Shmems are allocated internally and are only for CPU access. They can
be easily cached.
Venus have 4 sources of shmem allocations
- the ring buffer
- the reply stream
- the indirection submission upload cs
- one cs for each vn_command_buffer
The first one is allocated only once. The other three reallocate
occasionally. The frequencies depend on the workloads.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14179>
It suballocates from a shmem pool owned by vn_instance. The goals are
to speed up shmem allocations for VkCommandBuffer and to reduce the
number of BOs. Both are crucial when shmems are HOST3D BOs, because
they require roundtrips to the renderer to allocate and they take up KVM
memslots.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14179>
It provides shmem suballocations. It is designed to be used with
short-lived shmems. A long-lived shmem can hold on to some large
allocation while only using a likely small region of the large
allocation.
v2: cleanups suggested by Yiwei
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Ryan Neph <ryanneph@google.com> (v1)
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14179>
Move helpers built on top of vn_renderer.h to the new files.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14179>
The binary search can lead to infinite loop. Fixes
dEQP-VK.api.object_management.alloc_callback_fail.device where
vn_CreateBuffer can always fail.
Fixes: a74f2495ca ("venus: implement vn_buffer_get_max_buffer_size")
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14184>
This change defers the bo allocation for non-external mappable memory
direct allocation.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13874>
bo allocations for the below cases are after device memory allocation:
1. direct non-external mappable memory allocation
2. pool grow
3. exportable memory allocation
For (1) and (2), the bo is for mapping, which is a pure kernel operation
to happen later. So roundtrip waiting can be deferred until free memory.
For (3), the bo is for either fd export or mapping, which are both pure
kernel operations. So roundtrip waiting can also be deferred until free
memory.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13874>
The behavior we stick to is that the base_bo is always created for pool
memory so that to keep it alive during suballocates and frees.
This CL does the below:
1. rename pool_alloc to pool_suballocate and align the api interface
2. rename simple_alloc to pool_grow_alloc to make it pool specific
3. refactor pool_free and simple_free into a pair of pool_ref and
pool_unref to simplify that vkFreeMemory is only called after the
pool bo gets destroyed.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13874>
1. advertise high hit rate cache combinations, and we should limit the
caches to those only require device memory pool alloc
2. use size = 1 to ask for buffer memory requirements so that we do a
sanity check on our assumption of returned size and alignment. For
implementations don't meet our assumption, continue without cache.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13428>
This change estimates the max_buffer_size with quick sort. Try to
avoid some traffic upon device creation time, but not worth adding a
buffer simple create api to avoid the extra requirement query traffic
since this is temporary.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13428>
these were duplicated all over the place, and it's annoying to have to keep
duplicating them any time a new component includes the vulkan header
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13141>
We might reorder vkCmdBindDescriptorSets after
vkDestroyDescriptorSetLayout due to batching, which is likely invalid.
Keep the layouts alive with the sets to defer
vkDestroyDescriptorSetLayout.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13090>
The reference count does not go beyond 1 yet.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13090>
Memory ordering is hard. Add vn_refcount to take care of the details.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13090>
We should not drop VkPhysicalDeviceImageDrmFormatModifierInfoEXT when it
is provided.
Fixes dEQP-VK.drm_format_modifiers.export_import.*.
Fixes: efa185ed5c ("venus: rework external memory capability queries")
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13094>
When it comes to AHB, pExternalBufferInfo->handleType is already
overridden to the renderer handle type. Thus the AHB buffer prop path is
not used. However, this is not caught by cts, vvl or apps because the
host renderer memory features so far satisfy the ahb requirement.
Fixes: ebf0e45506 ("venus: add ahb image and buffer properties query support")
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13076>
It is not an error when a physical device is filtered out. But it is an
error when a physical device fails to initialize.
Fixed
dEQP-VK.api.device_init.create_instance_device_intentional_alloc_fail.
v2: separate success and fail paths
Fixes: 7dfac808 ("venus: no supported device is not an error")
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org> (v1)
Reviewed-by: Ryan Neph <ryanneph@google.com> (v1)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12723>
Move device enumeration to a new function.
v2: separate success and fail paths, rename functions
(suggested by Ryan)
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org> (v1)
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12723>
When the renderer supports largeRing, raise the ring buffer size to 64KB
and raise the direct submit threshold to 4KB. The numbers are randomly
picked.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12568>
Until we can assume large ring support.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12568>
The goal is to remove local_cs_data. To that end, vn_ring_submit is
modified to take a cs than a pointer.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12568>
We can no longer use
vn_call_vkGetVenusExperimentalFeatureData100000MESA which requires the
ring to be initialized.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12568>
Let vn_instance_get_reply_shmem_locked use the renderer if it is called
before the ring is initialized.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12568>
It submits a command stream to the renderer and waits.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12568>
