immediates_count and immediates_size are supposed to have the same
units, but it was only incrementing immediates_count by 1. While we're
here, also fix the case where constants are specified out-of-order.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10291>
Otherwise we can end up in situations like having divide-by-zero. If the optimization is smart enough
that we end up with a *constant* divide-by-zero, then the DXIL validator will fail to sign, which
can trigger fatal errors with CLOn12.
We want to run an initial translation of all kernels during program build, but at that point we don't
know the local size to be able to specify it through kernel specialization data.
v2: Metadata output of 0 is used to indicate that the size wasn't explicitly specific. Copy the
size to the metadata before overriding it to (1,1,1). If conf was explicitly specified,
update the metadata again (though nobody should be paying attention to it).
Closes: https://github.com/microsoft/OpenCLOn12/issues/20
Closes: https://github.com/darktable-org/darktable/issues/8700
Reviewed-By: Bill Kristiansen <billkris@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10303>
Certain games create and destroy lots of resources without binding them.
This can take quite a bit of time and even create unneeded
synchronization. However, we know that if a resource was never bound to
anything, it can be cached. This change does that.
Counting the number of uncached allocation with a tabletop simulator trace:
Before: 2967 uncached allocations over the replay
After: 24 uncached allocations over the replay
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10225>
This is going to make it easier to implement the custom border-color
extension.
While we're moving the code, tweak the memset code a bit, so we don't do
any float-ism in the int-case. It doesn't change anything functionally,
just makes it slightly clearer what's going on here.
Reviewed-by: Joshua Ashton <joshua@froggi.es>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10320>
tcs user vars are var_size[32], which isn't actually how many slots they need,
just how big the variable is (oops), so this needs to be divided
by MAX_PATCH_VERTICES to get the real slot count
slot mapping has always been broken for all tcs inputs, but this probably fixes
all of the related issues there, including unlimited crashes when playing Tomb Raider
Fixes: 2d98efd323 ("zink: pre-populate locations in variables")
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10269>
nir_convert_from_ssa and assign_io_locations both modify this unconditionally,
the latter of which possibly re-modifies variables in ways that can break the
slot map and cause stack overflows during vk driver pipeline compilation
Fixes: 2b4609b66c ("zink: run nir_convert_from_ssa last during compile")
Fixes: 2d98efd323 ("zink: pre-populate locations in variables")
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10269>
This cap should only ever be emitted for fragment-shaders, but we
accidentally emit it for all shaders. Let's tighten the check to avoid a
validator warning when emitting non-fragment shaders without support for
VK_EXT_shader_stencil_export.
Fixes: 8724d4fb36 ("zink: check shader stencil output")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10249>
With the new Vulkan Video extensions a bunch of enum values have been
added. The problem is that those are behind #ifdef so our generated
code complains that we using enum values not defined.
Since we're not using enum names but integer values, we can silence
all those by casting the enum type to int64_t.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10304>
The prog_to_nir->NIR-to-TGSI change ended up causing regressions on r300,
and svga against r300-class hardware, because nir_lower_uniforms_to_ubo()
introduced shifts that nir_lower_ubo_vec4() tried to reverse, but that NIR
couldn't prove are no-ops (since shifting up and back down may drop bits),
and the hardware can't do the integer ops.
Instead, make it so that nir_lower_uniforms_to_ubo can generate
nir_intrinsic_load_ubo_vec4 directly for !INTEGER hardware.
Fixes: cf3fc79cd0 ("st/mesa: Replace mesa_to_tgsi() with prog_to_nir() and nir_to_tgsi().")
Closes: #4602
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10194>
This has been confirmed to fix sporadic graphics corruption on Gen12
platforms for a number of workloads (including Heaven, Valley and
CS:GO among others). Corruption seems to occur during context switch
fairly consistently, but unfortunately this problem doesn't seem to be
documented. Until the hardware team comes up with a better
workaround, fix the problem by reemitting constants at the beginning
of each batch.
No corruption has been observed so far in GL due to preemption,
however this is a possibility to keep in mind, it may be necessary to
disable preemption in addition to this patch in order to fully address
this problem (see also 81201e4617).
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4412
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4454
Cc: mesa-stable
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Previously UBOs only supported static indices, and SSBOs only
supported dynamic indices. UBO support for descriptors was added
as an alternative to static indices, but the logic for detecting
descriptors to SSBOs couldn't just differentiate on constants vs not.
Add a helper which can differentiate cleanly across the board and
handle pre-created handles from descriptors, or static/dynamic raw
indices.
Reviewed-by: Enrico Galli <enrico.galli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10149>
Instead of doing all of the handle logic in the descriptor load, split
it so that the resource index is actually computed during resource_index
processing, and it's converted to a handle during the load_descriptor.
At the same time, add SSBO handling and dynamic indexing handling.
Reviewed-by: Enrico Galli <enrico.galli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10149>
