Rename it to VN_CS_ENCODER_INITIALIZER_LOCAL to make it clear that it is
only for local variables.
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/10437>
This will allow us to change vn_instance_submit_command without
regenerating the headers.
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/10437>
The combination of removing bottlenecks in userspace (userspace fences,
etc) and slow GPU results in hitting full ringbuffer on a306. Haven't
figured out a reasonable way to work around that in userspace until a
kernel fix is in place, so disable this one for now.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
Move the submit ioctl to it's own thread to unblock the driver thread
and let it move on to the next frame.
Note that I did experiment with doing the append_bo() parts
synchronously on the theory that we should be more likely to hit the
fast path if we did that part of submit merging before the bo was
potentially re-used in the next batch/submit. It helped some things
by a couple percent, but hurt more things.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
A more direct solution would be for bo's to have a reference to the
device. But bo's are ref/unrefd more frequently.
This avoids async submits unrefing a bo after the device handle-
table is freed.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
This is a bit ugly, but with userspace fences and deferred submits it is
better to poll until the timestamp buffer is ready. Otherwise we could
be triggering deferred submits to flush sooner than they normally would,
changing the behavior of what we are trying to measure.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
For submits flushed with (a) no required fence, and (b) no externally
visible effects (ie. imported/exported bo), we can defer flushing the
submit and merge it into a later submit.
This is a bit more work in userspace, but it cuts down the number of
submit ioctls. And a common case is that later submits overlap in the
bo's used (for example, blit upload to a buffer, which is then used in
the following draw pass), so it reduces the net amount of work needed
to be done in the kernel to handle the submit ioctl.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/19
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
For deferred submits, we still need to do the prep steps immediately,
but the ioctl construction can be deferred.. prepare for that by
splitting the prep out.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
Now that we have some bo state tracking for userspace fences, we can
build on this to add a way for the pipe implementation to defer a submit
flush in order to merge submits into a single ioctl.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
Move everything into a struct assocated with the pipe_fence_handle, so
that the drm layer can fill in the seqn/fd fences directly.
This will give us a comvenient place to insert a util_queue_fence in the
next commit.
While we're at it, extract the uint32_t fence (previously called
'timestamp' in place, a kgsl legacy) into a struct that encapsulates
both the kernel fence and the userspace fence.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
Without deferred flushes, we would track the last_fence any time when
there had be no rendering/etc since the last flush, to avoid creating
empty batches simply for the purpose of getting a fence. We can do the
same thing with deferred flushes, where the fence is created from the
frontend thread before we know what will be flushed, by repopulating
the pre-created fence.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
We only need fences for batches flushed via pctx->flush(). This will
serve as a useful signal about submits that can be merged.
v2. disable this optimization pre-a6xx until I can debug issues that
submit merging exposes
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
In the common case, a bo will have no more than a single fence attached.
We can inline the storage to avoid a separate allocation in this case.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
Add a per-fd_pipe fence "timeline" so we can detect cases where we don't
need to call into the kernel to determine if a fd_bo is still busy.
This reuses table_lock, rather than introducing a per-bo lock to protect
fence state updates because (a) the common / hotpath pattern is to
update fences on a lot of objects, but checking the fence state of a
single object is less common, and (b) because we already hold the table
lock in common spots where we need to check the bo's fence state (ie.
allocations from the bo-cache).
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
There are a couple cases where we want to use _NOSYNC, but at the same
time we want to ensure that rendering related to a bo is actually
flushed.
This doesn't do anything yet, but when we start deferring/merging
submits we'll need a way to trigger anything deferred to flush.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
Most of them were actually unused. The memory type (KMEM vs SMI) only
applied to very old a2xx era devices that had a small/fast stacked
memory (SMI) vs normal memory (KMEM). And the cache flags are ignored
(ie. everything is writecombine), but we can add new cache flags later
when they actually do something.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10444>
Use the same generators as used in anv driver so both Vulkan and OpenGL
drivers can share the same external memory objects.
Signed-off-by: Rohan Garg <rohan.garg@collabora.com>
Reviewed-by: Eleni Maria Stea <estea@igalia.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5594>
The s8x24 is a packed group in the format, and for packing we properly
write our s8 in the low bits of the 32, but for unpacking we tried
treating it as a byte array.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7776>
Before we can fix big endian, we kind of have to agree on how pixel data
should be laid out in memory.
For this we have to use packed data for !is_array declared formats and per
element values for is_array ones.
Also we need proper PACKED_1xn definitions for BE.
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7776>
So one less python generator. Based on anv (MR#8792) and radv
(MR#8900).
With this change v3dv doesn't have any more a custom python code
generator.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10484>
There is no point in having a write to an attachment enabled when there
is no corresponding output in the shader. Per VK spec it is an UB,
however a few apps depend on attachment not being changed if FS doesn't
have corresponding output.
Fixes water in Genshin Impact.
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10489>
It's already implemented and it works quite fine on GFX10+, so it
should be safe to enable it on GFX8-9.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10488>
Add
dEQP-VK.synchronization.op.single_queue.binary_semaphore.write_copy_buffer_read_ssbo_compute.buffer_16384
to the flake list.
v2:
- Add a new flake (jasuarez)
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10479>
The idea behind this assert is that if a buffer is in
bufmgr->handle_table it's because it has been shared from i965 to the
outside. This is when we add the drm FD associated to this BO to
bo->exports.
But we also import buffer from the outside into i965 and those buffers
don't have an associated drm FD added to bo->exports.
If you import the same buffer more than once, you'll run into this
assert.
v2: Also drop assert from brw_bo_gem_create_from_name() (Ian)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 57e4d0aa1c ("i965: fix export of GEM handles")
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10386>
This is spun on in cmd execution so make sure it doesn't get
optimised or cached out.
Fixes some timeouts in dEQP-VK.api.command_buffers.record_simul_use_primary
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10494>
Note that it's no longer sufficient to check for >=1 sampler/image
in a potential array, because unbounded arrays have 0 of them.
Reviewed-by: Enrico Galli <enrico.galli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10298>
For GL, we've previously mostly ignored the binding property for sampler variables
during the shader compilation step. For CL, our image bindings were always 0-based as well.
Now, for Vulkan, we are going to be getting explicit bindings and need to emit DXIL that
respects those bindings. Since Vulkan can also have both split and combined images and samplers,
we now need to be smarter about recognizing when NIR is trying to use a "sampler" as *both* an
image and sampler (in deref mode, the same variable will be deref'd as both image and sampler).
That "being smarter" bit comes next, but first, let's prep GL for building correct root
signatures and binding the resources correctly.
Reviewed-By: Bill Kristiansen <billkris@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10298>
Currently, a resized window results in an VK_ERROR_OUT_OF_DATE_KHR. But,
this seems to go against the spec, which suggests returning
VK_SUBOPTIMAL_KHR if the surface extent no longer matches the window
extent. (It does not *require* the WSI actually scale the contents, only
encourages doing so "if possible")
More to the point, VK_SUBOPTIMAL_KHR is the correct status to return if
the client *can* continue presenting to this swapchain, even if the
result will look wrong / "suboptimal". It's the client's decision on
when exactly to recreate the swapchain in this case, and the client may
very well end up wanting to continue presenting for a frame or two, to
prevent flooding the driver with swapchain recreations (which can result
in the window appearing visually "frozen" while resizing).
This change makes live resizing *significantly* smoother in e.g. mpv.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3196>
If the swapchain already does not match the current geometry at the time
of creation, set the initial status to VK_SUBOPTIMAL_KHR as a way to
inform clients to please resize the swapchain.
This solves a race condition where users have no way of knowing if the
surface extent changed in between the user querying the current extent
and creating the swapchain, and might end up missing a resize event
because of it, since XCB_PRESENT_CONFIGURE_NOTIFY is not yet being
handled.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3196>
This is similar to the transcode_etc flag in that it changes the ASTC
fallback (when present) to use DXT5 instead of RGBA8888. This reduces
the memory footprint of the app at the expense of a bit of correctness.
Because it's not quite correct, it's hidden behind a driconf option.
Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10476>
Fixes a double-free in dEQP-VK.wsi.display_control.register_device_event
where the fence that we destroyed got destroyed again. Leaving it unset
in the error path leaves the test's NULL in place.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10473>
Fixes dEQP-VK.spirv_assembly.instruction.compute.variable_pointers.dynamic_offset.select_descriptor_array
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9523>
Make it create a vec2 instead of a pointer, like what the NIR expects.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9523>
I mixed up the EXT and ARB version of the extensions; we actually do
require shaderStorageImageWriteWithoutFormat as well here.
Thanks to Ilia Mirkin for pointing this out.
It also seems I got really confused about what was required when writing
the docs, so let's fix that as well.
Fixes: 341332b23a ("zink: correct image cap checks")
Fixes: ecac7f3da1 ("docs: add missing zink-requirement")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10463>
This fixes linking failures with new GL45 linkage tests, no
regressions spotted on existing tests.
v2: add spec reference (Samuel)
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10373>
Sometimes, there's a surplus of semicolons, and there's enough to go
around for everyone. So maybe you'll save a few away for later. But in
the big pictures, it seems better to get rid of these so we don't
overrun the semicolon quota.
Or whatever, I'm just cleaning up here.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10466>
It might change the NGG code on GFX10+.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10449>
This commit makes mir_pack_swizzle set rep_low, rep_high and half in a
more explicit way, helping in the transition to a unified expand_mode
enum in the following commit.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9461>
This code is taken from src/freedreno/isa/decode.c.
Since we need a similar function in panfrost, it's probably good to move
it to utils.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Acked-by: Rob Clark <robclark@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9461>
This fixes a bunch of MSAA related CTS regressions. This restores
previous behaviour on GFX9 but it should be fixed properly.
Cc: 21.1 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10374>