cubit
Open-source SM120 (Blackwell) SASS assembler — encode, schedule & assemble GPU machine code, including all tensor-core instructions.
File Explorer
- ci.yml
- __init__.py
- assembler.py
- ENCODER_GUARDS.md
- MERC13_COMPANIONS.md
- nvcc-kloop-reference.md
- POSTFIX103_STALLFIX.md
- PYTHON_API_ARCH_SELECT.md
- sasstuning.md
- sm120_undocumented_instructions.md
- SM120_UPPER32_DEFAULTS.md
- imma_u8.sass
- ptxlower.rs
- qmma_fp8.sass
- qmma_gemv_nodesc.sass
- qmma_pipelined.sass
- qmma_pipelined_v2.sass
- qmma_sp_gemv_nodesc.sass
- test_pipelined.cu
- test_qmma_gemv.cu
- ctrl_class.rs
- decoder.rs
- directives.rs
- eiattr.rs
- elf.rs
- elf_builder.rs
- encoder.rs
- ir.rs
- lib.rs
- main.rs
- mercury.rs
- parser.rs
- pred_liveness.rs
- printer.rs
- prof.rs
- ptx_lower.rs
- ptx_map.rs
- ptx_opt.rs
- ptx_parse.rs
- python.rs
- ra.rs
- ra_full.rs
- reg_liveness.rs
- render.rs
- sass_file.rs
- sched.rs
- scheduling.rs
- scheduling_pass.rs
- stallfix.rs
- table.rs
- sm100a.json
- sm103a.json
- sm120.json
- sm121a.json
- SOURCE.json
- qmma_gemv.sass
- p08_redux-eb6f6d.ptx
- p15_half2-c568ac.ptx
- readshared2-af1179.ptx
- arch_vendoring.rs
- assembler.rs
- b10_harness_fixes.rs
- b11_render_parity.rs
- b4fill2_rows.rs
- b4fill3_rows.rs
- b4fill4_ldef_family.rs
- b4fill_256x.rs
- b8_prof.rs
- b9_ptx_lower.rs
- b9p12_intmisc_barred.rs
- b9p13_satmufu.rs
- b9p14_movstgvec.rs
- bug012_encodeverify50.rs
- bug014_sts128_x16.rs
- bug026_bra_label_panic.rs
- bug030_up_lop3.rs
- bug032_upred_gate.rs
- bug037_mma_align.rs
- bug038_memclass_census.rs
- bug040_regcount_shadow.rs
- bug041_atomg_el_desc.rs
- bug042_double_pred.rs
- bug043_asm_fail_closed.rs
- bug043_s2r_sysreg.rs
- bug044_param3_template.rs
- bug045_ulop3_updest.rs
- bug046_haz_boundary.rs
- bug048_iadd3_neg_tok5.rs
- bug049_narrowfit_renum.rs
- bug049r_imad_hi_perarch.rs
- bug050_dmma_geometry.rs
- bug052_qmma_f8f6f4.rs
- bug053_omma_sf.rs
- bug054_vote_any_all_alias.rs
- bug055_redg_el_record.rs
- bug057_isetp_urform.rs
- bug059_imnmx_sm103.rs
- bug061_zeroparam_capmerc.rs
- bug062_flag_latency.rs
- bug067_voteu_umov_shadow.rs
- bug068_decode_determinism.rs
- bug069_ulea_hi_5tok.rs
- bug070_stg256_enl2_imm.rs
- bug071_harvest_junk_imm.rs
- bug072_flo_popc_vote.rs
- bug073_addr_dual_imm.rs
- bug076_stg_desc_pair_parity.rs
- bug077_ldg_desc_pair_parity.rs
- bug078_atom_desc_pair_parity.rs
- bug079_ulea_family.rs
- bug080_guarded_atomic.rs
- bug081_decoder_corruption_families.rs
- bug082_dsetp_recanon.rs
- bug083_imad_cluster.rs
- bug084_ldg_sts_recanon.rs
- bug086_isetp_ex_recanon.rs
- bug088_wide_mem_reg_align.rs
- bug089_iadd3x_tailpred.rs
- bug090_ltc128b_encode.rs
- bug091_bra_dotlabel.rs
- bug092_uisetp_legacy_retention.rs
- bug094_lds_ldl_stl_cas.rs
- bug097_ld_st_desc.rs
- bug098_ldsm_recanon.rs
- bug099_rawaddr_ur.rs
- bug101_ldg256_span.rs
- bug102_ldc_batch_haz.rs
- bug103_105_utc_mma_ctrl.rs
- bug104_dsel2.rs
- bug108_widex_stall.rs
- bug110_raw_armer_set.rs
- bug111_i239_disposition.rs
- bug113_iadd3_nopred_disposition.rs
- bug115_ret_nodec.rs
- bug116_warpsync_render.rs
- bug117_ldg256_pinned_wait.rs
- bug118_use_before_def.rs
- bug122_ltc128b_row.rs
- bug123_ldg_policy_133.rs
- bug125_f2i_imm.rs
- bug126_ushf_width.rs
- bug127_multikernel_nondet.rs
- bug128_redg_pt_alias.rs
- bug129_gpr_bump_cap.rs
- bug130_bar_bump_cap.rs
- bug131_fadd_ftz_rz.rs
- bug132_mod_drop.rs
- bug135_ldc128_reject.rs
- bug137_uiadd3_drain.rs
- bug138_brxu_abs_remap.rs
- bug139_fit_lint.rs
- bug140_aggregate_fit.rs
- bug141_f64frag_imm.rs
- bug142_atom_redg_holes.rs
- bug143_sm103a_atom_rows.rs
- bug144_sm120_drain_rows.rs
- bug145_popc_rows.rs
- bug146_float_sci_parity.rs
- bug147_subr24w9_normalize.rs
- bug148_sm120_subr24w9.rs
- bug149_ldg_u32ur_hibyte.rs
- bug150_sm120_ldg_aruri64e.rs
- bug151_ldcu_subur1.rs
- bug152_ldcu_cauri_sm120.rs
- bug153_call_abs_noinc.rs
- bug154_syncs_phasechk.rs
- bug155_cas_subimm_windows.rs
- bug156_ari_imm_windows.rs
- bug157_syncs_ari_ur.rs
- bug158_redg_dari_sm120.rs
- bug159_mufu_rsq_qnan.rs
- bug160_desc_urz.rs
- bug161_atom_dari_imm_windows.rs
- bug162_ldc_cai_windows.rs
- bug163_sm120_cai_bank.rs
- bug164_rz_elide.rs
- bug165_sm120_ldc_widths.rs
- bug167_ldc_ari_widths.rs
- bug168_uplop3_lattice.rs
- bug169_sm120_uplop3_eco.rs
- bug170_lop3_opex1.rs
- bug171_up_guard_glyph.rs
- bug172_ldc_s16_widths.rs
- bug173_ldcu_cari_hygiene.rs
- bug174_rz_sentinel_glyph.rs
- bug175_ldc_cari_widths.rs
- bug176_redg_p_junk.rs
- bug177_fsel_qnan.rs
- bug178_label_baked.rs
- bug179_atoms_urz_sink.rs
- bug180_ldgsts_render.rs
- bug181_atoms_sm120_auri.rs
- bug182_cm16_sign_glyph.rs
- bug182_warpsync_fit_family.rs
- bug183_label_imm_field.rs
- bug184_lepc.rs
- bug185_idesc_derived.rs
- bug186_utcur_lane.rs
- bug187_ldc_idx255_law.rs
- bug188_urink_9bit.rs
- bug189_incpopc_dest.rs
- bug190_utccp_window_law.rs
- bug191_redux_add_alias.rs
- bug192_modorder.rs
- bug193_redux_and_elision.rs
- bug194_redux_s32_siblings.rs
- bug195_atom_scope_width.rs
- bug196_atom_pred_windows.rs
- bug197_sm100a_atomw.rs
- bug198_utc_atomsws_pred.rs
- bug199_atom_lane.rs
- bug200_pred_window_law.rs
- bug201_pred_residual.rs
- bug202_dsetp_tok2_explicit.rs
- bug203_atom_coverage.rs
- bug204_atoms_auri_watch.rs
- bug205_atom_min.rs
- bug206_redg_scope.rs
- bug211_atoms_s32_glyph.rs
- bug212_warpsync_fit_family.rs
- bugs_errata_sm120.rs
- encoding.rs
- gold_manifest.rs
- kand058_ufadd_imm_neg.rs
- merc_companions.rs
- mercury.rs
- mercury_gen.rs
- mercury_gold.rs
- mk28_elf.rs
- mk34_mfamily.rs
- mk35_records.rs
- mk40_store.rs
- mk41_xsetp.rs
- mk42_edge.rs
- mk44_plop3.rs
- mk45_cs2r.rs
- mk46_geo.rs
- mk47_lop3not.rs
- mk48_redg.rs
- mk49_atomg.rs
- mk50_edge_ldg.rs
- mk51_dfmaimm.rs
- mk52_usetp.rs
- mk53_ldgsts2.rs
- mk54_setp0210.rs
- mk55_multiwait.rs
- mk56_ldcgeo.rs
- mk57_gemask.rs
- mk58_ulop3not.rs
- mk59_d1wc47.rs
- mk60_redux0132.rs
- mk61_ulea2.rs
- mk62_region09.rs
- mk63_store63.rs
- mk64_ghost64.rs
- mk65_wsreg.rs
- mk66_cctl.rs
- mk66_ws_colall.rs
- mk67_hadd2_break.rs
- mk68_minis.rs
- mk69_xsetp_fix.rs
- mk70_foldminis.rs
- mk71_fold28_uxor.rs
- mk72_lop3xorur.rs
- mk73_xor_guard_reuse.rs
- mk74_carryminis.rs
- pred_liveness.rs
- ra_m41.rs
- ra_m42.rs
- ra_m43.rs
- ra_m43_descns.rs
- ra_m61.rs
- ra_m62.rs
- ra_m63.rs
- reg_liveness.rs
- render_m43.rs
- sched_m45.rs
- sched_m46.rs
- sched_m47.rs
- sched_m5_pins.rs
- stallfix_ss.rs
- check_public_surface.sh
- discover.py
- sync_table.py
- validate_isa.py
- .gitignore
- Cargo.lock
- Cargo.toml
- LICENSE
- pyproject.toml
- README.md
# Use via CDN
jsDelivrjsDelivr serves any public GitHub repository as a CDN with zero setup. Pick a version and a file to get a ready-to-paste link and snippet.
Link
Example
// repository documentation
Was this content helpful?
(0 ratings)