When caching NIR, it's cached as a raw object because we cache the
serialized NIR. When it's then loaded from the disk cache later, we
fail to deserialize it because raw objects are a special case. There
are two callers of vk_pipeline_cache_object_deserialize(), one of which
has a special case for raw objects and the other is called only when
we've checked that it isn't a raw object. The special cases are
pointless; raw objects should deserialize themselves.
Fixes: 591da98779 ("vulkan: Add a common VkPipelineCache implementation")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16281>
We already had a little workaround for v3dv where, for some if its meta
ops, it had to bind a depth/stenicil image as color. Instead of
special-casing binding depth/stencil as color, let's flip on the
drier_internal flag and get rid of most of the checks in that case.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16376>
This code just got duplicated a lot. There is still more, but the
remaining instances do a bit more than just removing other functions.
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16348>
I'm running into crashes because cache->cache_object is uninitialized.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 591da98779 ("vulkan: Add a common VkPipelineCache implementation")
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16275>
Due to both Lavapipe on Windows and Dozen, we need to support MSVC in
the shared Vulkan code. So let's make sure we compile with the
compatibility flags for it.
Techinically speaking, we also need this in the wsi subdir, because we
also compile wsi_common_win32.c with MSVC. But wsi_common_wayland.c
contains void-pointer arithmetic, causing compiler errors if we do.
Fixing that properly is a bit more involved, because Meson doesn't love
passing different compiler arguments per source-file. The alternative is
to remove the void-pointer arithmetic, but that seems a bit pointless as
this code will never be compiled on MSVC.
So, let's leave that one out for now. We can probably do better in the
future, but this gets us a step further.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6386
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16162>
Most of the time when the logging code is invoked, it means we're
already in an edge case. It should be as robust as possible, otherwise
we risk making hard to debug things even harder. To that end, instead
of blowing up if passed a NULL object on the list, handle it as
gracefully as we can.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16107>
This is a bit more accurate for what's going on and, while all Mesa
drivers today seem to be ok with extra bits, ensures we're passing a
valid Vulkan thing.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16145>
My understanding of the signal masks is that they control what stages
must complete before the semaphore is signaled. Using 0 theoretically
means the semaphore could be signaled immediately without waiting on
anything. Use ~0 instead to say it depends on everything.
Fixes: 97f0a4494b ("vulkan: implement legacy entrypoints on top of VK_KHR_synchronization2")
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16145>
alignof(void) is a non-standard GCC extension, and it doesn't compile on
MSVC. But since the Windows CI has been disabled due to stability
issues, a breakage snuk in nevertheless.
Since alignof(char) works the same as alignof(void), let's pass char
instead of void here. That hides the GCC weirdness without doing any
functional changes.
Fixes: 591da98779 ("vulkan: Add a common VkPipelineCache implementation")
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16134>
This is partly copied+pasted from ANV but is mostly new code with lots
of reference counting bugs fixed (I hope!). The new cache caches
"object" which derive from a base vk_pipeline_class_object struct. It
uses a kernel-style "ops" interface for virtual methods on these objects
to allow for easy destruction (when the reference count hits zero) as
well as serialization an deserialization interfaces. This should allow
drivers to cache basically whatever they want without having to think
too hard about the details.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13184>
Accessing queue->submit.thread_run without holding lock vk_queue.submit.mutex.
Elsewhere, vk_queue.submit.thread_run is accessed with vk_queue.mutex
held 1 out of 2 times (1 of these accesses strongly imply that it is necessary).
Found by Coverity.
CID: 1503334
Signed-off-by: Illia Abernikhin <illia.abernikhin@globallogic.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15736>
VkObjectType and VkDebugReportObjectTypeEXT has the same enum-values.
Why the Vulkan WG thought this was a good idea, beats me. But it's what
we have to live with now.
Anyway, instead of having a statement that implicitly casts two
different values from the former to the latter, let's fully relsove the
type as the former, and cast the value when using it instead.
Fixes: 41318a5819 ("vulkan: Use vk_object_base::type for debug_report")
Acked-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15547>
The v3dv kernel driver doesn't support timelines yet but we want
threaded submit and that requires WAIT_PENDING. Fortunately, it should
never sit in this loop for long in practice. The primary use-case is
sorting out dependencies and these checks will always trivially succeed
for non-shared semaphores because v3dv only has a single queue.
Acked-by: Alejandro Piñeiro <apinheiro@igalia.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15704>
What we really care about is if we're DEFERRED so we need to do a flush
and if there can be any other threads we might race against. We don't
really care about the timeline mode itself.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15566>
Instead of basing everything on the timeline mode, base it on the submit
mode of the queue. This makes a lot more sense since it's what we
really care about anyway.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15566>
Now that we have a threading mode in the device, we can set that based
on the environment variable instead of delaying it to submit time. This
allows us to avoid the static variable trickery we use to avoid reading
environment variables over and over again. We also move the enabling of
the submit thread up a level or two and give it a bit more obvious
condition.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15566>
This encapsulates all three possible submit modes: immediate, deferred,
and threaded. It's more clear than the has_thread boolean combined with
device-level checks.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15566>
Fix 'error C4576: a parenthesized type followed by an initializer
list is a non-standard explicit type conversion syntax' errors by
declaring an actual variable and returning it in
vk_image_view_subresource_range().
All those MSVC/c++ related-constraints are quite annoying to be honest,
but it looks like the D3D12 headers have been updated to plain C
recently, which will allow us to write the driver in C, and hopefully
get all this sort of issues behind us.
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14766>
VK_SYSTEM_ALLOCATION_SCOPE_COMMAND is used for transient allocations that
are not expected to live outside the vkXxx(). Use
VK_SYSTEM_ALLOCATION_SCOPE_OBJECT for cmd_entry allocations.
v2 (Jason Ekstrand):
- Also fix the manually typed entrypoints in vk_cmd_enqueue.c
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14406>
These check the command buffer level and enqueue the command if it's not
a primary but uses vk_device::command_dispatch_table for primaries.
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14406>
These helpers are used by vkCreateGraphicsPipelines to get the
VkPipelineRenderingCreateInfo and in vkCmdBeginCommandBuffer to get the
VkCommandBufferInheritanceRenderingInfo. This is required because the
Vulkan runtime code can't yet hook and modify calls made to driver-
provided functions. Instead, we just provide a helper to be used in leu
of vk_find_struct_const(). The structs themselves are stored in the
render pass so we can pass back a pointer and there's no need to
construct one on the stack or stuff it in the pipeline.
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14961>
This implements vkCmdBeginRenderPass, vkCmdEndRenderPass, and
vkCmdNextSubpass in terms of the new vkCmdBegin/EndRendering included in
VK_KHR_dynamic_rendering and Vulkan 1.3. All subpass dependencies and
implicit layout transitions are turned into actual barriers. It does
require VK_KHR_synchronization2 because it always uses the 64-bit
version of the pipeline stage and access bitfields.
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14961>
