Add unittest to test conversion of data from linear to
Tile-4 format based on bit swizzling conversion info
mentioned in Bspec.
Signed-off-by: Aditya Swarup <aditya.swarup@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13391>
Add memcpy function to convert linear data to Tile 4 format.
Tile 4 format consists of 4KB block divided into chunks of 512B.
Each 512B chunk/block is comprised of 8 64B blocks arranged in
Y-tile format.
Signed-off-by: Aditya Swarup <aditya.swarup@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13391>
Add unittests for linear to tiled and tiled to linear texture
conversions for Ytile. The test prints the source/output buffer
in hex format with debug flags to verify the result.
Linear to tile conversion fills the linear buffer with values
based on the OWORD index number i.e., OWORD3 will contain
all values filled as 0x03 and
OWORD3 = 0x03030303030303030303030303030303.
The Y-tile to Linear tile conversion uses a similar logic to place
the tiled values in a manner that will result in a linear buffer with
OWORDs filled according to index number as mentioned above.
Signed-off-by: Aditya Swarup <aditya.swarup@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13391>
Ys & Yf are both implemented in ISL now, we still have some Yf issues
to investigate. Instead of disabling them in ISL, we disable them in
the two drivers.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
A number of CTS tests are failing, like :
dEQP-VK.ycbcr.format.b8g8r8g8_422_unorm.vertex_optimal
Failures are reproduced on simulation.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
This commit adds the code for choosing where to start the miptail and
enables miptails by default unless the client driver passes
info->min_miptail_start_level >= info->levels.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
Since the input & output arguments can point to the same location in
memory, if you ever access the input after writing the output you're
in trouble.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
This commit just adds a miptail start field to isl_surf and wires it up
in the RENDER_SURFACE_STATE and 3DSTATE_DEPTH code. We also add a
minimum miptail LOD so that client drivers have a knob to control the
miptails a bit.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
We're about to add separate code for Yf/Ys and this helps keep the diff
reasonable.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
With these tilings, everything is aligned to a tile and the tiled
surface size calculations will handle the array stride for us. We need
to provide an accurate 4D size so that 3D and multisampled images get
tiled correctly.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
This makes things a tiny bit stickier in isl_calc_phys_total_extent_el
but will be worth it when we enable Yf and Ys.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
The ICL+ PRMs show that this is not a supported thing anymore for
Ys/Yf and there isn't really much use for it in Vulkan/Sparse.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
The tile size calculations use a clever bit of math to make them short
and simple. We add unit tests to assert that they identically match the
tables in the PRM.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
The Yf and Ys tilings change a bit between SKL and later generations so
we have to be able to distinguish between them.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
Since this value is 0, it doesn't change anything, but it's just good
practice like we did for Gfx12.5 right above.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23620>
At the time this was added to fix some test failures. But it seems that
the failures were happening due to missing cache flushes, so
this extra space is no longer neccessary.
Fixes: 37b4eacc ("intel/isl: Resize clear color buffer to full cacheline")
Signed-off-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24768>
This is a ADL-P workaround of a pre-production stepping, with RPL-P
already being sold we can remove this workaround.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24699>
The way this workaround is implemented, it is being applied to all
gfx 12 platforms(TGL, ADL, RKL, RPL, DG1, DG2 and MTL) but it was
supposed to be fixed in TGL B0.
Unfortunately I did not found any workaround number that would match it.
But as all released platforms don't ship to customers with revision == 0
this workaround was never being applied and can be safely removed.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24699>
Also use parent WA number of 14017240301 instead of 14017353530.
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22401>
This enables the DG2 render compression modifier in anv. When I tested
this against vkcube, I observed that the full resolve which happened at
the end of every frame was converted to a partial resolve, allowing the
framebuffer to retain compression.
According to Caleb Callaway's testing, enabling this modifier positively
impacts the FPS of the following game benchmarks:
- Strange Brigade.vk-g6 +12.78%
- Strange Brigade.dx12vk-g6 + 9.33%
- Shadow of the Tomb Raider.vk-g6-lx + 2.37%
- Dota 2 (replay Jul 2020).vk-g6 + 2.28%
Thanks to Felix Degrood for pointing out that Strange Brigade would
benefit from this optimization.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24120>
Replace the aux_usage field with two booleans: one for render
compression and one for media compression.
This more accurately describes how CCS_E is used on gfx12. On those
platforms, the FCV feature may be enabled or disabled, but ISL's
modifier table has been using the FCV aux-usage for every gfx12 render
compression modifier. Instead, set the newly-added render compression
boolean to true.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24120>
We're going to enable the DG2_RC_CCS modifier in anv. Add and use this
function to prepare for the new plane count that comes with that
modifier.
iris is left alone for now because it supports more modifiers than
isl_drm_modifier_get_score is aware of.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24120>
We're still seeing failures with render target reads of multisampled
images on Alchemist platforms, but Meteorlake doesn't appear to have
that issue. Enable MCS on Meteorlake.
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22802>
Without this change various OpenGL CTS tranform feedback tests were
failing.
Cc: 23.2 <mesa-stable>
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23823>
Rework:
* Jordan: Add uncached for all platforms (Requested by Francisco)
* Jordan: Use gen7 & gen8 values suggested by Francisco
* Jordan: Fix IVB and CHV MOCS mistakes pointed out by Francisco
Cc: 23.2 <mesa-stable>
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23823>
Depth/stencil surfaces cannot be linear but they can be 1D, so they
end up being tile64 when sparse (as we force every sparse resource to
be either tile64 or linear).
According to the "1D surfaces" page from BSpec, our driver treats 1D
surfaces as 2D surfaces with a height of 1 texel, since we don't
enable the corresponding bit from HAS_SLICE_CHICKEN7. And since we
support 2D surfaces, we should also support 1D.
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22974>
