Update skia to milestone 117 #138
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Change-Id: I4e2d6464dee9eb6e65605f282742e61fb324d277 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/727700 Reviewed-by: Nicolette Prevost <[email protected]> Commit-Queue: Michael Ludwig <[email protected]>
In std140 layout, float arrays are very space-inefficient. Change-Id: I397504654a6d83b2e54bf46f660c6c8b2dddb740 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/726948 Reviewed-by: Robert Phillips <[email protected]> Reviewed-by: Michael Ludwig <[email protected]> Auto-Submit: John Stiles <[email protected]> Commit-Queue: John Stiles <[email protected]>
This makes most modifier-flag checks a lot easier to parse. I've intentionally omitted accessors for the in/out bits for clarity (should modifiers.isOut() return true for an inout?). Change-Id: I8fbf5abea9978d06572a9c88600b1d797e12fcb2 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/727916 Auto-Submit: John Stiles <[email protected]> Commit-Queue: James Godfrey-Kittle <[email protected]> Reviewed-by: James Godfrey-Kittle <[email protected]>
Change-Id: I5e7ed806b7467f1df9e3ce4e66175f7a682312f0 Bug: skia:14317 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728697 Reviewed-by: Michael Ludwig <[email protected]> Commit-Queue: Kevin Lubick <[email protected]>
This was not actually used in any public headers. Change-Id: Ica1a2fd9d31e2377f5a673cf8596edf6bc965f2c Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728796 Reviewed-by: Michael Ludwig <[email protected]> Auto-Submit: John Stiles <[email protected]> Commit-Queue: John Stiles <[email protected]> Reviewed-by: Kevin Lubick <[email protected]>
Change-Id: Ie6969d47263258a5de48019c2dfc419a629a6a89 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/726056 Reviewed-by: Jim Van Verth <[email protected]> Commit-Queue: Nicolette Prevost <[email protected]>
This puts it on equal footing with its rival, SkBitmaskEnum.h. Change-Id: I0e976988db1d43b7422957733f95f6b6824b7121 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728699 Commit-Queue: John Stiles <[email protected]> Auto-Submit: John Stiles <[email protected]> Commit-Queue: Kevin Lubick <[email protected]> Reviewed-by: Kevin Lubick <[email protected]>
This took more rearranging than I expected mainly bc the original compute_size helper needed Caps and the Graphite Image doesn't carry them around. To achieve the end goal I had to: Move compute_size to TextureUtils.h (renamed to ComputeSize) Add TextureInfo::bytesPerPixel (and use it in ComputeSize) Add TextureProxy::uninstantiatedGpuMemorySize Change-Id: Ifb284586f37a5d22dd51031bc96f1f83b28df997 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/727276 Reviewed-by: Brian Osman <[email protected]> Reviewed-by: Jim Van Verth <[email protected]> Commit-Queue: Robert Phillips <[email protected]>
Change-Id: I093656dc30ed4746a9207e69f3bcc90f47232c67 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728737 Commit-Queue: Jorge Betancourt <[email protected]> Reviewed-by: Florin Malita <[email protected]>
This landed in https://crrev.com/c/4374637 and was not reverted in the months since. Change-Id: I63e2fa3f657cdc1be3ae54c2c8896c3196e99c30 Bug: skia:14163 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728698 Reviewed-by: Brian Osman <[email protected]> Commit-Queue: Kevin Lubick <[email protected]>
Client CLs: - http://ag/24151403 - http://cl/549704569 Change-Id: Ib94475992a4f6fb7323f92053c62542812777b9d Bug: skia:14317 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728701 Reviewed-by: Brian Osman <[email protected]> Commit-Queue: Kevin Lubick <[email protected]>
https://skia.googlesource.com/buildbot.git/+log/1de81d2a1fc2..2ca55949153a 2023-07-24 [email protected] Roll Skia Infra CIPD packages from ac6948eb5d9f to 1de81d2a1fc2 (10 revisions) If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/sk-tool-skia Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Tbr: [email protected] Change-Id: I53d92273abe55f619efb89300a07710d302ed8a5 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728821 Commit-Queue: skia-autoroll <[email protected]> Bot-Commit: skia-autoroll <[email protected]>
Bug: b/289896316 Change-Id: I4d2b0b534dce53ce7f3ec2bffe35348c5728aa4e Reviewed-on: https://skia-review.googlesource.com/c/skia/+/727816 Reviewed-by: Brian Osman <[email protected]> Reviewed-by: Michael Ludwig <[email protected]> Auto-Submit: Jim Van Verth <[email protected]> Commit-Queue: Brian Osman <[email protected]>
https://chromium.googlesource.com/vulkan-deps.git/+log/18e535f0ebf3..4c8d1ddb012a Also rolling transitive DEPS: https://skia.googlesource.com/external/github.com/KhronosGroup/SPIRV-Tools/+log/d52c39c37d4d7aece12e6177203cc3d733e8b772..ec90d2872acbb4802661c9186a1e0fce8924e97b https://chromium.googlesource.com/external/github.com/KhronosGroup/Vulkan-Tools/+log/e69a7c71c52c2e9316f4bd7f95af7a084a1a2870..ed6820d508fd24b1a50166a600dbf7ee8300e9d9 If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/vulkan-deps-skia-autoroll Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Cq-Include-Trybots: skia/skia.primary:Build-Debian10-Clang-x86_64-Debug-Dawn;skia/skia.primary:Build-Debian10-Clang-x86_64-Release-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-All-ANGLE Bug: None Tbr: [email protected] Change-Id: I9f1af15f558982ff6f2225e4d9e9c628a0deffe3 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728721 Commit-Queue: skia-autoroll <[email protected]> Bot-Commit: skia-autoroll <[email protected]>
For excessively large crop rects, isEmpty() was returning true when the width or height overflowed. isEmpty64() just checks left < right and top < bottom, which makes it consistent with SkRect::isEmpty(). The large crop rects are now allowed through, and when they are intersected with the desired output, become reasonable and usable. Previously they were causing the entire operation to be dropped. In the event that somehow we try to create a backing image for a layer irect that is this large, but was previously being rejected with the 32-bit isEmpty(), the width and height will still be negative and image allocation will fail, which is still guarded. Bug: chromium:1466359 Change-Id: I3045773e111bbb779826e155ea26da13212481bf Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728676 Reviewed-by: Robert Phillips <[email protected]> Reviewed-by: Brian Osman <[email protected]> Commit-Queue: Michael Ludwig <[email protected]>
SkEmptyFontMgr is the last resort fallback for font managers. Prior to this change if one was needed it would be created as a static local variable and an sk_sp to it returned and then the sk_sp was also stored as a static local variable. This meant that the SkEmptyFontMgr static local could be destroyed before the sk_sp static local variable. This is the Static Initialization Order Fiasco but for destruction. To avoid these problems create the single SkEmptyFontMgr on the heap if it is needed and make it always owned by sk_sp. The order of destruction is no longer a problem as the last sk_sp to be destroyed will delete the SkEmptyFontMgr. Change-Id: Ia32678ec449dbfc4674a276193b636c29dbaf9c2 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728738 Reviewed-by: Leandro Lovisolo <[email protected]> Commit-Queue: Ben Wagner <[email protected]> Auto-Submit: Ben Wagner <[email protected]> Commit-Queue: Leandro Lovisolo <[email protected]>
https://swiftshader.googlesource.com/SwiftShader.git/+log/66d6b0dd0c39..9fbca2df22a8 2023-07-24 [email protected] LLVMReactor: Support LLVM 17+ 2023-07-24 [email protected] LLVMReactor: Remove CreateFreeze() call If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/swiftshader-skia-autoroll Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in SwiftShader: https://bugs.chromium.org/p/swiftshader/issues/entry To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Cq-Include-Trybots: skia/skia.primary:Test-Debian10-Clang-GCE-GPU-SwiftShader-x86_64-Debug-All-SwiftShader Bug: None Tbr: [email protected] Change-Id: Ida93676f600e1f9ff0d0c229791d608b34c9ac90 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728913 Commit-Queue: skia-autoroll <[email protected]> Bot-Commit: skia-autoroll <[email protected]>
https://skia.googlesource.com/buildbot.git/+log/1de81d2a1fc2..2ca55949153a 2023-07-24 [email protected] Roll Skia Infra CIPD packages from ac6948eb5d9f to 1de81d2a1fc2 (10 revisions) If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/infra-skia Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in Skia Infra: https://bugs.chromium.org/p/skia/issues/entry To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Bug: None Tbr: [email protected] Change-Id: Ib5d95e1a84a810d3882eb8332bfebb8dbfa8a32c Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728914 Bot-Commit: skia-autoroll <[email protected]> Commit-Queue: skia-autoroll <[email protected]>
https://chromium.googlesource.com/angle/angle.git/+log/430a4f559cbc..2d999f744809 2023-07-24 [email protected] Android: Add a way to prepare traces without running tests 2023-07-24 [email protected] Vulkan: Remove dead code 2023-07-24 [email protected] Metal: Don't flush on eglBindTexImage. 2023-07-24 [email protected] Android.bp generation: vulkan-headers LICENSE.txt -> .md 2023-07-24 [email protected] Roll SwiftShader from 4a260c12b8c1 to 66d6b0dd0c39 (6 revisions) 2023-07-24 [email protected] Roll Chromium from 986ed21b8935 to 6eba95b5d89b (527 revisions) If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/angle-skia-autoroll Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in ANGLE: https://bugs.chromium.org/p/angleproject/issues/entry To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Cq-Include-Trybots: skia/skia.primary:Build-Debian10-Clang-x86_64-Release-ANGLE;skia/skia.primary:Build-Mac-Clang-arm64-Release-ANGLE;skia/skia.primary:Test-Win10-Clang-AlphaR2-GPU-RadeonR9M470X-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-Golo-GPU-QuadroP400-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC8i5BEK-GPU-IntelIris655-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUCD34010WYKH-GPU-IntelHD4400-x86_64-Debug-All-ANGLE Tbr: [email protected] Change-Id: If80a537f88ce49b4f4f775000c6daae33ba893a2 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728827 Commit-Queue: skia-autoroll <[email protected]> Bot-Commit: skia-autoroll <[email protected]>
https://chromium.googlesource.com/vulkan-deps.git/+log/4c8d1ddb012a..6f1c3384ecb6 If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/vulkan-deps-skia-autoroll Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Cq-Include-Trybots: skia/skia.primary:Build-Debian10-Clang-x86_64-Debug-Dawn;skia/skia.primary:Build-Debian10-Clang-x86_64-Release-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-All-ANGLE Bug: None Tbr: [email protected] Change-Id: I1a4b92ada50261b3c946fead8d367d46378b6e91 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728831 Commit-Queue: skia-autoroll <[email protected]> Bot-Commit: skia-autoroll <[email protected]>
Change-Id: I0af36ff2cf738a88f23104d90165d7b5d417fdf2 Bug: b/291097600 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/726057 Commit-Queue: Nicolette Prevost <[email protected]> Reviewed-by: Jim Van Verth <[email protected]>
This map was only used to associate a field with an interface block when emitting field-access expressions. This allowed us to give a unique name to anonymous interface-blocks when their field accesses occurred. Instead of having a map, we can use the type of the interface block as a key for this lookup instead, since each interface block must have a unique type. Change-Id: Ief0beed539ff215c0a8f2fe39f62b86983a2de55 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728916 Reviewed-by: Brian Osman <[email protected]> Auto-Submit: John Stiles <[email protected]> Commit-Queue: Brian Osman <[email protected]>
Array types currently do not compile cleanly in std430 layout. For
instance, consider the float[] type. If we apply std430 layout to SSBOs
and std140 layout to UBOs (as the latter is required), a different array
stride will be inferred for this type if it is declared inside a UBO
block and a SSBO block in the same program (16 in std140 and 4 in
std430).
This difference in layout is currently not handled well by SkSL,
especially when array types appear in user-defined function parameter
signatures. For example, consider this program:
layout(...) uniform UBO {
float uboArray[4];
};
layout(...) buffer SSBO {
float ssboArray[4];
};
void helper(float[4] arg) {...}
half4 main() {
...
helper(uboArray);
helper(ssboArray);
...
}
This is a valid SkSL program but it is unclear what code this should
generate in the WGSL and SPIR-V backends, both of which need to
determine:
1. What is the stride of `arg`?
2. What is the stride of `ssboArray`?
3. If `arg` and `ssboArray` have different strides what are the
argument passing semantics of `helper(ssboArray)`?
4. What is the stride of a temporary binding, e.g.
float[4] myArray = ssboArray;
myArray = uboArray;
The SPIR-V backend passes function arguments by pointer but the pointer
types of two different arrays with different strides are incompatible,
so a better way to pass an array as an argument is to copy its
elements into a temporary composite, which is not how SPIR-V codegen
works today.
WGSL requires that different array strides be declared via different
type signatures. `array<f32, N>` always has a stride of 4 and is
incompatible with UBO layout, which requires `float[N]` to be converted
to `array<vec4<f32>, N>` or `array<S, N>` where `S` is a struct with 16
byte alignment. What should the function parameter signature of `helper`
be in WGSL?
It makes sense for SPIR-V and WGSL codegen to have a consistent story
to determining stride if we want to support packed array layouts with
std430. Until then, use std140 layout with SSBOs.
Bug: skia:14639
Change-Id: I9fc154dabadf384fdd926786a5fbade959a3e935
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728956
Commit-Queue: Arman Uguray <[email protected]>
Reviewed-by: John Stiles <[email protected]>
Change-Id: Iba793a1c9c13af393aca5e343e31f3fad78cbba4 Bug: skia:14317 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728703 Reviewed-by: Michael Ludwig <[email protected]>
Note the default implementation of SkColorFilterBase::addToKey is handled in the kNoOp switch case. Change-Id: I963098148a1b9e1b40e5ff53ddc885e23f5e8517 Bug: skia:13983 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728917 Commit-Queue: Kevin Lubick <[email protected]> Reviewed-by: Michael Ludwig <[email protected]>
We now generate valid code for the storage-buffer golden output test case. This also fixes the multiple anonymous-interface-block test cases; this isn't important (we don't use it in Skia), but it came naturally from the implementation. Bug: skia:13092 Change-Id: I75d33e1d63977473261225ec06659c37f80772b6 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/729256 Reviewed-by: Arman Uguray <[email protected]> Auto-Submit: John Stiles <[email protected]> Commit-Queue: John Stiles <[email protected]>
https://skia.googlesource.com/buildbot.git/+log/2ca55949153a..983f600aab02 2023-07-25 [email protected] [CT] Email CT Admins only on failures 2023-07-25 [email protected] [cd] Ignore "other" file mode in assertFileMatches 2023-07-25 [email protected] Roll Skia Infra CIPD packages from 1de81d2a1fc2 to 2ca55949153a (1 revision) If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/sk-tool-skia Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Tbr: [email protected] Change-Id: I13fe1a5015bb914546f2f8bfabbe4af1787308d2 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/729378 Commit-Queue: skia-autoroll <[email protected]> Bot-Commit: skia-autoroll <[email protected]>
Also enforce IWYU on it. Change-Id: I2dc4331cc43bc1224530be91038b378c9cdd8a02 Bug: skia:14317 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728700 Commit-Queue: Kevin Lubick <[email protected]> Reviewed-by: Michael Ludwig <[email protected]>
Bug: skia:14638 Change-Id: I72fa72d4ecd16d839108d7c0fc22a04be30cb7bc Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728736 Reviewed-by: Brian Osman <[email protected]> Commit-Queue: Michael Ludwig <[email protected]> Reviewed-by: Robert Phillips <[email protected]>
Moved the compute test file declarations out of the metal test list and into their own filegroup in preparation for enabling them in more backends. Moved the compute error tests to the shared error sources as they are not backend-specific. Bug: b/262428625 Change-Id: Ia0cbac78aa8cafb957d1c5393467cb17219a9e15 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/729096 Reviewed-by: John Stiles <[email protected]> Commit-Queue: Arman Uguray <[email protected]>
This flag was added to allow us to transition from SkVM to SkRP. We don't plan on supporting a build of Skia that omits SkSL, so there's no reason to disable SK_ENABLE_SKSL_IN_RASTER_PIPELINE. (This would cause various effects to stop painting.) Bug: b/294209201 Change-Id: I61b9868413266f3b6e7f65408f0188eebee01f05 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736880 Commit-Queue: John Stiles <[email protected]> Auto-Submit: John Stiles <[email protected]> Reviewed-by: Kevin Lubick <[email protected]> Reviewed-by: Brian Osman <[email protected]> Commit-Queue: Brian Osman <[email protected]>
This flag did not have any effect; we use SkRP for software rasterization. Change-Id: I0147c81acef4947cabd7dc03450a79e0bba17e32 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736882 Reviewed-by: Kevin Lubick <[email protected]> Reviewed-by: Brian Osman <[email protected]> Commit-Queue: John Stiles <[email protected]> Auto-Submit: John Stiles <[email protected]>
There are several CPU backend SkOpts modules that actually belong in SkGraphics, rather than trying to find isolated homes for them. If anything, it's the other opts that need distinct locations. Change-Id: I5b76a1a87976476775921879c62e06908b74fc1f Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736976 Commit-Queue: Brian Osman <[email protected]> Reviewed-by: Kevin Lubick <[email protected]>
This was an odd helper struct. It's much clearer to just call SkGraphics::Init (as is done in most of our other tools and client code). Last client usage removed in cl/554487694 Change-Id: Idc30ae2da278ca2a13319001cd6ceaab65f3887f Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736883 Commit-Queue: Brian Osman <[email protected]> Reviewed-by: Herb Derby <[email protected]>
sk_sp<T> has a destructor which we do not want to register. Change-Id: I1c5fdfed21a9a480eb6c57bb718829ea5d30a68b Reviewed-on: https://skia-review.googlesource.com/c/skia/+/728797 Commit-Queue: John Stiles <[email protected]> Auto-Submit: John Stiles <[email protected]> Reviewed-by: Herb Derby <[email protected]> Reviewed-by: Ben Wagner <[email protected]>
A new machine won't have `python` bound to anything. Change-Id: I67ce37dea2ecfb3deaca4712804b04dc8d39ae36 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736996 Commit-Queue: Kevin Lubick <[email protected]> Auto-Submit: John Stiles <[email protected]> Reviewed-by: Kevin Lubick <[email protected]>
Modify the Bazel side build rules so that the CXX core-lib target contains its symbols from libc++. The Bazel build performs the build of the CXX Rust interface library based on clang arguments '-std=c++17' '-stdlib=libc++' so that it requires libc++ symbols to be present. Depending on platform and linker, the GN+ninja build cannot provide those (example ld linker and GNU libstd++ on Linux). We address this by making the CXX core-lib build include the libc++ symbols on Linux. This may lead to a partial duplication of C++ standard library functionality, which may be acceptable for the viewer executable or other test executables. In Chromium, Skia + Fontations + Blink's font stack is built using the GN based Rust builds that the Chrome security team developed - so this problem will not apply there. Bug: b/40045269 Change-Id: I1a6654a8dda99fcc15781fbb7414d5bf889e43e4 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/719916 Commit-Queue: Dominik Röttsches <[email protected]> Reviewed-by: Kevin Lubick <[email protected]> Commit-Queue: Dominik Röttsches <[email protected]>
If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/debugger-app-base-skia Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Tbr: [email protected] Change-Id: I98909dcc97a4bc9846f426e0efcb7927082563e2 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736803 Bot-Commit: skia-autoroll <[email protected]> Commit-Queue: skia-autoroll <[email protected]>
Like Metal, WGSL only supports matrix-op-scalar natively for multiplication, so we need to rewrite these expressions into matrix-op-matrix by splatting the scalar across every matrix element. This fixes one test. I've also adjusted the test slightly to increase its coverage a bit. Previously it only tested assignment-expressions (`matrix op= scalar`) but neglected to test `scalar-op-matrix` or `matrix = matrix op scalar`. I've rewritten the expressions to cover these forms as well. Change-Id: I41dd4b6fc1ef93a6bbc2b5ed6424147500fdeb9b Reviewed-on: https://skia-review.googlesource.com/c/skia/+/737056 Auto-Submit: John Stiles <[email protected]> Commit-Queue: Brian Osman <[email protected]> Reviewed-by: Brian Osman <[email protected]>
See https://skia-review.googlesource.com/c/skia/+/727257 for details Bug: b/40045064 Bug: b/40045066 Change-Id: I2294e49be894d7b470e6a69b4a7ec06ec44439cd Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736881 Commit-Queue: Brian Osman <[email protected]> Reviewed-by: Kevin Lubick <[email protected]>
This is used by Graphite, so a polyfill is necessary here. The logic was cloned from MetalCodeGenerator and then updated to WGSL syntax. Change-Id: I9e87f880ba8925347345c75b7cd76fd27110714c Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736879 Commit-Queue: John Stiles <[email protected]> Auto-Submit: John Stiles <[email protected]> Reviewed-by: Brian Osman <[email protected]>
This CL adds support for registering GMs that are unknown at compile time, such as those that depend on command-line flags. See comments in //gm/gm.h for rationale, and //gm/BazelGMRunner.cpp for sample use. The follow-up CL (https://skia-review.googlesource.com/c/skia/+/734379) introduces some dynamically registered GMs. Bug: b/40045301 Change-Id: I532391cc500aac6b4501d0d1f56f879c478aa469 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/734163 Commit-Queue: Leandro Lovisolo <[email protected]> Reviewed-by: Kevin Lubick <[email protected]>
Previously, we disallowed ternary expressions with a void result, but the error message was slightly unclear; we also disallowed ternary expressions that contained an array, but only in ES2. It turns out that WebGL2 explicitly disallows ternaries that evaluate to void or to arrays or s-of-a's. I've improved the error message in the void case, and made the array case invalid all the time instead of just in ES2. (https://registry.khronos.org/webgl/specs/latest/2.0/#5.26) Change-Id: I6dbdc98a89dd127d5e7dcdbac51ad17130a9b661 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736358 Commit-Queue: John Stiles <[email protected]> Reviewed-by: Brian Osman <[email protected]> Commit-Queue: Brian Osman <[email protected]> Auto-Submit: John Stiles <[email protected]>
Change-Id: I35ad18eae732e4f06bdecd552455741fc6d2af5a Reviewed-on: https://skia-review.googlesource.com/c/skia/+/737176 Commit-Queue: John Stiles <[email protected]> Commit-Queue: Brian Osman <[email protected]> Auto-Submit: Brian Osman <[email protected]> Reviewed-by: John Stiles <[email protected]>
Until we have a workaround in the GLSL code generator, we will need to disable this test in WebGL. Bug: b/294893925 Change-Id: Id08a3f029a91bad9a444dce628a4fe09de8ab7cc Reviewed-on: https://skia-review.googlesource.com/c/skia/+/737180 Commit-Queue: Kevin Lubick <[email protected]> Reviewed-by: Kevin Lubick <[email protected]> Auto-Submit: John Stiles <[email protected]> Commit-Queue: John Stiles <[email protected]>
https://chromium.googlesource.com/vulkan-deps.git/+log/33af718d939e..c46d48f777b7 If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/vulkan-deps-skia-autoroll Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Cq-Include-Trybots: skia/skia.primary:Build-Debian10-Clang-x86_64-Debug-Dawn;skia/skia.primary:Build-Debian10-Clang-x86_64-Release-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-All-ANGLE Bug: None Tbr: [email protected] Change-Id: I08c961332e8e236814a92b6bffac21a7299a8010 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736674 Commit-Queue: skia-autoroll <[email protected]> Bot-Commit: skia-autoroll <[email protected]>
Follow-up to https://skia-review.googlesource.com/c/skia/+/736256 (I had already started this CL when I noticed the bug in the original SkOpts, and never fixed this code before landing). Change-Id: I6114cd220e131ad26a12e62d0fd61551621ba561 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736361 Auto-Submit: Brian Osman <[email protected]> Commit-Queue: John Stiles <[email protected]> Commit-Queue: Brian Osman <[email protected]> Reviewed-by: John Stiles <[email protected]>
https://skia.googlesource.com/buildbot.git/+log/89658e42bb91..333a87d1ef8a 2023-08-07 [email protected] [machineserver] Fix Intel CPU parsing bug. 2023-08-07 [email protected] ephemeral_storage.go: Fix test_machine_monitor panic on Windows. If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/sk-tool-skia Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Tbr: [email protected] Change-Id: I7d5542fe78134ece44689b33cc48752d6dc4e36f Reviewed-on: https://skia-review.googlesource.com/c/skia/+/736810 Bot-Commit: skia-autoroll <[email protected]> Commit-Queue: skia-autoroll <[email protected]>
https://skia.googlesource.com/buildbot.git/+log/0e52994bf1b6..333a87d1ef8a 2023-08-07 [email protected] [machineserver] Fix Intel CPU parsing bug. 2023-08-07 [email protected] ephemeral_storage.go: Fix test_machine_monitor panic on Windows. 2023-08-07 [email protected] initial refactor to support text properties per text box 2023-08-07 [email protected] Reland "Reland "Reland "Reland "Testing whitespace change with GitWatcher"""" 2023-08-07 [email protected] [autoroll] Creating docker child 2023-08-07 [email protected] Roll Skia Infra CIPD packages from ed824b45206d to 0e52994bf1b6 (4 revisions) If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/infra-skia Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in Skia Infra: https://bugs.chromium.org/p/skia/issues/entry To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Bug: chromium:1470711 Tbr: [email protected] Change-Id: Ic7a39aa9f9d93894752bef02c6e4bbe9c6c9cddd Reviewed-on: https://skia-review.googlesource.com/c/skia/+/737296 Commit-Queue: skia-autoroll <[email protected]> Bot-Commit: skia-autoroll <[email protected]>
https://chromium.googlesource.com/angle/angle.git/+log/135a24fc3706..f7d7be8d2ff0 2023-08-07 [email protected] Metal: upstream "UBO convert only whole block". 2023-08-07 [email protected] Vulkan: Retain loadOp when there is a resolve attachment 2023-08-07 [email protected] Simplify struct LinkedUniform a bit more 2023-08-07 [email protected] load/save entire std::vector<LinkedUniform> with one call. 2023-08-07 [email protected] Move name and mappedName out of LinkedUniform struct 2023-08-07 [email protected] Add Gitmodules to angle. 2023-08-07 [email protected] Vulkan: Remove type indices with host-visible bit 2023-08-07 [email protected] Change LinkedUniform::arraySizes from std::vector to unsigned int 2023-08-07 [email protected] Use offscreen DisplayVk on ChromeOS 2023-08-07 [email protected] Validate program binaries are the same CPU bit-ness. 2023-08-07 [email protected] Roll VK-GL-CTS from 00cccd7cf562 to cedd20620c84 (11 revisions) 2023-08-07 [email protected] Metal: disable fastmath less often. 2023-08-07 [email protected] ANGLETest: fix crashes when switching between GLESDriverTypes If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/angle-skia-autoroll Please CC [email protected],[email protected] on the revert to ensure that a human is aware of the problem. To file a bug in ANGLE: https://bugs.chromium.org/p/angleproject/issues/entry To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Cq-Include-Trybots: skia/skia.primary:Build-Debian10-Clang-x86_64-Release-ANGLE;skia/skia.primary:Build-Mac-Clang-arm64-Release-ANGLE;skia/skia.primary:Test-Win10-Clang-AlphaR2-GPU-RadeonR9M470X-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-Golo-GPU-QuadroP400-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC8i5BEK-GPU-IntelIris655-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUCD34010WYKH-GPU-IntelHD4400-x86_64-Debug-All-ANGLE Tbr: [email protected] Change-Id: Id4ba41b582628ea0fcc446a498c4e35d85769d13 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/737257 Commit-Queue: skia-autoroll <[email protected]> Bot-Commit: skia-autoroll <[email protected]>
Change-Id: Ibbd0cfa6df5ef13448e59c0e0d772597c049fee1 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/737933 Reviewed-by: Chris Mumford <[email protected]> Auto-Submit: Rakshit Sharma <[email protected]>
Change-Id: I658a4d46085bd47dd3cfbe58162a96b49e5b9248 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/738222 Reviewed-by: Chris Mumford <[email protected]>
As per the linked bugs, this was causing a severe performance regression. Change-Id: Ibacf842408cee1eb434795437c8bbe0dc5244c8e Bug: b/291711510, chromium:1475906 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/750403 Reviewed-by: Greg Daniel <[email protected]> Commit-Queue: Kevin Lubick <[email protected]> (cherry picked from commit 02fa147) Reviewed-on: https://skia-review.googlesource.com/c/skia/+/751517
Change-Id: I080ba2191ede25c1c848cf94f05169334084eb32 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/786340 Reviewed-by: Heather Miller <[email protected]>
…kia-m117 # Conflicts: # DEPS # RELEASE_NOTES.md # gn/core.gni # gn/utils.gni # infra/bots/jobs.json # src/gpu/graphite/mtl/MtlCaps.mm # src/utils/win/SkDWrite.cpp
5 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description of Change
Updating the version to the next one.
API Changes
Removed:
Added:
Behavioral Changes
None.
Required SkiaSharp PR
mono/SkiaSharp#3047
PR Checklist
skiasharpat time of PR