Heavily based on the already existing for the v3d OpenGL driver, but
without references, and with some extra OOM checks (Vulkan CTS has
several OOM tests).
With this commit v3dv_bo_alloc and v3dv_bo_free became frontends to
the bo_cache. The former tries to get a BO from the cache if possible,
and the latter stores the BO on the cache if possible. The former also
adds a new parameter to point if the BO to allocate is private.
As v3d we are only caching private BOs, those created by the driver
for internal use (like CLs, tile_alloc, etc). They are the ones with
the highest change of being reused (for example, CL BOs are always
4KB, so they can always be reused). User-created BOs can have any
size, including some very large ones for buffers and images, which
makes them far less likely to be reused and would add a lot of memory
pressure if we decided to cache them.
In any case, in practice, we found that we could get a performance
improvement by caching also user-created BOs, but that would need more
care and an analysis to decide which ones makes sense. Would also
require to change how the cached BOs are stored by size. Right now
there are an array of list_head, that doesn't work well with big
BOs. If done, that would be handled on a separate commit.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
We had done all the plumbing for this but EZ can be disabled in 3 places
and we were never setting the enable bit in the configuration bits packet.
Also, configuration bits must not enable EZ if this has been disabled in
the RCL for the whole frame, which we do if we don't have a depth
attachment at all.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Otherwise cloned BO lists point to the original list objects and not
the cloned ones, and that will confuse anything that tries to iterate over
them, such as list_length(), leading to infinite loops.
Fixes (in debug mode):
dEQP-VK.api.command_buffers.render_pass_continue
In that test we clone a full CL job from a secondary, and without this,
the BO lists in its CL lists will point to the bo_list field in the
original job, leading to an infinite loop as we assert the expected size
of these lists at queue submit time in handle_cl_job.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Asserting on them makes it easier to identify applications and tests that
try to use unimplemented features.
Also, there are some APIs that relate to optional features we don't
(or can't) support, such as sparse, so for these we just provide
the trivial implementation.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
We were assuming that if the command buffer state doesn't have any
attachments (as per the attachment count) the attachment state array
should not be allocated, however, during meta operations it is
possible that the attachment state grows (since meta operations can
emit render passes of their own). In that case, we would grow the
state for the meta operation but then pop the previous attachment
count and we would leak the state.
An example of that is a secondary command buffer which has no
attachment state by default since it doesn't execute a render pass
begin, but that executes one in a meta operation (for
vkCmdClearAttachments for example).
Fix this by making the attachment count an allocation count instead
and not popping it once we finish a meta operation. Also, always free
the state so long as there is a valid pointer, and assert that the
allocated count is not zero in that case.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
The main change we are introducing here is that now we allow secondary
command buffers that execute in a render pass to have a job list with
more than one job.
The main issue with vkCmdClearAttachments is that we currently need
this to spawn multiple jobs to clear multilayered framebuffers, as we
need to setup a different 2D framebuffer for each layer to clear and
therefore emit a different RCL for each. We could avoid this
completely by used layered rendering with the "clear rect" path to
redirect the clear rects to appropriate layers of the primary
framebuffer, however, our hardware only supports layered rendering
with geometry shaders, which we don't support at present.
Because vkCmdClearAttachments relies on having framebuffer state
available (something we would not need if we used the geometry shader
implementation), if this is not available in the secondary we need to
postpone emission of the command until the secondary is executed
inside a primary. We do this by using a new CPU job
V3DV_JOB_TYPE_CPU_CLEAR_ATTACHMENTS that is processed during
vkCmdExecuteCommands by calling vkCmdClearAttachments directly in the
primary.
As a consequence of these changes, it is now possible that a secondary
command buffer that runs inside a render pass have any kind of job in
its job list, including partial CLs that need to be branched to and
full CLs that need to be submitted to the GPU as is, so we introduced
a new GPU job type V3DV_JOB_TYPE_GPU_CL_SECONDARY to identify partial
CLs.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Event waits can be safely moved before a render pass start, since
event setting and resetting commands cannot happen inside one. We
don't need to go that far, but we can use this to record the wait
in its own separate job and then execute this job before the
binning commands recorded in the secondary command buffer when
we execute the secondary into a primary.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
There are basically two types of scenarios to consider:
- Secondary command buffers that run inside a render pass.
- Secondary command buffers that run outside a render pass.
For the former we want to record their commands into a binning command
list that we can branch to when executed into a primary command
buffer. This means this kind of command buffers don't spawn new jobs,
just the default one where they record the binning commands which
won't include the frame setup, which will be provided by the primary
they will be executed in.
For the latter we don't require anything special, we just record as
many jobs as we need as usual and link that job list from the primary
job list when executed.
This handles most scenarios except:
- vkCmdWaitForEvents
- VkCmdClearAttachments
Both of these can spawn new jobs inside a render pass, which is not
what we want for secondary command buffers. We will address this is
follow-up patches.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
We should always do this. So far we have been getting away with this
because we overallocate at v3dv_job_start_frame, but that won't do
for secondary command buffers for example, it is also unreliable
if CLs grow past that initial allocation.
In the future, we might want to fix our emit macros so they do the
allocation check implicitly, which would simplify the code and would
make this process a lot less error prone.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
During a command buffer reset we call cmd_buffer_init(), which will
add the command buffer to the pool, so make sure we remove it first
and that we use a safe iterator when resetting a pool.
Fixes:
dEQP-VK.api.command_buffers.pool_reset_reuse
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
In order to properly check (and possibly compile) shader variants we
need a pipeline and a compatible descriptor set. So far we were trying
to do that check as early as possible, so we were trying to do it at
CmdBindPipeline or CmdBindDescriptorSets, and a combination of dirty
flags. This showed to not cover all the corners cases, and made the
code complex, as needed to handle cases where the descriptors were not
yet available, and return early. The latter also meant that we were
running several checks that failed in the middle.
This commit moves the variant check to CmdDraw, when we should have a
pipeline and compatible descriptor sets, and simplifies and makes more
strict the existing code.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
This reverts a previous half-attempt at an implementation of events
using a BO to hold the event state, and provides a full
implementation. V3D doesn't have any built-in GPU functionality to
wait on any kind of events, so we need to implement this in the driver
an therefore we no longer need to use a BO for the event state.
Instead, we implement GPU waits by using a CPU job for the wait
operation that spawns a wait thread if the wait operation doesn't have
all its events signaled by the time it is processed. To implement the
semantics of the wait correctly, any jobs in the same command buffer
that come after the wait will not be emitted until the wait thread
completes.
If a submit spawns any wait threads for a command buffer we can't
signal any semaphores for it until all the wait threads complete and
we know that all the jobs for those command buffers have been
submitted. The same applies to the submit fence, if present.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
This is generally very difficult to handle properly everywhere, but
at least this is good enough to make the few CTS tests for this happy.
Fixes (on Rpi4):
dEQP-VK.wsi.xlib.swapchain.simulate_oom.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
There is a hw bug by which the only way to clear the depth/stencil
tile buffers is to emit a clear of all tile buffers, so if we have
to do any such clears, we just emit a single clear of all tile
buffers and skip doing any per-buffer clears, even for color buffers,
since they would be redundant.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
We were declaring the destroy callback function as taking a pointer for the
vulkan object handle and relying on an implicit conversion to the Vulkan
handle type, however that would be incorrect on 32-bit platforms, where
non-dispatchable Vulkan objects (the kind that we may allocate privately during
command buffer recording), are defined as uint64_t, so the signature of the
destry callback type doesn't match the signature of the actual Vulkan
function, leading to bogus results. Fix that by using uint64_t instead.
This fixes compilation warnings and also crashes in some tests when
compiling and executing natively in Rpi4.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
We were pre-packing the constants from the pipeline state and then
always emitting that at draw time, ignoring dynamic state. This makes
it so we don't prepack at pipeline creation time and we always emit
the correct constants directly the command buffer dynamic state.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
So far we were just asserting or aborting if any of the internal
method used during the pipeline creation failed.
We needed to change the return value of several methods, in order to
bubble up the proper memory allocation error.
Note that as the pipeline creation is complex and splitted in several
methods, if an error happens in the middle, it returns back, and rely
on the higher level to call PipelineDestroy. This method needs to take
into account that some of the resources could have not been allocated
when freeing it.
Also note that v3dv_get_shader_variant is used during the pipeline
bind, as with the new resources bound, we need to check if we need to
recompile a new variant. At that moment we are not creating a new
vulkan object so we can really return a OOM error. For now we just
assert on that case.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
This uses the framework to register private commmand buffer objects
that get freed automatically when the command buffer is destroyed by
the application.
This change also moves the descriptor set pool that the meta blit path
uses to allocate descriptors for the blit source textures, from the
device to the command buffer, so we can have a descriptor pool per
command buffer. This is necessary to ensure correct behavior when
doing multi-threaded command buffer recording (alternatively, we would
have to lock around the descriptor set allocation code, which would be
undesirable).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
This allows the driver to register private Vulkan objects it creates as part
of command buffer recording (usually for meta operations) in the command
buffer, so they can be destroyed together with it.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
This needs to be updated everytime we bind a new pipeline, but we can
bind a pipeline and not have an actual job yet, so we want to postpone
this until we actually need to emit CFG_BITS, during the pre-draw
setup.
Also, rename the update helper to be about the job rather the command
buffer, since it is updating state that we track per job.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
We were asserting that we had a valid subpass index, but we can have
meta operations that run outside a render pass, such as for blitting.
If we allow this, then we also need to account for the fact that
pipelines can be bound outside a render pass too.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
While very limited in scope, this might be the most efficient way to blit
when applicable. In fact, we might also want to use this for the image copy
commands when possible instead of the TLB.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
The design for queries in Vulkan requires that some commands execute
in the GPU as part of a command buffer. Unfortunately, V3D doesn't
really have supprt for this, which means that we need to execute them
in the CPU but we still need to make it look as if they happened
inside the comamnd buffer from the point of view of the user, which
adds certain hassle.
The above means that in some cases we need to do CPU waits for certain
parts of the command buffer to execute so we can then run the CPU
code. For exmaple, we need to wait before executing a query resets
just in case the GPU is using them, and we have to do a CPU wait wait
for previous GPU jobs to complete before copying query results if the
user has asked us to do that. In the future, we may want to have
submission thread instead so we don't block the main thread in these
scenarios.
Because we now need to execute some tasks in the CPU as part of a
command buffer, this introduces the concept of job types, there is one
type for all GPU jobs, and then we have one type for each kind of job
that needs to execute in the CPU. CPU jobs are executed by the queue
in order just like GPU jobs, only that they are exclusively CPU tasks.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Most of our state doesn't carry over across jobs, so it needs to be re-emitted.
For example, if we have two render passes running back to back using the
same pipeline, the application could decide to only bind the vertex buffer
or/and the pipeline just once, but as soon as we record the second render
pass and create a new job for it we will need to re-emit the shader state
record for it just because it is a new job.
We could probably only do this for jobs inside a render pass, since those
are the only ones that actually draw something and need to care about
dirty state, however, there is no harm in doing this for all jobs, for the
same reason.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
When testing if we could merge the new subpass into the previous one
We were taking the subpass index from the state (which isn't updated
to the new subpass until a bit later when the job for the new subpass
has been settled). This means that we were doing the merge checks with
the previous subpass, not the current one.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
There are some texture operations (like mipmap query levels) that
doesn't require a sampler. In fact, you should ignore it. So we need
to take it into account when combining the
indexes. nir_tex_instr_src_index is returning a negative value to
identify that case, but as we are using a uint32_t to pack both values
(for convenience, easy to pack/unpack the hash table key), we just use
a uint value big enough to be a wrong sampler id.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
This requires that we emit a specific draw command and that we emit
the base instance if not zero right before the instanced draw call.
Notice that we were already doing this for instanced indexed draw
calls, so here we are only adding this for non-indexed draw calls.
We also need to flag whether the vertex shader reads the base instance
in the shader record (which it will if it reads uses gl_InstanceIndex,
as that is lowered in Vulkan to base_instance + instance_id).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
OpenGL doesn't have the concept of individual texture and sampler, so
texture and sampler indexes have the same value. v3d compiler uses
this assumption, so for example, the texture info at the v3d key
include values that you need to use the texture format and the sampler
to fill (like the return_size).
One option would be to adapt the v3d compiler to handle both, but then
we would need to adapt to the lowerings it uses, like nir_lower_tex,
that also take the same assumption.
We deal with this on the Vulkan driver, by reassigning the texture and
sampler index to a combined one. We add a hash table to map the
combined texture idx and sampler idx to this combined idx, and a
simple array to the opposite map. On the driver we work with the
separate indices to fill up the data, while the v3d compiler works
with the combined one.
As mentioned, this is needed to properly fill up the texture return
size, so as we are here, we fix that. This gets tests like the
following working:
dEQP-VK.glsl.texture_gather.basic.2d.depth32f.base_level.level_2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
We must update our check for whether the render area is tile-aligned for
each subpass, since the hardware will update tile sizes for each RCL.
Fixes:
dEQP-VK.renderpass.suballocation.attachment_allocation.roll.8
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
This is the same as the subpass start version, only that it won't
emit subpass clears. This is necessary when resuming a subpass
from a partial clear to make sure we don't try to clear subpass
attachments again.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Since a meta partial clear starts a new render pass, we need to store
all state that can be changed with vkCmdBeginRenderPass.
Also, since the meta clear pipeline sets dynamic state, we also
have to restore that.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
The problem with this is that TLB clears always clear and store full
tiles, so if our render area is not perfectly aligned to tile boundaries
we end up clearing all pixels in tiles that are only partially covered.
In this scenario we have to avoid using TLB clears and instead fallback
to clearing by rendering a scissored quad in the clear color, like we do
for partial clears in vkCmdClearAttachments.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
