We can approximate (xy + 127) / 255 with (xy + 255) / 256.
On ARM this divide-by-255 is a single instruction, one of the two we use today
to do a perfect divide-by-255 (#if 0). This cuts div-255 in half, or a full
mul-div-255 by a third. The U16(255) constant can even be created in a single
instruction without hitting memory, which is as good as it gets.
Here's a nice little example:
0000000000000000 <sk_premul_8bit>:
0: f8408404 ldr x4, [x0], #8 // Load the next stage.
4: 2e23c000 umull v0.8h, v0.8b, v3.8b // r = r * a
8: 6f02e6b0 movi v16.2d, #0xff00ff00ff00ff // create U16(255)
c: 2e23c021 umull v1.8h, v1.8b, v3.8b // g = g * a
10: 2e23c042 umull v2.8h, v2.8b, v3.8b // b = b * a
14: 0e304000 addhn v0.8b, v0.8h, v16.8h // r = div255(r)
18: 0e304021 addhn v1.8b, v1.8h, v16.8h // g = div255(g)
1c: 0e304042 addhn v2.8b, v2.8h, v16.8h // b = div255(b)
20: d61f0080 br x4 // JUMP!
Change-Id: I4224ed3844abf6c67d9e42b67444a60f4aee8f08
Reviewed-on: https://skia-review.googlesource.com/40121
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Florin Malita <fmalita@chromium.org>
We're going to hide the path renderer bitfield behind a testing-only flag,
but clients may want to suppress this path renderer in particular, so a
dedicated flag makes sense.
Bug: skia:
Change-Id: I6cc4049243b5d3c129a824dfe7c1a65a70a05ccd
Reviewed-on: https://skia-review.googlesource.com/40083
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
pre-CL before trying to add a hint-rect field to allow shaders to "optimize"
their stages for a given restriction in device space
- e.g. if the shader's intrinsic domain is contained, it won't need to tile/clamp
Bug: skia:
Change-Id: Ia2da557691da25f31e4b9e3f53c3bc6709b89083
Reviewed-on: https://skia-review.googlesource.com/40224
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Reed <reed@google.com>
This (plus SW) ensures that we can always draw any path. Changed the path
renderer command line flags so that 'default' means everything but CCPR,
and 'All' really means 'every path renderer'.
Bug: skia:
Change-Id: I2801d1c50ab939f23efb4600e0f90e6add2891b7
Reviewed-on: https://skia-review.googlesource.com/40221
Reviewed-by: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Bug: skia:
Change-Id: I0372f7782124c096370b42624b6c0a3aff8a90b0
Reviewed-on: https://skia-review.googlesource.com/40101
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
The only reason we were keeping SkJumper_constants around is that it was
hard to get float/integer iota vectors on arm64 without relocations.
Now that we're compiling arm64 normally as part of Skia, we don't have
to worry about relocations.
This means we can kill the struct and stop passing around that pointer.
Change-Id: I013c6a735947f3db2bc87f2bfa38b7520d2e2fce
Reviewed-on: https://skia-review.googlesource.com/40200
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Herb Derby <herb@google.com>
39e781220c..b79b3f9627
$ git log 39e781220..b79b3f962 --date=short --no-merges --format='%ad %ae %s'
2017-07-21 oetuaho Assign symbol ids in RemoveDynamicIndexing
2017-08-28 yizhou.jiang Skip end2end test case due to failure on D3D Intel GPU
2017-08-25 mradev Relax multi-view end-point validation on detach
Created with:
roll-dep skia/third_party/externals/angle2
Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md
If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls
CQ_INCLUDE_TRYBOTS=skia.primary:Perf-Win10-MSVC-AlphaR2-GPU-RadeonR9M470X-x86_64-Debug-ANGLE,Perf-Win10-MSVC-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-ANGLE,Perf-Win10-MSVC-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-ANGLE,Perf-Win10-MSVC-NUCD34010WYKH-GPU-IntelHD4400-x86_64-Debug-ANGLE,Perf-Win10-MSVC-ShuttleC-GPU-GTX960-x86_64-Debug-ANGLE,Perf-Win10-MSVC-ZBOX-GPU-GTX1070-x86_64-Debug-ANGLE,Test-Win10-MSVC-AlphaR2-GPU-RadeonR9M470X-x86_64-Debug-ANGLE,Test-Win10-MSVC-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-ANGLE,Test-Win10-MSVC-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-ANGLE,Test-Win10-MSVC-NUCD34010WYKH-GPU-IntelHD4400-x86_64-Debug-ANGLE,Test-Win10-MSVC-ShuttleC-GPU-GTX960-x86_64-Debug-ANGLE,Test-Win10-MSVC-ZBOX-GPU-GTX1070-x86_64-Debug-ANGLE
TBR=djsollen@google.com
Change-Id: I6c7f43e779c9c9496ddf6fdb46f272e81e78f491
Reviewed-on: https://skia-review.googlesource.com/40180
Commit-Queue: angle-deps-roller . <angle-deps-roller@chromium.org>
Reviewed-by: angle-deps-roller . <angle-deps-roller@chromium.org>
Reduce path validation interfaces,
deferring the harder work of rewriting
the callers until later.
R=reed@google.com,enne@chromium.org
Change-Id: Iea56f1cd1be93bb1d96b50836a9bd3cd4872ad23
Reviewed-on: https://skia-review.googlesource.com/37541
Commit-Queue: Cary Clark <caryclark@google.com>
Reviewed-by: Adrienne Walker <enne@chromium.org>
Our interlaced approach works pretty well for x86, but on ARM we're a
lot better off deinterlacing in loads and reinterlacing in stores.
This leaves the stages mostly looking like the float stages, and cuts
out some awkward parts from the code generation.
Diffs are all invisible. Performance is noticeably better for some
blend modes like Overlay.
Change-Id: Ie599e823602bfd14552de78df44a621aea66e1a2
Reviewed-on: https://skia-review.googlesource.com/40100
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Florin Malita <fmalita@chromium.org>
We don't really use anything very ARMv8 specific in the 8-bit NEON
stages, so we can just naturally extend what we're doing to ARMv7 too.
Note that unlike the float stages, we're not requiring VFPv4 either,
just NEON. VFPv4 is for FMA and F16<->F32 conversion, both of which are
unnecessary for the integer pipeline.
GMs and perf improvement are similar to the previous ARMv8 change.
Change-Id: Id618801ea1920564c1deee144a640a4133c4505f
Reviewed-on: https://skia-review.googlesource.com/39840
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Reviewed-by: Herb Derby <herb@google.com>
This mainly consists of rm origin from GrSurface and the wrapBackEnd*
methods and then re-adding an explicit origin parameter to all the
GrGpu methods that need it.
TBR=bsalomon@google.com
Change-Id: I4248b2a4749ef844da4233ce53b0dc504bc9eb74
Reviewed-on: https://skia-review.googlesource.com/30280
Commit-Queue: Robert Phillips <robertphillips@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
This is a partial revert of https://skia-review.googlesource.com/c/20280,
adding back L32 support for 4f gradients and switching off the legacy
impl.
If it sticks, we should be able to completely delete the legacy gradient
code.
Change-Id: Iaa6d722ea4dfd8e4d959bb76eca815a30ca7098f
Reviewed-on: https://skia-review.googlesource.com/31425
Reviewed-by: Herb Derby <herb@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
This reverts commit 6d13575108.
Now with guards for "errors" like this:
external/skia/src/jumper/SkJumper_stages_8bit.cpp:240:50: error:
'memcpy' called with size bigger than buffer
case 12: memcpy(&v, src, 12*sizeof(T)); break;
This code is unreachable and generally removed by Clang's optimizer
anyway... as far as I can tell the code generation diff is arbitrary.
Change-Id: I6216567caaa6166f71258bd25343a09e93892a10
Reviewed-on: https://skia-review.googlesource.com/39961
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
This reverts commit 08133583d5.
Reason for revert: Blocking Android Autoroller on compile error.
Original change's description:
> 8-bit jumper on armv8
>
> The GM diffs are all minor and what you'd expect.
>
> I did a quick performance sanity check, which also looks fine.
>
> $ out/ok bench rp filter:search=Modulate
> [blendmode_rect_Modulate] 30.2ms @0 32ms @95 32ms @100
> [blendmode_mask_Modulate] 12.6ms @0 12.6ms @95 14.5ms @100
> ~~~>
> [blendmode_rect_Modulate] 11.2ms @0 11.7ms @95 12.4ms @100
> [blendmode_mask_Modulate] 10.5ms @0 23.6ms @95 23.9ms @100
>
> This isn't even really the fastest we can make 8-bit go on ARMv8;
> it's actually much more natural to work de-interlaced there. Lots
> of room to follow up.
>
> Change-Id: I86b1099f6742bcb0b8b4fa153e85eaba9567cbf7
> Reviewed-on: https://skia-review.googlesource.com/39740
> Reviewed-by: Florin Malita <fmalita@chromium.org>
> Commit-Queue: Mike Klein <mtklein@chromium.org>
TBR=mtklein@chromium.org,herb@google.com,fmalita@chromium.org,reed@google.com
Change-Id: I71425d8b7fbb66be5cb50025871dd81358111da4
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/39980
Reviewed-by: Derek Sollenberger <djsollen@google.com>
Commit-Queue: Derek Sollenberger <djsollen@google.com>
a42a4e5ebf..ff161f83ce
$ git log a42a4e5eb..ff161f83c --date=short --no-merges --format='%ad %ae %s'
2017-08-26 jmadill Fix EP generation event output format for uint.
Created with:
roll-dep skia/third_party/externals/angle2
Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md
If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls
CQ_INCLUDE_TRYBOTS=skia.primary:Perf-Win10-MSVC-AlphaR2-GPU-RadeonR9M470X-x86_64-Debug-ANGLE,Perf-Win10-MSVC-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-ANGLE,Perf-Win10-MSVC-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-ANGLE,Perf-Win10-MSVC-NUCD34010WYKH-GPU-IntelHD4400-x86_64-Debug-ANGLE,Perf-Win10-MSVC-ShuttleC-GPU-GTX960-x86_64-Debug-ANGLE,Perf-Win10-MSVC-ZBOX-GPU-GTX1070-x86_64-Debug-ANGLE,Test-Win10-MSVC-AlphaR2-GPU-RadeonR9M470X-x86_64-Debug-ANGLE,Test-Win10-MSVC-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-ANGLE,Test-Win10-MSVC-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-ANGLE,Test-Win10-MSVC-NUCD34010WYKH-GPU-IntelHD4400-x86_64-Debug-ANGLE,Test-Win10-MSVC-ShuttleC-GPU-GTX960-x86_64-Debug-ANGLE,Test-Win10-MSVC-ZBOX-GPU-GTX1070-x86_64-Debug-ANGLE
TBR=djsollen@google.com
Change-Id: I2a87310d1a5a95da10cbc73adeed07a2894cb4ab
Reviewed-on: https://skia-review.googlesource.com/39822
Reviewed-by: angle-deps-roller . <angle-deps-roller@chromium.org>
Commit-Queue: angle-deps-roller . <angle-deps-roller@chromium.org>
The GM diffs are all minor and what you'd expect.
I did a quick performance sanity check, which also looks fine.
$ out/ok bench rp filter:search=Modulate
[blendmode_rect_Modulate] 30.2ms @0 32ms @95 32ms @100
[blendmode_mask_Modulate] 12.6ms @0 12.6ms @95 14.5ms @100
~~~>
[blendmode_rect_Modulate] 11.2ms @0 11.7ms @95 12.4ms @100
[blendmode_mask_Modulate] 10.5ms @0 23.6ms @95 23.9ms @100
This isn't even really the fastest we can make 8-bit go on ARMv8;
it's actually much more natural to work de-interlaced there. Lots
of room to follow up.
Change-Id: I86b1099f6742bcb0b8b4fa153e85eaba9567cbf7
Reviewed-on: https://skia-review.googlesource.com/39740
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
This was undone in an editor, but didn't get commited.
Change-Id: Id410de0a066b3fe65119d0b72c6f29a1f91d1ebb
Reviewed-on: https://skia-review.googlesource.com/39700
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
1) Replace a couple commas with semicolons.
2) Make sure to zero a couple vectors.
1) has no effect on code generation.
2) does add a bunch of self-vxorps, but they're cheap and we already do
the equivalent for <AVX SSE code, and they're in not very
performance-critical routines. We could circle back and guard these
with !defined(JUMPER_IS_OFFLINE) if we really need the vectors to start
uninitialized for speed.
CQ_INCLUDE_TRYBOTS=skia.primary:Build-Debian9-Clang-x86_64-Release-Fast
Change-Id: I1a13f3eb28d664dbc345d71c3adbc62be5ff7c45
Reviewed-on: https://skia-review.googlesource.com/39661
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Just recalculates these points instead of investing the code
complexity and memory bandwidth in passing them around.
Bug: skia:
Change-Id: Ib6fc7af49363469de91b4c59baf238015045d79c
Reviewed-on: https://skia-review.googlesource.com/39540
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
This avoids the dangerous overload problem of
growToInclude(0, 0)
matching to (const SkPoint[], count) rather than growToInclude(x, y)
Bug: skia:
Change-Id: Iaba8b1a579638ff363fde62e4e3004052dd2b2ac
Reviewed-on: https://skia-review.googlesource.com/39501
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Mike Reed <reed@google.com>
The baseline compiled into Skia is now pretty much identical.
Minor diffs due to the offline code using -ffp-contract=fast, and the
baseline not. Explicit calls to fma() are still FMAs, but we're no
longer letting the compiler uncover FMAs we didn't explicitly call out.
If this goes well, we should be able to turn on the 8-bit pipeline.
Change-Id: I8f73157cfce7373574c20f6435fe86b46477afa9
Reviewed-on: https://skia-review.googlesource.com/39520
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
This was created by looking at warnings produced by clang's
-Wzero-as-null-pointer-constant. This updates most issues in
Skia code. However, there are places where GL and Vulkan want
pointer values which are explicitly 0, external headers which
use NULL directly, and possibly more uses in un-compiled
sources (for other platforms).
Change-Id: Id22fbac04d5c53497a53d734f0896b4f06fe8345
Reviewed-on: https://skia-review.googlesource.com/39521
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
Cleans up calc_inverse_transpose_power_basis_matrix and makes it
publicly visible.
Bug: skia:
Change-Id: I568c2b8518c74931962ece23ed07490e7e10e94b
Reviewed-on: https://skia-review.googlesource.com/39180
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
Whether JUMPER is defined is starting to get a little overloaded:
- are we compiling offline (defined) or as part of Skia (!defined)?
- are we using Clang vector extensions (defined) or scalars (!defined)?
This splits JUMPER into these two separate concerns:
- JUMPER_IS_OFFLINE
- JUMPER_IS_SCALAR, JUMPER_IS_NEON, JUMPER_IS_AVX2, etc.
The upshot is that we'll now use Clang vector extensions when available
for our "portable" baseline. On x86-64 and ARMv8 compiled by Clang,
we're guaranteed to pick up SSE2 and NEON respectively. Our -Fast
bot should even get all the way to AVX2.
Another CL will do some refactoring in SkJumper to remove the redundant
copies of guaranteed vector code on x86-64 and ARMv8. I didn't want to
do that here yet to demonstrate that there is zero effect on the .S
files from this CL.
Change-Id: Ib5e8f00b35e8721b2cc7180e294840ffaf9dddce
Reviewed-on: https://skia-review.googlesource.com/39500
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Enough ccpr-specific geometry code is in flight that it feels like it
should have its own file.
Bug: skia:
Change-Id: I99ef620a7dc35178cf774b3a4ec6159d46f401c7
Reviewed-on: https://skia-review.googlesource.com/39162
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Motivated by wanting to speed-up A8 blits in general (and at the moment, aarect blits). More to come in these areas.
Bug: skia:
Change-Id: I45e8ef951b8e89a825af72b1918049be10920137
Reviewed-on: https://skia-review.googlesource.com/39401
Reviewed-by: Yuqian Li <liyuqian@google.com>
Commit-Queue: Mike Reed <reed@google.com>
