fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-05-05 18:18:06 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

ac/nir/tess: unify computing LDS output patch size, minimize LDS bank conflicts

Browse source 
This unifies the duplicated LDS output patch size computation between
hs_output_lds_offset and ac_nir_compute_tess_wg_info.

"+ 4" to the output patch stride minimizes LDS bank conflicts by making
the beginning of each patch start on a different LDS bank.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34780>
This commit is contained in:
Marek Olšák 2025-04-25 20:02:48 -04:00 • committed by Marge Bot
parent 37dc376395
commit c732306c5a
1 changed files with 32 additions and 28 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

60
src/amd/common/nir/ac_nir_lower_tess_io_to_mem.c
View file

@ -430,20 +430,33 @@ hs_output_lds_map_io_location(nir_shader *shader,
}
}
static unsigned
get_lds_output_vertex_size(const ac_nir_tess_io_info *io_info)
{
return util_bitcount64(io_info->lds_output_mask & ~TESS_LVL_MASK) * 16;
}
static unsigned
get_lds_pervertex_output_patch_size(const ac_nir_tess_io_info *io_info, unsigned tcs_vertices_out)
{
return tcs_vertices_out * get_lds_output_vertex_size(io_info);
}
static unsigned
get_lds_output_patch_stride(const ac_nir_tess_io_info *io_info, unsigned tcs_vertices_out)
{
unsigned lds_perpatch_output_patch_size = (util_bitcount64(io_info->lds_output_mask & TESS_LVL_MASK) +
util_bitcount(io_info->lds_patch_output_mask)) * 16;
/* Add 4 to the output patch size to minimize LDS bank conflicts. */
return get_lds_pervertex_output_patch_size(io_info, tcs_vertices_out) +
lds_perpatch_output_patch_size + 4;
}
static nir_def *
hs_output_lds_offset(nir_builder *b, lower_tess_io_state *st, unsigned location, unsigned component,
nir_def *vertex_index, nir_def *io_offset)
{
const uint64_t per_vertex_mask = st->io_info.lds_output_mask & ~TESS_LVL_MASK;
const uint64_t tf_mask = st->io_info.lds_output_mask & TESS_LVL_MASK;
const uint32_t patch_out_mask = st->io_info.lds_patch_output_mask;
unsigned tcs_num_reserved_outputs = util_bitcount64(per_vertex_mask);
unsigned tcs_num_reserved_patch_outputs = util_bitcount64(tf_mask) + util_bitcount(patch_out_mask);
unsigned output_vertex_size = tcs_num_reserved_outputs * 16u;
unsigned pervertex_output_patch_size = b->shader->info.tess.tcs_vertices_out * output_vertex_size;
unsigned output_patch_stride = pervertex_output_patch_size + tcs_num_reserved_patch_outputs * 16u;
unsigned tcs_vertices_out = b->shader->info.tess.tcs_vertices_out;
nir_def *off = NULL;
if (io_offset) {
@ -456,24 +469,21 @@ hs_output_lds_offset(nir_builder *b, lower_tess_io_state *st, unsigned location,
}
nir_def *rel_patch_id = nir_load_tess_rel_patch_id_amd(b);
nir_def *patch_offset = nir_imul_imm(b, rel_patch_id, output_patch_stride);
nir_def *patch_offset = nir_imul_imm(b, rel_patch_id,
get_lds_output_patch_stride(&st->io_info, tcs_vertices_out));
nir_def *tcs_in_vtxcnt = nir_load_patch_vertices_in(b);
nir_def *tcs_num_patches = nir_load_tcs_num_patches_amd(b);
nir_def *input_patch_size = nir_imul(b, tcs_in_vtxcnt, nir_load_lshs_vertex_stride_amd(b));
nir_def *output_patch0_offset = nir_imul(b, input_patch_size, tcs_num_patches);
nir_def *output_patch_offset = nir_iadd_nuw(b, patch_offset, output_patch0_offset);
nir_def *lds_offset;
if (vertex_index) {
nir_def *vertex_index_off = nir_imul_imm(b, vertex_index, output_vertex_size);
if (vertex_index)
off = nir_iadd_nuw(b, off, nir_imul_imm(b, vertex_index, get_lds_output_vertex_size(&st->io_info)));
else
off = nir_iadd_imm_nuw(b, off, get_lds_pervertex_output_patch_size(&st->io_info, tcs_vertices_out));
off = nir_iadd_nuw(b, off, vertex_index_off);
lds_offset = nir_iadd_nuw(b, off, output_patch_offset);
} else {
off = nir_iadd_imm_nuw(b, off, pervertex_output_patch_size);
lds_offset = nir_iadd_nuw(b, off, output_patch_offset);
}
nir_def *lds_offset = nir_iadd_nuw(b, off, output_patch_offset);
/* The beginning of LDS is reserved for the tess level group vote. */
return nir_iadd_imm_nuw(b, lds_offset, AC_TESS_LEVEL_VOTE_LDS_BYTES);
@ -1515,15 +1525,9 @@ ac_nir_compute_tess_wg_info(const struct radeon_info *info, const ac_nir_tess_io
unsigned num_remapped_tess_level_outputs, unsigned *num_patches_per_wg,
unsigned *hw_lds_size)
{
unsigned num_tcs_output_cp = tcs_vertices_out;
unsigned lds_output_vertex_size = util_bitcount64(io_info->lds_output_mask & ~TESS_LVL_MASK) * 16;
unsigned lds_perpatch_output_patch_size = (util_bitcount64(io_info->lds_output_mask & TESS_LVL_MASK) +
util_bitcount(io_info->lds_patch_output_mask)) * 16;
unsigned lds_per_patch = num_tcs_input_cp * lds_input_vertex_size +
num_tcs_output_cp * lds_output_vertex_size +
lds_perpatch_output_patch_size;
unsigned num_patches = ac_compute_num_tess_patches(info, num_tcs_input_cp, num_tcs_output_cp,
get_lds_output_patch_stride(io_info, tcs_vertices_out);
unsigned num_patches = ac_compute_num_tess_patches(info, num_tcs_input_cp, tcs_vertices_out,
io_info->highest_remapped_vram_output,
MAX2(io_info->highest_remapped_vram_patch_output,
num_remapped_tess_level_outputs),