That GM has a path with 16k points, arranged in a rectangle. When we try
to draw it in Ganesh, we first check to see if it's a (closed) rectangle.
We eventually decide that it is a rectangle, but not closed, so it gets
passed along. Then we construct a GrShape, which tries to simplify the
shape, again asking if it's a rectangle...
Each isRect query was iterating over the 16k verbs, and for each one,
calling atVerb(i). That, internally, calls verbs(), which calls validate()
in debug builds. So we were walking all 16k points 16k times (to ensure
the bounds were correct). The end result is that the GM took over 11
seconds to draw, and now takes 3 ms.
Bug: skia:
Change-Id: If5f7a067b8c25f049dc64275d94a42ae4a50f6a9
Reviewed-on: https://skia-review.googlesource.com/34723
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
The original code had an optimization when sigma < 2 that averaged
pixels instead of Gaussian bluring them. This CL adds that
behavior back to the new implementation.
Add the flag SK_LEGACY_USE_GAUSS_FOR_SMALL_RADII to control the
new behavior.
BUG=chromium:745290
Change-Id: I6e7417ce7f42949f88376b549743a9f047667f09
Reviewed-on: https://skia-review.googlesource.com/34624
Commit-Queue: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Bug: skia:
Change-Id: I3f3c5f9df43e48f975b09f0bafadf03a15afec51
Reviewed-on: https://skia-review.googlesource.com/34745
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Some of these are pretty trivial, but a few really moved the needle. In
total, this cut the time in the parser by about 10% in my profiling.
The early returns from peek and checkNext are probably the biggest wins,
everything else is mostly about cutting down on string copying.
Bug: skia:
Change-Id: I3da0a0cfbc557e8cef480cb3a7d301c2b3f8802a
Reviewed-on: https://skia-review.googlesource.com/34720
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This reverts commit f6975ea5fb.
Reason for revert:
Did not take into account there are layout test problems too. Adding a legacy flag.
Original change's description:
> Revert "Revert "Add an average mode for sigma < 2""
>
> We are going to have the google3 auto roller build the CL,
> and add the fixes to the failing tests to that CL. Because
> these are scuba changes, this should be simpler.
>
> This reverts commit 8e7503195c.
>
> Reason for revert: <INSERT REASONING HERE>
>
> Original change's description:
> > Revert "Add an average mode for sigma < 2"
> >
> > This reverts commit ba8275148a.
> >
> > Reason for revert: layout tests, Google3 equivalent failing.
> >
> > Original change's description:
> > > Add an average mode for sigma < 2
> > >
> > > The original code had an optimization when sigma < 2 that averaged
> > > pixels instead of Gaussian bluring them. This CL adds that
> > > behavior back to the new implementat.
> > >
> > > BUG=chromium:745290
> > > Change-Id: I35b7de2138a859d546439cc2053b4b599a94ebe1
> > > Reviewed-on: https://skia-review.googlesource.com/34180
> > > Commit-Queue: Herb Derby <herb@google.com>
> > > Reviewed-by: Mike Klein <mtklein@chromium.org>
> >
> > TBR=mtklein@chromium.org,mtklein@google.com,herb@google.com
> >
> > Change-Id: Ie8f38b042acec42260913cb7bed5538a2c957e9b
> > No-Presubmit: true
> > No-Tree-Checks: true
> > No-Try: true
> > Bug: chromium:745290
> > Reviewed-on: https://skia-review.googlesource.com/34640
> > Reviewed-by: Mike Klein <mtklein@google.com>
> > Commit-Queue: Mike Klein <mtklein@google.com>
>
> TBR=mtklein@chromium.org,mtklein@google.com,herb@google.com
>
> Change-Id: I5f7b706a2f8c8d52b3576c638ecddf12a1652031
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: chromium:745290
> Reviewed-on: https://skia-review.googlesource.com/34641
> Reviewed-by: Herb Derby <herb@google.com>
> Commit-Queue: Herb Derby <herb@google.com>
TBR=mtklein@chromium.org,mtklein@google.com,herb@google.com
Change-Id: I9064f8494e5a1a4a78dd545697e22f52985c0fa1
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:745290
Reviewed-on: https://skia-review.googlesource.com/34623
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Herb Derby <herb@google.com>
We are going to have the google3 auto roller build the CL,
and add the fixes to the failing tests to that CL. Because
these are scuba changes, this should be simpler.
This reverts commit 8e7503195c.
Reason for revert: <INSERT REASONING HERE>
Original change's description:
> Revert "Add an average mode for sigma < 2"
>
> This reverts commit ba8275148a.
>
> Reason for revert: layout tests, Google3 equivalent failing.
>
> Original change's description:
> > Add an average mode for sigma < 2
> >
> > The original code had an optimization when sigma < 2 that averaged
> > pixels instead of Gaussian bluring them. This CL adds that
> > behavior back to the new implementat.
> >
> > BUG=chromium:745290
> > Change-Id: I35b7de2138a859d546439cc2053b4b599a94ebe1
> > Reviewed-on: https://skia-review.googlesource.com/34180
> > Commit-Queue: Herb Derby <herb@google.com>
> > Reviewed-by: Mike Klein <mtklein@chromium.org>
>
> TBR=mtklein@chromium.org,mtklein@google.com,herb@google.com
>
> Change-Id: Ie8f38b042acec42260913cb7bed5538a2c957e9b
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: chromium:745290
> Reviewed-on: https://skia-review.googlesource.com/34640
> Reviewed-by: Mike Klein <mtklein@google.com>
> Commit-Queue: Mike Klein <mtklein@google.com>
TBR=mtklein@chromium.org,mtklein@google.com,herb@google.com
Change-Id: I5f7b706a2f8c8d52b3576c638ecddf12a1652031
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:745290
Reviewed-on: https://skia-review.googlesource.com/34641
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Herb Derby <herb@google.com>
Bug: skia:
Change-Id: Ifbccc6d6880522177ac7a0ae2183be64a3ebfe50
Reviewed-on: https://skia-review.googlesource.com/34681
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
This reverts commit ba8275148a.
Reason for revert: layout tests, Google3 equivalent failing.
Original change's description:
> Add an average mode for sigma < 2
>
> The original code had an optimization when sigma < 2 that averaged
> pixels instead of Gaussian bluring them. This CL adds that
> behavior back to the new implementat.
>
> BUG=chromium:745290
> Change-Id: I35b7de2138a859d546439cc2053b4b599a94ebe1
> Reviewed-on: https://skia-review.googlesource.com/34180
> Commit-Queue: Herb Derby <herb@google.com>
> Reviewed-by: Mike Klein <mtklein@chromium.org>
TBR=mtklein@chromium.org,mtklein@google.com,herb@google.com
Change-Id: Ie8f38b042acec42260913cb7bed5538a2c957e9b
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:745290
Reviewed-on: https://skia-review.googlesource.com/34640
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Bug: skia:
Change-Id: I65cd5761bb01e648db706878b902e6b6ed67fab9
Reviewed-on: https://skia-review.googlesource.com/33761
Reviewed-by: Eric Boren <borenet@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
On desktop, this saves just over 5% of the time in the SkSL compiler.
As written, the code will now build either way, so it's much easier to
switch back (or even have some platforms use SkString, if that's ever
required).
Bug: skia:
Change-Id: I634f26a4f6fcb404e59bda6a5c6a21a9c6d73c0b
Reviewed-on: https://skia-review.googlesource.com/34381
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
The original code had an optimization when sigma < 2 that averaged
pixels instead of Gaussian bluring them. This CL adds that
behavior back to the new implementat.
BUG=chromium:745290
Change-Id: I35b7de2138a859d546439cc2053b4b599a94ebe1
Reviewed-on: https://skia-review.googlesource.com/34180
Commit-Queue: Herb Derby <herb@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>
BUG=chromium:745290
Change-Id: I78cabf988115598e14a7ce39faf96e3a697a1a8f
Reviewed-on: https://skia-review.googlesource.com/34382
Commit-Queue: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>
This reduces jank when flinging a page of emoji on Android (I have separately shared out a spreadsheet). Arguably, the quality loss should be acceptable since we shouldn't be down-scaling that much (magnitude-wise not frequency-wise). Ideally, we would add a raster mode where we would draw at the correct size w/o generating the mip maps.
Change-Id: Id91d638db1b2457567e54c264ed3fa5d10316976
Reviewed-on: https://skia-review.googlesource.com/33763
Commit-Queue: Robert Phillips <robertphillips@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Change-Id: I88f3e56971e9844ab2ff74edb0718e6b6e9c6559
Reviewed-on: https://skia-review.googlesource.com/34260
Reviewed-by: Mike Klein <mtklein@chromium.org>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Bug: skia:
Change-Id: Iab823b01b6f5f85bbb47c4ac69c6ce396dabf497
Reviewed-on: https://skia-review.googlesource.com/34000
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
We can record multiple frames in an .skp by recording SkCanvas::flush().
This should make SkPictures, SkLiteDL, and .skp files all record flush().
Change-Id: I6cf6e0e4ef993530d9f92fa168a53702ffce7d5e
Reviewed-on: https://skia-review.googlesource.com/34081
Reviewed-by: Derek Sollenberger <djsollen@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
No-Try: true
Change-Id: I41dfddf21f8ed80582d5dc88150e1fbaf98326e6
Reviewed-on: https://skia-review.googlesource.com/33803
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ben Wagner <benjaminwagner@google.com>
Chromium uses the lowp code, we have to stage the changes.
TBR=
Change-Id: I45e97a51eca285c9afc71926bbf736a03d0d146c
Reviewed-on: https://skia-review.googlesource.com/33765
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
We use SK_SUPPORT_LEGACY_DELTA_AA to guard the golden image change.
Such flag is defined for Android, Chrome, and Google3 so our auto-rollers
should all be OK.
TBR: bungeman@google.com
Bug: skia:6947
Change-Id: Ic2705e82f4f7f15ec08499254dce75b93d41727e
Reviewed-on: https://skia-review.googlesource.com/33762
Reviewed-by: Yuqian Li <liyuqian@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
Do multiply (mul) and add while tracking that the
calculation does not overflow, which can be checked with
ok().
The new unit test shows a couple examples.
Author: Herb Derby <herb@google.com>
Change-Id: I7e67671d2488d67f21d47d9618736a6bae8f23c3
Reviewed-on: https://skia-review.googlesource.com/33721
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Herb Derby <herb@google.com>
We only need to clip a temporary x to ensure that we don't blit
beyond clip. Storing such clipped x is problematic because it
may make our edges unsorted.
The added unit test would fail without this fix.
Bug: skia:6947
Change-Id: I6c21d7c7c097e50fef18ab151921d6c07c089318
Reviewed-on: https://skia-review.googlesource.com/33420
Commit-Queue: Yuqian Li <liyuqian@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
I'm betting big on ok bench. This is a forcing function.
Change-Id: I8c359b7d712e16f8f0cbb90591801e0014073288
Reviewed-on: https://skia-review.googlesource.com/33660
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
log2 didn't exist on Android until API 18, and it's doing double
precision math. I think this integer version is what we want?
Bug: skia:
Change-Id: I4909153c56a266688355349cda5d553b69f5c942
Reviewed-on: https://skia-review.googlesource.com/33680
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This basically unrolls all loops, handling twice as many pixels in a
stride. We now pass around 4 native registers instead of just 2.
I've temporarily disabled AVX2 mask loads and stores. It shouldn't be
hard to turn them back on, but I'd want to test on AVX2 hardware first.
Change-Id: I0907070f086a0650167456c149a479c1d96b8a2d
Reviewed-on: https://skia-review.googlesource.com/33361
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
