c78be5da30 ("intel/fs: lower ray query intrinsics") introduced a
helper function using nir_(push|pop)_if which invalidated dominance &
block_index for the replacement of nir_intrinsic_rt_trace_ray.
We can still keep dominance/block_index metadata for the lowering of
nir_intrinsic_rt_execute_callable though.
This change uses 2 different lowering function with correct metadata
preservation.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: c78be5da30 ("intel/fs: lower ray query intrinsics")
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15910>
The intel_perf_query path used for performance queries on GL was
passing a bogus "end" pointer to intel_perf_query_result_accumulate(),
causing it to accumulate garbage values. This was causing the values
of many performance counters to be corrupted.
The "end" pointer was incorrect because the current code was assuming
that different OA reports were located TOTAL_QUERY_DATA_SIZE bytes
apart, which is a hard-coded preprocessor define. However recent
(Gfx12+) hardware generations use a variable query size determined by
the query layout. Use the size derived from it instead, and remove
the stale define.
Fixes: 3c51325025 ("intel/perf: switch query code to use query layout")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15783>
This enables a lower power mode in the sampler hardware in certain
common scenarios. On Tigerlake, SAMPLER_MODE is not programmable by
userspace but the kernel already sets this bit for us.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15628>
While this register still exists, it's no longer a per-context register.
Instead, on Gfx12+, SAMPLER_MODE exists per dual-subslice and is
accessed as a "multicast" register, where you write control which
version is accessed by the "steering control register".
At any rate, userspace cannot write it any longer, and so there's not
much point to it existing in our genxml (which was missing most of the
fields anyway).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15628>
This allows the sampler to perform faster filtering of 8-bit UNORM
textures by filtering them at a different precision. The filtering
is intended to still be OpenGL and DirectX spec compliant.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15628>
DG2 can only do sample_d and sample_d_c on 1D and 2D surfaces. The
maximum number of gradient components and coordinate components should
be 2. In spite of this limitation, the Bspec lists a mysterious R
component before the min_lod, so the maximum coordinate components is 3.
Fixes the following Vulkan CTS failures on DG2:
dEQP-VK.glsl.texture_functions.texturegradclamp.isampler1d_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.isampler2d_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler1d_fixed_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler1d_float_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler2d_fixed_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler2d_float_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.usampler1d_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.usampler2d_fragment
The Fixes: tag below is a bit misleading. This commit fixes some test
cases similar to ones fixed by the Fixes: commit. I just want to make
sure this commit gets applied everywhere that commit was also applied.
Fixes: 635ed58e52 ("intel/compiler: Lower txd for 3D samplers on XeHP.")
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15781>
DG2 can only do sample_d and sample_d_c on 1D and 2D surfaces. Cube
maps and 3D surfaces were already handled, but 1D array and 2D array
surfaces were not.
Fixes the following Vulkan CTS failures on DG2:
dEQP-VK.glsl.texture_functions.texturegradclamp.isampler1darray_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.isampler2darray_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler1darray_fixed_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler1darray_float_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler2darray_fixed_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler2darray_float_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.usampler1darray_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.usampler2darray_fragment
The Fixes: tag below is a bit misleading. This commit adds another
lowering, similar to the one in the Fixes: commit, that probably should
have been added at the same time. I just want to make sure this commit
gets applied everywhere that commit was also applied.
Fixes: 635ed58e52 ("intel/compiler: Lower txd for 3D samplers on XeHP.")
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15681>
according to KHR_gl_texture_3D_image:
If <target> is EGL_GL_TEXTURE_3D_KHR, <buffer> must be the name of a
complete, nonzero, GL_TEXTURE_3D (or equivalent in GL extensions) target
texture object, cast
into the type EGLClientBuffer. <attr_list> should specify the mipmap
level (EGL_GL_TEXTURE_LEVEL_KHR) and z-offset (EGL_GL_TEXTURE_ZOFFSET_KHR)
which will be used as the EGLImage source; the specified mipmap level must
be part of <buffer>, and the specified z-offset must be smaller than the
depth of the specified mipmap level.
thus a 2d view of a 3d surface is not only legal, it's part of the spec and
must be supported when available
cc: mesa-stable
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15584>
Now that we're using 3DSTATE_BINDING_TABLE_POOL_ALLOC to set the base
address for the binding table pool separately from surface states, we
don't actually need to update surface state base address anymore.
Instead, we can just set STATE_BASE_ADDRESS once at context creation,
and never bother updating it again, saving some heavyweight flushes
and freeing us from the need for address offsetting trickery.
This patch was originally written by Jason Ekstrand, with fixes from
Lionel Landwerlin, but was targeting Icelake. Doing it there requires
additional changes (15:5 -> 18:8 binding table pointer formats) which
also involve some trade-offs, whereas the XeHP change is purely a win,
so we'll do it here first.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15616>
You should probably resize the sources array before accessing entries
that might be out of bounds. inst->resize_sources() always allocates
enough space for at least 3 sources, so this is really only an issue
when there are 4+ sources.
Fixes: a920979d4f ("intel/fs: Use split sends for surface writes on gen9+")
Fixes: 4f86a70599 ("intel/fs: Lower DW untyped r/w messages to LSC when available")
Fixes: d372abe397 ("intel/fs: Add surface OWORD BLOCK opcodes")
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Rohan Garg <rohan.garg@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15632>
In general, an atomic intrinsic may perform separate atomics for every
enabled SIMD channel, as each channel may operate on different memory.
However, an extremely common case is for all channels to access the same
memory location. In this case, we can simply perform a reduction/scan
across the subgroup, and perform one atomic for the whole subgroup,
rather than one per channel. For example, if an intrinsic says to take
the minimum value of the existing memory and the value in each channel,
we can do a thread-local minimum of all enabled channels, then do a
single atomic to take the minimum of that and the existing memory.
Our hardware doesn't optimize the case where multiple channels ask for
atomics on the same memory location; it assumes the compiler will do so.
nir_opt_uniform_atomics() uses divergence analysis to detect this case,
adds the necessary subgroup operations, and moves the atomic inside a
conditional that disables all but a single invocation. It even detects
cases where the shader code already performs this kind of optimization,
and avoids doing it a second time.
This may not be the optimal solution for us. In the backend, we could
detect this case and emit send(1) instructions with NoMask, rather than
generating if...send(16)...endif, and a lot of unnecessary ALU ops. But
it's simple to do, reuses the same path as ACO, and still provides most
of the benefit by cutting up to 16x atomics down to a single atomic,
which is more merciful to the memory bus.
Improves performance of Shadow of the Tomb Raider by 5.5% on XeHP.
Improves performance of a customer-internal benchmark on XeHP at
3840x2160 and low settings by approximately 30%.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15484>
We haven't exposed this intrinsic as it doesn't directly correspond to
anything in SPIR-V. However, it's used internally by some NIR passes,
namely nir_opt_uniform_atomics().
We reuse most of the infrastructure in brw_find_live_channel, but with
LZD/ADD instead of FBL. A new SHADER_OPCODE_FIND_LAST_LIVE_CHANNEL is
like SHADER_OPCODE_FIND_LIVE_CHANNEL but from the other side.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15484>
Instead of reusing the in/out slot mechanism, use a separated NIR
variable mode. This will make easier later to implement staging the
output in shared memory (and storing all at the end to the URB).
Note to get 64-bit type support we currently rely on the
brw_nir_lower_mem_access_bit_sizes() pass.
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15022>
