GE_CNTL is the equivalent of IA_MULTI_VGT_PARAM on GFX9 and older.
Calling this function for every draw shouldn't really hurt in practice
because only non-NGG pipelines need this.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18344>
The number of tessellation patches that is computed from the number
of patch control points might change dynamically too.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18344>
The number of patch control points (TCS) and the number of patches
(TCS/TES) is read from user SGPRs.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18344>
This introduces two new user SGPRS:
- tcs_offchip_layout: input patch size and number of patches in TCS
- tes_num_patches: number of patches in TES
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18344>
It's the maximum possible value. This is to ensure that compilers
don't optimize away barriers, like in ACO when workgroup_size is less
than or equal to wave_size, s_barrier is considered a no-op.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18344>
This will be used to compile different tessellation shaders when
patch control points is dynamic.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18344>
To be consistent with the LDS shader config for LS, and this will
be emitted from the cmdbuf for dynamic patch control points.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18344>
OpenGL 3.0 requires RGTC support, and until we have emulation in place,
we should document that requirement.
Fixes: d50e8554b9 ("zink: add feature-documentation")
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18549>
This is currently always freed after the upload but this will allow
to upload the PS epilog later.
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/18363>
One that creates the radv_shader_part object and one that uploads it.
This will be used to upload the PS epilogs as part of the existing
pipeline BO.
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/18363>
Since RADV switched to 100% dynamic rendering, image transitions for
attachments not used in a subpass are correctly performed.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18542>
With GPL, we will have to keep the shader binary in the library for
uploading it later, so it's easier to have a pointer in radv_shader.
The shader binary will be freed when the library is destroyed.
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/18515>
Move subgroup_id, that's only used by CS for verx10 < 125, as part of
the payload too -- even though is not, strictly speaking.
Note the thread execution of Task/Mesh is similar enough, so we make
their common struct inherit from cs_thread_payload.
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18176>
Move the setup into the FS thread payload constructor. Consolidate
payload setup for that in brw_fs_thread_payload.cpp file.
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18176>
The option struct passed to nir_lower_blend doesn't have a "blending
disabled" flag. Unless blending is skipped due to logic ops or
framebuffer formats, nir_lower_blend always blends, even if the blend
mode is "replace" (corresponding to the API level blend disable).
That's mostly okay, since NIR can optimize out the code, at the expense
of a little compile time. However, there's a catch: nir_lower_blend
emits fsat at the start of the shader (for UNORM framebuffers, or
fsat_signed for SNORM). We can expect hardware to saturate the input to
store_output itself, so these operations are redundant, but it's tricky
to optimize these instructions out otherwise. Don't even try: detect the
replace blend mode and don't call nir_blend in that case. Colour masking
is still applied as usual.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18535>
We only have to zero out the handles on failure, which is missed. For
pipelines, we can just do it at the beginning for simplicity.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18555>
This is quite rare but still helps few tesseract shaders and
is quite straightforward.
shader-db with RV530:
total instructions in shared programs: 135671 -> 135646 (-0.02%)
instructions in affected programs: 322 -> 297 (-7.76%)
helped: 13
HURT: 0
Reviewed-by: Filip Gawin <filip@gawin.net>
Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18288>
Assuming they write different channels of the same destination
and they share at least one source or one of the sources is
RC_FILE_NONE.
shader-db with RV530:
total instructions in shared programs: 136033 -> 135673 (-0.26%)
instructions in affected programs: 22987 -> 22627 (-1.57%)
total temps in shared programs: 18977 -> 18965 (-0.06%)
temps in affected programs: 74 -> 62 (-16.22%)
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6188
Reviewed-by: Filip Gawin <filip@gawin.net>
Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18288>
The previous pass can actually create some new movs that are eligible
for copy propagation. There is some minor inctruction and temps
improvement but the biggest win is a gained gnome-shell shader.
Nine more still fail with too many instructions though.
GAINED: shaders/gnome-shell-42/6-1.shader_test FS
shader-db with RV530:
total instructions in shared programs: 136382 -> 135538 (-0.62%)
instructions in affected programs: 31021 -> 30177 (-2.72%)
total temps in shared programs: 18939 -> 18937 (-0.01%)
temps in affected programs: 563 -> 561 (-0.36%)
Reviewed-by: Filip Gawin <filip@gawin.net>
Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18288>
