... which is still needed to keep the end2end tests happy
when one thread creates and destroys instances multiple times
(which happens in the MultiThreadedShutdown test).
Test: GfxstreamEnd2EndTests
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36471>
The header says:
enum ADataSpace AHardwareBuffer_getDataSpace(const AHardwareBuffer* _Nonnull buffer)
__INTRODUCED_IN(__ANDROID_API_V__);
which is API level 35.
Reviewed-by: Jason Macnak <jmacnak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36338>
information
This information (otherwise invisible to the guest) may be used in the
future in some other guest-only functionality
Reviewed-by: Gurchetan Singh <gurchetansingh@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36330>
... in GfxstreamConnectionManager, to acquire the threadLocalInstance
of the connectionManager. Using thread_local attribute storage to store
some CPP primitives (i.e. unique_ptr) can cause issues on some
platforms.
Better to do all TLS as explicit as possible, using the available
utilities from common src/util. Don't need to wrap it in
a std::unique_ptr either, as the instance is just managed by the
destructor in the TLS interface.
Reviewed-by: Gurchetan Singh <gurchetansingh@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36302>
... and defer getDeviceQueue impl to vk_common and trim down impls in
gfxstream.
gfxstream advertises, and selects queues/queueFamilies from what the
real device on the host advertises. During createDevice(), it needs to
allocate the queue objects to support this.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36227>
... And rename LinuxVirtGpu* -> DrmVirtGpu*
The characteristic of this virtgpu implementation is that it works
through the DRI from Linux. Yes, this is traditionally "Linux" specific,
but some platforms such as QNX, have started to incorporate parts of the
"DRM framework", on a platform that otherwise still is not "Linux". This
is just a more generally applicable naming to this implementation.
Reviewed-by: Gurchetan Singh <gurchetansingh@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36229>
VkDeviceMemory is always 64-bit, and %p on 32-bit is, well, 32-bit,
breaking the build.
There doesn't seem to be a good way to printf a Vulkan handle
cross-platform-ly, and it's unlikely to actually be useful,
so just don't print it at all.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36231>
Reduce number of possible allocations and remove unnecessary
memory initialization which will be overwritten immediately.
Test: dEQP-GLES31.functional.ssbo.layout.*
Reviewed-by: Marcin Radomski <dextero@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35768>
This should help vk-icd-loader skip libraries of the wrong bit width.
Reviewed-by: Eric Engestrom <eric@igalia.com>
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35567>
There's no need for a per driver HMI implementation since the
vk_icdGetInstanceProcAddr implementation can well populate the required
entrypoints for Android icd.
Changes have to be done in this single commit for simplicity. Otherwise,
I would have to create a separate android shared library in the runtime
like how vk_instance is handled today, so that the target is able to
check per driver enablement def. However, after all drivers have
migrated over within this MR, we still have to clean those up. So I
decided to just do those in a single commit instead.
v2: avoid preloading u_gralloc in vulkan hal open
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> (v1)
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35561>
Per <hardware/hwvulkan.h>, the hw_device_t::close() function is called
upon driver unloading. The behavior has been like this since Android 10.
Acked-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35561>
Will be used by Cuttlefish CI/CD, but also commonly used by
gfxstream developers.
meson setup gfxstream-build -Dvulkan-drivers="gfxstream" -Dgallium-drivers="" -Dvirtgpu_kumquat=true -Dopengl=false -Drust_std=2021
In distros, virtgpu_kumquat is not expected to be packaged since
it is a CI/CD tool.
Reviewed-by: Aaron Ruby <aruby@qnx.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35210>
So Goldfish doesn't really have /dev/dri/renderN128
in it's system image, but somehow this worked in the
past?
Probably since VirtGpuDevice::getInstance() never
returned a nullptr in the past, and now it does.
Add nullptr checks to fix the crash.
TEST=emulator -feature GuestAngle -gpu lavapipe boots
to UI on main
Reviewed-by: Marcin Radomski <dextero@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35580>
When `VkFramebufferCreateInfo::flags` has
`VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT`,
`VkFramebufferCreateInfo::pAttachments` should be ignored.
Test: cvd create \
--gpu_mode=gfxstream_guest_angle_host_swiftshader \
--gpu_renderer_features="VulkanSnapshots:enabled"
Reviewed-by: Marcin Radomski <dextero@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35580>
For Linux guests, remove syncFd semaphores entirely, as they are waited
on/signaled using the guest-side syncFd operations, and coarse
queueWaits only.
Reviewed-by: Gurchetan Singh <gurchetansingh@google.com>
Reviewed-by: Marcin Radomski <dextero@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34751>
Otherwise, errors such as:
GrallocEmulated.cpp:101:29: error: use of undeclared identifier 'VIRGL_FORMAT_R8G8B8A8_UNORM'
101 | .virglFormat = VIRGL_FORMAT_R8G8B8A8_UNORM,
| ^
GrallocEmulated.cpp:117:29: error: use of undeclared identifier 'VIRGL_FORMAT_B8G8R8A8_UNORM'
117 | .virglFormat = VIRGL_FORMAT_B8G8R8A8_UNORM,
happen in Android builds. Add it to VirtGpu.h to fix this.
Reviewed-by: Marcin Radomski <dextero@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35416>
Swapchain mutable format extension issues were resolved by converting
clear colors on the guest side but that introduced regressions on other
clear related CTS test. Limit clear color modification to images with
Android native buffer structure, to filter swapchain images better.
Test: dEQP-VK.image.*
Reviewed-by: Marcin Radomski <dextero@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35416>
... as VkSnapshotApiCallInfo* is a pointer into a
std::vector<VkSnapshotApiCallInfo> which can be invalidated during
multi threaded VK usage when the std::vector<> is resized.
Test: bazel test common/end2end:gfxstream_end2end_tests \
--graphics_drivers=gles_angle_vulkan_swiftshader \
--test_arg="--gtest_filter=*MultiThreadedResetCommandBuffer*VulkanSnapshots*" \
--test_arg="--gtest_repeat=100" \
--test_output=streamed
Reviewed-by: Marcin Radomski <dextero@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35416>
- mSize should be used rather than kHostVisibleHeapSize.
Since kHostVisibleHeapSize is bigger, OOB errors occured.
- Freeing memory did not work. Look up via the offset
rather pointer. Complain loudly if offset isn't found.
- Align to 4096 bytes (page size), like the old AEMU suballocator
flow.
TEST=set deferredMapping = 0 and test via Cuttlefish
Reviewed-by: Marcin Radomski <dextero@google.com>
Reviewed-by: Aaron Ruby <aruby@qnx.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35323>
.. for api level 1.1, 1.2 and 1.3.
Also allow VK_KHR_external_memory_fd for non-Android Linux, as the
related functionalities are only implemented for that case.
Bug: 389637419
Test: dEQP smoke tests && CI
Reviewed-by: Marcin Radomski <dextero@google.com>
Reviewed-by: Aaron Ruby <aruby@qnx.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35323>
The logic is implemented on the guest side as we need special handling
for swapchain and device memory handle types.
Memory handles for coherent memory allocations on the guest can map into
a single handle on the host side, which makes it infeasible to pass
through the extension functions with handle remapping. Swapchain handles
are not passed to the host driver, so we need to keep a separate table
for them as well. Instead of separating the logic based on the handle
type, we manage all the private data set/get calls on the guest side
without encoding the commands to the host.
Test: dEQP-VK.api.object_management.private_data.*
Test: dEQP-VK.wsi.android.swapchain.private_data.*
Reviewed-by: Marcin Radomski <dextero@google.com>
Reviewed-by: Aaron Ruby <aruby@qnx.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35323>
