__AVX2__ will not be defined if you omit -mavx2. Android does this
intentionally for x86 builds. (No mobile CPU supports AVX2 AFAIK.)
This should fix the Android roll.
Change-Id: Ib94c862641abc11fbb46863afc53bcc049f362ad
Reviewed-on: https://skia-review.googlesource.com/7633
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Reviewed-by: Matt Sarett <msarett@google.com>
Some of my local changes slipped into the tree.
BUG=skia:
Change-Id: I6e452f12608db6b80c5887ccd3cac199cfbf9494
Reviewed-on: https://skia-review.googlesource.com/7625
Reviewed-by: Matt Sarett <msarett@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Matt Sarett <msarett@google.com>
These new names reflect the actual pixels stored in each register.
Change-Id: I8e626196cd8bcbef622e4fb87ac3566a79d3573a
Reviewed-on: https://skia-review.googlesource.com/7624
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Matt Sarett <msarett@google.com>
Like *sse2*, *avx*, etc, these are x86-specific and
even there must be compiled with particular flags.
Change-Id: Ibbc004b9ba5da6da70449177423a785d2c644846
Reviewed-on: https://skia-review.googlesource.com/7623
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Added by humper in 2013, it is not
referenced anywhere I can find.
R=reed@google.com
Change-Id: I35b3891b29b2fddb132647b5a1e337a0fc68e79e
Reviewed-on: https://skia-review.googlesource.com/7610
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Cary Clark <caryclark@google.com>
I'm warming back up to the idea of very careful use of SkOpts_hsw.
But if we're going to do that, we need a strict header discipline.
No header can be assumed to be safe without vetting, and most aren't.
Today there's only one function defined in SkOpts_hsw, so this CL mostly
rewrites that convolve_vertically() to use no headers beyond immintrin.h
and stdint.h, both safe. It shared very little code with the others
anyway, so we're not losing anything by putting it directly into
SkOpts_hsw.cpp. I have also streamlined the implementation considerably
to improve maintainability and readability.
Change-Id: Ia03daae660e54125a0d2e2988464cfc930349e80
Reviewed-on: https://skia-review.googlesource.com/7611
Reviewed-by: Matt Sarett <msarett@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Change-Id: Ic285b24b384bbf284cc680fe770433dd4d643833
Reviewed-on: https://skia-review.googlesource.com/7561
Reviewed-by: Robert Phillips <robertphillips@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This was found by a tip of tree build of clang.
Change-Id: I9aa3fecd47ae69350e8d58e796f960a991b7759e
Reviewed-on: https://skia-review.googlesource.com/7586
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Change-Id: I04bcaf91aa7a43e0563e332c1fe2836d762a04d4
Reviewed-on: https://skia-review.googlesource.com/7520
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
BUG=skia:6119
Change-Id: I9756b5aeaced1f21a65063470ccb013c7b856f28
Reviewed-on: https://skia-review.googlesource.com/7505
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
BUG=skia:
Change-Id: If10fe8b3691c8a71b52774f763a08120ef20dfd8
Reviewed-on: https://skia-review.googlesource.com/7502
Reviewed-by: Ravi Mistry <rmistry@google.com>
Commit-Queue: Eric Boren <borenet@google.com>
This reverts commit 8ccbbb455c.
Reason for revert: Chrome perf regression
BUG=chrome:684483
Original change's description:
> Remove SkImage::MakeTextureFromPixmap
>
> Another step towards reducing the number of texture upload paths.
>
> BUG=skia:
>
> Change-Id: Ica185e7334f52dc9ebf87e21fe6f60589ef87bb3
> Reviewed-on: https://skia-review.googlesource.com/7346
> Commit-Queue: Brian Osman <brianosman@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
>
TBR=bsalomon@google.com,robertphillips@google.com,brianosman@google.com,cblume@google.com
# Not skipping CQ checks because original CL landed > 1 day ago.
BUG=skia:
Change-Id: I8bf6970b764018f3ebcac8123e4d6a8ad05317ff
Reviewed-on: https://skia-review.googlesource.com/7484
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This reverts commit fdb1bdf1aa.
Reason for revert: Breaking the roll (see e.g. https://codereview.chromium.org/2652833006)
Original change's description:
> Add geometric implementation for ambient shadows
>
> BUG=skia:6119
>
> Change-Id: I3140522f223c35fc059a33b593064897485dff7c
> Reviewed-on: https://skia-review.googlesource.com/7273
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Jim Van Verth <jvanverth@google.com>
>
TBR=jvanverth@google.com,bsalomon@google.com,robertphillips@google.com,reviews@skia.org,reed@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:6119
Change-Id: I307f6084c40e5ba305c38d15b6b433465295553c
Reviewed-on: https://skia-review.googlesource.com/7480
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
BUG=skia:6119
Change-Id: I3140522f223c35fc059a33b593064897485dff7c
Reviewed-on: https://skia-review.googlesource.com/7273
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
This reverts commit bfd97c5935.
Reason for revert: Windows bots crashing.
Original change's description:
> Remove sentinal GLContext used in early days of Vulkan in our tools
>
> I've ran this through multiple variations of dm and nanobench on Linux
> and no longer see any crash/hangs. I forget what the original repo case
> for the bug was, but I'm at least not seeing it now with updated drivers.
>
> BUG=skia:
>
> Change-Id: I6b7129a4c4d67938baa35d2e2c720cb078fc4c18
> Reviewed-on: https://skia-review.googlesource.com/7441
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Greg Daniel <egdaniel@google.com>
>
TBR=egdaniel@google.com,bsalomon@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Change-Id: I94a16b2b13d09182232061b0aebd0d58df96db8a
Reviewed-on: https://skia-review.googlesource.com/7453
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
R=mtklein@google.com
BUG=chromium:684061
Change-Id: I210ca93e00ec6fa5d166c0fde5cdcff0f60c26ca
Reviewed-on: https://skia-review.googlesource.com/7450
Commit-Queue: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>
I've ran this through multiple variations of dm and nanobench on Linux
and no longer see any crash/hangs. I forget what the original repo case
for the bug was, but I'm at least not seeing it now with updated drivers.
BUG=skia:
Change-Id: I6b7129a4c4d67938baa35d2e2c720cb078fc4c18
Reviewed-on: https://skia-review.googlesource.com/7441
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
I envision this eventually looking like the following:
Fast Path 1: memcpy
Fast Path 2: Legacy RGBA/BGRA unpremul/premul
Fast Path 3: SkColorSpaceXform
Fallback: Raster pipeline
BUG=skia:6021
Change-Id: I3f0e36b08b051ab4ba3949f04474f20d0c20a963
Reviewed-on: https://skia-review.googlesource.com/7435
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Matt Sarett <msarett@google.com>
This helps pick it out on profiles and when debugging.
This will have a negligible impact on speed in most cases (small memsets are always fast, large memsets dwarf function call overhead), and clients like Chrome that use link-time optimization should see exactly zero impact.
This does however affect code size: it makes libskia 43K smaller on my laptop.
Change-Id: I55bdc75b7429a63024ab3b7d55dc1f886bde102a
Reviewed-on: https://skia-review.googlesource.com/6589
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Herb Derby <herb@google.com>
Instead of allocating memory for glyph images eagerly by embedding
the memory in the glyph cache, allocate memory dynamically on need.
TBR=bungeman@google.com
BUG=chromium:684366
Change-Id: If32bbc4d2608c976b93868feb519dcfa1212ce59
Reviewed-on: https://skia-review.googlesource.com/7433
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Herb Derby <herb@google.com>
This is causing problems with image-cacherator-from-ctable gm.
In that gm, we use an index8->index8 readPixels() to take
a subset.
I think we need to keep this as long as we support index8 in the
generator/cacherator. Though it's really strange because
we don't really bother to verify that color tables are the same
and we don't try to overwrite the dst color table...
BUG=skia:
Change-Id: Icf008caa571c4b81a52a23b8b7d9226789626ef5
Reviewed-on: https://skia-review.googlesource.com/7424
Commit-Queue: Matt Sarett <msarett@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This opens the door for swapping all the effects over to taking GrTextureProxies.
Change-Id: I3b03ba93a68f9945c9a8fee008fd170ed57616eb
Reviewed-on: https://skia-review.googlesource.com/7344
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Hope that's right... nothing but the Android roll builds this code.
Change-Id: Ib86099deaaa5a67a480d6dc0eece8552e58aae1c
Reviewed-on: https://skia-review.googlesource.com/7428
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
