Instead of counting descriptors to be allocated for the entire command
buffer in a pre-pass, the bindgroup state tracker is used to allocate
only dirty bindgroups upon recording draw/dispatch. If the heap has no
more room and must be changed, bindgroups will be re-created according
to the BGL.
A future change will address the CPU descriptors and removal of the
pre-pass.
BUG=dawn:256,dawn:307
Change-Id: I6603de17cfda713bd4512c46e1c93618ca01bb7b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/13400
Commit-Queue: Bryan Bernhart <bryan.bernhart@intel.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
These are not supported on some older OpenGL, OpenGL ES, and iOS
devices.
Bug: dawn:343
Change-Id: I70def749ae57fcfe2895f8556674dd241941d3d3
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/16163
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Jiawei Shao <jiawei.shao@intel.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This Toggle is set if MTLFeatureSet_iOS_GPUFamily3_v1 is not supported.
Bug: dawn:342
Change-Id: Ia5f43e87fdd2c13eaffe9557cb0ce9a06dec3b29
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/16180
Reviewed-by: Jiawei Shao <jiawei.shao@intel.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
BUG=dawn:344
Change-Id: Ifa9e1e3167ecfe7d38c16f393cec0443ea1589f2
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/16164
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
If these extern variables are initialized after DAWN_INSTANTIATE_TEST,
they will be zero. Change them to be function calls instead.
Since they're function calls, fold in arguments from ForceToggles to
enable/disable toggles.
Bug: dawn:341
Change-Id: I1aeaa1e535a0a003977e8ce7ab3d5278c5d81281
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/16162
Reviewed-by: Mark Henderson <mehe@google.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Austin Eng <enga@chromium.org>
Roll third_party/SPIRV-Tools/ 4a80497a8..8910ea5f1 (4 commits)
4a80497a88..8910ea5f1c
$ git log 4a80497a8..8910ea5f1 --date=short --no-merges --format='%ad %ae %s'
2020-02-23 nicolasweber Fix Wrange-loop-analysis warnings in SPIRV-Tools. (#3201)
2020-02-21 geofflang Add missing dependencies when generating spvtools_core_tables (#3199)
2020-02-21 afdx Brief guide to writing a spirv-fuzz fuzzer pass (#3190)
2020-02-21 47594367+rg3igalia Fix ignored const qualifier warning in static_cast (#3197)
Roll third_party/glslang/ c12493ff6..07e1a0a67 (1 commit)
c12493ff69..07e1a0a67a
$ git log c12493ff6..07e1a0a67 --date=short --no-merges --format='%ad %ae %s'
2020-02-22 rex.xu Fix an issue of SPV generation for imageAtomicStore.
Roll third_party/shaderc/ 738f1655a..1059f43a3 (1 commit)
$ git log 738f1655a..1059f43a3 --date=short --no-merges --format='%ad %ae %s'
2020-02-21 rharrison Rolling 6 dependencies and update expectations (#987)
Roll third_party/spirv-cross/ f19fdb94d..c5f7b5575 (5 commits)
f19fdb94d7..c5f7b55756
$ git log f19fdb94d..c5f7b5575 --date=short --no-merges --format='%ad %ae %s'
2020-02-24 post MSL: Add C API for force native arrays.
2020-02-24 post MSL: Add native array test for composite array initialization.
2020-02-24 post MSL: Reintroduce workaround for constant arrays being passed by value.
2020-02-24 post MSL: Reinstate workaround for returning arrays.
2020-02-24 post MSL: Add a workaround path to force native arrays for everything.
Created with:
roll-dep third_party/SPIRV-Tools third_party/glslang third_party/shaderc third_party/spirv-cross third_party/spirv-headers
Change-Id: Ic9f459c21a9656093b5ec27e19f0744182734c7c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/16120
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Previously we would always assume that if the driver supported a Vulkan
version it would also support extensions that were promoted in that
version. This is not a spec requirement, so instead try to load the core
entrypoints, and only if the version is not available, load the
extension entrypoints.
Also renames VulkanFunction members that are from promoted extension to
not have a vendor prefix.
Also tag the promoted extensions that are the same in a core version as
available when that core version is available. This simplifies checking
for features in the Vulkan backend.
Bug:
Change-Id: I0817c01b8838ba26070858abb0cbed030e3291df
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/16040
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: David Turner <digit@google.com>
This relands commit 0bbfec1f7f which
was reverted in 21e5074dcd.
The original CL broke the Chromium roll because drm/drm_fourcc.h
could not be found. Still not sure why this is the case since it
seems to be present on all of my CrOS test machines, but at the end
of the day, I realized that I don't even need this header in the
first place.
This CL removes the header and relands the rest of the original CL.
Bug: chromium:996470
Change-Id: I77d6b1692094b7798f3c5d9c2b50219e674c8a8c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/16060
Commit-Queue: Brian Ho <hob@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
This reverts commit 0bbfec1f7f.
Reason for revert: Makes the roll into Chromium fail with the following:
[314/21578] CXX obj/third_party/dawn/dawn_white_box_tests_sources/VulkanImageWrappingTestsDmaBuf.o
FAILED: obj/third_party/dawn/dawn_white_box_tests_sources/VulkanImageWrappingTestsDmaBuf.o
../../build/toolchain/clang_code_coverage_wrapper.py --target-os=chromeos --files-to-instrument=../....(too long)
../../third_party/dawn/src/tests/white_box/VulkanImageWrappingTestsDmaBuf.cpp:28:10: fatal error: 'drm/drm_fourcc.h' file not found
#include <drm/drm_fourcc.h>
^~~~~~~~~~~~~~~~~~
1 error generated.
Original change's description:
> Create VulkanImageWrappingTests for dma-buf images
>
> This CL branches the existing VulkanImageWrappingTests to separate
> tests for OpaqueFD-backed amd DmaBuf-backed external images. On
> Chrome OS of Dawn, we no longer interop using opaque FDs, so these
> tests were failing in the end2end test suite.
>
> The new VulkanImageWrappingTestsDmaBuf tests are essentially 1:1
> mappings of their counterparts in the Opaque FD version. The only
> difference is that we allocate memory directly on the device using
> GBM instead of creating a VkImage (which will likely call some GBM
> methods under the hood) and then extracting the FD using a Vulkan
> extension. We then communicate this to Dawn via the DmaBuf
> ExternalImageDescriptor.
>
> Also, this fixes VulkanImageWrappingUsageTests::LargeImage on AMD
> devices (assuming the extension is implemented) as we can now
> specify DRM modifiers.
>
> Bug: chromium:996470
> Change-Id: I2b3c57d7f5ff14131d415e99a09d32d2f16b3e54
> Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15800
> Commit-Queue: Brian Ho <hob@chromium.org>
> Reviewed-by: Austin Eng <enga@chromium.org>
TBR=cwallez@chromium.org,enga@chromium.org,hob@chromium.org
Change-Id: Idb45586c608ce20432142834a4f14d42c76d3b3b
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:996470
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/16001
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This CL branches the existing VulkanImageWrappingTests to separate
tests for OpaqueFD-backed amd DmaBuf-backed external images. On
Chrome OS of Dawn, we no longer interop using opaque FDs, so these
tests were failing in the end2end test suite.
The new VulkanImageWrappingTestsDmaBuf tests are essentially 1:1
mappings of their counterparts in the Opaque FD version. The only
difference is that we allocate memory directly on the device using
GBM instead of creating a VkImage (which will likely call some GBM
methods under the hood) and then extracting the FD using a Vulkan
extension. We then communicate this to Dawn via the DmaBuf
ExternalImageDescriptor.
Also, this fixes VulkanImageWrappingUsageTests::LargeImage on AMD
devices (assuming the extension is implemented) as we can now
specify DRM modifiers.
Bug: chromium:996470
Change-Id: I2b3c57d7f5ff14131d415e99a09d32d2f16b3e54
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15800
Commit-Queue: Brian Ho <hob@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Also fixes some warnings when compiling with GCC
Bug: dawn:333
Change-Id: Ib597bb3b950476279a1e20e3556765ec9f1db697
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15960
Commit-Queue: Kai Ninomiya <kainino@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Headers only INTERFACE library with generated headers don't work in CMake
because the GENERATED property is local to a directory. Instead we make a
STATIC library with a Dummy cpp file.
INTERFACE libraries can only have INTERFACE sources so the sources get added
to the dependant's list of sources. If these dependents are in another
directory, they don't see the GENERATED property and fail to configure
because the file doesn't exist on disk.
Use this trick for both dawn_headers and dawncpp_headers that are header
only libraries with generated headers.
Bug: dawn:333
Change-Id: Ib0d6dcc5f351a638d1c5360214c0ce14a28fee3e
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15921
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
The symbol was not marked as static and would cause linking errors
because it would be defined in multiple translation units. Replace it
with a more traditional C-style #define.
Bug:
Change-Id: I19151884b7e8e171f829ffa47b1d119aff12ff99
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15740
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This also changes the name to be consistent with the option being set
in the spvc API.
BUG=dawn:335
Change-Id: I6f7431095493874e1fef0856e563f7f1225cfc21
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15780
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
There appears to be issues with the CTS + spvc that are preventing
chromium to roll, so I am reverting spvc by default, until the CTS
issues are resolved.
BUG=dawn:337
Change-Id: I171ee5325b9afbf5d240a469009433105caf6ddb
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15840
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Adds CMake support for:
- Generating the Dawn headers and C++ wrappers
- libdawn_wire
- libdawn_native with the Metal backend for now
- All the examples.
Bug: dawn:333
Change-Id: I6ffbe090b0bd21d6a805c03a507ad51fda0275ca
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15720
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Dan Sinclair <dsinclair@chromium.org>
This reverts commit 97c3be2699.
Reason for revert: Breaks linking on MSVC
Original change's description:
> Use libshaderc_spvc as a source_set, so complete_static_lib applies properly
>
> Brings the list of dependencies for a standalone Dawn app down from:
>
> obj/libdawn_native.a
> obj/libdawn_utils.a
> obj/src/dawn/libdawn_proc.a
> obj/third_party/shaderc/libshaderc_spvc/spvc.o
> obj/third_party/shaderc/libshaderc_spvc/spvc_private.o
> obj/third_party/shaderc/libshaderc_spvc/spvcir_pass.o
> obj/third_party/SPIRV-Tools/libspvtools_opt.a
>
> to
>
> obj/libdawn_native.a
> obj/src/dawn/libdawn_proc.a
>
> Bug: dawn:327
> Change-Id: I74654b304a9cb5f2aff19e72aa6a8bf1eb708c15
> Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15481
> Reviewed-by: Kai Ninomiya <kainino@chromium.org>
> Reviewed-by: Ryan Harrison <rharrison@chromium.org>
> Commit-Queue: Kai Ninomiya <kainino@chromium.org>
TBR=kainino@chromium.org,rharrison@chromium.org
Change-Id: Ib6d17eccebf371b71e74f1857b50bb3cfd67595a
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: dawn:327
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15722
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Kai Ninomiya <kainino@chromium.org>
Falling-back to direct allocation ensures allocation failure returns OOM.
If no OOM, the resource could be left then used while in an invalid state.
BUG=chromium:1045811,chromium:1047220,chromium:1047048
Change-Id: I927962b1dc6a7422a7d6eac114d82f28a42794a2
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15600
Commit-Queue: Bryan Bernhart <bryan.bernhart@intel.com>
Reviewed-by: Austin Eng <enga@chromium.org>
Brings the list of dependencies for a standalone Dawn app down from:
obj/libdawn_native.a
obj/libdawn_utils.a
obj/src/dawn/libdawn_proc.a
obj/third_party/shaderc/libshaderc_spvc/spvc.o
obj/third_party/shaderc/libshaderc_spvc/spvc_private.o
obj/third_party/shaderc/libshaderc_spvc/spvcir_pass.o
obj/third_party/SPIRV-Tools/libspvtools_opt.a
to
obj/libdawn_native.a
obj/src/dawn/libdawn_proc.a
Bug: dawn:327
Change-Id: I74654b304a9cb5f2aff19e72aa6a8bf1eb708c15
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15481
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Kai Ninomiya <kainino@chromium.org>
The SampleUtils uses CreateSwapChain with a nullptr surface. This is
currently valid with implementation-based swapchains so the argument
should be tagged as optional.
Bug: dawn:269
Change-Id: Ic00d5a67fb038d2771174bb36f99b66b84f1a252
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15680
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Overriding VK_DEFINE_NON_DISPATCHABLE_HANDLE changes the function
signatures of Vulkan functions, changing their ABI and making us
incompatible with real drivers. This removes that magic, and replaces it
with an explicit wrapper, VkHandle, which has much of the same
functionality as the original VkNonDispatchableHandle.
It adds definitions for dawn_native::vulkan::VkBuffer et al, which
shadow the native ::VkBuffer et al. This retains type safety throughout
the Vulkan backend without changing every single usage.
Notably, the following things had to change:
- An explicit conversion from VkBuffer* to ::VkBuffer* is needed for
arrays. This is implemented as a reinterpret_cast, which is still
safe as the new VkHandle still has the same memory layout properties
as VkNonDispatchableHandle did.
- When pointing to a VkHandle as an output pointer, it's now necessary
to explicitly get the native ::VkBuffer (via operator*) and point to
it.
Previously reviewed on:
https://dawn-review.googlesource.com/c/dawn/+/15580
Bug: chromium:1046362
Change-Id: I7d34ec38a805025f92165ea9a7ee07ae5c182076
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15641
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Kai Ninomiya <kainino@chromium.org>
This reverts commit 4e17d5c248.
Reason for revert: broken on chromeos
Original change's description:
> Remove VK_DEFINE_NON_DISPATCHABLE_HANDLE magic, use explicit VkHandle wrapper
>
> Overriding VK_DEFINE_NON_DISPATCHABLE_HANDLE changes the function
> signatures of Vulkan functions, changing their ABI and making us
> incompatible with real drivers. This removes that magic, and replaces it
> with an explicit wrapper, VkHandle, which has much of the same
> functionality as the original VkNonDispatchableHandle.
>
> It adds definitions for dawn_native::vulkan::VkBuffer et al, which
> shadow the native ::VkBuffer et al. This retains type safety throughout
> the Vulkan backend without changing every single usage.
>
> Notably, the following things had to change:
> - An explicit conversion from VkBuffer* to ::VkBuffer* is needed for
> arrays. This is implemented as a reinterpret_cast, which is still
> safe as the new VkHandle still has the same memory layout properties
> as VkNonDispatchableHandle did.
> - When pointing to a VkHandle as an output pointer, it's now necessary
> to explicitly get the native ::VkBuffer (via operator*) and point to it.
>
> Bug: chromium:1046362
> Change-Id: I9c5691b6e295aca1b46d4e3d0203956e4d570285
> Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15580
> Reviewed-by: Austin Eng <enga@chromium.org>
> Reviewed-by: Kai Ninomiya <kainino@chromium.org>
> Commit-Queue: Kai Ninomiya <kainino@chromium.org>
TBR=cwallez@chromium.org,kainino@chromium.org,enga@chromium.org
Change-Id: I500df2e34fd0f245ad04c517ff028ddd7bb5a2bf
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:1046362
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15620
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Kai Ninomiya <kainino@chromium.org>
Overriding VK_DEFINE_NON_DISPATCHABLE_HANDLE changes the function
signatures of Vulkan functions, changing their ABI and making us
incompatible with real drivers. This removes that magic, and replaces it
with an explicit wrapper, VkHandle, which has much of the same
functionality as the original VkNonDispatchableHandle.
It adds definitions for dawn_native::vulkan::VkBuffer et al, which
shadow the native ::VkBuffer et al. This retains type safety throughout
the Vulkan backend without changing every single usage.
Notably, the following things had to change:
- An explicit conversion from VkBuffer* to ::VkBuffer* is needed for
arrays. This is implemented as a reinterpret_cast, which is still
safe as the new VkHandle still has the same memory layout properties
as VkNonDispatchableHandle did.
- When pointing to a VkHandle as an output pointer, it's now necessary
to explicitly get the native ::VkBuffer (via operator*) and point to it.
Bug: chromium:1046362
Change-Id: I9c5691b6e295aca1b46d4e3d0203956e4d570285
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15580
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Kai Ninomiya <kainino@chromium.org>
BUG=dawn:325
Change-Id: I66bed7fec65cadecc956878bf11fc204b66ad195
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15500
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Roll third_party/SPIRV-Tools/ 323a81fc5..1b3441036 (9 commits)
323a81fc5e..1b3441036a
$ git log 323a81fc5..1b3441036 --date=short --no-merges --format='%ad %ae %s'
2020-01-24 syoussefi Fix chromium build (#3152)
2020-01-24 dneto Clarify mapping of target env to SPIR-V version (#3150)
2020-01-24 greg Use dummy switch instead of dummy loop in MergeReturn pass. (#3151)
2020-01-23 alanbaker Fix structured exit validation (#3141)
2020-01-23 dneto Add spvParseVulkanEnv (#3142)
2020-01-23 jaebaek Handle conflict between debug info and existing validation rule (#3104)
2020-01-23 syoussefi Use spirv-headers' BUILD.gn (#3148)
2020-01-23 syoussefi Roll external/spirv-headers/ af64a9e82..dc77030ac (4 commits) (#3147)
2020-01-21 afdx spirv-fuzz: Refactoring and type-related fixes (#3144)
Roll third_party/glslang/ b0ada8035..07a55839e (15 commits)
b0ada80356..07a55839ee
$ git log b0ada8035..07a55839e --date=short --no-merges --format='%ad %ae %s'
2020-01-27 rharrison Use correct enum type in case statement
2020-01-27 cepheus Build: Fix more build warnings caused by PR #2038.
2020-01-26 cepheus Build warning: Fix#2062, missing enum value in a switch.
2020-01-08 sk Public: replaced tabs with spaces
2020-01-08 sk CInterface: replaced tabs with spaces
2019-12-30 sk CInterface: added static asserts to compare C/C++ enums
2019-12-30 sk Added LAST_ELEMENT_MARKER for every enum used in the C interface
2019-12-26 sk CInterface: replaced SH_ prefix with GLSLANG_
2019-12-25 sk CInterface: added glslang_program_SPIRV_get_ptr()
2019-12-25 sk CInterface: added files to CMakeLists.txt
2019-12-25 sk CInterface: added _BIT suffix to glslang_messages_t and glslang_reflection_options_t, fixed const-correctness
2019-12-25 sk CInterface: reformatted according to .clang-format rules
2019-12-24 sk Added original glslang_c_interface implementation by Viktor Latypov
2019-12-24 laddoc Add Tess machine dependent built-in variables initialization for GLES 3.2
2019-10-18 timo.suoranta Fixes for gcc 9 / -Werror=deprecated-copy
Roll third_party/shaderc/ a4f9acedf..0681dd61d (2 commits)
a4f9acedf3..0681dd61d3
$ git log a4f9acedf..0681dd61d --date=short --no-merges --format='%ad %ae %s'
2020-01-27 rharrison Remove deprecated APIs (#974)
2020-01-24 dneto Handle new Glslang profile enum in switch (#973)
Roll third_party/spirv-cross/ f9818f080..68bf0f824 (6 commits)
f9818f0804..68bf0f824c
$ git log f9818f080..68bf0f824 --date=short --no-merges --format='%ad %ae %s'
2020-01-27 post Compile fix on older compilers.
2020-01-27 post GLSL: Support GL_ARB_enchanced_layouts for XFB.
2020-01-25 cdavis MSL: Move inline uniform blocks to the end of the argument buffer.
2019-12-16 cdavis MSL: Support inline uniform blocks in argument buffers.
2020-01-23 post Make SmallVector noexcept.
2020-01-22 42098783+barath121 Typo at line 324
Roll third_party/spirv-headers/ 204cd131c..dc77030ac (2 commits)
204cd131c4..dc77030acc
$ git log 204cd131c..dc77030ac --date=short --no-merges --format='%ad %ae %s'
2020-01-20 dneto Fix the license to match LICENSE
2020-01-20 syoussefi Add BUILD.gn
Created with:
roll-dep third_party/SPIRV-Tools third_party/glslang third_party/shaderc third_party/spirv-cross third_party/spirv-headers
Change-Id: I696c1facf47a03e03e09c72c4d44449a37c1b971
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15480
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
The state-tracking of the webgpu.h swapchain is a bit complicated
because contrary to implementation-based swapchains, they have more
guarantees and a "replacing mechanism". For example instead of hoping
the implementation-based swapchain resize automatically, the
surface-based swapchain needs to be replaced by a new swapchain and
invalidated.
This mechanism of invalidation also needs to be triggered when the last
reference to the surface is lost because we don't want to risk the
application destroying the window from under us.
Adds tests for all the cases of invalidation I could think of apart from
device loss.
Bug: dawn:269
Change-Id: Id515dbb640e13c6e30bb1f1e93b8e54f1e2bba4b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15400
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
If Device creation fails, several things are just partially initialized
and the destroy sequence crashes dereferencing null data.
This commit marks the Vulkan device as lost until after it is created.
This avoids parts of the destroy sequence which are unecessary since
the Device was never successfully created and no commands are in flight.
Bug: chromium:1043095
Change-Id: I8e121709fa19b215e118a615b639380d1db1f3f2
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15460
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
In the webgpu-headers PR it was decided that in the C header
WGPUChainedStruct would be included as a member instead of members being
inlined.
See https://github.com/webgpu-native/webgpu-headers/pull/30
Bug: dawn:160
Change-Id: I8caf91f3106578077c80778621a632411da44423
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15441
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This commit fixes two problems:
1) It includes testing/libfuzzer/libfuzzer_exports.h which exports
the LLVMFuzzerInitialize symbol. On Mac, not doing so causes it
to be removed by the linker.
2) It makes it so that dawn_use_swiftshader is enabled by default
when fuzzing, and enabling swiftshader also enables the Vulkan
backend.
Bug: chromium:1042426
Change-Id: Ic0ae0d59c91476b0079148f33a4b1ebd1cc6743c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/15462
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Kai Ninomiya <kainino@chromium.org>
Bryan Bernhart
Austin Eng
Ryan Harrison
Natasha Lee
Corentin Wallez
Brian Ho
Kai Ninomiya
Reid Kleckner
Mark Henderson