mesa/src/amd
Marek Olšák 42e98f115a radv: always use the ngg_lds_layout SGPR
This is a prerequisite for NGG lowering passes to return LDS vertex and
scratch sizes, which will lead to further simplifications. That will
require calling gfx10_get_ngg_info after radv_postprocess_nir, which means
LDS offsets are unknown when the passes are called.

This makes the 2 values no longer compile-time constants.

A later commit will remove NGG_LDS_LAYOUT_SCRATCH_BASE (the passes will
determine it), so only NGG_LDS_LAYOUT_GS_OUT_VERTEX_BASE will come from
an SGPR, though that could be removed too (non-trivially) or handled as
a relocation.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35351>
2025-06-28 08:20:26 +00:00
..
addrlib amd/addrlib: remove the DCC page fault workaround 2025-04-01 03:23:22 -04:00
ci radv/ci: document flakes 2025-06-27 09:09:52 +00:00
common ac/nir/lower_ngg_gs: split lower_ngg_gs_intrinsic into gathering and lowering 2025-06-28 08:20:26 +00:00
compiler aco/spill: Don't spill scratch_rsrc-related temps 2025-06-26 11:02:53 +00:00
drm-shim amd/drm-shim: add gfx1201 2025-03-10 11:21:36 +00:00
gmlib amd/gmlib: remove the executable bit 2025-06-12 07:44:27 +00:00
lanczoslib radeonsi/vpe: enhance scaling quality 2025-06-12 07:44:26 +00:00
llvm ac/llvm: use ds_bpermute_b32 for GFX12 wave64 2025-06-14 13:59:11 +00:00
registers amd: Rename GFX1103_R1/R2 to PHOENIX/2 2024-11-20 02:14:40 +00:00
vpelib amd/vpelib: Fix CodeQL issues Pt1 2025-05-16 11:33:08 +08:00
vulkan radv: always use the ngg_lds_layout SGPR 2025-06-28 08:20:26 +00:00
meson.build radeonsi/vpe: enhance scaling quality 2025-06-12 07:44:26 +00:00