We scale the actual rendering by patching the viewport state. This is
helped by a HW bit to make the viewport index equal to the view index,
so that we can have a different scaling per-view.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20304>
FDM is implemented pretty much entirely inside the driver, by patching
various structures for each bin. This adds the core infrastructure to
sample the density map, compute the scaled bin sizes we will use, create
patchpoints, and apply them at the start of each bin before executing
the IB2.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20304>
In order to patch the command stream on the gpu, we need two features:
1. The ability to use a read-write BO instead of a read-only one, when
patching might be performed.
2. The ability to get the iova of the current position after reserving
some number of dwords, even with externally-allocated command
streams.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20304>
This is similar to old gens which couldn't support loading from GMEM
automatically. It will be needed for loads with a fragment density map,
because we need to scale the image when loading to GMEM.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20304>
Different uses in various registers and the texture descriptor have
different shifts, and we already had a few ugly workarounds to handle
this. Remove the foot-gun by specifying it in bytes and letting users
handle the shift themselves using the correct macro.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20304>
Otherwise accesses to non-0 views of input attachments may be considered
out-of-bounds and return 0. This should've been removed when enabling
multiview for GMEM, not sure how it was missed.
Fixes: def56b531c ("tu: Support GMEM with layered rendering and multiview")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20304>
to comply with ES2+ line clipping rules, guardband clipping should be
used so that the rasterizer will clip lines without using clip planes
fixes (llvmpipe):
dEQP-GLES*.functional.clipping.line.wide_line_clip_viewport_center
dEQP-GLES*.functional.clipping.line.wide_line_clip_viewport_corner
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17284>
Below is the summary from the power validation.. "it looks like the only
workload where I see savings from DCC is PLT and it is only about 65mW
which is just run to run variation. For Idle I am seeing ~280mW increase
in power, ~200mW increase for power_VideoCall, and ~80mW increase for VP"
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22771>
Matching the original %016lx, and the "0x" prefix which is still
in the format string.
Fixes: 53b77a8102 ("anv: remove 48bit address space checks")
Signed-off-by: Christopher Snowhill <kode54@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22882>
Commit bb8e31b7ed ("anv: avoid hardcoding instruction VA constant in
shaders") had a slight negative impact on shaders (Red Dead Redemption
2 in particular). Dropping a few shaders from SIMD32 to SIMD16.
With this change, it brings back all the dropped SIMD32 shaders.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22872>
When an ACTIVE batch takes over the active writer role from a SUBMITTED
batch, the written BO has the syncobj from the latter even though the
writer is the former. This is correct and an intended state, but it
means that then we can't gate the syncobj cleanup in agx_batch_cleanup
on being the active writer, since the SUBMITTED batch won't be.
Fixes: asahi/mesa#18
Signed-off-by: Asahi Lina <lina@asahilina.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22891>
This is just a sanity check, I haven't actually hit this case but if we
ever do something is very broken (e.g. BO refcounting bug).
Signed-off-by: Asahi Lina <lina@asahilina.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22891>
If depth_writemask is set, we need to write depth regardless of whether we run
the depth test, to write out the fixed-function fragment depth. This will matter
when we start honouring these flags.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22891>
This code was originally based on the Panfrost implementation, but has been
improved in a number of ways.
1. Transform feedback programs are dispatched generically with Gallium calls,
rather than emitting something hardware-specific. This is cleaner and
portable to future GPUs.
2. Transform feedback with indexed draws is now fixed, by lowering to an index
buffer pull.
3. Transform feedback with buffer overflows is now fixed, by correctly
bounds checking in transform feedback programs.
4. Transform feedback with strips/fans/loops are fixed, by correctly
tessellating to the underlying primitives as required by OpenGL.
5. Transform feedback with QUADS is fixed, by tessellating to triangles as
required by OpenGL.
That said, the code is still not in its final form.
1. It still does not support indirect draws. This will require a substantial
overhaul to do tracking on the GPU instead of the CPU. Currently we force
unroll indirect draws (slow but kosher in GL, treif in Vulkan). This isn't
hard to solve but I'm not going to duplicate the code until the algorithms
are otherwise complete because it's a lot easier to hack on the CPU versions
than the GPU versions.
2. It still does not support primitive restart. This has especially nasty
interactions with transform feedback. Again we force unroll to non-primitive
restart forms, again slow but kosher in GL but treif in Vulkan. This is a lot
harder to deal with. I sketched out something really nasty in my notebook
(hinging on efficient GPU prefix sums) but I'm not in a hurry to type this
out.
3. There will be interactions with geometry and tessellation shaders and I don't
think I can get the core code here future-proofed without actually bringing
up the new shader stages.
As such, this is a hard fork of the panfrost code for now, I'm not trying to
share the code (although it *would* clear out almost all of panfrost's transform
feedback related piglit failures).
Passes dEQP-GLES3.functional.transform_feedback.* and most of the relevant
piglits.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22891>
This shortcuts all headaches about how big this should be. It does increase
memory usage a bit if there are lots of shader variants compiled, but this
should be tolerable, and can be optimized later if so required. Thanks to the
previous commit, the disk cache size should be unaffected.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22891>
Only store the push ranges we actually need, not all of them. This should save
some disk space, while insulating us to MAX_PUSH_RANGES changes.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22891>
RA asserts this, but by then if you've messed it up, the failure is inscrutable.
Let's check it in the validator for more pleasant debugging.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22891>
We were being sloppy with the sizes before. It mostly worked out, but there were
some corner cases where we would end up with mixed sized collects and that won't
end well for us. Let's rework the logic to make all the sizes explicit in NIR --
32-bit for depth and 16-bit stencil -- and then do the needed promotions to make
it happen in the AGX IR side.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22891>
In case .x isn't read, it'll be null which has the wrong size and will fail
the validation added later in this series. We fix this by padding with sized
undefs (something that exists of defined size but undefined value) rather than
nothingness (of undefined size).
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22891>
Sometimes a shader might index with a non-power-of-two stride. For example, if
it's indexing into an array of structures where the structure size is not a
power of two, we'll get a multiply with a constant as opposed to a shift. We
want to handle these cases, too. To do so, we generalize our pattern matching to
look for any kind of multiply (with our new helper), rather than hardcoding
logic for ishl. This eliminates right-shifts in a pile of compute shaders, which
makes me happy from a "I read lots of shader assembly when debugging"
perspective.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22891>
