This first tries to JIT while hoisting all constants,
and if that fails, tries again hoisting no constants.
I figure this is one of those 80/20 deals for how to
handle constant hoisting and register pressure. This
probably mostly moots doing anything fancy like using
memory operands with AVX or lane operands with NEON.
This _doesn't_ moot hoisting the NEON tbl arguments,
which is not yet done here, but probably my next CL.
Change-Id: Id09d5cdddcdb45207bdfc914a5a3128a481a26f3
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/229058
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Even if a JIT ultimately doesn't end up hoisting any values, it's going
to want this information while it decides. Writing it in one place also
ensures we only get it wrong in one place...
I'm no_ extending the lifetime of hoisted instructions here in Builder.
That's something to leave to the backend so they have the flexibility of
which of these values to hoist, if any. If they don't hoist, they'll
need to know when the value dies.
Moving this information back here lets the test expectation goldens
reflect the hoist bit again too. Kind of nice.
Change-Id: Ib165ca898a97c1d822cb28fe24f15bae4d570a17
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/229024
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
When Chrome has a LUM16F texture they tell Skia it is R16F. Although this has been working for them so far it causes trouble with some upcoming changes.
Change-Id: I2473f70e4f725128f143c2dfb08adb79f3c7c166
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228565
Reviewed-by: Brian Salomon <bsalomon@google.com>
Adds the option to use a multisampled (or mixed sampled) atlas, and
uses the sample mask and stencil buffer instead of coverage counts.
Bug: skia:
Change-Id: I9fb76d17895ae25208124f6c27e37977ac31b5eb
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227428
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
We can always move data around so that an FMA is possible using no more
registers than we would otherwise, and on x86, evne using no more
instructions.
The basic idea here is that if we can't reuse one of the inputs to
destructively host the FMA instruction, the next best thing is to copy
one of the arguments into tmp() and accumulate the FMA there.
Once the FMA has happened, we just need to copy that result to dst().
We can of course skip that copy if dst() == tmp(). On x86 we never need
that copy; dst() and tmp() are picked using the same logic except that
dst may alias one of its inputs, and we only fall into this case after
we've already found it doesn't. So we can just assert dst() == tmp()
rather than check it like we do on ARM.
It's subtle, but I think sound.
I'm using logical-or to copy registers around. This is a little lazy,
but maybe not as lazy as it looks: on ARM that is _the_ way to copy
registers. There's a vmovdqa instruction I could use on x86, TBD.
All paths through this new code were being exercised on ARM, but we
didn't have anything hitting the tmp case on x86, so I've added a new
unit test that hits the corner cases of both implementations.
Change-Id: I5422414fc50c64d491b4933b4b580b784596f291
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228630
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
When SkBaseDevice switched to drawGlyphRunList(), we lost the ability to
detect a) constant-Y text and b) default-positioned text.
As a result, the emitted SVG contains lots of redundant/repeating glyph
positions.
This CL enhances SVGTextBuilder to detect and consolidate constant-Y
glyph positions.
Also restore a useful whitespace unit test.
Change-Id: I50568aef1955f75898ebab41441ad5fe418dac43
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228563
Commit-Queue: Florin Malita <fmalita@chromium.org>
Reviewed-by: Mike Reed <reed@google.com>
We have pack(x,y,imm) = x | (y<<imm) assuming (x & (y<<imm)) == 0.
If we can destroy x, sli (shift-left-insert) lets us implement that
as x |= y << imm. This happens quite often, so you'll see sequences
of pack that used to look like this
shl v4.4s, v2.4s, #8
orr v1.16b, v4.16b, v1.16b
shl v2.4s, v0.4s, #8
orr v0.16b, v2.16b, v3.16b
shl v2.4s, v0.4s, #16
orr v0.16b, v2.16b, v1.16b
now look like this
sli v1.4s, v2.4s, #8
sli v3.4s, v0.4s, #8
sli v1.4s, v3.4s, #16
We can do this thanks to the new simultaneous register assignment
and instruction selection I added. We used to never hit this case.
Change-Id: I75fa3defc1afd38779b3993887ca302a0885c5b1
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228611
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Trying to keep most of the structural parts shared between x86_64 and
aarch64. Not sure if this will stay factored like this long-term, but
the last version felt like there was a bit too much redundancy, and I
don't want to write things like register management more often than have
to.
Change-Id: Ieeb21f433715a730c41c85d657c5b33fa4702696
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228608
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
While most callers of decomposeScale only depend on the scale factors
that are returned, image filter decomposition depends on the remaining
matrix as well. Hence, the following necessary updates to work with
the new behavior of decomposeScale:
* Update imagefilter DAG sample to reflect prescaling
* Correct embedded matrix in SkApplyCTMToFilter
* Add comment to clipRectBounds() clarifying coordinate spaces and image filters
But, we want to have decomposeScale using preScale() because it then
produces a remainder matrix that can be used as the transform for the
image filter draw, instead of wrapping the image filter in an
SkMatrixImageFilter as currently done by SkApplyCTMToFilter.
Bug: skia:7211
Change-Id: If14570afb4189cebc75f3815e8ccdde05cb074e1
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228438
Reviewed-by: Mike Reed <reed@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
- get rid of variadic Assembler::byte()... not used very often
- rename Assembler::byte(ptr,n) to bytes()
- align with 0 bytes, get rid of nop()
Change-Id: I7564d3bad00e3f0d1c7a80153c445966914fccf0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228601
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This reverts commit 2f6c8af73a.
Reason for revert: Failing on some bots
Original change's description:
> Add Ganesh support for LUM16F
>
> When Chrome has a LUM16F texture they tell Skia it is R16F. Although this has been working for them so far it causes trouble with some upcoming changes.
>
> Change-Id: Ic2143ec69f33a17cb1cb64f6ebc39bffd94a5e68
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228557
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Robert Phillips <robertphillips@google.com>
TBR=bsalomon@google.com,robertphillips@google.com
Change-Id: Ie52150f4f5255a2db697b9f0e810eff2042eadec
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228560
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
When Chrome has a LUM16F texture they tell Skia it is R16F. Although this has been working for them so far it causes trouble with some upcoming changes.
Change-Id: Ic2143ec69f33a17cb1cb64f6ebc39bffd94a5e68
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228557
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Bug: skia:
Change-Id: I1c801a26727b72f36d76e1a1c21cd0e571107f8c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228558
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
This is a reland of 558b639225
PS2... oh, right, not everything supports AVX2.
Original change's description:
> more JIT refactoring
>
> This re-enables AVX2 JIT with simultaneous register assignment and
> instruction selection. You can see it working in a very basic way in
> how we choose instructions and registers for Op::mad_f32.
>
> Constants are still broadcast, here inside the loop instead of hoisted.
> I think it'll probably end up best to use constants directly from memory
> (as in vpshufb's masks), falling back to these in-loop broadcasts when
> that can't work.
>
> Change-Id: If17d51b9960f08da3612e51ac04424e996bf83d4
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228366
> Commit-Queue: Mike Klein <mtklein@google.com>
> Reviewed-by: Mike Klein <mtklein@google.com>
Cq-Include-Trybots: skia.primary:Test-Mac10.13-Clang-VMware7.1-CPU-AVX-x86_64-Debug-All-NativeFonts
Change-Id: I6f99d275040abe6210a980fc544f7f22c3b85727
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228476
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This reverts commit 558b639225.
Reason for revert: broke some perf bots
Original change's description:
> more JIT refactoring
>
> This re-enables AVX2 JIT with simultaneous register assignment and
> instruction selection. You can see it working in a very basic way in
> how we choose instructions and registers for Op::mad_f32.
>
> Constants are still broadcast, here inside the loop instead of hoisted.
> I think it'll probably end up best to use constants directly from memory
> (as in vpshufb's masks), falling back to these in-loop broadcasts when
> that can't work.
>
> Change-Id: If17d51b9960f08da3612e51ac04424e996bf83d4
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228366
> Commit-Queue: Mike Klein <mtklein@google.com>
> Reviewed-by: Mike Klein <mtklein@google.com>
TBR=mtklein@google.com
Change-Id: Id6cd5acd873499bb394009489d77e7636ecbc9c6
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228462
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This re-enables AVX2 JIT with simultaneous register assignment and
instruction selection. You can see it working in a very basic way in
how we choose instructions and registers for Op::mad_f32.
Constants are still broadcast, here inside the loop instead of hoisted.
I think it'll probably end up best to use constants directly from memory
(as in vpshufb's masks), falling back to these in-loop broadcasts when
that can't work.
Change-Id: If17d51b9960f08da3612e51ac04424e996bf83d4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228366
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
This is a reland of f42de9e1e5
Original change's description:
> Interpreter: Bounds check array access, add bool return from run
>
> Out of bounds access with constant indices is a compile error.
> At runtime, causes the interpreter to fail. Made several other
> conditions trigger the same failure logic, and updated all
> uses of the interpreter to validate success.
>
> Change-Id: I3720b3c83903220b010ec574121fc64dbe102378
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228256
> Commit-Queue: Brian Osman <brianosman@google.com>
> Reviewed-by: Mike Reed <reed@google.com>
Change-Id: I8849de815f7efb730ac9c55b6edd296cb9ca7599
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228353
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
With this change I also removed GrPixelConfig as param to supportedReadPixelsColorType().
This meant some updates had to be made to Vulkan and Metal to make sure they return the
right GrColorType.
Bug: skia:6718
Change-Id: I71b6360489cf499692c7b777e5915090fad05c56
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228349
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Change-Id: Ic30f6e0345e851ea8a942996b9eaf2c894455e3e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228236
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Mike Reed <reed@google.com>
I find myself passing around parallel vectors of Builder::Instructions
and deaths so often that it just makes more sense practically to store
them together. It's a little awkward that the values are only useful
after calling done(), but I can live with that.
Get a little more careful about mutation, passing Builder::Instructions
by const&. Instead of extending lifetimes of live hoisted
instructions, just check for them in maybe_recycle_register() instead.
Change-Id: I1cb9e25c1a7c46a250c2271334821be8535353bf
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228367
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Add a test case that previously asserted (due to incorrect code-gen
underflowing the stack).
Change-Id: I9df7a08b9ac5c7b5bc246129f3383dc723173351
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228351
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This reverts commit f42de9e1e5.
Reason for revert: All the SANs
Original change's description:
> Interpreter: Bounds check array access, add bool return from run
>
> Out of bounds access with constant indices is a compile error.
> At runtime, causes the interpreter to fail. Made several other
> conditions trigger the same failure logic, and updated all
> uses of the interpreter to validate success.
>
> Change-Id: I3720b3c83903220b010ec574121fc64dbe102378
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228256
> Commit-Queue: Brian Osman <brianosman@google.com>
> Reviewed-by: Mike Reed <reed@google.com>
TBR=mtklein@google.com,brianosman@google.com,ethannicholas@google.com,reed@google.com
Change-Id: I434601960d54fbd7d00e2af2dc6269a83a768c5b
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228352
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Out of bounds access with constant indices is a compile error.
At runtime, causes the interpreter to fail. Made several other
conditions trigger the same failure logic, and updated all
uses of the interpreter to validate success.
Change-Id: I3720b3c83903220b010ec574121fc64dbe102378
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228256
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Mike Reed <reed@google.com>
This reverts commit 9725638fb1.
Reason for revert: Chrome roll
Original change's description:
> Make rest of GrGpu::wrapBackend* methods take a GrColorType
>
> This CL is intended to further wean Ganesh off of using the GrBackendTexture's pixel config
>
> Bug: skia:6718
> Change-Id: I593c0c73922fb76045e379214e20adb1f17ea215
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227780
> Commit-Queue: Robert Phillips <robertphillips@google.com>
> Reviewed-by: Greg Daniel <egdaniel@google.com>
TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com
Change-Id: Id71acf1dec63c288a858fccd7109c84cf3cc6f0a
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:6718
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228337
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This CL is intended to further wean Ganesh off of using the GrBackendTexture's pixel config
Bug: skia:6718
Change-Id: I593c0c73922fb76045e379214e20adb1f17ea215
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227780
Commit-Queue: Robert Phillips <robertphillips@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
I'm slowly refactoring my way to where hoisting and register assignment
are done in backend-specific ways, but this liveness analysis is always
going to be useful for each backend.
Use deaths() to restore friendly ☠️ dead code markers in test dumps.
Change-Id: I3ab94665bbbbf0788b0b27e00d644eba927dff47
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228113
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Change-Id: I7fd3a8f6c02217d011f353ef602718a537bb87b2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228116
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
This is the last surface desc flag, so remove flags from GrSurfaceDesc.
Bug: skia:6718
Change-Id: Id3ed66b161289927b62f40bfb1f6482cf544deda
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227858
Reviewed-by: Robert Phillips <robertphillips@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Add Program::dropJIT() to allow us to proactively drop
any JIT code forcing fallback on the interpreter,
and use it to test both on JIT-supported platforms.
Other platforms will just test the interpreter twice.
Change-Id: I607d00ef3c648e66a0b3a1374b11aa82dbfff70c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227424
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Herb Derby <herb@google.com>
Arg strides are the reason JIT happens lazily in Program::eval() today
instead of proactively in Builder::done() or Program's constructor. It
also just really doesn't make sense to delay this information... it's
not like you can change it up sanely between calls to eval().
The argument index now comes implicitly from the order of calling arg().
This may seem logically independent, but it prevents a weird situation
where you could use the same argument index twice with different
strides... not sure what that would mean.
Change-Id: I0f5d46e94a1ca112a72675c5492f17c0dd825ce0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227390
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Removes some usage of GrSurfaceDesc.
Bug: skia:6718
Change-Id: Icc4f93aba0e5c49a801b4c7bbfcba76a6e30c538
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227776
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
1) It only applies when a texture is created, not when recycled from cache
2) It is all textures or none, not a flag GrSurfaceDesc
3) It is implemented by GrGpu clearing the texture after creation if
such a thing is supported in underlying API. Otherwise, GrResourceProvider
must provide pre-zeroed mip levels.
4) Works for MIP mapped textures (all levels without initial data are cleared)
This could cause performance regressions in WebGL until we re-add the
ability to clear using glCear() in GL. Doing that requires making the "can
clear using GrGpu" caps query be per-format. Deferring doing that until
GrPixelConfig work is farther along.
Bug: skia:6718
Change-Id: I234715b9faaf61e8b44d54464497a17cd553585d
start
Change-Id: Ib84a8c3ece010cc3164b18895107e78484cbf76b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/226977
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Make them query caps for supported read/write info and do CPU
conversions before uploading/after reading.
Removes use of GrColor so in theory could be used to test
non-8888 color types.
Bug: skia:6718
Change-Id: Icf9d0b778348a4e960fbfec49e1308b21e45a051
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227497
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This is a step towards reducing our reliance-on/use-of the GrPixelConfig stored in the GrBackendTexture.
TBR=egdaniel@google.com
Bug: skia:6718
Change-Id: I316a98416c51f273e6ab578f9cbaea5f7adfe331
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227639
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This reverts commit 5572737d95.
Reason for revert: Adreno 4xx bots hitting compatibility assert in InitialTextureClear test
Original change's description:
> Pass GrColorType to the GrGpu::wrapRenderableBackendTexture chain of calls
>
> This is a step towards reducing our reliance-on/use-of the GrPixelConfig stored in the GrBackendTexture.
>
> Bug: skia:6718
> Change-Id: I2170032bfbbb57423c4bb0d901ad014c61d38131
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/223701
> Commit-Queue: Robert Phillips <robertphillips@google.com>
> Reviewed-by: Greg Daniel <egdaniel@google.com>
TBR=egdaniel@google.com,robertphillips@google.com
Change-Id: I24cf6b3aa0dfca2e935a36592860ad91171b21a7
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:6718
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227637
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
This is a step towards reducing our reliance-on/use-of the GrPixelConfig stored in the GrBackendTexture.
Bug: skia:6718
Change-Id: I2170032bfbbb57423c4bb0d901ad014c61d38131
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/223701
Commit-Queue: Robert Phillips <robertphillips@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
Change-Id: If581c8ceeaa76985535cb7b6772742f0011cfe8e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227436
Commit-Queue: Hal Canary <halcanary@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
Auto-Submit: Hal Canary <halcanary@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
