This file is auto-generated by tools/update_cpu_features.zig.
std.Target.amdgcn.all_features
constant
std.Target.amdgcn.cpu
namespace
bonairecarrizofijigenericgeneric_hsagfx1010gfx1011gfx1012gfx1013gfx1030gfx1031gfx1032gfx1033gfx1034gfx1035gfx1036gfx10_1_genericgfx10_3_genericgfx1100gfx1101gfx1102gfx1103gfx1150gfx1151gfx1152gfx1153gfx11_genericgfx1200gfx1201gfx1250gfx1251gfx12_genericgfx600gfx601gfx602gfx700gfx701gfx702gfx703gfx704gfx705gfx801gfx802gfx803gfx805gfx810gfx900gfx902gfx904gfx906gfx908gfx909gfx90agfx90cgfx942gfx950gfx9_4_genericgfx9_generichainanhawaiiicelandkabinikaverimullinsolandpitcairnpolaris10polaris11stoneytahititongatongaproverde
std.Target.amdgcn.cpu.bonaire
constant
CpuModel
std.Target.amdgcn.cpu.carrizo
constant
CpuModel
std.Target.amdgcn.cpu.fiji
constant
CpuModel
std.Target.amdgcn.cpu.generic
constant
CpuModel
std.Target.amdgcn.cpu.generic_hsa
constant
CpuModel
std.Target.amdgcn.cpu.gfx10_1_generic
constant
CpuModel
std.Target.amdgcn.cpu.gfx10_3_generic
constant
CpuModel
std.Target.amdgcn.cpu.gfx1010
constant
CpuModel
std.Target.amdgcn.cpu.gfx1011
constant
CpuModel
std.Target.amdgcn.cpu.gfx1012
constant
CpuModel
std.Target.amdgcn.cpu.gfx1013
constant
CpuModel
std.Target.amdgcn.cpu.gfx1030
constant
CpuModel
std.Target.amdgcn.cpu.gfx1031
constant
CpuModel
std.Target.amdgcn.cpu.gfx1032
constant
CpuModel
std.Target.amdgcn.cpu.gfx1033
constant
CpuModel
std.Target.amdgcn.cpu.gfx1034
constant
CpuModel
std.Target.amdgcn.cpu.gfx1035
constant
CpuModel
std.Target.amdgcn.cpu.gfx1036
constant
CpuModel
std.Target.amdgcn.cpu.gfx11_generic
constant
CpuModel
std.Target.amdgcn.cpu.gfx1100
constant
CpuModel
std.Target.amdgcn.cpu.gfx1101
constant
CpuModel
std.Target.amdgcn.cpu.gfx1102
constant
CpuModel
std.Target.amdgcn.cpu.gfx1103
constant
CpuModel
std.Target.amdgcn.cpu.gfx1150
constant
CpuModel
std.Target.amdgcn.cpu.gfx1151
constant
CpuModel
std.Target.amdgcn.cpu.gfx1152
constant
CpuModel
std.Target.amdgcn.cpu.gfx1153
constant
CpuModel
std.Target.amdgcn.cpu.gfx12_generic
constant
CpuModel
std.Target.amdgcn.cpu.gfx1200
constant
CpuModel
std.Target.amdgcn.cpu.gfx1201
constant
CpuModel
std.Target.amdgcn.cpu.gfx1250
constant
CpuModel
std.Target.amdgcn.cpu.gfx1251
constant
CpuModel
std.Target.amdgcn.cpu.gfx600
constant
CpuModel
std.Target.amdgcn.cpu.gfx601
constant
CpuModel
std.Target.amdgcn.cpu.gfx602
constant
CpuModel
std.Target.amdgcn.cpu.gfx700
constant
CpuModel
std.Target.amdgcn.cpu.gfx701
constant
CpuModel
std.Target.amdgcn.cpu.gfx702
constant
CpuModel
std.Target.amdgcn.cpu.gfx703
constant
CpuModel
std.Target.amdgcn.cpu.gfx704
constant
CpuModel
std.Target.amdgcn.cpu.gfx705
constant
CpuModel
std.Target.amdgcn.cpu.gfx801
constant
CpuModel
std.Target.amdgcn.cpu.gfx802
constant
CpuModel
std.Target.amdgcn.cpu.gfx803
constant
CpuModel
std.Target.amdgcn.cpu.gfx805
constant
CpuModel
std.Target.amdgcn.cpu.gfx810
constant
CpuModel
std.Target.amdgcn.cpu.gfx9_4_generic
constant
CpuModel
std.Target.amdgcn.cpu.gfx9_generic
constant
CpuModel
std.Target.amdgcn.cpu.gfx900
constant
CpuModel
std.Target.amdgcn.cpu.gfx902
constant
CpuModel
std.Target.amdgcn.cpu.gfx904
constant
CpuModel
std.Target.amdgcn.cpu.gfx906
constant
CpuModel
std.Target.amdgcn.cpu.gfx908
constant
CpuModel
std.Target.amdgcn.cpu.gfx909
constant
CpuModel
std.Target.amdgcn.cpu.gfx90a
constant
CpuModel
std.Target.amdgcn.cpu.gfx90c
constant
CpuModel
std.Target.amdgcn.cpu.gfx942
constant
CpuModel
std.Target.amdgcn.cpu.gfx950
constant
CpuModel
std.Target.amdgcn.cpu.hainan
constant
CpuModel
std.Target.amdgcn.cpu.hawaii
constant
CpuModel
std.Target.amdgcn.cpu.iceland
constant
CpuModel
std.Target.amdgcn.cpu.kabini
constant
CpuModel
std.Target.amdgcn.cpu.kaveri
constant
CpuModel
std.Target.amdgcn.cpu.mullins
constant
CpuModel
std.Target.amdgcn.cpu.oland
constant
CpuModel
std.Target.amdgcn.cpu.pitcairn
constant
CpuModel
std.Target.amdgcn.cpu.polaris10
constant
CpuModel
std.Target.amdgcn.cpu.polaris11
constant
CpuModel
std.Target.amdgcn.cpu.stoney
constant
CpuModel
std.Target.amdgcn.cpu.tahiti
constant
CpuModel
std.Target.amdgcn.cpu.tonga
constant
CpuModel
std.Target.amdgcn.cpu.tongapro
constant
CpuModel
std.Target.amdgcn.cpu.verde
constant
CpuModel
std.Target.amdgcn.CpuFeature
alias
Alias of std.Target.Cpu.Feature
std.Target.amdgcn.CpuModel
alias
Alias of std.Target.Cpu.Model
std.Target.amdgcn.Feature
container
Fields
@"1024_addressable_vgprs"
@"16_bit_insts"
@"45_bit_num_records_buffer_resource"
@"64_bit_literals"
a16
add_min_max_insts
add_no_carry_insts
add_sub_u64_insts
addressablelocalmemorysize163840
addressablelocalmemorysize32768
addressablelocalmemorysize327680
addressablelocalmemorysize65536
agent_scope_fine_grained_remote_memory_atomics
allocate1_5xvgprs
aperture_regs
architected_flat_scratch
architected_sgprs
ashr_pk_insts
assembler_permissive_wavesize
atomic_buffer_global_pk_add_f16_insts
atomic_buffer_global_pk_add_f16_no_rtn_insts
atomic_buffer_pk_add_bf16_inst
atomic_csub_no_rtn_insts
atomic_ds_pk_add_16_insts
atomic_fadd_no_rtn_insts
atomic_fadd_rtn_insts
atomic_flat_pk_add_16_insts
atomic_fmin_fmax_flat_f32
atomic_fmin_fmax_flat_f64
atomic_fmin_fmax_global_f32
atomic_fmin_fmax_global_f64
atomic_global_pk_add_bf16_inst
auto_waitcnt_before_barrier
back_off_barrier
bf16_cvt_insts
bf16_pk_insts
bf16_trans_insts
bf8_cvt_scale_insts
bitop3_insts
block_vgpr_csr
bvh_dual_bvh_8_insts
ci_insts
clusters
cube_insts
cumode
cvt_fp8_vop1_bug
cvt_norm_insts
cvt_pk_f16_f32_inst
cvt_pknorm_vop2_insts
cvt_pknorm_vop3_insts
d16_write_vgpr32
default_component_broadcast
default_component_zero
dl_insts
dot10_insts
dot11_insts
dot12_insts
dot13_insts
dot1_insts
dot2_insts
dot3_insts
dot4_insts
dot5_insts
dot6_insts
dot7_insts
dot8_insts
dot9_insts
dpp
dpp8
dpp_64bit
dpp_src1_sgpr
ds128
ds_src2_insts
emulated_system_scope_atomics
extended_image_insts
f16bf16_to_fp6bf6_cvt_scale_insts
f32_to_f16bf16_cvt_sr_insts
fast_denormal_f32
fast_fmaf
flat_address_space
flat_atomic_fadd_f32_inst
flat_buffer_global_fadd_f64_inst
flat_for_global
flat_global_insts
flat_gvs_mode
flat_inst_offsets
flat_scratch
flat_scratch_insts
flat_segment_offset_bug
fma_mix_bf16_insts
fma_mix_insts
fmacf64_inst
fmaf
fp4_cvt_scale_insts
fp64
fp6bf6_cvt_scale_insts
fp8_conversion_insts
fp8_cvt_scale_insts
fp8_insts
fp8e5m3_insts
full_rate_64_ops
g16
gcn3_encoding
gds
get_wave_id_inst
gfx10
gfx10_3_insts
gfx10_a_encoding
gfx10_b_encoding
gfx10_insts
gfx11
gfx11_insts
gfx12
gfx1250_insts
gfx12_insts
gfx7_gfx8_gfx9_insts
gfx8_insts
gfx9
gfx90a_insts
gfx940_insts
gfx950_insts
gfx9_insts
globally_addressable_scratch
gws
half_rate_64_ops
ieee_minimum_maximum_insts
image_gather4_d16_bug
image_insts
image_store_d16_bug
inst_fwd_prefetch_bug
int_clamp_insts
inv_2pi_inline_imm
kernarg_preload
lds_barrier_arrive_atomic
lds_branch_vmem_war_hazard
lds_misaligned_bug
ldsbankcount16
ldsbankcount32
lerp_inst
load_store_opt
lshl_add_u64_inst
mad_intra_fwd_bug
mad_mac_f32_insts
mad_mix_insts
mad_u32_inst
mai_insts
max_hard_clause_length_32
max_hard_clause_length_63
max_private_element_size_16
max_private_element_size_4
max_private_element_size_8
mcast_load_insts
memory_atomic_fadd_f32_denormal_support
mfma_inline_literal_bug
mimg_r128
min3_max3_pkf16
minimum3_maximum3_f16
minimum3_maximum3_f32
minimum3_maximum3_pkf16
movrel
msaa_load_dst_sel_bug
negative_scratch_offset_bug
negative_unaligned_scratch_offset_bug
no_data_dep_hazard
no_sdst_cmpx
nsa_clause_bug
nsa_encoding
nsa_to_vmem_bug
offset_3f_bug
packed_fp32_ops
packed_tid
partial_nsa_encoding
permlane16_swap
permlane32_swap
pk_add_min_max_insts
pk_fmac_f16_inst
point_sample_accel
precise_memory
priv_enabled_trap2_nop_bug
prng_inst
promote_alloca
prt_strict_null
pseudo_scalar_trans
qsad_insts
r128_a16
real_true16
relaxed_buffer_oob_mode
required_export_priority
requires_cov6
restricted_soffset
s_memrealtime
s_memtime_inst
s_wakeup_barrier_inst
sad_insts
safe_cu_prefetch
safe_smem_prefetch
salu_float
scalar_atomics
scalar_dwordx3_loads
scalar_flat_scratch_insts
scalar_stores
sdwa
sdwa_mav
sdwa_omod
sdwa_out_mods_vopc
sdwa_scalar
sdwa_sdst
sea_islands
setprio_inc_wg_inst
setreg_vgpr_msb_fixup
sgpr_init_bug
shader_cycles_hi_lo_registers
shader_cycles_register
si_scheduler
smem_to_vector_write_hazard
southern_islands
sramecc
sramecc_support
tanh_insts
tensor_cvt_lut_insts
tgsplit
transpose_load_f4f6_insts
trap_handler
trig_reduced_range
true16
unaligned_access_mode
unaligned_buffer_access
unaligned_ds_access
unaligned_scratch_access
unpacked_d16_vmem
unsafe_ds_offset_folding
user_sgpr_init16_bug
valu_trans_use_hazard
vcmpx_exec_war_hazard
vcmpx_permlane_hazard
vgpr_align2
vgpr_index_mode
vmem_pref_insts
vmem_to_lds_load_insts
vmem_to_scalar_write_hazard
vmem_write_vgpr_in_order
volcanic_islands
vop3_literal
vop3p
vopd
vscnt
wait_xcnt
waits_before_system_scope_stores
wavefrontsize16
wavefrontsize32
wavefrontsize64
xf32_insts
xnack
xnack_support
std.Target.amdgcn.featureSet
constant
std.Target.amdgcn.featureSetHas
constant
std.Target.amdgcn.featureSetHasAll
constant
std.Target.amdgcn.featureSetHasAny
constant
std.Target.amdgcn.std
alias
Alias of std
Source code
std/Target/amdgcn.zig
//! This file is auto-generated by tools/update_cpu_features.zig.
const std = @import("../std.zig");
const CpuFeature = std.Target.Cpu.Feature;
const CpuModel = std.Target.Cpu.Model;
pub const Feature = enum {
@"1024_addressable_vgprs",
@"16_bit_insts",
@"45_bit_num_records_buffer_resource",
@"64_bit_literals",
a16,
add_min_max_insts,
add_no_carry_insts,
add_sub_u64_insts,
addressablelocalmemorysize163840,
addressablelocalmemorysize32768,
addressablelocalmemorysize327680,
addressablelocalmemorysize65536,
agent_scope_fine_grained_remote_memory_atomics,
allocate1_5xvgprs,
aperture_regs,
architected_flat_scratch,
architected_sgprs,
ashr_pk_insts,
assembler_permissive_wavesize,
atomic_buffer_global_pk_add_f16_insts,
atomic_buffer_global_pk_add_f16_no_rtn_insts,
atomic_buffer_pk_add_bf16_inst,
atomic_csub_no_rtn_insts,
atomic_ds_pk_add_16_insts,
atomic_fadd_no_rtn_insts,
atomic_fadd_rtn_insts,
atomic_flat_pk_add_16_insts,
atomic_fmin_fmax_flat_f32,
atomic_fmin_fmax_flat_f64,
atomic_fmin_fmax_global_f32,
atomic_fmin_fmax_global_f64,
atomic_global_pk_add_bf16_inst,
auto_waitcnt_before_barrier,
back_off_barrier,
bf16_cvt_insts,
bf16_pk_insts,
bf16_trans_insts,
bf8_cvt_scale_insts,
bitop3_insts,
block_vgpr_csr,
bvh_dual_bvh_8_insts,
ci_insts,
clusters,
cube_insts,
cumode,
cvt_fp8_vop1_bug,
cvt_norm_insts,
cvt_pk_f16_f32_inst,
cvt_pknorm_vop2_insts,
cvt_pknorm_vop3_insts,
d16_write_vgpr32,
default_component_broadcast,
default_component_zero,
dl_insts,
dot10_insts,
dot11_insts,
dot12_insts,
dot13_insts,
dot1_insts,
dot2_insts,
dot3_insts,
dot4_insts,
dot5_insts,
dot6_insts,
dot7_insts,
dot8_insts,
dot9_insts,
dpp,
dpp8,
dpp_64bit,
dpp_src1_sgpr,
ds128,
ds_src2_insts,
emulated_system_scope_atomics,
extended_image_insts,
f16bf16_to_fp6bf6_cvt_scale_insts,
f32_to_f16bf16_cvt_sr_insts,
fast_denormal_f32,
fast_fmaf,
flat_address_space,
flat_atomic_fadd_f32_inst,
flat_buffer_global_fadd_f64_inst,
flat_for_global,
flat_global_insts,
flat_gvs_mode,
flat_inst_offsets,
flat_scratch,
flat_scratch_insts,
flat_segment_offset_bug,
fma_mix_bf16_insts,
fma_mix_insts,
fmacf64_inst,
fmaf,
fp4_cvt_scale_insts,
fp64,
fp6bf6_cvt_scale_insts,
fp8_conversion_insts,
fp8_cvt_scale_insts,
fp8_insts,
fp8e5m3_insts,
full_rate_64_ops,
g16,
gcn3_encoding,
gds,
get_wave_id_inst,
gfx10,
gfx10_3_insts,
gfx10_a_encoding,
gfx10_b_encoding,
gfx10_insts,
gfx11,
gfx11_insts,
gfx12,
gfx1250_insts,
gfx12_insts,
gfx7_gfx8_gfx9_insts,
gfx8_insts,
gfx9,
gfx90a_insts,
gfx940_insts,
gfx950_insts,
gfx9_insts,
globally_addressable_scratch,
gws,
half_rate_64_ops,
ieee_minimum_maximum_insts,
image_gather4_d16_bug,
image_insts,
image_store_d16_bug,
inst_fwd_prefetch_bug,
int_clamp_insts,
inv_2pi_inline_imm,
kernarg_preload,
lds_barrier_arrive_atomic,
lds_branch_vmem_war_hazard,
lds_misaligned_bug,
ldsbankcount16,
ldsbankcount32,
lerp_inst,
load_store_opt,
lshl_add_u64_inst,
mad_intra_fwd_bug,
mad_mac_f32_insts,
mad_mix_insts,
mad_u32_inst,
mai_insts,
max_hard_clause_length_32,
max_hard_clause_length_63,
max_private_element_size_16,
max_private_element_size_4,
max_private_element_size_8,
mcast_load_insts,
memory_atomic_fadd_f32_denormal_support,
mfma_inline_literal_bug,
mimg_r128,
min3_max3_pkf16,
minimum3_maximum3_f16,
minimum3_maximum3_f32,
minimum3_maximum3_pkf16,
movrel,
msaa_load_dst_sel_bug,
negative_scratch_offset_bug,
negative_unaligned_scratch_offset_bug,
no_data_dep_hazard,
no_sdst_cmpx,
nsa_clause_bug,
nsa_encoding,
nsa_to_vmem_bug,
offset_3f_bug,
packed_fp32_ops,
packed_tid,
partial_nsa_encoding,
permlane16_swap,
permlane32_swap,
pk_add_min_max_insts,
pk_fmac_f16_inst,
point_sample_accel,
precise_memory,
priv_enabled_trap2_nop_bug,
prng_inst,
promote_alloca,
prt_strict_null,
pseudo_scalar_trans,
qsad_insts,
r128_a16,
real_true16,
relaxed_buffer_oob_mode,
required_export_priority,
requires_cov6,
restricted_soffset,
s_memrealtime,
s_memtime_inst,
s_wakeup_barrier_inst,
sad_insts,
safe_cu_prefetch,
safe_smem_prefetch,
salu_float,
scalar_atomics,
scalar_dwordx3_loads,
scalar_flat_scratch_insts,
scalar_stores,
sdwa,
sdwa_mav,
sdwa_omod,
sdwa_out_mods_vopc,
sdwa_scalar,
sdwa_sdst,
sea_islands,
setprio_inc_wg_inst,
setreg_vgpr_msb_fixup,
sgpr_init_bug,
shader_cycles_hi_lo_registers,
shader_cycles_register,
si_scheduler,
smem_to_vector_write_hazard,
southern_islands,
sramecc,
sramecc_support,
tanh_insts,
tensor_cvt_lut_insts,
tgsplit,
transpose_load_f4f6_insts,
trap_handler,
trig_reduced_range,
true16,
unaligned_access_mode,
unaligned_buffer_access,
unaligned_ds_access,
unaligned_scratch_access,
unpacked_d16_vmem,
unsafe_ds_offset_folding,
user_sgpr_init16_bug,
valu_trans_use_hazard,
vcmpx_exec_war_hazard,
vcmpx_permlane_hazard,
vgpr_align2,
vgpr_index_mode,
vmem_pref_insts,
vmem_to_lds_load_insts,
vmem_to_scalar_write_hazard,
vmem_write_vgpr_in_order,
volcanic_islands,
vop3_literal,
vop3p,
vopd,
vscnt,
wait_xcnt,
waits_before_system_scope_stores,
wavefrontsize16,
wavefrontsize32,
wavefrontsize64,
xf32_insts,
xnack,
xnack_support,
};
pub const featureSet = CpuFeature.FeatureSetFns(Feature).featureSet;
pub const featureSetHas = CpuFeature.FeatureSetFns(Feature).featureSetHas;
pub const featureSetHasAny = CpuFeature.FeatureSetFns(Feature).featureSetHasAny;
pub const featureSetHasAll = CpuFeature.FeatureSetFns(Feature).featureSetHasAll;
pub const all_features = blk: {
@setEvalBranchQuota(2000);
const len = @typeInfo(Feature).@"enum".field_names.len;
std.debug.assert(len <= CpuFeature.Set.needed_bit_count);
var result: [len]CpuFeature = undefined;
result[@backingInt(Feature.@"1024_addressable_vgprs")] = .{
.llvm_name = "1024-addressable-vgprs",
.description = "Has 1024 addressable VGPRs",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.@"16_bit_insts")] = .{
.llvm_name = "16-bit-insts",
.description = "Has i16/f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.@"45_bit_num_records_buffer_resource")] = .{
.llvm_name = "45-bit-num-records-buffer-resource",
.description = "The buffer resource (V#) supports 45-bit num_records",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.@"64_bit_literals")] = .{
.llvm_name = "64-bit-literals",
.description = "Can use 64-bit literals with single DWORD instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.a16)] = .{
.llvm_name = "a16",
.description = "Support A16 for 16-bit coordinates/gradients/lod/clamp/mip image operands",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.add_min_max_insts)] = .{
.llvm_name = "add-min-max-insts",
.description = "Has v_add_{min|max}_{i|u}32 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.add_no_carry_insts)] = .{
.llvm_name = "add-no-carry-insts",
.description = "Have VALU add/sub instructions without carry out",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.add_sub_u64_insts)] = .{
.llvm_name = "add-sub-u64-insts",
.description = "Has v_add_u64 and v_sub_u64 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.addressablelocalmemorysize163840)] = .{
.llvm_name = "addressablelocalmemorysize163840",
.description = "The size of local memory in bytes",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.addressablelocalmemorysize32768)] = .{
.llvm_name = "addressablelocalmemorysize32768",
.description = "The size of local memory in bytes",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.addressablelocalmemorysize327680)] = .{
.llvm_name = "addressablelocalmemorysize327680",
.description = "The size of local memory in bytes",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.addressablelocalmemorysize65536)] = .{
.llvm_name = "addressablelocalmemorysize65536",
.description = "The size of local memory in bytes",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.agent_scope_fine_grained_remote_memory_atomics)] = .{
.llvm_name = "agent-scope-fine-grained-remote-memory-atomics",
.description = "Agent (device) scoped atomic operations, excluding those directly supported by PCIe (i.e. integer atomic add, exchange, and compare-and-swap), are functional for allocations in host or peer device memory.",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.allocate1_5xvgprs)] = .{
.llvm_name = "allocate1_5xvgprs",
.description = "Has 50% more physical VGPRs and 50% larger allocation granule",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.aperture_regs)] = .{
.llvm_name = "aperture-regs",
.description = "Has Memory Aperture Base and Size Registers",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.architected_flat_scratch)] = .{
.llvm_name = "architected-flat-scratch",
.description = "Flat Scratch register is a readonly SPI initialized architected register",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.architected_sgprs)] = .{
.llvm_name = "architected-sgprs",
.description = "Enable the architected SGPRs",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.ashr_pk_insts)] = .{
.llvm_name = "ashr-pk-insts",
.description = "Has Arithmetic Shift Pack instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.assembler_permissive_wavesize)] = .{
.llvm_name = "assembler-permissive-wavesize",
.description = "allow parsing wave32 and wave64 variants of instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.atomic_buffer_global_pk_add_f16_insts)] = .{
.llvm_name = "atomic-buffer-global-pk-add-f16-insts",
.description = "Has buffer_atomic_pk_add_f16 and global_atomic_pk_add_f16 instructions that can return original value",
.dependencies = featureSet(&[_]Feature{
.flat_global_insts,
}),
};
result[@backingInt(Feature.atomic_buffer_global_pk_add_f16_no_rtn_insts)] = .{
.llvm_name = "atomic-buffer-global-pk-add-f16-no-rtn-insts",
.description = "Has buffer_atomic_pk_add_f16 and global_atomic_pk_add_f16 instructions that don't return original value",
.dependencies = featureSet(&[_]Feature{
.flat_global_insts,
}),
};
result[@backingInt(Feature.atomic_buffer_pk_add_bf16_inst)] = .{
.llvm_name = "atomic-buffer-pk-add-bf16-inst",
.description = "Has buffer_atomic_pk_add_bf16 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.atomic_csub_no_rtn_insts)] = .{
.llvm_name = "atomic-csub-no-rtn-insts",
.description = "Has buffer_atomic_csub and global_atomic_csub instructions that don't return original value",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.atomic_ds_pk_add_16_insts)] = .{
.llvm_name = "atomic-ds-pk-add-16-insts",
.description = "Has ds_pk_add_bf16, ds_pk_add_f16, ds_pk_add_rtn_bf16, ds_pk_add_rtn_f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.atomic_fadd_no_rtn_insts)] = .{
.llvm_name = "atomic-fadd-no-rtn-insts",
.description = "Has buffer_atomic_add_f32 and global_atomic_add_f32 instructions that don't return original value",
.dependencies = featureSet(&[_]Feature{
.flat_global_insts,
}),
};
result[@backingInt(Feature.atomic_fadd_rtn_insts)] = .{
.llvm_name = "atomic-fadd-rtn-insts",
.description = "Has buffer_atomic_add_f32 and global_atomic_add_f32 instructions that return original value",
.dependencies = featureSet(&[_]Feature{
.flat_global_insts,
}),
};
result[@backingInt(Feature.atomic_flat_pk_add_16_insts)] = .{
.llvm_name = "atomic-flat-pk-add-16-insts",
.description = "Has flat_atomic_pk_add_f16 and flat_atomic_pk_add_bf16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.atomic_fmin_fmax_flat_f32)] = .{
.llvm_name = "atomic-fmin-fmax-flat-f32",
.description = "Has flat memory instructions for atomicrmw fmin/fmax for float",
.dependencies = featureSet(&[_]Feature{
.flat_address_space,
}),
};
result[@backingInt(Feature.atomic_fmin_fmax_flat_f64)] = .{
.llvm_name = "atomic-fmin-fmax-flat-f64",
.description = "Has flat memory instructions for atomicrmw fmin/fmax for double",
.dependencies = featureSet(&[_]Feature{
.flat_address_space,
}),
};
result[@backingInt(Feature.atomic_fmin_fmax_global_f32)] = .{
.llvm_name = "atomic-fmin-fmax-global-f32",
.description = "Has global/buffer instructions for atomicrmw fmin/fmax for float",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.atomic_fmin_fmax_global_f64)] = .{
.llvm_name = "atomic-fmin-fmax-global-f64",
.description = "Has global/buffer instructions for atomicrmw fmin/fmax for float",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.atomic_global_pk_add_bf16_inst)] = .{
.llvm_name = "atomic-global-pk-add-bf16-inst",
.description = "Has global_atomic_pk_add_bf16 instruction",
.dependencies = featureSet(&[_]Feature{
.flat_global_insts,
}),
};
result[@backingInt(Feature.auto_waitcnt_before_barrier)] = .{
.llvm_name = "auto-waitcnt-before-barrier",
.description = "Hardware automatically inserts waitcnt before barrier",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.back_off_barrier)] = .{
.llvm_name = "back-off-barrier",
.description = "Hardware supports backing off s_barrier if an exception occurs",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.bf16_cvt_insts)] = .{
.llvm_name = "bf16-cvt-insts",
.description = "Has bf16 conversion instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.bf16_pk_insts)] = .{
.llvm_name = "bf16-pk-insts",
.description = "Has bf16 packed instructions (fma, add, mul, max, min)",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.bf16_trans_insts)] = .{
.llvm_name = "bf16-trans-insts",
.description = "Has bf16 transcendental instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.bf8_cvt_scale_insts)] = .{
.llvm_name = "bf8-cvt-scale-insts",
.description = "Has bf8 conversion scale instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.bitop3_insts)] = .{
.llvm_name = "bitop3-insts",
.description = "Has v_bitop3_b32/v_bitop3_b16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.block_vgpr_csr)] = .{
.llvm_name = "block-vgpr-csr",
.description = "Use block load/store for VGPR callee saved registers",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.bvh_dual_bvh_8_insts)] = .{
.llvm_name = "bvh-dual-bvh-8-insts",
.description = "Has image_bvh_dual_intersect_ray and image_bvh8_intersect_ray instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.ci_insts)] = .{
.llvm_name = "ci-insts",
.description = "Additional instructions for CI+",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.clusters)] = .{
.llvm_name = "clusters",
.description = "Has clusters of workgroups support",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.cube_insts)] = .{
.llvm_name = "cube-insts",
.description = "Has v_cube* instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.cumode)] = .{
.llvm_name = "cumode",
.description = "Enable CU wavefront execution mode",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.cvt_fp8_vop1_bug)] = .{
.llvm_name = "cvt-fp8-vop1-bug",
.description = "FP8/BF8 VOP1 form of conversion to F32 is unreliable",
.dependencies = featureSet(&[_]Feature{
.fp8_conversion_insts,
}),
};
result[@backingInt(Feature.cvt_norm_insts)] = .{
.llvm_name = "cvt-norm-insts",
.description = "Has v_cvt_norm* instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.cvt_pk_f16_f32_inst)] = .{
.llvm_name = "cvt-pk-f16-f32-inst",
.description = "Has cvt_pk_f16_f32 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.cvt_pknorm_vop2_insts)] = .{
.llvm_name = "cvt-pknorm-vop2-insts",
.description = "Has v_cvt_pk_norm_*f32 instructions/Has v_cvt_pk_norm_*_f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.cvt_pknorm_vop3_insts)] = .{
.llvm_name = "cvt-pknorm-vop3-insts",
.description = "Has v_cvt_pk_norm_*f32 instructions/Has v_cvt_pk_norm_*_f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.d16_write_vgpr32)] = .{
.llvm_name = "d16-write-vgpr32",
.description = "D16 instructions potentially have 32-bit data dependencies",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.default_component_broadcast)] = .{
.llvm_name = "default-component-broadcast",
.description = "BUFFER/IMAGE store instructions set unspecified components to x component (GFX12)",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.default_component_zero)] = .{
.llvm_name = "default-component-zero",
.description = "BUFFER/IMAGE store instructions set unspecified components to zero (before GFX12)",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dl_insts)] = .{
.llvm_name = "dl-insts",
.description = "Has v_fmac_f32 and v_xnor_b32 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot10_insts)] = .{
.llvm_name = "dot10-insts",
.description = "Has v_dot2_f32_f16 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot11_insts)] = .{
.llvm_name = "dot11-insts",
.description = "Has v_dot4_f32_fp8_fp8, v_dot4_f32_fp8_bf8, v_dot4_f32_bf8_fp8, v_dot4_f32_bf8_bf8 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot12_insts)] = .{
.llvm_name = "dot12-insts",
.description = "Has v_dot2_f32_bf16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot13_insts)] = .{
.llvm_name = "dot13-insts",
.description = "Has v_dot2c_f32_bf16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot1_insts)] = .{
.llvm_name = "dot1-insts",
.description = "Has v_dot4_i32_i8 and v_dot8_i32_i4 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot2_insts)] = .{
.llvm_name = "dot2-insts",
.description = "Has v_dot2_i32_i16, v_dot2_u32_u16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot3_insts)] = .{
.llvm_name = "dot3-insts",
.description = "Has v_dot8c_i32_i4 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot4_insts)] = .{
.llvm_name = "dot4-insts",
.description = "Has v_dot2c_i32_i16 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot5_insts)] = .{
.llvm_name = "dot5-insts",
.description = "Has v_dot2c_f32_f16 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot6_insts)] = .{
.llvm_name = "dot6-insts",
.description = "Has v_dot4c_i32_i8 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot7_insts)] = .{
.llvm_name = "dot7-insts",
.description = "Has v_dot4_u32_u8, v_dot8_u32_u4 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot8_insts)] = .{
.llvm_name = "dot8-insts",
.description = "Has v_dot4_i32_iu8, v_dot8_i32_iu4 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dot9_insts)] = .{
.llvm_name = "dot9-insts",
.description = "Has v_dot2_f16_f16, v_dot2_bf16_bf16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dpp)] = .{
.llvm_name = "dpp",
.description = "Support DPP (Data Parallel Primitives) extension",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dpp8)] = .{
.llvm_name = "dpp8",
.description = "Support DPP8 (Data Parallel Primitives) extension",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dpp_64bit)] = .{
.llvm_name = "dpp-64bit",
.description = "Support DPP (Data Parallel Primitives) extension in DP ALU",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.dpp_src1_sgpr)] = .{
.llvm_name = "dpp-src1-sgpr",
.description = "Support SGPR for Src1 of DPP instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.ds128)] = .{
.llvm_name = "enable-ds128",
.description = "Use ds_{read|write}_b128",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.ds_src2_insts)] = .{
.llvm_name = "ds-src2-insts",
.description = "Has ds_*_src2 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.emulated_system_scope_atomics)] = .{
.llvm_name = "emulated-system-scope-atomics",
.description = "System scope atomics unsupported by the PCI-e are emulated in HW via CAS loop and functional.",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.extended_image_insts)] = .{
.llvm_name = "extended-image-insts",
.description = "Support mips != 0, lod != 0, gather4, and get_lod",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.f16bf16_to_fp6bf6_cvt_scale_insts)] = .{
.llvm_name = "f16bf16-to-fp6bf6-cvt-scale-insts",
.description = "Has f16bf16 to fp6bf6 conversion scale instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.f32_to_f16bf16_cvt_sr_insts)] = .{
.llvm_name = "f32-to-f16bf16-cvt-sr-insts",
.description = "Has f32 to f16bf16 conversion scale instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fast_denormal_f32)] = .{
.llvm_name = "fast-denormal-f32",
.description = "Enabling denormals does not cause f32 instructions to run at f64 rates",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fast_fmaf)] = .{
.llvm_name = "fast-fmaf",
.description = "Assuming f32 fma is at least as fast as mul + add",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.flat_address_space)] = .{
.llvm_name = "flat-address-space",
.description = "Support flat address space",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.flat_atomic_fadd_f32_inst)] = .{
.llvm_name = "flat-atomic-fadd-f32-inst",
.description = "Has flat_atomic_add_f32 instruction",
.dependencies = featureSet(&[_]Feature{
.flat_address_space,
}),
};
result[@backingInt(Feature.flat_buffer_global_fadd_f64_inst)] = .{
.llvm_name = "flat-buffer-global-fadd-f64-inst",
.description = "Has flat, buffer, and global instructions for f64 atomic fadd",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.flat_for_global)] = .{
.llvm_name = "flat-for-global",
.description = "Force to generate flat instruction for global",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.flat_global_insts)] = .{
.llvm_name = "flat-global-insts",
.description = "Have global_* flat memory instructions",
.dependencies = featureSet(&[_]Feature{
.flat_address_space,
}),
};
result[@backingInt(Feature.flat_gvs_mode)] = .{
.llvm_name = "flat-gvs-mode",
.description = "Have GVS addressing mode with flat_* instructions",
.dependencies = featureSet(&[_]Feature{
.flat_address_space,
}),
};
result[@backingInt(Feature.flat_inst_offsets)] = .{
.llvm_name = "flat-inst-offsets",
.description = "Flat instructions have immediate offset addressing mode",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.flat_scratch)] = .{
.llvm_name = "enable-flat-scratch",
.description = "Use scratch_* flat memory instructions to access scratch",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.flat_scratch_insts)] = .{
.llvm_name = "flat-scratch-insts",
.description = "Have scratch_* flat memory instructions",
.dependencies = featureSet(&[_]Feature{
.flat_address_space,
}),
};
result[@backingInt(Feature.flat_segment_offset_bug)] = .{
.llvm_name = "flat-segment-offset-bug",
.description = "GFX10 bug where inst_offset is ignored when flat instructions access global memory",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fma_mix_bf16_insts)] = .{
.llvm_name = "fma-mix-bf16-insts",
.description = "Has v_fma_mix_f32_bf16, v_fma_mixlo_bf16, v_fma_mixhi_bf16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fma_mix_insts)] = .{
.llvm_name = "fma-mix-insts",
.description = "Has v_fma_mix_f32, v_fma_mixlo_f16, v_fma_mixhi_f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fmacf64_inst)] = .{
.llvm_name = "fmacf64-inst",
.description = "Has v_fmac_f64 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fmaf)] = .{
.llvm_name = "fmaf",
.description = "Enable single precision FMA (not as fast as mul+add, but fused)",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fp4_cvt_scale_insts)] = .{
.llvm_name = "fp4-cvt-scale-insts",
.description = "Has fp4 conversion scale instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fp64)] = .{
.llvm_name = "fp64",
.description = "Enable double precision operations",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fp6bf6_cvt_scale_insts)] = .{
.llvm_name = "fp6bf6-cvt-scale-insts",
.description = "Has fp6 and bf6 conversion scale instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fp8_conversion_insts)] = .{
.llvm_name = "fp8-conversion-insts",
.description = "Has fp8 and bf8 conversion instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fp8_cvt_scale_insts)] = .{
.llvm_name = "fp8-cvt-scale-insts",
.description = "Has fp8 conversion scale instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fp8_insts)] = .{
.llvm_name = "fp8-insts",
.description = "Has fp8 and bf8 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.fp8e5m3_insts)] = .{
.llvm_name = "fp8e5m3-insts",
.description = "Has fp8 e5m3 format support",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.full_rate_64_ops)] = .{
.llvm_name = "full-rate-64-ops",
.description = "Most fp64 instructions are full rate",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.g16)] = .{
.llvm_name = "g16",
.description = "Support G16 for 16-bit gradient image operands",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gcn3_encoding)] = .{
.llvm_name = "gcn3-encoding",
.description = "Encoding format for VI",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gds)] = .{
.llvm_name = "gds",
.description = "Has Global Data Share",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.get_wave_id_inst)] = .{
.llvm_name = "get-wave-id-inst",
.description = "Has s_get_waveid_in_workgroup instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx10)] = .{
.llvm_name = "gfx10",
.description = "GFX10 GPU generation",
.dependencies = featureSet(&[_]Feature{
.@"16_bit_insts",
.a16,
.add_no_carry_insts,
.addressablelocalmemorysize65536,
.aperture_regs,
.atomic_fmin_fmax_flat_f32,
.atomic_fmin_fmax_flat_f64,
.atomic_fmin_fmax_global_f32,
.atomic_fmin_fmax_global_f64,
.ci_insts,
.cube_insts,
.cvt_norm_insts,
.cvt_pknorm_vop2_insts,
.cvt_pknorm_vop3_insts,
.default_component_zero,
.dpp,
.dpp8,
.extended_image_insts,
.fast_denormal_f32,
.fast_fmaf,
.flat_global_insts,
.flat_inst_offsets,
.flat_scratch_insts,
.fma_mix_insts,
.fp64,
.g16,
.gds,
.gfx10_insts,
.gfx8_insts,
.gfx9_insts,
.gws,
.image_insts,
.int_clamp_insts,
.inv_2pi_inline_imm,
.lerp_inst,
.max_hard_clause_length_63,
.mimg_r128,
.movrel,
.no_data_dep_hazard,
.no_sdst_cmpx,
.pk_fmac_f16_inst,
.qsad_insts,
.s_memrealtime,
.s_memtime_inst,
.sad_insts,
.sdwa,
.sdwa_omod,
.sdwa_scalar,
.sdwa_sdst,
.unaligned_buffer_access,
.unaligned_ds_access,
.unaligned_scratch_access,
.vmem_to_lds_load_insts,
.vmem_write_vgpr_in_order,
.vop3_literal,
.vop3p,
.vscnt,
}),
};
result[@backingInt(Feature.gfx10_3_insts)] = .{
.llvm_name = "gfx10-3-insts",
.description = "Additional instructions for GFX10.3",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx10_a_encoding)] = .{
.llvm_name = "gfx10_a-encoding",
.description = "Has BVH ray tracing instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx10_b_encoding)] = .{
.llvm_name = "gfx10_b-encoding",
.description = "Encoding format GFX10_B",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx10_insts)] = .{
.llvm_name = "gfx10-insts",
.description = "Additional instructions for GFX10+",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx11)] = .{
.llvm_name = "gfx11",
.description = "GFX11 GPU generation",
.dependencies = featureSet(&[_]Feature{
.@"16_bit_insts",
.a16,
.add_no_carry_insts,
.addressablelocalmemorysize65536,
.aperture_regs,
.atomic_fmin_fmax_flat_f32,
.atomic_fmin_fmax_global_f32,
.ci_insts,
.cube_insts,
.cvt_norm_insts,
.cvt_pknorm_vop2_insts,
.cvt_pknorm_vop3_insts,
.default_component_zero,
.dpp,
.dpp8,
.extended_image_insts,
.fast_denormal_f32,
.fast_fmaf,
.flat_global_insts,
.flat_inst_offsets,
.flat_scratch_insts,
.fma_mix_insts,
.fp64,
.g16,
.gds,
.gfx10_3_insts,
.gfx10_a_encoding,
.gfx10_b_encoding,
.gfx10_insts,
.gfx11_insts,
.gfx8_insts,
.gfx9_insts,
.gws,
.int_clamp_insts,
.inv_2pi_inline_imm,
.lerp_inst,
.max_hard_clause_length_32,
.mimg_r128,
.movrel,
.no_data_dep_hazard,
.no_sdst_cmpx,
.pk_fmac_f16_inst,
.qsad_insts,
.sad_insts,
.true16,
.unaligned_buffer_access,
.unaligned_ds_access,
.unaligned_scratch_access,
.vmem_write_vgpr_in_order,
.vop3_literal,
.vop3p,
.vopd,
.vscnt,
}),
};
result[@backingInt(Feature.gfx11_insts)] = .{
.llvm_name = "gfx11-insts",
.description = "Additional instructions for GFX11+",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx12)] = .{
.llvm_name = "gfx12",
.description = "GFX12 GPU generation",
.dependencies = featureSet(&[_]Feature{
.@"16_bit_insts",
.a16,
.add_no_carry_insts,
.agent_scope_fine_grained_remote_memory_atomics,
.aperture_regs,
.atomic_fmin_fmax_flat_f32,
.atomic_fmin_fmax_global_f32,
.ci_insts,
.default_component_broadcast,
.dpp,
.dpp8,
.fast_denormal_f32,
.fast_fmaf,
.flat_global_insts,
.flat_inst_offsets,
.flat_scratch_insts,
.fma_mix_insts,
.fp64,
.g16,
.gfx10_3_insts,
.gfx10_a_encoding,
.gfx10_b_encoding,
.gfx10_insts,
.gfx11_insts,
.gfx12_insts,
.gfx8_insts,
.gfx9_insts,
.ieee_minimum_maximum_insts,
.int_clamp_insts,
.inv_2pi_inline_imm,
.max_hard_clause_length_32,
.mimg_r128,
.minimum3_maximum3_f16,
.minimum3_maximum3_f32,
.movrel,
.no_data_dep_hazard,
.no_sdst_cmpx,
.pk_fmac_f16_inst,
.true16,
.unaligned_buffer_access,
.unaligned_ds_access,
.unaligned_scratch_access,
.vop3_literal,
.vop3p,
.vopd,
.vscnt,
}),
};
result[@backingInt(Feature.gfx1250_insts)] = .{
.llvm_name = "gfx1250-insts",
.description = "Additional instructions for GFX1250+",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx12_insts)] = .{
.llvm_name = "gfx12-insts",
.description = "Additional instructions for GFX12+",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx7_gfx8_gfx9_insts)] = .{
.llvm_name = "gfx7-gfx8-gfx9-insts",
.description = "Instructions shared in GFX7, GFX8, GFX9",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx8_insts)] = .{
.llvm_name = "gfx8-insts",
.description = "Additional instructions for GFX8+",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx9)] = .{
.llvm_name = "gfx9",
.description = "GFX9 GPU generation",
.dependencies = featureSet(&[_]Feature{
.@"16_bit_insts",
.a16,
.add_no_carry_insts,
.aperture_regs,
.ci_insts,
.cube_insts,
.cvt_norm_insts,
.cvt_pknorm_vop2_insts,
.cvt_pknorm_vop3_insts,
.default_component_zero,
.dpp,
.fast_denormal_f32,
.fast_fmaf,
.flat_global_insts,
.flat_inst_offsets,
.flat_scratch_insts,
.fp64,
.gcn3_encoding,
.gfx7_gfx8_gfx9_insts,
.gfx8_insts,
.gfx9_insts,
.gws,
.int_clamp_insts,
.inv_2pi_inline_imm,
.lerp_inst,
.negative_scratch_offset_bug,
.qsad_insts,
.r128_a16,
.s_memrealtime,
.s_memtime_inst,
.sad_insts,
.scalar_atomics,
.scalar_flat_scratch_insts,
.scalar_stores,
.sdwa,
.sdwa_omod,
.sdwa_scalar,
.sdwa_sdst,
.unaligned_buffer_access,
.unaligned_ds_access,
.unaligned_scratch_access,
.vgpr_index_mode,
.vmem_to_lds_load_insts,
.vmem_write_vgpr_in_order,
.vop3p,
.wavefrontsize64,
.xnack_support,
}),
};
result[@backingInt(Feature.gfx90a_insts)] = .{
.llvm_name = "gfx90a-insts",
.description = "Additional instructions for GFX90A+",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx940_insts)] = .{
.llvm_name = "gfx940-insts",
.description = "Additional instructions for GFX940+",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gfx950_insts)] = .{
.llvm_name = "gfx950-insts",
.description = "Additional instructions for GFX950+",
.dependencies = featureSet(&[_]Feature{
.ashr_pk_insts,
.bf8_cvt_scale_insts,
.cvt_pk_f16_f32_inst,
.f16bf16_to_fp6bf6_cvt_scale_insts,
.f32_to_f16bf16_cvt_sr_insts,
.fp4_cvt_scale_insts,
.fp6bf6_cvt_scale_insts,
.fp8_cvt_scale_insts,
.minimum3_maximum3_f32,
.minimum3_maximum3_pkf16,
.permlane16_swap,
.permlane32_swap,
}),
};
result[@backingInt(Feature.gfx9_insts)] = .{
.llvm_name = "gfx9-insts",
.description = "Additional instructions for GFX9+",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.globally_addressable_scratch)] = .{
.llvm_name = "globally-addressable-scratch",
.description = "FLAT instructions can access scratch memory for any thread in any wave",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.gws)] = .{
.llvm_name = "gws",
.description = "Has Global Wave Sync",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.half_rate_64_ops)] = .{
.llvm_name = "half-rate-64-ops",
.description = "Most fp64 instructions are half rate instead of quarter",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.ieee_minimum_maximum_insts)] = .{
.llvm_name = "ieee-minimum-maximum-insts",
.description = "Has v_minimum/maximum_f16/f32/f64, v_minimummaximum/maximumminimum_f16/f32 and v_pk_minimum/maximum_f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.image_gather4_d16_bug)] = .{
.llvm_name = "image-gather4-d16-bug",
.description = "Image Gather4 D16 hardware bug",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.image_insts)] = .{
.llvm_name = "image-insts",
.description = "Support image instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.image_store_d16_bug)] = .{
.llvm_name = "image-store-d16-bug",
.description = "Image Store D16 hardware bug",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.inst_fwd_prefetch_bug)] = .{
.llvm_name = "inst-fwd-prefetch-bug",
.description = "S_INST_PREFETCH instruction causes shader to hang",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.int_clamp_insts)] = .{
.llvm_name = "int-clamp-insts",
.description = "Support clamp for integer destination",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.inv_2pi_inline_imm)] = .{
.llvm_name = "inv-2pi-inline-imm",
.description = "Has 1 / (2 * pi) as inline immediate",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.kernarg_preload)] = .{
.llvm_name = "kernarg-preload",
.description = "Hardware supports preloading of kernel arguments in user SGPRs.",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.lds_barrier_arrive_atomic)] = .{
.llvm_name = "lds-barrier-arrive-atomic",
.description = "Has LDS barrier-arrive atomic instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.lds_branch_vmem_war_hazard)] = .{
.llvm_name = "lds-branch-vmem-war-hazard",
.description = "Switching between LDS and VMEM-tex not waiting VM_VSRC=0",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.lds_misaligned_bug)] = .{
.llvm_name = "lds-misaligned-bug",
.description = "Some GFX10 bug with multi-dword LDS and flat access that is not naturally aligned in WGP mode",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.ldsbankcount16)] = .{
.llvm_name = "ldsbankcount16",
.description = "The number of LDS banks per compute unit.",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.ldsbankcount32)] = .{
.llvm_name = "ldsbankcount32",
.description = "The number of LDS banks per compute unit.",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.lerp_inst)] = .{
.llvm_name = "lerp-inst",
.description = "Has v_lerp_u8 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.load_store_opt)] = .{
.llvm_name = "load-store-opt",
.description = "Enable SI load/store optimizer pass",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.lshl_add_u64_inst)] = .{
.llvm_name = "lshl-add-u64-inst",
.description = "Has v_lshl_add_u64 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.mad_intra_fwd_bug)] = .{
.llvm_name = "mad-intra-fwd-bug",
.description = "MAD_U64/I64 intra instruction forwarding bug",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.mad_mac_f32_insts)] = .{
.llvm_name = "mad-mac-f32-insts",
.description = "Has v_mad_f32/v_mac_f32/v_madak_f32/v_madmk_f32 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.mad_mix_insts)] = .{
.llvm_name = "mad-mix-insts",
.description = "Has v_mad_mix_f32, v_mad_mixlo_f16, v_mad_mixhi_f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.mad_u32_inst)] = .{
.llvm_name = "mad-u32-inst",
.description = "Has v_mad_u32 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.mai_insts)] = .{
.llvm_name = "mai-insts",
.description = "Has mAI instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.max_hard_clause_length_32)] = .{
.llvm_name = "max-hard-clause-length-32",
.description = "Maximum number of instructions in an explicit S_CLAUSE is 32",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.max_hard_clause_length_63)] = .{
.llvm_name = "max-hard-clause-length-63",
.description = "Maximum number of instructions in an explicit S_CLAUSE is 63",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.max_private_element_size_16)] = .{
.llvm_name = "max-private-element-size-16",
.description = "Maximum private access size may be 16",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.max_private_element_size_4)] = .{
.llvm_name = "max-private-element-size-4",
.description = "Maximum private access size may be 4",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.max_private_element_size_8)] = .{
.llvm_name = "max-private-element-size-8",
.description = "Maximum private access size may be 8",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.mcast_load_insts)] = .{
.llvm_name = "mcast-load-insts",
.description = "Has multicast load instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.memory_atomic_fadd_f32_denormal_support)] = .{
.llvm_name = "memory-atomic-fadd-f32-denormal-support",
.description = "global/flat/buffer atomic fadd for float supports denormal handling",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.mfma_inline_literal_bug)] = .{
.llvm_name = "mfma-inline-literal-bug",
.description = "MFMA cannot use inline literal as SrcC",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.mimg_r128)] = .{
.llvm_name = "mimg-r128",
.description = "Support 128-bit texture resources",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.min3_max3_pkf16)] = .{
.llvm_name = "min3-max3-pkf16",
.description = "Has v_pk_min3_num_f16 and v_pk_max3_num_f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.minimum3_maximum3_f16)] = .{
.llvm_name = "minimum3-maximum3-f16",
.description = "Has v_minimum3_f16 and v_maximum3_f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.minimum3_maximum3_f32)] = .{
.llvm_name = "minimum3-maximum3-f32",
.description = "Has v_minimum3_f32 and v_maximum3_f32 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.minimum3_maximum3_pkf16)] = .{
.llvm_name = "minimum3-maximum3-pkf16",
.description = "Has v_pk_minimum3_f16 and v_pk_maximum3_f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.movrel)] = .{
.llvm_name = "movrel",
.description = "Has v_movrel*_b32 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.msaa_load_dst_sel_bug)] = .{
.llvm_name = "msaa-load-dst-sel-bug",
.description = "MSAA loads not honoring dst_sel bug",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.negative_scratch_offset_bug)] = .{
.llvm_name = "negative-scratch-offset-bug",
.description = "Negative immediate offsets in scratch instructions with an SGPR offset page fault on GFX9",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.negative_unaligned_scratch_offset_bug)] = .{
.llvm_name = "negative-unaligned-scratch-offset-bug",
.description = "Scratch instructions with a VGPR offset and a negative immediate offset that is not a multiple of 4 read wrong memory on GFX10",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.no_data_dep_hazard)] = .{
.llvm_name = "no-data-dep-hazard",
.description = "Does not need SW waitstates",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.no_sdst_cmpx)] = .{
.llvm_name = "no-sdst-cmpx",
.description = "V_CMPX does not write VCC/SGPR in addition to EXEC",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.nsa_clause_bug)] = .{
.llvm_name = "nsa-clause-bug",
.description = "MIMG-NSA in a hard clause has unpredictable results on GFX10.1",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.nsa_encoding)] = .{
.llvm_name = "nsa-encoding",
.description = "Support NSA encoding for image instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.nsa_to_vmem_bug)] = .{
.llvm_name = "nsa-to-vmem-bug",
.description = "MIMG-NSA followed by VMEM fail if EXEC_LO or EXEC_HI equals zero",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.offset_3f_bug)] = .{
.llvm_name = "offset-3f-bug",
.description = "Branch offset of 3f hardware bug",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.packed_fp32_ops)] = .{
.llvm_name = "packed-fp32-ops",
.description = "Support packed fp32 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.packed_tid)] = .{
.llvm_name = "packed-tid",
.description = "Workitem IDs are packed into v0 at kernel launch",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.partial_nsa_encoding)] = .{
.llvm_name = "partial-nsa-encoding",
.description = "Support partial NSA encoding for image instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.permlane16_swap)] = .{
.llvm_name = "permlane16-swap",
.description = "Has v_permlane16_swap_b32 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.permlane32_swap)] = .{
.llvm_name = "permlane32-swap",
.description = "Has v_permlane32_swap_b32 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.pk_add_min_max_insts)] = .{
.llvm_name = "pk-add-min-max-insts",
.description = "Has v_pk_add_{min|max}_{i|u}16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.pk_fmac_f16_inst)] = .{
.llvm_name = "pk-fmac-f16-inst",
.description = "Has v_pk_fmac_f16 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.point_sample_accel)] = .{
.llvm_name = "point-sample-accel",
.description = "Has point sample acceleration feature",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.precise_memory)] = .{
.llvm_name = "precise-memory",
.description = "Enable precise memory mode",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.priv_enabled_trap2_nop_bug)] = .{
.llvm_name = "priv-enabled-trap2-nop-bug",
.description = "Hardware that runs with PRIV=1 interpreting 's_trap 2' as a nop bug",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.prng_inst)] = .{
.llvm_name = "prng-inst",
.description = "Has v_prng_b32 instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.promote_alloca)] = .{
.llvm_name = "promote-alloca",
.description = "Enable promote alloca pass",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.prt_strict_null)] = .{
.llvm_name = "enable-prt-strict-null",
.description = "Enable zeroing of result registers for sparse texture fetches",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.pseudo_scalar_trans)] = .{
.llvm_name = "pseudo-scalar-trans",
.description = "Has Pseudo Scalar Transcendental instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.qsad_insts)] = .{
.llvm_name = "qsad-insts",
.description = "Has v_qsad* instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.r128_a16)] = .{
.llvm_name = "r128-a16",
.description = "Support gfx9-style A16 for 16-bit coordinates/gradients/lod/clamp/mip image operands, where a16 is aliased with r128",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.real_true16)] = .{
.llvm_name = "real-true16",
.description = "Use true 16-bit registers",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.relaxed_buffer_oob_mode)] = .{
.llvm_name = "relaxed-buffer-oob-mode",
.description = "Disable strict out-of-bounds buffer guarantees. An OOB access may potentially cause an adjacent access to be treated as if it were also OOB",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.required_export_priority)] = .{
.llvm_name = "required-export-priority",
.description = "Export priority must be explicitly manipulated on GFX11.5",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.requires_cov6)] = .{
.llvm_name = "requires-cov6",
.description = "Target Requires Code Object V6",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.restricted_soffset)] = .{
.llvm_name = "restricted-soffset",
.description = "Has restricted SOffset (immediate not supported).",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.s_memrealtime)] = .{
.llvm_name = "s-memrealtime",
.description = "Has s_memrealtime instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.s_memtime_inst)] = .{
.llvm_name = "s-memtime-inst",
.description = "Has s_memtime instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.s_wakeup_barrier_inst)] = .{
.llvm_name = "s-wakeup-barrier-inst",
.description = "Has s_wakeup_barrier instruction.",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.sad_insts)] = .{
.llvm_name = "sad-insts",
.description = "Has v_sad* instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.safe_cu_prefetch)] = .{
.llvm_name = "safe-cu-prefetch",
.description = "VMEM CU scope prefetches do not fail on illegal address",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.safe_smem_prefetch)] = .{
.llvm_name = "safe-smem-prefetch",
.description = "SMEM prefetches do not fail on illegal address",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.salu_float)] = .{
.llvm_name = "salu-float",
.description = "Has SALU floating point instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.scalar_atomics)] = .{
.llvm_name = "scalar-atomics",
.description = "Has atomic scalar memory instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.scalar_dwordx3_loads)] = .{
.llvm_name = "scalar-dwordx3-loads",
.description = "Has 96-bit scalar load instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.scalar_flat_scratch_insts)] = .{
.llvm_name = "scalar-flat-scratch-insts",
.description = "Have s_scratch_* flat memory instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.scalar_stores)] = .{
.llvm_name = "scalar-stores",
.description = "Has store scalar memory instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.sdwa)] = .{
.llvm_name = "sdwa",
.description = "Support SDWA (Sub-DWORD Addressing) extension",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.sdwa_mav)] = .{
.llvm_name = "sdwa-mav",
.description = "Support v_mac_f32/f16 with SDWA (Sub-DWORD Addressing) extension",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.sdwa_omod)] = .{
.llvm_name = "sdwa-omod",
.description = "Support OMod with SDWA (Sub-DWORD Addressing) extension",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.sdwa_out_mods_vopc)] = .{
.llvm_name = "sdwa-out-mods-vopc",
.description = "Support clamp for VOPC with SDWA (Sub-DWORD Addressing) extension",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.sdwa_scalar)] = .{
.llvm_name = "sdwa-scalar",
.description = "Support scalar register with SDWA (Sub-DWORD Addressing) extension",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.sdwa_sdst)] = .{
.llvm_name = "sdwa-sdst",
.description = "Support scalar dst for VOPC with SDWA (Sub-DWORD Addressing) extension",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.sea_islands)] = .{
.llvm_name = "sea-islands",
.description = "SEA_ISLANDS GPU generation",
.dependencies = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.atomic_fmin_fmax_flat_f32,
.atomic_fmin_fmax_flat_f64,
.atomic_fmin_fmax_global_f32,
.atomic_fmin_fmax_global_f64,
.ci_insts,
.cube_insts,
.cvt_pknorm_vop2_insts,
.default_component_zero,
.ds_src2_insts,
.extended_image_insts,
.fp64,
.gds,
.gfx7_gfx8_gfx9_insts,
.gws,
.image_insts,
.lerp_inst,
.mad_mac_f32_insts,
.mimg_r128,
.movrel,
.qsad_insts,
.s_memtime_inst,
.sad_insts,
.trig_reduced_range,
.unaligned_buffer_access,
.vmem_write_vgpr_in_order,
.wavefrontsize64,
}),
};
result[@backingInt(Feature.setprio_inc_wg_inst)] = .{
.llvm_name = "setprio-inc-wg-inst",
.description = "Has s_setprio_inc_wg instruction.",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.setreg_vgpr_msb_fixup)] = .{
.llvm_name = "setreg-vgpr-msb-fixup",
.description = "S_SETREG to MODE clobbers VGPR MSB bits, requires fixup",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.sgpr_init_bug)] = .{
.llvm_name = "sgpr-init-bug",
.description = "VI SGPR initialization bug requiring a fixed SGPR allocation size",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.shader_cycles_hi_lo_registers)] = .{
.llvm_name = "shader-cycles-hi-lo-registers",
.description = "Has SHADER_CYCLES_HI/LO hardware registers",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.shader_cycles_register)] = .{
.llvm_name = "shader-cycles-register",
.description = "Has SHADER_CYCLES hardware register",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.si_scheduler)] = .{
.llvm_name = "si-scheduler",
.description = "Enable SI Machine Scheduler",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.smem_to_vector_write_hazard)] = .{
.llvm_name = "smem-to-vector-write-hazard",
.description = "s_load_dword followed by v_cmp page faults",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.southern_islands)] = .{
.llvm_name = "southern-islands",
.description = "SOUTHERN_ISLANDS GPU generation",
.dependencies = featureSet(&[_]Feature{
.addressablelocalmemorysize32768,
.atomic_fmin_fmax_global_f32,
.atomic_fmin_fmax_global_f64,
.cube_insts,
.cvt_pknorm_vop2_insts,
.default_component_zero,
.ds_src2_insts,
.extended_image_insts,
.fp64,
.gds,
.gws,
.image_insts,
.ldsbankcount32,
.lerp_inst,
.mad_mac_f32_insts,
.mimg_r128,
.movrel,
.s_memtime_inst,
.sad_insts,
.trig_reduced_range,
.vmem_write_vgpr_in_order,
.wavefrontsize64,
}),
};
result[@backingInt(Feature.sramecc)] = .{
.llvm_name = "sramecc",
.description = "Enable SRAMECC",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.sramecc_support)] = .{
.llvm_name = "sramecc-support",
.description = "Hardware supports SRAMECC",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.tanh_insts)] = .{
.llvm_name = "tanh-insts",
.description = "Has v_tanh_f32/f16 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.tensor_cvt_lut_insts)] = .{
.llvm_name = "tensor-cvt-lut-insts",
.description = "Has v_perm_pk16* instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.tgsplit)] = .{
.llvm_name = "tgsplit",
.description = "Enable threadgroup split execution",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.transpose_load_f4f6_insts)] = .{
.llvm_name = "transpose-load-f4f6-insts",
.description = "Has ds_load_tr4/tr6 and global_load_tr4/tr6 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.trap_handler)] = .{
.llvm_name = "trap-handler",
.description = "Trap handler support",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.trig_reduced_range)] = .{
.llvm_name = "trig-reduced-range",
.description = "Requires use of fract on arguments to trig instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.true16)] = .{
.llvm_name = "true16",
.description = "True 16-bit operand instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.unaligned_access_mode)] = .{
.llvm_name = "unaligned-access-mode",
.description = "Enable unaligned global, local and region loads and stores if the hardware supports it",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.unaligned_buffer_access)] = .{
.llvm_name = "unaligned-buffer-access",
.description = "Hardware supports unaligned global loads and stores",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.unaligned_ds_access)] = .{
.llvm_name = "unaligned-ds-access",
.description = "Hardware supports unaligned local and region loads and stores",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.unaligned_scratch_access)] = .{
.llvm_name = "unaligned-scratch-access",
.description = "Support unaligned scratch loads and stores",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.unpacked_d16_vmem)] = .{
.llvm_name = "unpacked-d16-vmem",
.description = "Has unpacked d16 vmem instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.unsafe_ds_offset_folding)] = .{
.llvm_name = "unsafe-ds-offset-folding",
.description = "Force using DS instruction immediate offsets on SI",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.user_sgpr_init16_bug)] = .{
.llvm_name = "user-sgpr-init16-bug",
.description = "Bug requiring at least 16 user+system SGPRs to be enabled",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.valu_trans_use_hazard)] = .{
.llvm_name = "valu-trans-use-hazard",
.description = "Hazard when TRANS instructions are closely followed by a use of the result",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vcmpx_exec_war_hazard)] = .{
.llvm_name = "vcmpx-exec-war-hazard",
.description = "V_CMPX WAR hazard on EXEC (V_CMPX issue ONLY)",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vcmpx_permlane_hazard)] = .{
.llvm_name = "vcmpx-permlane-hazard",
.description = "TODO: describe me",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vgpr_align2)] = .{
.llvm_name = "vgpr-align2",
.description = "VGPR and AGPR tuple operands require even alignment",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vgpr_index_mode)] = .{
.llvm_name = "vgpr-index-mode",
.description = "Has VGPR mode register indexing",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vmem_pref_insts)] = .{
.llvm_name = "vmem-pref-insts",
.description = "Has flat_prefect_b8 and global_prefetch_b8 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vmem_to_lds_load_insts)] = .{
.llvm_name = "vmem-to-lds-load-insts",
.description = "The platform has memory to lds instructions (global_load w/lds bit set, buffer_load w/lds bit set or global_load_lds. This does not include scratch_load_lds.",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vmem_to_scalar_write_hazard)] = .{
.llvm_name = "vmem-to-scalar-write-hazard",
.description = "VMEM instruction followed by scalar writing to EXEC mask, M0 or SGPR leads to incorrect execution.",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vmem_write_vgpr_in_order)] = .{
.llvm_name = "vmem-write-vgpr-in-order",
.description = "VMEM instructions of the same type write VGPR results in order",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.volcanic_islands)] = .{
.llvm_name = "volcanic-islands",
.description = "VOLCANIC_ISLANDS GPU generation",
.dependencies = featureSet(&[_]Feature{
.@"16_bit_insts",
.addressablelocalmemorysize65536,
.ci_insts,
.cube_insts,
.cvt_pknorm_vop2_insts,
.default_component_zero,
.dpp,
.ds_src2_insts,
.extended_image_insts,
.fast_denormal_f32,
.flat_address_space,
.fp64,
.gcn3_encoding,
.gds,
.gfx7_gfx8_gfx9_insts,
.gfx8_insts,
.gws,
.image_insts,
.int_clamp_insts,
.inv_2pi_inline_imm,
.lerp_inst,
.mad_mac_f32_insts,
.mimg_r128,
.movrel,
.qsad_insts,
.s_memrealtime,
.s_memtime_inst,
.sad_insts,
.scalar_stores,
.sdwa,
.sdwa_mav,
.sdwa_out_mods_vopc,
.trig_reduced_range,
.unaligned_buffer_access,
.vgpr_index_mode,
.vmem_write_vgpr_in_order,
.wavefrontsize64,
}),
};
result[@backingInt(Feature.vop3_literal)] = .{
.llvm_name = "vop3-literal",
.description = "Can use one literal in VOP3",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vop3p)] = .{
.llvm_name = "vop3p",
.description = "Has VOP3P packed instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vopd)] = .{
.llvm_name = "vopd",
.description = "Has VOPD dual issue wave32 instructions",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.vscnt)] = .{
.llvm_name = "vscnt",
.description = "Has separate store vscnt counter",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.wait_xcnt)] = .{
.llvm_name = "wait-xcnt",
.description = "Has s_wait_xcnt instruction",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.waits_before_system_scope_stores)] = .{
.llvm_name = "waits-before-system-scope-stores",
.description = "Target requires waits for loads and atomics before system scope stores",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.wavefrontsize16)] = .{
.llvm_name = "wavefrontsize16",
.description = "The number of threads per wavefront",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.wavefrontsize32)] = .{
.llvm_name = "wavefrontsize32",
.description = "The number of threads per wavefront",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.wavefrontsize64)] = .{
.llvm_name = "wavefrontsize64",
.description = "The number of threads per wavefront",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.xf32_insts)] = .{
.llvm_name = "xf32-insts",
.description = "Has instructions that support xf32 format, such as v_mfma_f32_16x16x8_xf32 and v_mfma_f32_32x32x4_xf32",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.xnack)] = .{
.llvm_name = "xnack",
.description = "Enable XNACK support",
.dependencies = featureSet(&[_]Feature{}),
};
result[@backingInt(Feature.xnack_support)] = .{
.llvm_name = "xnack-support",
.description = "Hardware supports XNACK",
.dependencies = featureSet(&[_]Feature{}),
};
const ti = @typeInfo(Feature);
for (&result, 0..) |*elem, i| {
elem.index = i;
elem.name = ti.@"enum".field_names[i];
}
break :blk result;
};
pub const cpu = struct {
pub const bonaire: CpuModel = .{
.name = "bonaire",
.llvm_name = "bonaire",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.sea_islands,
}),
};
pub const carrizo: CpuModel = .{
.name = "carrizo",
.llvm_name = "carrizo",
.features = featureSet(&[_]Feature{
.fast_fmaf,
.half_rate_64_ops,
.ldsbankcount32,
.unpacked_d16_vmem,
.volcanic_islands,
.xnack_support,
}),
};
pub const fiji: CpuModel = .{
.name = "fiji",
.llvm_name = "fiji",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.unpacked_d16_vmem,
.volcanic_islands,
}),
};
pub const generic: CpuModel = .{
.name = "generic",
.llvm_name = "generic",
.features = featureSet(&[_]Feature{}),
};
pub const generic_hsa: CpuModel = .{
.name = "generic_hsa",
.llvm_name = "generic-hsa",
.features = featureSet(&[_]Feature{
.flat_address_space,
}),
};
pub const gfx1010: CpuModel = .{
.name = "gfx1010",
.llvm_name = "gfx1010",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.ds_src2_insts,
.flat_segment_offset_bug,
.get_wave_id_inst,
.gfx10,
.inst_fwd_prefetch_bug,
.lds_branch_vmem_war_hazard,
.lds_misaligned_bug,
.ldsbankcount32,
.mad_mac_f32_insts,
.negative_unaligned_scratch_offset_bug,
.nsa_clause_bug,
.nsa_encoding,
.nsa_to_vmem_bug,
.offset_3f_bug,
.scalar_atomics,
.scalar_flat_scratch_insts,
.scalar_stores,
.smem_to_vector_write_hazard,
.vcmpx_exec_war_hazard,
.vcmpx_permlane_hazard,
.vmem_to_scalar_write_hazard,
.xnack_support,
}),
};
pub const gfx1011: CpuModel = .{
.name = "gfx1011",
.llvm_name = "gfx1011",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.ds_src2_insts,
.flat_segment_offset_bug,
.get_wave_id_inst,
.gfx10,
.inst_fwd_prefetch_bug,
.lds_branch_vmem_war_hazard,
.lds_misaligned_bug,
.ldsbankcount32,
.mad_mac_f32_insts,
.negative_unaligned_scratch_offset_bug,
.nsa_clause_bug,
.nsa_encoding,
.nsa_to_vmem_bug,
.offset_3f_bug,
.scalar_atomics,
.scalar_flat_scratch_insts,
.scalar_stores,
.smem_to_vector_write_hazard,
.vcmpx_exec_war_hazard,
.vcmpx_permlane_hazard,
.vmem_to_scalar_write_hazard,
.xnack_support,
}),
};
pub const gfx1012: CpuModel = .{
.name = "gfx1012",
.llvm_name = "gfx1012",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.ds_src2_insts,
.flat_segment_offset_bug,
.get_wave_id_inst,
.gfx10,
.inst_fwd_prefetch_bug,
.lds_branch_vmem_war_hazard,
.lds_misaligned_bug,
.ldsbankcount32,
.mad_mac_f32_insts,
.negative_unaligned_scratch_offset_bug,
.nsa_clause_bug,
.nsa_encoding,
.nsa_to_vmem_bug,
.offset_3f_bug,
.scalar_atomics,
.scalar_flat_scratch_insts,
.scalar_stores,
.smem_to_vector_write_hazard,
.vcmpx_exec_war_hazard,
.vcmpx_permlane_hazard,
.vmem_to_scalar_write_hazard,
.xnack_support,
}),
};
pub const gfx1013: CpuModel = .{
.name = "gfx1013",
.llvm_name = "gfx1013",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.ds_src2_insts,
.flat_segment_offset_bug,
.get_wave_id_inst,
.gfx10,
.gfx10_a_encoding,
.inst_fwd_prefetch_bug,
.lds_branch_vmem_war_hazard,
.lds_misaligned_bug,
.ldsbankcount32,
.mad_mac_f32_insts,
.negative_unaligned_scratch_offset_bug,
.nsa_clause_bug,
.nsa_encoding,
.nsa_to_vmem_bug,
.offset_3f_bug,
.scalar_atomics,
.scalar_flat_scratch_insts,
.scalar_stores,
.smem_to_vector_write_hazard,
.vcmpx_exec_war_hazard,
.vcmpx_permlane_hazard,
.vmem_to_scalar_write_hazard,
.xnack_support,
}),
};
pub const gfx1030: CpuModel = .{
.name = "gfx1030",
.llvm_name = "gfx1030",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.gfx10,
.gfx10_3_insts,
.gfx10_a_encoding,
.gfx10_b_encoding,
.ldsbankcount32,
.nsa_encoding,
.shader_cycles_register,
}),
};
pub const gfx1031: CpuModel = .{
.name = "gfx1031",
.llvm_name = "gfx1031",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.gfx10,
.gfx10_3_insts,
.gfx10_a_encoding,
.gfx10_b_encoding,
.ldsbankcount32,
.nsa_encoding,
.shader_cycles_register,
}),
};
pub const gfx1032: CpuModel = .{
.name = "gfx1032",
.llvm_name = "gfx1032",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.gfx10,
.gfx10_3_insts,
.gfx10_a_encoding,
.gfx10_b_encoding,
.ldsbankcount32,
.nsa_encoding,
.shader_cycles_register,
}),
};
pub const gfx1033: CpuModel = .{
.name = "gfx1033",
.llvm_name = "gfx1033",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.gfx10,
.gfx10_3_insts,
.gfx10_a_encoding,
.gfx10_b_encoding,
.ldsbankcount32,
.nsa_encoding,
.shader_cycles_register,
}),
};
pub const gfx1034: CpuModel = .{
.name = "gfx1034",
.llvm_name = "gfx1034",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.gfx10,
.gfx10_3_insts,
.gfx10_a_encoding,
.gfx10_b_encoding,
.ldsbankcount32,
.nsa_encoding,
.shader_cycles_register,
}),
};
pub const gfx1035: CpuModel = .{
.name = "gfx1035",
.llvm_name = "gfx1035",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.gfx10,
.gfx10_3_insts,
.gfx10_a_encoding,
.gfx10_b_encoding,
.ldsbankcount32,
.nsa_encoding,
.shader_cycles_register,
}),
};
pub const gfx1036: CpuModel = .{
.name = "gfx1036",
.llvm_name = "gfx1036",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.gfx10,
.gfx10_3_insts,
.gfx10_a_encoding,
.gfx10_b_encoding,
.ldsbankcount32,
.nsa_encoding,
.shader_cycles_register,
}),
};
pub const gfx10_1_generic: CpuModel = .{
.name = "gfx10_1_generic",
.llvm_name = "gfx10-1-generic",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.ds_src2_insts,
.flat_segment_offset_bug,
.get_wave_id_inst,
.gfx10,
.inst_fwd_prefetch_bug,
.lds_branch_vmem_war_hazard,
.lds_misaligned_bug,
.ldsbankcount32,
.mad_mac_f32_insts,
.negative_unaligned_scratch_offset_bug,
.nsa_clause_bug,
.nsa_encoding,
.nsa_to_vmem_bug,
.offset_3f_bug,
.requires_cov6,
.scalar_atomics,
.scalar_flat_scratch_insts,
.scalar_stores,
.smem_to_vector_write_hazard,
.vcmpx_exec_war_hazard,
.vcmpx_permlane_hazard,
.vmem_to_scalar_write_hazard,
.xnack_support,
}),
};
pub const gfx10_3_generic: CpuModel = .{
.name = "gfx10_3_generic",
.llvm_name = "gfx10-3-generic",
.features = featureSet(&[_]Feature{
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.gfx10,
.gfx10_3_insts,
.gfx10_a_encoding,
.gfx10_b_encoding,
.ldsbankcount32,
.nsa_encoding,
.requires_cov6,
.shader_cycles_register,
}),
};
pub const gfx1100: CpuModel = .{
.name = "gfx1100",
.llvm_name = "gfx1100",
.features = featureSet(&[_]Feature{
.allocate1_5xvgprs,
.architected_flat_scratch,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.back_off_barrier,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot12_insts,
.dot5_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.flat_atomic_fadd_f32_inst,
.gfx11,
.image_insts,
.ldsbankcount32,
.mad_intra_fwd_bug,
.memory_atomic_fadd_f32_denormal_support,
.msaa_load_dst_sel_bug,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.priv_enabled_trap2_nop_bug,
.real_true16,
.shader_cycles_register,
.user_sgpr_init16_bug,
.valu_trans_use_hazard,
.vcmpx_permlane_hazard,
}),
};
pub const gfx1101: CpuModel = .{
.name = "gfx1101",
.llvm_name = "gfx1101",
.features = featureSet(&[_]Feature{
.allocate1_5xvgprs,
.architected_flat_scratch,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.back_off_barrier,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot12_insts,
.dot5_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.flat_atomic_fadd_f32_inst,
.gfx11,
.image_insts,
.ldsbankcount32,
.mad_intra_fwd_bug,
.memory_atomic_fadd_f32_denormal_support,
.msaa_load_dst_sel_bug,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.priv_enabled_trap2_nop_bug,
.real_true16,
.shader_cycles_register,
.valu_trans_use_hazard,
.vcmpx_permlane_hazard,
}),
};
pub const gfx1102: CpuModel = .{
.name = "gfx1102",
.llvm_name = "gfx1102",
.features = featureSet(&[_]Feature{
.architected_flat_scratch,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.back_off_barrier,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot12_insts,
.dot5_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.flat_atomic_fadd_f32_inst,
.gfx11,
.image_insts,
.ldsbankcount32,
.mad_intra_fwd_bug,
.memory_atomic_fadd_f32_denormal_support,
.msaa_load_dst_sel_bug,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.priv_enabled_trap2_nop_bug,
.real_true16,
.shader_cycles_register,
.user_sgpr_init16_bug,
.valu_trans_use_hazard,
.vcmpx_permlane_hazard,
}),
};
pub const gfx1103: CpuModel = .{
.name = "gfx1103",
.llvm_name = "gfx1103",
.features = featureSet(&[_]Feature{
.architected_flat_scratch,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.back_off_barrier,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot12_insts,
.dot5_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.flat_atomic_fadd_f32_inst,
.gfx11,
.image_insts,
.ldsbankcount32,
.mad_intra_fwd_bug,
.memory_atomic_fadd_f32_denormal_support,
.msaa_load_dst_sel_bug,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.priv_enabled_trap2_nop_bug,
.real_true16,
.shader_cycles_register,
.valu_trans_use_hazard,
.vcmpx_permlane_hazard,
}),
};
pub const gfx1150: CpuModel = .{
.name = "gfx1150",
.llvm_name = "gfx1150",
.features = featureSet(&[_]Feature{
.architected_flat_scratch,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.back_off_barrier,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot12_insts,
.dot5_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.dpp_src1_sgpr,
.flat_atomic_fadd_f32_inst,
.gfx11,
.image_insts,
.ldsbankcount32,
.memory_atomic_fadd_f32_denormal_support,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.point_sample_accel,
.real_true16,
.required_export_priority,
.salu_float,
.shader_cycles_register,
.vcmpx_permlane_hazard,
}),
};
pub const gfx1151: CpuModel = .{
.name = "gfx1151",
.llvm_name = "gfx1151",
.features = featureSet(&[_]Feature{
.allocate1_5xvgprs,
.architected_flat_scratch,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.back_off_barrier,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot12_insts,
.dot5_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.dpp_src1_sgpr,
.flat_atomic_fadd_f32_inst,
.gfx11,
.image_insts,
.ldsbankcount32,
.memory_atomic_fadd_f32_denormal_support,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.point_sample_accel,
.real_true16,
.required_export_priority,
.salu_float,
.shader_cycles_register,
.vcmpx_permlane_hazard,
}),
};
pub const gfx1152: CpuModel = .{
.name = "gfx1152",
.llvm_name = "gfx1152",
.features = featureSet(&[_]Feature{
.architected_flat_scratch,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.back_off_barrier,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot12_insts,
.dot5_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.dpp_src1_sgpr,
.flat_atomic_fadd_f32_inst,
.gfx11,
.image_insts,
.ldsbankcount32,
.memory_atomic_fadd_f32_denormal_support,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.point_sample_accel,
.real_true16,
.required_export_priority,
.salu_float,
.shader_cycles_register,
.vcmpx_permlane_hazard,
}),
};
pub const gfx1153: CpuModel = .{
.name = "gfx1153",
.llvm_name = "gfx1153",
.features = featureSet(&[_]Feature{
.architected_flat_scratch,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.back_off_barrier,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot12_insts,
.dot5_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.dpp_src1_sgpr,
.flat_atomic_fadd_f32_inst,
.gfx11,
.image_insts,
.ldsbankcount32,
.memory_atomic_fadd_f32_denormal_support,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.real_true16,
.required_export_priority,
.salu_float,
.shader_cycles_register,
.vcmpx_permlane_hazard,
}),
};
pub const gfx11_generic: CpuModel = .{
.name = "gfx11_generic",
.llvm_name = "gfx11-generic",
.features = featureSet(&[_]Feature{
.architected_flat_scratch,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.back_off_barrier,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot12_insts,
.dot5_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.flat_atomic_fadd_f32_inst,
.gfx11,
.image_insts,
.ldsbankcount32,
.mad_intra_fwd_bug,
.memory_atomic_fadd_f32_denormal_support,
.msaa_load_dst_sel_bug,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.priv_enabled_trap2_nop_bug,
.real_true16,
.required_export_priority,
.requires_cov6,
.shader_cycles_register,
.user_sgpr_init16_bug,
.valu_trans_use_hazard,
.vcmpx_permlane_hazard,
}),
};
pub const gfx1200: CpuModel = .{
.name = "gfx1200",
.llvm_name = "gfx1200",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.allocate1_5xvgprs,
.architected_flat_scratch,
.architected_sgprs,
.atomic_buffer_global_pk_add_f16_insts,
.atomic_buffer_pk_add_bf16_inst,
.atomic_ds_pk_add_16_insts,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.atomic_flat_pk_add_16_insts,
.atomic_global_pk_add_bf16_inst,
.back_off_barrier,
.bvh_dual_bvh_8_insts,
.cube_insts,
.cvt_norm_insts,
.cvt_pknorm_vop2_insts,
.cvt_pknorm_vop3_insts,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot11_insts,
.dot12_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.dpp_src1_sgpr,
.extended_image_insts,
.flat_atomic_fadd_f32_inst,
.fp8_conversion_insts,
.gfx12,
.image_insts,
.ldsbankcount32,
.lerp_inst,
.memory_atomic_fadd_f32_denormal_support,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.pseudo_scalar_trans,
.qsad_insts,
.restricted_soffset,
.sad_insts,
.salu_float,
.scalar_dwordx3_loads,
.shader_cycles_hi_lo_registers,
.vcmpx_permlane_hazard,
.waits_before_system_scope_stores,
}),
};
pub const gfx1201: CpuModel = .{
.name = "gfx1201",
.llvm_name = "gfx1201",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.allocate1_5xvgprs,
.architected_flat_scratch,
.architected_sgprs,
.atomic_buffer_global_pk_add_f16_insts,
.atomic_buffer_pk_add_bf16_inst,
.atomic_ds_pk_add_16_insts,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.atomic_flat_pk_add_16_insts,
.atomic_global_pk_add_bf16_inst,
.back_off_barrier,
.bvh_dual_bvh_8_insts,
.cube_insts,
.cvt_norm_insts,
.cvt_pknorm_vop2_insts,
.cvt_pknorm_vop3_insts,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot11_insts,
.dot12_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.dpp_src1_sgpr,
.extended_image_insts,
.flat_atomic_fadd_f32_inst,
.fp8_conversion_insts,
.gfx12,
.image_insts,
.ldsbankcount32,
.lerp_inst,
.memory_atomic_fadd_f32_denormal_support,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.pseudo_scalar_trans,
.qsad_insts,
.restricted_soffset,
.sad_insts,
.salu_float,
.scalar_dwordx3_loads,
.shader_cycles_hi_lo_registers,
.vcmpx_permlane_hazard,
.waits_before_system_scope_stores,
}),
};
pub const gfx1250: CpuModel = .{
.name = "gfx1250",
.llvm_name = "gfx1250",
.features = featureSet(&[_]Feature{
.@"1024_addressable_vgprs",
.@"45_bit_num_records_buffer_resource",
.@"64_bit_literals",
.add_min_max_insts,
.add_sub_u64_insts,
.addressablelocalmemorysize327680,
.architected_flat_scratch,
.architected_sgprs,
.ashr_pk_insts,
.atomic_buffer_global_pk_add_f16_insts,
.atomic_buffer_pk_add_bf16_inst,
.atomic_ds_pk_add_16_insts,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.atomic_flat_pk_add_16_insts,
.atomic_fmin_fmax_flat_f64,
.atomic_fmin_fmax_global_f64,
.atomic_global_pk_add_bf16_inst,
.bf16_cvt_insts,
.bf16_pk_insts,
.bf16_trans_insts,
.bitop3_insts,
.clusters,
.cube_insts,
.cumode,
.cvt_norm_insts,
.cvt_pk_f16_f32_inst,
.cvt_pknorm_vop2_insts,
.cvt_pknorm_vop3_insts,
.d16_write_vgpr32,
.dl_insts,
.dot7_insts,
.dot8_insts,
.dpp_src1_sgpr,
.emulated_system_scope_atomics,
.flat_atomic_fadd_f32_inst,
.flat_buffer_global_fadd_f64_inst,
.flat_gvs_mode,
.fma_mix_bf16_insts,
.fmacf64_inst,
.fp8_conversion_insts,
.fp8e5m3_insts,
.gfx12,
.gfx1250_insts,
.globally_addressable_scratch,
.kernarg_preload,
.lds_barrier_arrive_atomic,
.ldsbankcount32,
.lerp_inst,
.lshl_add_u64_inst,
.mad_u32_inst,
.max_hard_clause_length_63,
.mcast_load_insts,
.memory_atomic_fadd_f32_denormal_support,
.min3_max3_pkf16,
.minimum3_maximum3_pkf16,
.packed_fp32_ops,
.packed_tid,
.permlane16_swap,
.pk_add_min_max_insts,
.prng_inst,
.pseudo_scalar_trans,
.qsad_insts,
.restricted_soffset,
.s_wakeup_barrier_inst,
.sad_insts,
.salu_float,
.scalar_dwordx3_loads,
.setprio_inc_wg_inst,
.setreg_vgpr_msb_fixup,
.shader_cycles_hi_lo_registers,
.sramecc_support,
.tanh_insts,
.tensor_cvt_lut_insts,
.transpose_load_f4f6_insts,
.vcmpx_permlane_hazard,
.vgpr_align2,
.vmem_pref_insts,
.wait_xcnt,
.wavefrontsize32,
.xnack,
.xnack_support,
}),
};
pub const gfx1251: CpuModel = .{
.name = "gfx1251",
.llvm_name = "gfx1251",
.features = featureSet(&[_]Feature{
.@"1024_addressable_vgprs",
.@"45_bit_num_records_buffer_resource",
.@"64_bit_literals",
.add_min_max_insts,
.add_sub_u64_insts,
.addressablelocalmemorysize327680,
.architected_flat_scratch,
.architected_sgprs,
.ashr_pk_insts,
.atomic_buffer_global_pk_add_f16_insts,
.atomic_buffer_pk_add_bf16_inst,
.atomic_ds_pk_add_16_insts,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.atomic_flat_pk_add_16_insts,
.atomic_fmin_fmax_flat_f64,
.atomic_fmin_fmax_global_f64,
.atomic_global_pk_add_bf16_inst,
.bf16_cvt_insts,
.bf16_pk_insts,
.bf16_trans_insts,
.bitop3_insts,
.clusters,
.cube_insts,
.cumode,
.cvt_norm_insts,
.cvt_pk_f16_f32_inst,
.cvt_pknorm_vop2_insts,
.cvt_pknorm_vop3_insts,
.d16_write_vgpr32,
.dl_insts,
.dot7_insts,
.dot8_insts,
.dpp_64bit,
.dpp_src1_sgpr,
.emulated_system_scope_atomics,
.flat_atomic_fadd_f32_inst,
.flat_buffer_global_fadd_f64_inst,
.flat_gvs_mode,
.fma_mix_bf16_insts,
.fmacf64_inst,
.fp8_conversion_insts,
.fp8e5m3_insts,
.gfx12,
.gfx1250_insts,
.globally_addressable_scratch,
.kernarg_preload,
.lds_barrier_arrive_atomic,
.ldsbankcount32,
.lerp_inst,
.lshl_add_u64_inst,
.mad_u32_inst,
.max_hard_clause_length_63,
.mcast_load_insts,
.memory_atomic_fadd_f32_denormal_support,
.min3_max3_pkf16,
.minimum3_maximum3_pkf16,
.packed_fp32_ops,
.packed_tid,
.permlane16_swap,
.pk_add_min_max_insts,
.prng_inst,
.pseudo_scalar_trans,
.qsad_insts,
.restricted_soffset,
.s_wakeup_barrier_inst,
.sad_insts,
.salu_float,
.scalar_dwordx3_loads,
.setprio_inc_wg_inst,
.shader_cycles_hi_lo_registers,
.sramecc_support,
.tanh_insts,
.tensor_cvt_lut_insts,
.transpose_load_f4f6_insts,
.vcmpx_permlane_hazard,
.vgpr_align2,
.vmem_pref_insts,
.wait_xcnt,
.wavefrontsize32,
.xnack,
.xnack_support,
}),
};
pub const gfx12_generic: CpuModel = .{
.name = "gfx12_generic",
.llvm_name = "gfx12-generic",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.allocate1_5xvgprs,
.architected_flat_scratch,
.architected_sgprs,
.atomic_buffer_global_pk_add_f16_insts,
.atomic_buffer_pk_add_bf16_inst,
.atomic_ds_pk_add_16_insts,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.atomic_flat_pk_add_16_insts,
.atomic_global_pk_add_bf16_inst,
.back_off_barrier,
.bvh_dual_bvh_8_insts,
.cube_insts,
.cvt_norm_insts,
.cvt_pknorm_vop2_insts,
.cvt_pknorm_vop3_insts,
.d16_write_vgpr32,
.dl_insts,
.dot10_insts,
.dot11_insts,
.dot12_insts,
.dot7_insts,
.dot8_insts,
.dot9_insts,
.dpp_src1_sgpr,
.extended_image_insts,
.flat_atomic_fadd_f32_inst,
.fp8_conversion_insts,
.gfx12,
.image_insts,
.ldsbankcount32,
.lerp_inst,
.memory_atomic_fadd_f32_denormal_support,
.nsa_encoding,
.packed_tid,
.partial_nsa_encoding,
.pseudo_scalar_trans,
.qsad_insts,
.requires_cov6,
.restricted_soffset,
.sad_insts,
.salu_float,
.scalar_dwordx3_loads,
.shader_cycles_hi_lo_registers,
.vcmpx_permlane_hazard,
.waits_before_system_scope_stores,
}),
};
pub const gfx600: CpuModel = .{
.name = "gfx600",
.llvm_name = "gfx600",
.features = featureSet(&[_]Feature{
.fast_fmaf,
.half_rate_64_ops,
.southern_islands,
}),
};
pub const gfx601: CpuModel = .{
.name = "gfx601",
.llvm_name = "gfx601",
.features = featureSet(&[_]Feature{
.southern_islands,
}),
};
pub const gfx602: CpuModel = .{
.name = "gfx602",
.llvm_name = "gfx602",
.features = featureSet(&[_]Feature{
.southern_islands,
}),
};
pub const gfx700: CpuModel = .{
.name = "gfx700",
.llvm_name = "gfx700",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.sea_islands,
}),
};
pub const gfx701: CpuModel = .{
.name = "gfx701",
.llvm_name = "gfx701",
.features = featureSet(&[_]Feature{
.fast_fmaf,
.half_rate_64_ops,
.ldsbankcount32,
.sea_islands,
}),
};
pub const gfx702: CpuModel = .{
.name = "gfx702",
.llvm_name = "gfx702",
.features = featureSet(&[_]Feature{
.fast_fmaf,
.ldsbankcount16,
.sea_islands,
}),
};
pub const gfx703: CpuModel = .{
.name = "gfx703",
.llvm_name = "gfx703",
.features = featureSet(&[_]Feature{
.ldsbankcount16,
.sea_islands,
}),
};
pub const gfx704: CpuModel = .{
.name = "gfx704",
.llvm_name = "gfx704",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.sea_islands,
}),
};
pub const gfx705: CpuModel = .{
.name = "gfx705",
.llvm_name = "gfx705",
.features = featureSet(&[_]Feature{
.ldsbankcount16,
.sea_islands,
}),
};
pub const gfx801: CpuModel = .{
.name = "gfx801",
.llvm_name = "gfx801",
.features = featureSet(&[_]Feature{
.fast_fmaf,
.half_rate_64_ops,
.ldsbankcount32,
.unpacked_d16_vmem,
.volcanic_islands,
.xnack_support,
}),
};
pub const gfx802: CpuModel = .{
.name = "gfx802",
.llvm_name = "gfx802",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.sgpr_init_bug,
.unpacked_d16_vmem,
.volcanic_islands,
}),
};
pub const gfx803: CpuModel = .{
.name = "gfx803",
.llvm_name = "gfx803",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.unpacked_d16_vmem,
.volcanic_islands,
}),
};
pub const gfx805: CpuModel = .{
.name = "gfx805",
.llvm_name = "gfx805",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.sgpr_init_bug,
.unpacked_d16_vmem,
.volcanic_islands,
}),
};
pub const gfx810: CpuModel = .{
.name = "gfx810",
.llvm_name = "gfx810",
.features = featureSet(&[_]Feature{
.image_gather4_d16_bug,
.image_store_d16_bug,
.ldsbankcount16,
.volcanic_islands,
.xnack_support,
}),
};
pub const gfx900: CpuModel = .{
.name = "gfx900",
.llvm_name = "gfx900",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.ds_src2_insts,
.extended_image_insts,
.gds,
.gfx9,
.image_gather4_d16_bug,
.image_insts,
.ldsbankcount32,
.mad_mac_f32_insts,
.mad_mix_insts,
}),
};
pub const gfx902: CpuModel = .{
.name = "gfx902",
.llvm_name = "gfx902",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.ds_src2_insts,
.extended_image_insts,
.gds,
.gfx9,
.image_gather4_d16_bug,
.image_insts,
.ldsbankcount32,
.mad_mac_f32_insts,
.mad_mix_insts,
}),
};
pub const gfx904: CpuModel = .{
.name = "gfx904",
.llvm_name = "gfx904",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.ds_src2_insts,
.extended_image_insts,
.fma_mix_insts,
.gds,
.gfx9,
.image_gather4_d16_bug,
.image_insts,
.ldsbankcount32,
.mad_mac_f32_insts,
}),
};
pub const gfx906: CpuModel = .{
.name = "gfx906",
.llvm_name = "gfx906",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot7_insts,
.ds_src2_insts,
.extended_image_insts,
.fma_mix_insts,
.gds,
.gfx9,
.half_rate_64_ops,
.image_gather4_d16_bug,
.image_insts,
.ldsbankcount32,
.mad_mac_f32_insts,
.sramecc_support,
}),
};
pub const gfx908: CpuModel = .{
.name = "gfx908",
.llvm_name = "gfx908",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.atomic_buffer_global_pk_add_f16_no_rtn_insts,
.atomic_fadd_no_rtn_insts,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot3_insts,
.dot4_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.ds_src2_insts,
.extended_image_insts,
.fma_mix_insts,
.gds,
.gfx9,
.half_rate_64_ops,
.image_gather4_d16_bug,
.image_insts,
.ldsbankcount32,
.mad_mac_f32_insts,
.mai_insts,
.mfma_inline_literal_bug,
.pk_fmac_f16_inst,
.sramecc_support,
}),
};
pub const gfx909: CpuModel = .{
.name = "gfx909",
.llvm_name = "gfx909",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.ds_src2_insts,
.extended_image_insts,
.gds,
.gfx9,
.image_gather4_d16_bug,
.image_insts,
.ldsbankcount32,
.mad_mac_f32_insts,
.mad_mix_insts,
}),
};
pub const gfx90a: CpuModel = .{
.name = "gfx90a",
.llvm_name = "gfx90a",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.atomic_buffer_global_pk_add_f16_insts,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.atomic_fmin_fmax_flat_f64,
.atomic_fmin_fmax_global_f64,
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot3_insts,
.dot4_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.dpp_64bit,
.flat_buffer_global_fadd_f64_inst,
.fma_mix_insts,
.fmacf64_inst,
.full_rate_64_ops,
.gfx9,
.gfx90a_insts,
.image_insts,
.kernarg_preload,
.ldsbankcount32,
.mad_mac_f32_insts,
.mai_insts,
.packed_fp32_ops,
.packed_tid,
.pk_fmac_f16_inst,
.sramecc_support,
.vgpr_align2,
}),
};
pub const gfx90c: CpuModel = .{
.name = "gfx90c",
.llvm_name = "gfx90c",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.ds_src2_insts,
.extended_image_insts,
.gds,
.gfx9,
.image_gather4_d16_bug,
.image_insts,
.ldsbankcount32,
.mad_mac_f32_insts,
.mad_mix_insts,
}),
};
pub const gfx942: CpuModel = .{
.name = "gfx942",
.llvm_name = "gfx942",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.agent_scope_fine_grained_remote_memory_atomics,
.architected_flat_scratch,
.atomic_buffer_global_pk_add_f16_insts,
.atomic_ds_pk_add_16_insts,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.atomic_flat_pk_add_16_insts,
.atomic_fmin_fmax_flat_f64,
.atomic_fmin_fmax_global_f64,
.atomic_global_pk_add_bf16_inst,
.back_off_barrier,
.cvt_fp8_vop1_bug,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot3_insts,
.dot4_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.dpp_64bit,
.flat_atomic_fadd_f32_inst,
.flat_buffer_global_fadd_f64_inst,
.fma_mix_insts,
.fmacf64_inst,
.fp8_insts,
.full_rate_64_ops,
.gfx9,
.gfx90a_insts,
.gfx940_insts,
.kernarg_preload,
.ldsbankcount32,
.lshl_add_u64_inst,
.mai_insts,
.memory_atomic_fadd_f32_denormal_support,
.packed_fp32_ops,
.packed_tid,
.pk_fmac_f16_inst,
.sramecc_support,
.vgpr_align2,
.xf32_insts,
}),
};
pub const gfx950: CpuModel = .{
.name = "gfx950",
.llvm_name = "gfx950",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize163840,
.agent_scope_fine_grained_remote_memory_atomics,
.architected_flat_scratch,
.atomic_buffer_global_pk_add_f16_insts,
.atomic_buffer_pk_add_bf16_inst,
.atomic_ds_pk_add_16_insts,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.atomic_flat_pk_add_16_insts,
.atomic_fmin_fmax_flat_f64,
.atomic_fmin_fmax_global_f64,
.atomic_global_pk_add_bf16_inst,
.back_off_barrier,
.bf16_cvt_insts,
.bitop3_insts,
.dl_insts,
.dot10_insts,
.dot12_insts,
.dot13_insts,
.dot1_insts,
.dot2_insts,
.dot3_insts,
.dot4_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.dpp_64bit,
.flat_atomic_fadd_f32_inst,
.flat_buffer_global_fadd_f64_inst,
.fma_mix_insts,
.fmacf64_inst,
.fp8_conversion_insts,
.fp8_insts,
.full_rate_64_ops,
.gfx9,
.gfx90a_insts,
.gfx940_insts,
.gfx950_insts,
.kernarg_preload,
.ldsbankcount32,
.lshl_add_u64_inst,
.mai_insts,
.memory_atomic_fadd_f32_denormal_support,
.packed_fp32_ops,
.packed_tid,
.pk_fmac_f16_inst,
.prng_inst,
.sramecc_support,
.vgpr_align2,
}),
};
pub const gfx9_4_generic: CpuModel = .{
.name = "gfx9_4_generic",
.llvm_name = "gfx9-4-generic",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.agent_scope_fine_grained_remote_memory_atomics,
.architected_flat_scratch,
.atomic_buffer_global_pk_add_f16_insts,
.atomic_ds_pk_add_16_insts,
.atomic_fadd_no_rtn_insts,
.atomic_fadd_rtn_insts,
.atomic_flat_pk_add_16_insts,
.atomic_fmin_fmax_flat_f64,
.atomic_fmin_fmax_global_f64,
.atomic_global_pk_add_bf16_inst,
.back_off_barrier,
.dl_insts,
.dot10_insts,
.dot1_insts,
.dot2_insts,
.dot3_insts,
.dot4_insts,
.dot5_insts,
.dot6_insts,
.dot7_insts,
.dpp_64bit,
.flat_atomic_fadd_f32_inst,
.flat_buffer_global_fadd_f64_inst,
.fma_mix_insts,
.fmacf64_inst,
.full_rate_64_ops,
.gfx9,
.gfx90a_insts,
.gfx940_insts,
.kernarg_preload,
.ldsbankcount32,
.lshl_add_u64_inst,
.mai_insts,
.memory_atomic_fadd_f32_denormal_support,
.packed_fp32_ops,
.packed_tid,
.pk_fmac_f16_inst,
.requires_cov6,
.sramecc_support,
.vgpr_align2,
}),
};
pub const gfx9_generic: CpuModel = .{
.name = "gfx9_generic",
.llvm_name = "gfx9-generic",
.features = featureSet(&[_]Feature{
.addressablelocalmemorysize65536,
.ds_src2_insts,
.extended_image_insts,
.gds,
.gfx9,
.image_gather4_d16_bug,
.image_insts,
.ldsbankcount32,
.mad_mac_f32_insts,
.requires_cov6,
}),
};
pub const hainan: CpuModel = .{
.name = "hainan",
.llvm_name = "hainan",
.features = featureSet(&[_]Feature{
.southern_islands,
}),
};
pub const hawaii: CpuModel = .{
.name = "hawaii",
.llvm_name = "hawaii",
.features = featureSet(&[_]Feature{
.fast_fmaf,
.half_rate_64_ops,
.ldsbankcount32,
.sea_islands,
}),
};
pub const iceland: CpuModel = .{
.name = "iceland",
.llvm_name = "iceland",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.sgpr_init_bug,
.unpacked_d16_vmem,
.volcanic_islands,
}),
};
pub const kabini: CpuModel = .{
.name = "kabini",
.llvm_name = "kabini",
.features = featureSet(&[_]Feature{
.ldsbankcount16,
.sea_islands,
}),
};
pub const kaveri: CpuModel = .{
.name = "kaveri",
.llvm_name = "kaveri",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.sea_islands,
}),
};
pub const mullins: CpuModel = .{
.name = "mullins",
.llvm_name = "mullins",
.features = featureSet(&[_]Feature{
.ldsbankcount16,
.sea_islands,
}),
};
pub const oland: CpuModel = .{
.name = "oland",
.llvm_name = "oland",
.features = featureSet(&[_]Feature{
.southern_islands,
}),
};
pub const pitcairn: CpuModel = .{
.name = "pitcairn",
.llvm_name = "pitcairn",
.features = featureSet(&[_]Feature{
.southern_islands,
}),
};
pub const polaris10: CpuModel = .{
.name = "polaris10",
.llvm_name = "polaris10",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.unpacked_d16_vmem,
.volcanic_islands,
}),
};
pub const polaris11: CpuModel = .{
.name = "polaris11",
.llvm_name = "polaris11",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.unpacked_d16_vmem,
.volcanic_islands,
}),
};
pub const stoney: CpuModel = .{
.name = "stoney",
.llvm_name = "stoney",
.features = featureSet(&[_]Feature{
.image_gather4_d16_bug,
.image_store_d16_bug,
.ldsbankcount16,
.volcanic_islands,
.xnack_support,
}),
};
pub const tahiti: CpuModel = .{
.name = "tahiti",
.llvm_name = "tahiti",
.features = featureSet(&[_]Feature{
.fast_fmaf,
.half_rate_64_ops,
.southern_islands,
}),
};
pub const tonga: CpuModel = .{
.name = "tonga",
.llvm_name = "tonga",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.sgpr_init_bug,
.unpacked_d16_vmem,
.volcanic_islands,
}),
};
pub const tongapro: CpuModel = .{
.name = "tongapro",
.llvm_name = "tongapro",
.features = featureSet(&[_]Feature{
.ldsbankcount32,
.sgpr_init_bug,
.unpacked_d16_vmem,
.volcanic_islands,
}),
};
pub const verde: CpuModel = .{
.name = "verde",
.llvm_name = "verde",
.features = featureSet(&[_]Feature{
.southern_islands,
}),
};
};