Re-order things in panfrost_direct_draw() so we have all non-HW specific
stuff done first, and then the descriptor emission.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Makes panfrost_direct_draw() shorter and thus easier to read.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Not strictly needed for CSF, but anything we can do to make
panfrost_direct_draw() shorter is good to take.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
First of all, we want to prefix those with jm_, so we can later add
CSF-specific job emission helpers. We also take this as an opportunity
to clarify what these helpers emit exactly: the draw section of a job
descriptor (suffixed with _draw) or the job descriptor itself (suffixed
with _job).
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Add a jm_emit_fragment_job() helper that does the fragment job emission
so we can re-use the rest of emit_fragment_job() for CSF too.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Prefix the function with jm_, so it's clear its JM-specific helper.
While at it, replace emit by push, since the descriptor emission
happens before that, all this function does is pushing jobs to the
job chain.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
In preparation of CSF support addition, we move any JM related bits out
of panfrost_batch into a panfrost_jm_batch struct that's embedded in
panfrost_batch inside an anonymous union. This way, we can easily
specialize things for CSF without polluting the panfrost_batch object
with CSF-specific fields.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
We will use those when adding CSF support.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
We will use the new panfrost_vertex_attribute_stride() when adding CSF
support.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Factor out the logic so we can use the helper when adding CSF support.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Suggested-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Store varying pointers in panfrost_batch. This will make the
job-frontend split easier by reducing the number of arguments we pass
to functions.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
They are accessible through
batch->{attribs,attrib_bufs}[PIPE_SHADER_VERTEX], which is updated in
panfrost_update_state_3d(), if needed.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
So we can later use it from job-frontend specific hooks without having
to pass the pointer around.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
This per-arch method will take care of preparing the batch for submission
and submitting it.
This allows us to get rid of a bunch of methods in panfrost_vtable, and
turn some indirect function calls into direct calls.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
We always pass ctx->syncobj, which can be extracted from the batch
passed to panfrost_batch_submit_jobs(), so let's just kill this
argument.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
in_sync is always zero, so let's just get rid of the parameter.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Move the panfrost_emit_tile_map() call before any of the per-gen
calls in panfrost_batch_submit(). This is in preparation of moving
the per-gen batch submission logic to a single hook instead of
having multiple indirect calls, and given the tile map doesn't
depend on any of the states modified by the per-gen calls, moving
it before them shouldn't be an issue.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Now that emit_fragment_job() stores the fragment job GPU pointer
in panfrost_batch, we can move this step in panfrost_batch_submit(),
which will help with v10/CSF support.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
In preparation of v10 support, where the fragment job is not passed
as a descriptor, but issued directly on the command stream, we make
->emit_fragment_job() a void method, and store the fragment job
descriptor in the batch.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Now that we count draws and compute jobs, we can change some of the
tests to be HW-agnostic, so they can be re-used when we introduce
support for CSF hardware.
Anything that's related to job submission keeps using the scoreboard
information, because this code will stay JM-specific.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
We want to move the kernel driver specific submission logic to
pan_cmdstream.c where we can define per-gen functions. In order to
do that, we need panfrost_has_fragment_job() to be public.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
We will use this information to make some helpers HW-agnostic so they can
work on CSF hardware.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
We turn the 10000 jobs limit into a 10000 draw limit because:
- on HW supporting IDVS a draw is just a single job, and having a common
limit for all HW is simpler
- draw_count < 10000 fits in the max jobs limit if we assume the worst
case scenario (3 jobs per draw)
- CI seems to be happy (no spurious timeouts after this change)
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
This way we have a HW-agnostic way to know how many draws were issued,
instead of looking into the job chain builder guts.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Tiler context emission will differ on v10. Given pan_emit_tiler_ctx()
was only used in the gallium driver, and its implementation is
relatively simple, inline the code in panfrost_batch_get_bifrost_tiler().
This way we will avoid the churn caused by the function prototype change.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Tiler heap emission will differ on v10. Since pan_emit_tiler_heap()
is only used by the gallium driver, and it's simple enough to be
inlined, move the code to panfrost_batch_get_bifrost_tiler().
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
We are about to add arch-specific context initialization/cleanup hooks.
Rename the existing context_init hook to avoid confusion (this hook
was just populating the pipe_context vtable).
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
panfrost_batch_submit() already has an has_fragment_job() at the
beginning of the function. Store the result in a local has_frag variable
and replace the open-coded test by an 'if (has_frag)'.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26249>
Using the unsized data field is incorrect. The data is located behind
the entire drm_i915_query_perf_config structure.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Rohan Garg <rohan.garg@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26285>
Vulkan shaderdb stats with pattern dEQP-VK.image.*.with_format.*.*:
total instructions in shared programs: 35993 -> 33245 (-7.63%)
instructions in affected programs: 21153 -> 18405 (-12.99%)
helped: 394
HURT: 1
Instructions are helped.
total uniforms in shared programs: 8550 -> 7418 (-13.24%)
uniforms in affected programs: 5136 -> 4004 (-22.04%)
helped: 399
HURT: 0
Uniforms are helped.
total max-temps in shared programs: 6014 -> 5905 (-1.81%)
max-temps in affected programs: 473 -> 364 (-23.04%)
helped: 58
HURT: 0
Max-temps are helped.
total nops in shared programs: 1515 -> 1504 (-0.73%)
nops in affected programs: 46 -> 35 (-23.91%)
helped: 14
HURT: 2
Inconclusive result (%-change mean confidence interval includes 0).
FWIW, that one HURT on the instructions count is for just one
instruction.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25726>
Since v71, broadcom hw include specific packing/conversion
instructions, so this commit adds opcodes to be able to make use of
them, specially for image stores:
* pack_2x16_to_unorm_2x8 (on backend vftounorm8/vftosnorm8):
2x16-bit floating point to 2x8-bit unorm/snorm
* f2unorm_16/f2snorm_16 (on backend ftounorm16/ftosnorm16):
floating point to 16-bit unorm/snorm
* pack_2x16_to_unorm_2x10/pack_2x16_to_unorm_10_2 (on backend
vftounorm10lo/vftounorm10hi): used to convert a floating point to
a r10g10b10a2 unorm
* pack_32_to_r11g11b10 (on backend v11fpack): packs 2 2x16 FP into
R11G11B10.
* pack_uint_32_to_r10g10b10a2 (on backend v10pack): pack 2 2x16
integer into R10G10B10A2
* pack_4x16_to_4x8 (on backend v8pack): packs 2 2x16 bit integer
into 4x8 bits.
* pack_2x32_to_2x16 (on backend vpack): 2x32 bit to 2x16 integer
pack
For the latter, it can be easly confused with the existing
pack_32_2x16_split. But note that this one receives two 16bit integer,
and packs them on a 32bit integer. But broadcom opcode takes two 32bit
integer, takes the lower halfword, and packs them as 2x16 on a 32bit
integer.
Interestingly broadcom also defines a similar one that packs the
higher halfword. Not used yet.
Note that at this point we use agnostic names, even if we add a _v3d
suffix as they are only available for broadcom, in order to follow
current NIR conventions.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25726>
This is loosely based on PAL. This seems to fix 3D PRT support with
RADV on Polaris10. THIN means the tile is a 2D slice. THICK means the
tile is a 3D box.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26260>
No need to handle f2f16 specially for OpenGL, and we can vectorize
f2f16 when using ACO.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25990>
To fix si_compute_blit created nir code compilation with ACO.
Two 16bit vector ops are used in it:
con 16x2 %11 = u2u16 %10.xy
con 16x2 %25 = f2f16 %22.xy
which is not supported by ACO yet.
PS. now ACO supports vec2 f2f16.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25990>
The code in `emit_load_var` that will attempt to read indirect inputs
expects the entire array of inputs to be there. Additionally the code
that populates `bld->inputs_array` will populate the array using the count
of `inputs_read`, without ensuring the inputs it copies are the ones read.
This change populates `bld->inputs_array` with the entire contents of bld->inputs
so indirect reads will always match up.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26153>
