Change-Id: I6506d7d51dc3b25a7dbcea4ac273f51cf05f3a89
Reviewed-on: https://skia-review.googlesource.com/6330
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Change-Id: I59a687569e2c7b69cdf65ebc9656c7fbffcc32e0
Motivation: to be used by SkXPS.
Reviewed-on: https://skia-review.googlesource.com/6355
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Hal Canary <halcanary@google.com>
Add one more empty contour check.
R=kjlubick@google.com
BUG=skia:6073
Change-Id: Ie971875d37388f6bd86a0bdaeaf9230d64aca38d
Reviewed-on: https://skia-review.googlesource.com/6356
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Cary Clark <caryclark@google.com>
There are only a couple of these, primarily focused on index8 srcs and 565 dsts. The burden's starting to outweigh the benefit. No one on the team knows MIPS assembly.
If we're going to try this again, I'd rather we try some sort of SkNx / compiler-intrinsic based approach, probably targeting MIPS SIMD (MSA), not this older instruction set.
We already ignore these files for 64-bit MIPS. This just closes the loop on 32-bit MIPS.
CQ_INCLUDE_TRYBOTS=skia.primary:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD,Build-Ubuntu-Clang-mips64el-Debug-Android,Build-Ubuntu-Clang-mips64el-Release-Android,Build-Ubuntu-Clang-mipsel-Debug-Android,Build-Ubuntu-Clang-mipsel-Release-Android
BUG=skia:6065
Change-Id: Iecac15b56f59625b2e743ea36e7791b90bb0b422
Reviewed-on: https://skia-review.googlesource.com/6353
Reviewed-by: Leon Scroggins <scroggo@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Change-Id: Ib831b9a6bcf4f37c0f077b26f68b1cefef81bb73
Reviewed-on: https://skia-review.googlesource.com/6351
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Motivation: Since I am activly refactoring SkXPS, I will be needing to
add and subtract sources. This is the only way to keep chrome and skia
in sync.
TODO(halcanry): follow up in Chrome build files.
BUG=skia:3495
Change-Id: I2bfcbfce7628c34f9a623c6fa806374a10cdbb75
Reviewed-on: https://skia-review.googlesource.com/6350
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Hal Canary <halcanary@google.com>
Change-Id: I68670e5ceb06716e9928ee58485d63e157c7aca7
Reviewed-on: https://skia-review.googlesource.com/6345
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Tag helper image as sRGB in magnifier image filter GM, so we can see
this working.
BUG=skia:
Change-Id: I8057dc332d09e1d508ad8462aaf0749b307f480f
Reviewed-on: https://skia-review.googlesource.com/6347
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Make the test factories use sk_sp.
Change-Id: Idba630b84deb2848f2203a80fd72e1efa5fc6acf
Reviewed-on: https://skia-review.googlesource.com/6342
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Most of the fields weren't really needed.
BUG=skia:
Change-Id: If0dbcb707b3a45d044656cbadbe118aa5583be2e
Reviewed-on: https://skia-review.googlesource.com/6338
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
BUG=skia:
Change-Id: I7508f27d4da6cec7acb801aa389f70c30cb5bfb5
Reviewed-on: https://skia-review.googlesource.com/6340
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Overlooked this in a recent CL
BUG=skia:
Change-Id: I050f8843cc2a8caadc2a46f1ea833b74b47f76b7
Reviewed-on: https://skia-review.googlesource.com/6341
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This CL fixes 3 fuzzer bugs:
the rowBytes computation is really unfortunate in ReadRawPixels when width or height is negative
In two instances uints were being read through the underlying SkReader32 rather than going through the readUInt virtual
In skip SkAlign4 can wrap around and return an answer of 0 for large size_ts!!
Change-Id: I657e3afdb4804627604b59abc8008ccdfe47eab2
Reviewed-on: https://skia-review.googlesource.com/6332
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
