Marcin Ślusarz
ea92bd8d44
intel/compiler: mask GS URB handles at thread payload construction
...
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Marcin Ślusarz
815eee10e0
intel/compiler/mesh: implement IO for xe2
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Marcin Ślusarz
ee4214de6e
intel/compiler/mesh: fix position of output URB handle for xe2
...
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Francisco Jerez
7f3dc4505d
intel/fs: Delete manual 'inst->mlen' calculations from all uses of logical URB reads.
...
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Francisco Jerez
53d1d793cb
intel/fs: Delete manual 'inst->mlen' calculations from all uses of logical URB writes.
...
Rework:
* Marcin: update emit_urb_indirect_vec4_write
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Francisco Jerez
34a2c9ce35
intel/fs: Specify number of data components of logical URB writes via control immediate.
...
This is what most logical SEND messages do when they take a variable
number of components. 'inst->mlen' is expected to be zero for logical
SEND opcodes, which are expected to behave like plain arithmetic
operations, so certain automated transformations (like SIMD lowering)
can manipulate them without opcode-specific special-casing.
Guessing the number of components from 'inst->mlen' has other
disadvantages, because it requires duplicating the logic that infers
the message payload size in every use of the instruction -- Instead we
can just do the computation once during logical send lowering. In
addition on LNL platform this causes the 'inst->mlen' field of URB
writes to have units inconsistent with every other SEND instruction,
which is likely to lead to confusion and bugs down the road.
Rework:
* Marcin: update emit_urb_indirect_vec4_write
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Francisco Jerez
74c9973c0b
intel/fs/xe2+: Fix URB writes with 0 data components.
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Caio Oliveira
c89597085a
intel/compiler/xe2: Update TCS ICP handle code to support SIMD16
...
Rework:
* Use ffs(grf_size_bytes) (s-b Ken)
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Caio Oliveira
f0fcb778b4
intel/compiler/xe2: Fix URB writes in TCS
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Caio Oliveira
0c03018abf
intel/compiler/xe2: URB fence uses LSC now
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Ian Romanick
623465e26d
intel/compiler/xe2: Update fs_visitor::emit_urb_writes to not assume SIMD8
...
v2: Account for 512b physical registers which causes the URB handle to be in FIXED_GFR 2 instead of 1.
XXX - Use fs_builder::vgrf() instead of open-coded dispatch_width calculations.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Kenneth Graunke
4fffdbbfa2
intel/fs: Fix Xe2 URB read/lowering with per-slot offsets
...
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Jordan Justen
c28539a2fe
intel/compiler: Use enum xe2_lsc_cache_load on xe2
...
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Jordan Justen
8d27b327f2
intel/compiler: Add enum xe2_lsc_cache_load
...
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Jordan Justen
09fc9ff114
intel/compiler: Use enum xe2_lsc_cache_store on xe2
...
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Jordan Justen
c54eff2e03
intel/compiler: Add enum xe2_lsc_cache_store
...
Rework:
* Rohan: Fix enum value for L1WB_L3WB
* Fix write-through comments (Ken)
Ref: bspec 71167
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Marcin Ślusarz
1245020282
intel/compiler: add initial support for URB_LOGICAL_SRC_CHANNEL_MASK to lower_urb_write_logical_send_xe2
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Marcin Ślusarz
9c90377962
intel/compiler: add lsc_msg_desc_wcmask
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Ian Romanick
feec9166cd
intel/compiler/xe2: Handle new URB write messages
...
Rework:
* idr v1: Fix compilation error.
* idr v2: Add support for per-channel offsets.
* idr v3: get_lowered_simd_width is 16 on Xe2+.
* idr v4: Add disassembly support. Add validation support.
* Sqaushed in changes Marcin Ślusarz's patches:
* "intel/compiler: skip adding 0 to payload address"
* "intel/compiler/xe2: drop masking off top 8 bits of URB handle"
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Ian Romanick
fa53a7d241
intel/compiler/xe2: Handle new URB read messages
...
Rework:
* Sqaushed in changes Marcin Ślusarz's patches:
* "intel/compiler: skip adding 0 to payload address"
* "intel/compiler/xe2: drop masking off top 8 bits of URB handle"
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195 >
2023-09-27 23:57:25 +00:00
Konstantin Seurer
4552e594f1
vulkan: Remove vk_get_physical_device_core_1_*_feature_ext
...
It's unused.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24773 >
2023-09-27 23:02:29 +00:00
Konstantin Seurer
05796b29f5
hasvk: Use the common GetPhysicalDeviceFeatures2 implementation
...
Reviewed-by: Julia Tatz <tatz.j@northeastern.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24773 >
2023-09-27 23:02:29 +00:00
Rhys Perry
65afc8bebf
nir/algebraic: optimize u2u32(a >> 32)
...
fossil-db (navi21):
Totals from 352 (0.44% of 79330) affected shaders:
Instrs: 271816 -> 271240 (-0.21%); split: -0.28%, +0.07%
CodeSize: 1546520 -> 1544448 (-0.13%); split: -0.23%, +0.09%
SpillVGPRs: 832 -> 827 (-0.60%); split: -1.08%, +0.48%
Latency: 4037120 -> 4021748 (-0.38%); split: -0.41%, +0.03%
InvThroughput: 1369540 -> 1362066 (-0.55%); split: -0.59%, +0.04%
VClause: 6476 -> 6471 (-0.08%); split: -0.12%, +0.05%
SClause: 6798 -> 6794 (-0.06%)
Copies: 44828 -> 44630 (-0.44%); split: -0.89%, +0.45%
Branches: 8845 -> 8844 (-0.01%); split: -0.05%, +0.03%
PreSGPRs: 14684 -> 14659 (-0.17%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25409 >
2023-09-27 22:13:01 +00:00
Rhys Perry
bcdac65ca3
nir/lower_int64: fix find_lsb(0)
...
If the high 32 bits were zero, this would be umin(find_lsb(lo), 31). This
evaluates to 31 if lo is also zero, instead of -1.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Fixes: 9293d8e64b ("nir: Add find_lsb lowering to nir_lower_int64.")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25409 >
2023-09-27 22:13:01 +00:00
wangra
e90f9d8e21
tu/kgsl: Fix bitfield of DITHER_MODE_MRT6
...
The enum `adreno_rb_dither_mode` needs 2 bits, change the `high` to 13 to make sure 2 bits are used for `DITHER_MODE_MRT6`
Fixes: e03259974e ("freedreno: Generate headers from xml files")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25439 >
2023-09-27 21:45:40 +00:00
Jordan Justen
b5eb96384c
anv: Print warning that Xe2 is not supported rather than failing
...
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25411 >
2023-09-27 21:11:18 +00:00
Jordan Justen
30ba269070
anv: Build for Xe2
...
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25411 >
2023-09-27 21:11:18 +00:00
Jordan Justen
65684b0c7f
anv: Disable Ray Tracing on xe2 until our compiler supports Xe2 RT
...
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25411 >
2023-09-27 21:11:18 +00:00
Jordan Justen
3f1b4dd434
anv/blorp: Use anv_genX to set device->blorp.exec
...
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25411 >
2023-09-27 21:11:18 +00:00
Jordan Justen
3bb96643f7
iris: Build for Xe2
...
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25411 >
2023-09-27 21:11:18 +00:00
Lionel Landwerlin
ee4062666c
iris: add missing workaround for 3DSTATE_LINE_STIPPLE
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25425 >
2023-09-27 20:32:50 +00:00
Lionel Landwerlin
a28ff995bb
anv: add missing workaround for 3DSTATE_LINE_STIPPLE
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25425 >
2023-09-27 20:32:50 +00:00
Lionel Landwerlin
c59179e6bb
anv/iris: widen Wa_14015946265 to Gfx11+
...
We missed out that ICL+ added a programming requiring a CS_STALL.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25426 >
2023-09-27 19:56:24 +00:00
Connor Abbott
302907e347
tu: Expose VK_KHR_maintenance5
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
b0f1cb8f31
freedreno/ci: Skip dEQP-VK.info.device_extensions
...
Copied from anv and radv.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
e7d4bc6095
tu: Add maintenance5 properties
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
a86f2132b3
tu: Check for DEVICE_LOST in vkGetEventStatus()
...
Required by maintenance5.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
b14a8dc2a7
tu: Support VkPipelineCreateFlags2CreateInfoKHR
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
53dbfab87b
tu: Use new buffer usage flags
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
4fd12a18db
tu: Implement vkGetRenderingAreaGranularityKHR
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
2fb42fbd42
tu: Implement vkGetImageSubresourceLayout2KHR and vkGetDeviceImageSubresourceLayoutKHR
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
53e518bcda
tu: Implement vkCmdBindIndexBuffer2KHR
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
26a88fc062
tu: Allow VK_WHOLE_SIZE in tu_CmdBindVertexBuffers2EXT pSizes
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
9669b48b4a
tu/clear_blit: Allow VK_REMAINING_ARRAY_LAYERS as layerCount
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
b9f673e9d4
tu/clear_blit: Fix staging image view layer count
...
This seems to be a mistake that didn't affect anything.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
c2c39a47a6
freedreno/fdl: Support PIPE_FORMAT_R5G5B5A1_UNORM on a6xx
...
This is just a different swap of the same format.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
dd99b58aa8
tu: Support clearing A8_UNORM
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
60b1b18eb9
freedreno/fdl: Use A8_UNORM HW format for sampling
...
We've already been using this for 3d blits on turnip and it works fine.
There's no need to emulate it, and it won't work on turnip where we
can't swizzle the border colors.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Connor Abbott
08581e2e92
zink: Rework color clamping and conversion
...
Before this commit, zink_format_clamp_channel_color() ignored the format
swizzle, so it was assuming that for emulated formats like alpha,
alpha-luminance etc. that the color had already been swizzled to match
the internal format rather than the emulated one. It's somewhat confusing
that passing in e.g. A8_UNORM actually means R8_UNORM, and led to a bug
when using VK_FORMAT_A8_UNORM for texture border colors because we
didn't swizzle it back. It also wouldn't have worked for media formats
like R10X6G10X6 due to the void channel in the middle.
In order to fix this, we need to untangle the mess in its users.
For convert_color() used when clearing, this means we now need to clamp
and then swizzle instead of swizzle and then clamp, and we can drop the
hack for A8_UNORM.
For texture border colors, the state tracker duplicates colors for the
emulated formats to help drivers, which zink was previously relying on,
but fixing zink_format_clamp_channel_color() breaks this because it
assumes that those duplicated colors are useless and clamps them.
However, because we know the format we can just swizzle the border color
ourself, which convert_color() was already doing. So, we pull that out
into a common zink_convert_color() function that handles both clamping
and format emulation, and have both clearing and border color handling
use it.
This fixes A8_UNORM in turnip+zink once we enable it.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25001 >
2023-09-27 19:07:22 +00:00
Erico Nunes
e3e64e3dbe
v3dv: allow headless device without display device
...
When trying to create a headless Vulkan instance without a display
device available, device enumeration would fail.
Part of the physical device creation code already accounts for a missing
display device, allow device enumeration to continue in that case too so
headless instances can be created.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25313 >
2023-09-27 18:46:31 +00:00