kernel                                                         instr  narrow   loads     ds     mac   vgpr  th_max     smem
kernel_add_id                                                    146       0       2      0       2    240    1024        0
kernel_arange_f32                                                110       0       0      0       0    185    1024        0
kernel_argmax_f32                                                261       0       1      3       1    242    1024        0
kernel_argsort_f32_i32                                         FALLA
kernel_argsort_f32_i32_asc                                       278       0       4     14       0    186    1024        0
kernel_argsort_f32_i32_desc                                      280       0       4     14       0    186    1024        0
kernel_argsort_merge_f32_i32                                   FALLA
kernel_argsort_merge_f32_i32_asc                                 743       0      14      0       4    227    1024        0
kernel_argsort_merge_f32_i32_desc                                743       0      14      0       6    227    1024        0
kernel_bin_fuse_f16_f16_f16                                    FALLA
kernel_bin_fuse_f16_f16_f16_4                                  FALLA
kernel_bin_fuse_f32_f32_f32                                    FALLA
kernel_bin_fuse_f32_f32_f32_4                                  FALLA
kernel_bin_fuse_impl                                           FALLA
kernel_col2im_1d                                               FALLA
kernel_col2im_1d_bf16                                            264       0       0      0       1    183    1024        0
kernel_col2im_1d_f16                                             257       0       0      0       2    182    1024        0
kernel_col2im_1d_f32                                             256       0       1      0       2    182    1024        0
kernel_concat                                                  FALLA
kernel_concat_bf16                                               264       1       1      0       1    183    1024        0
kernel_concat_f16                                                268       1       1      0       2    206    1024        0
kernel_concat_f32                                                269       0       1      0       1    225    1024        0
kernel_concat_i16                                                268       1       1      0       2    206    1024        0
kernel_concat_i32                                                269       0       1      0       1    225    1024        0
kernel_concat_i64                                                269       0       1      0       1    225    1024        0
kernel_concat_i8                                                 269       1       1      0       1    225    1024        0
kernel_conv_2d                                                 FALLA
kernel_conv_2d_dw                                              FALLA
kernel_conv_2d_dw_f16_f32                                        494       1       2      0       2    211    1024        0
kernel_conv_2d_dw_f32_f32                                        492       0       2      0       1    210    1024        0
kernel_conv_2d_dw_tiled                                        FALLA
kernel_conv_2d_dw_tiled_f16_f32                                  485       1       2      0       0    216    1024        0
kernel_conv_2d_dw_tiled_f32_f32                                  482       0       2      0       0    208    1024        0
kernel_conv_2d_f16_f32                                          1230       1       2      0      15    187    1024        0
kernel_conv_2d_f32_f32                                          1226       0       2      0      14    187    1024        0
kernel_conv_3d                                                 FALLA
kernel_conv_3d_f16_f32                                           882       1       3      0      12    187    1024        0
kernel_conv_3d_f32_f32                                           877       0       1      0      13    187    1024        0
kernel_conv_transpose_1d                                       FALLA
kernel_conv_transpose_1d_f16_f32                                 266       0       0      0       1    212    1024        0
kernel_conv_transpose_1d_f32_f32                                 267       0       0      0       1    212    1024        0
kernel_conv_transpose_2d                                       FALLA
kernel_conv_transpose_2d_f16_f32                                 765       0       1      2      10    214    1024        0
kernel_conv_transpose_2d_f32_f32                                 766       0       2      2       9    214    1024        0
kernel_count_equal                                             FALLA
kernel_count_equal_i32                                         FALLA
kernel_cpy_bf16_bf16                                             942       1       1      0      11    204    1024        0
kernel_cpy_bf16_f32                                              941       1       1      0      10    198    1024        0
kernel_cpy_contig_bf16_bf16                                       98       1       1      0       0    188    1024        0
kernel_cpy_contig_bf16_f32                                       100       0       0      0       1    188    1024        0
kernel_cpy_contig_f16_f16                                         96       1       1      0       1    188    1024        0
kernel_cpy_contig_f16_f32                                         97       0       0      0       1    188    1024        0
kernel_cpy_contig_f32_bf16                                       109       0       1      0       1    188    1024        0
kernel_cpy_contig_f32_f16                                        103       0       1      0       1    188    1024        0
kernel_cpy_contig_f32_f32                                         97       0       1      0       0    188    1024        0
kernel_cpy_contig_f32_i32                                        104       0       1      0       0    188    1024        0
kernel_cpy_contig_i32_f32                                        101       0       1      0       0    188    1024        0
kernel_cpy_contig_i32_i32                                         97       0       1      0       0    188    1024        0
kernel_cpy_contig_t_t                                          FALLA
kernel_cpy_f16_f16                                               936       1       2      0      13    188    1024        0
kernel_cpy_f16_f32                                               937       1       1      0      13    188    1024        0
kernel_cpy_f32_bf16                                              946       0       1      0      10    214    1024        0
kernel_cpy_f32_f16                                               936       0       1      0      13    188    1024        0
kernel_cpy_f32_f32                                               937       0       2      0      12    188    1024        0
kernel_cpy_f32_i32                                               937       0       1      0      12    188    1024        0
kernel_cpy_f32_iq4_nl                                           3292       0     174      0      22    228    1024        0
kernel_cpy_f32_q                                               FALLA
kernel_cpy_f32_q1_0                                             1359       8      48      0      10    211     512        0
kernel_cpy_f32_q2_0                                             2301      64     144      0      10    251    1024        0
kernel_cpy_f32_q4_0                                             1450       0      43      0      15    188    1024        0
kernel_cpy_f32_q4_1                                             1487       0      40      0      15    252    1024        0
kernel_cpy_f32_q5_0                                             1601       0      43      0      15    188    1024        0
kernel_cpy_f32_q5_1                                             1531       0      40      0      15    189    1024        0
kernel_cpy_f32_q8_0                                             1448       0      40      0      15    188    1024        0
kernel_cpy_f32_turbo2                                           3698       8     280      0     139    232     256        0
kernel_cpy_f32_turbo3                                           4031      16     288      0     139    248     256        0
kernel_cpy_f32_turbo4                                           4367       8     280      0     139    252     256        0
kernel_cpy_i32_f32                                               937       0       1      0      12    188    1024        0
kernel_cpy_i32_i32                                               936       0       2      0      12    188    1024        0
kernel_cpy_q1_0_f16                                             1038       2       2      0      10    205    1024        0
kernel_cpy_q1_0_f32                                             1022       2       2      0      10    192    1024        0
kernel_cpy_q2_0_f16                                             1047       4       4      0      10    217    1024        0
kernel_cpy_q2_0_f32                                             1029       4       4      0      10    199    1024        0
kernel_cpy_q4_0_f16                                             1055       9       9      0      25    222    1024        0
kernel_cpy_q4_0_f32                                             1026       9       9      0      26    193    1024        0
kernel_cpy_q4_1_f16                                             1057      10      10      0      25    224    1024        0
kernel_cpy_q4_1_f32                                             1023      10      10      0      26    191    1024        0
kernel_cpy_q5_0_f16                                             1127       4       4      0      25    248    1024        0
kernel_cpy_q5_0_f32                                             1115       4       4      0      25    182    1024        0
kernel_cpy_q5_1_f16                                             1135       4       5      0      25    182    1024        0
kernel_cpy_q5_1_f32                                             1104       4       5      0      25    182    1024        0
kernel_cpy_q8_0_f16                                             1028      16      16      0      10    197    1024        0
kernel_cpy_q8_0_f32                                             1014      16      16      0      10    183    1024        0
kernel_cpy_q_f32                                               FALLA
kernel_cpy_t_t                                                 FALLA
kernel_cpy_turbo2_f32                                           3003       8     144      0     139    232     256        0
kernel_cpy_turbo3_f32                                           3077      16     152      0     139    252     256        0
kernel_cpy_turbo4_f32                                           3009       8     144      0     139    252     256        0
kernel_cumsum_add                                              FALLA
kernel_cumsum_add_f32                                            452       0       1      0       4    204    1024        0
kernel_cumsum_blk                                              FALLA
kernel_cumsum_blk_f32                                            597       1       2      5       6    208    1024        0
kernel_diag_f32                                                  142       0       0      0       1    220    1024        0
kernel_dsv4_hc_comb_f32                                          266       0       2      0       4    182    1024        0
kernel_dsv4_hc_post_f32                                          364       0       7      0       1    186    1024        0
kernel_dsv4_hc_pre_f32                                           188       0       5      0       1    224    1024        0
kernel_flash_attn_ext                                          FALLA
kernel_flash_attn_ext_bf16_dk112_dv112                         FALLA
kernel_flash_attn_ext_bf16_dk128_dv128                         FALLA
kernel_flash_attn_ext_bf16_dk192_dv128                         FALLA
kernel_flash_attn_ext_bf16_dk192_dv192                         FALLA
kernel_flash_attn_ext_bf16_dk256_dv256                         FALLA
kernel_flash_attn_ext_bf16_dk320_dv256                         FALLA
kernel_flash_attn_ext_bf16_dk512_dv512                         FALLA
kernel_flash_attn_ext_bf16_dk576_dv512                         FALLA
kernel_flash_attn_ext_blk                                      FALLA
kernel_flash_attn_ext_f16_dk112_dv112                          FALLA
kernel_flash_attn_ext_f16_dk128_dv128                          FALLA
kernel_flash_attn_ext_f16_dk192_dv128                          FALLA
kernel_flash_attn_ext_f16_dk192_dv192                          FALLA
kernel_flash_attn_ext_f16_dk256_dv256                          FALLA
kernel_flash_attn_ext_f16_dk320_dv256                          FALLA
kernel_flash_attn_ext_f16_dk512_dv512                          FALLA
kernel_flash_attn_ext_f16_dk576_dv512                          FALLA
kernel_flash_attn_ext_f32_dk112_dv112                          FALLA
kernel_flash_attn_ext_f32_dk128_dv128                          FALLA
kernel_flash_attn_ext_f32_dk192_dv128                          FALLA
kernel_flash_attn_ext_f32_dk192_dv192                          FALLA
kernel_flash_attn_ext_f32_dk256_dv256                          FALLA
kernel_flash_attn_ext_f32_dk320_dv256                          FALLA
kernel_flash_attn_ext_f32_dk512_dv512                          FALLA
kernel_flash_attn_ext_f32_dk576_dv512                          FALLA
kernel_flash_attn_ext_pad                                      FALLA
kernel_flash_attn_ext_pf_amd                                   FALLA
kernel_flash_attn_ext_q4_0_dk112_dv112                         FALLA
kernel_flash_attn_ext_q4_0_dk128_dv128                         FALLA
kernel_flash_attn_ext_q4_0_dk192_dv128                         FALLA
kernel_flash_attn_ext_q4_0_dk192_dv192                         FALLA
kernel_flash_attn_ext_q4_0_dk256_dv256                         FALLA
kernel_flash_attn_ext_q4_0_dk320_dv256                         FALLA
kernel_flash_attn_ext_q4_0_dk512_dv512                         FALLA
kernel_flash_attn_ext_q4_0_dk576_dv512                         FALLA
kernel_flash_attn_ext_q4_1_dk112_dv112                         FALLA
kernel_flash_attn_ext_q4_1_dk128_dv128                         FALLA
kernel_flash_attn_ext_q4_1_dk192_dv128                         FALLA
kernel_flash_attn_ext_q4_1_dk192_dv192                         FALLA
kernel_flash_attn_ext_q4_1_dk256_dv256                         FALLA
kernel_flash_attn_ext_q4_1_dk320_dv256                         FALLA
kernel_flash_attn_ext_q4_1_dk512_dv512                         FALLA
kernel_flash_attn_ext_q4_1_dk576_dv512                         FALLA
kernel_flash_attn_ext_q5_0_dk112_dv112                         FALLA
kernel_flash_attn_ext_q5_0_dk128_dv128                         FALLA
kernel_flash_attn_ext_q5_0_dk192_dv128                         FALLA
kernel_flash_attn_ext_q5_0_dk192_dv192                         FALLA
kernel_flash_attn_ext_q5_0_dk256_dv256                         FALLA
kernel_flash_attn_ext_q5_0_dk320_dv256                         FALLA
kernel_flash_attn_ext_q5_0_dk512_dv512                         FALLA
kernel_flash_attn_ext_q5_0_dk576_dv512                         FALLA
kernel_flash_attn_ext_q5_1_dk112_dv112                         FALLA
kernel_flash_attn_ext_q5_1_dk128_dv128                         FALLA
kernel_flash_attn_ext_q5_1_dk192_dv128                         FALLA
kernel_flash_attn_ext_q5_1_dk192_dv192                         FALLA
kernel_flash_attn_ext_q5_1_dk256_dv256                         FALLA
kernel_flash_attn_ext_q5_1_dk320_dv256                         FALLA
kernel_flash_attn_ext_q5_1_dk512_dv512                         FALLA
kernel_flash_attn_ext_q5_1_dk576_dv512                         FALLA
kernel_flash_attn_ext_q8_0_dk112_dv112                         FALLA
kernel_flash_attn_ext_q8_0_dk128_dv128                         FALLA
kernel_flash_attn_ext_q8_0_dk192_dv128                         FALLA
kernel_flash_attn_ext_q8_0_dk192_dv192                         FALLA
kernel_flash_attn_ext_q8_0_dk256_dv256                         FALLA
kernel_flash_attn_ext_q8_0_dk320_dv256                         FALLA
kernel_flash_attn_ext_q8_0_dk512_dv512                         FALLA
kernel_flash_attn_ext_q8_0_dk576_dv512                         FALLA
kernel_flash_attn_ext_tile_amd                                 FALLA
kernel_flash_attn_ext_vec                                      FALLA
kernel_flash_attn_ext_vec_amd                                  FALLA
kernel_flash_attn_ext_vec_bf16_dk128_dv128                     FALLA
kernel_flash_attn_ext_vec_bf16_dk192_dv128                     FALLA
kernel_flash_attn_ext_vec_bf16_dk192_dv192                     FALLA
kernel_flash_attn_ext_vec_bf16_dk256_dv256                     FALLA
kernel_flash_attn_ext_vec_bf16_dk320_dv256                     FALLA
kernel_flash_attn_ext_vec_bf16_dk32_dv32                       FALLA
kernel_flash_attn_ext_vec_bf16_dk512_dv512                     FALLA
kernel_flash_attn_ext_vec_bf16_dk576_dv512                     FALLA
kernel_flash_attn_ext_vec_bf16_dk64_dv64                       FALLA
kernel_flash_attn_ext_vec_bf16_dk96_dv96                       FALLA
kernel_flash_attn_ext_vec_f16_dk128_dv128                      FALLA
kernel_flash_attn_ext_vec_f16_dk192_dv128                      FALLA
kernel_flash_attn_ext_vec_f16_dk192_dv192                      FALLA
kernel_flash_attn_ext_vec_f16_dk256_dv256                      FALLA
kernel_flash_attn_ext_vec_f16_dk320_dv256                      FALLA
kernel_flash_attn_ext_vec_f16_dk32_dv32                        FALLA
kernel_flash_attn_ext_vec_f16_dk512_dv512                      FALLA
kernel_flash_attn_ext_vec_f16_dk576_dv512                      FALLA
kernel_flash_attn_ext_vec_f16_dk64_dv64                        FALLA
kernel_flash_attn_ext_vec_f16_dk96_dv96                        FALLA
kernel_flash_attn_ext_vec_f32_dk128_dv128                      FALLA
kernel_flash_attn_ext_vec_f32_dk192_dv128                      FALLA
kernel_flash_attn_ext_vec_f32_dk192_dv192                      FALLA
kernel_flash_attn_ext_vec_f32_dk256_dv256                      FALLA
kernel_flash_attn_ext_vec_f32_dk320_dv256                      FALLA
kernel_flash_attn_ext_vec_f32_dk32_dv32                        FALLA
kernel_flash_attn_ext_vec_f32_dk512_dv512                      FALLA
kernel_flash_attn_ext_vec_f32_dk576_dv512                      FALLA
kernel_flash_attn_ext_vec_f32_dk64_dv64                        FALLA
kernel_flash_attn_ext_vec_f32_dk96_dv96                        FALLA
kernel_flash_attn_ext_vec_q4_0_dk128_dv128                     FALLA
kernel_flash_attn_ext_vec_q4_0_dk192_dv128                     FALLA
kernel_flash_attn_ext_vec_q4_0_dk192_dv192                     FALLA
kernel_flash_attn_ext_vec_q4_0_dk256_dv256                     FALLA
kernel_flash_attn_ext_vec_q4_0_dk320_dv256                     FALLA
kernel_flash_attn_ext_vec_q4_0_dk32_dv32                       FALLA
kernel_flash_attn_ext_vec_q4_0_dk512_dv512                     FALLA
kernel_flash_attn_ext_vec_q4_0_dk576_dv512                     FALLA
kernel_flash_attn_ext_vec_q4_0_dk64_dv64                       FALLA
kernel_flash_attn_ext_vec_q4_0_dk96_dv96                       FALLA
kernel_flash_attn_ext_vec_q4_1_dk128_dv128                     FALLA
kernel_flash_attn_ext_vec_q4_1_dk192_dv128                     FALLA
kernel_flash_attn_ext_vec_q4_1_dk192_dv192                     FALLA
kernel_flash_attn_ext_vec_q4_1_dk256_dv256                     FALLA
kernel_flash_attn_ext_vec_q4_1_dk320_dv256                     FALLA
kernel_flash_attn_ext_vec_q4_1_dk32_dv32                       FALLA
kernel_flash_attn_ext_vec_q4_1_dk512_dv512                     FALLA
kernel_flash_attn_ext_vec_q4_1_dk576_dv512                     FALLA
kernel_flash_attn_ext_vec_q4_1_dk64_dv64                       FALLA
kernel_flash_attn_ext_vec_q4_1_dk96_dv96                       FALLA
kernel_flash_attn_ext_vec_q5_0_dk128_dv128                     FALLA
kernel_flash_attn_ext_vec_q5_0_dk192_dv128                     FALLA
kernel_flash_attn_ext_vec_q5_0_dk192_dv192                     FALLA
kernel_flash_attn_ext_vec_q5_0_dk256_dv256                     FALLA
kernel_flash_attn_ext_vec_q5_0_dk320_dv256                     FALLA
kernel_flash_attn_ext_vec_q5_0_dk32_dv32                       FALLA
kernel_flash_attn_ext_vec_q5_0_dk512_dv512                     FALLA
kernel_flash_attn_ext_vec_q5_0_dk576_dv512                     FALLA
kernel_flash_attn_ext_vec_q5_0_dk64_dv64                       FALLA
kernel_flash_attn_ext_vec_q5_0_dk96_dv96                       FALLA
kernel_flash_attn_ext_vec_q5_1_dk128_dv128                     FALLA
kernel_flash_attn_ext_vec_q5_1_dk192_dv128                     FALLA
kernel_flash_attn_ext_vec_q5_1_dk192_dv192                     FALLA
kernel_flash_attn_ext_vec_q5_1_dk256_dv256                     FALLA
kernel_flash_attn_ext_vec_q5_1_dk320_dv256                     FALLA
kernel_flash_attn_ext_vec_q5_1_dk32_dv32                       FALLA
kernel_flash_attn_ext_vec_q5_1_dk512_dv512                     FALLA
kernel_flash_attn_ext_vec_q5_1_dk576_dv512                     FALLA
kernel_flash_attn_ext_vec_q5_1_dk64_dv64                       FALLA
kernel_flash_attn_ext_vec_q5_1_dk96_dv96                       FALLA
kernel_flash_attn_ext_vec_q8_0_dk128_dv128                     FALLA
kernel_flash_attn_ext_vec_q8_0_dk192_dv128                     FALLA
kernel_flash_attn_ext_vec_q8_0_dk192_dv192                     FALLA
kernel_flash_attn_ext_vec_q8_0_dk256_dv256                     FALLA
kernel_flash_attn_ext_vec_q8_0_dk320_dv256                     FALLA
kernel_flash_attn_ext_vec_q8_0_dk32_dv32                       FALLA
kernel_flash_attn_ext_vec_q8_0_dk512_dv512                     FALLA
kernel_flash_attn_ext_vec_q8_0_dk576_dv512                     FALLA
kernel_flash_attn_ext_vec_q8_0_dk64_dv64                       FALLA
kernel_flash_attn_ext_vec_q8_0_dk96_dv96                       FALLA
kernel_flash_attn_ext_vec_reduce                               FALLA
kernel_fwht_f32                                                FALLA
kernel_fwht_f32_128                                              271       0       4      0       1    186    1024        0
kernel_fwht_f32_128_w64                                          236       0       2      0       1    224    1024        0
kernel_fwht_f32_256                                              391       0       8      0       1    224    1024        0
kernel_fwht_f32_256_w64                                          302       0       4      0       1    186    1024        0
kernel_fwht_f32_512                                              644       0      16      0       1    205    1024        0
kernel_fwht_f32_512_w64                                          449       0       8      0       1    186    1024        0
kernel_fwht_f32_64                                               215       0       2      0       1    186    1024        0
kernel_fwht_f32_64_w64                                           203       0       1      0       1    186    1024        0
kernel_gated_delta_net_f32_1                                   FALLA
kernel_gated_delta_net_f32_2                                   FALLA
kernel_gated_delta_net_f32_4                                   FALLA
kernel_gated_delta_net_impl                                    FALLA
kernel_geglu                                                   FALLA
kernel_geglu_erf                                               FALLA
kernel_geglu_erf_f16                                             160       0       0      0       7    182    1024        0
kernel_geglu_erf_f32                                             160       0       2      0       6    182    1024        0
kernel_geglu_f16                                                 258       0       0      0      17    182    1024        0
kernel_geglu_f32                                                 258       0       2      0      16    182    1024        0
kernel_geglu_quick                                             FALLA
kernel_geglu_quick_f16                                           152       0       0      0       1    186    1024        0
kernel_geglu_quick_f32                                           151       0       2      0       0    186    1024        0
kernel_get_rows_bf16                                             207       0       1      0       2    206    1024        0
kernel_get_rows_f                                              FALLA
kernel_get_rows_f16                                              205       0       1      0       1    187    1024        0
kernel_get_rows_f32                                              204       0       2      0       1    187    1024        0
kernel_get_rows_i32                                              204       0       2      0       1    187    1024        0
kernel_get_rows_iq1_m                                           2451      12      13      0       5    252    1024        0
kernel_get_rows_iq1_s                                           2406       9      10      0       4    206    1024        0
kernel_get_rows_iq2_s                                           2446       6       7      0       1    187    1024        0
kernel_get_rows_iq2_xs                                          1457       5       6      0       3    241    1024        0
kernel_get_rows_iq2_xxs                                          939       7       8      0       3    241    1024        0
kernel_get_rows_iq3_s                                            955       8       9      0       1    187    1024        0
kernel_get_rows_iq3_xxs                                          700      10      11      0       5    241    1024        0
kernel_get_rows_iq4_nl                                           340       1      18      0       1    248    1024        0
kernel_get_rows_iq4_xs                                           375       1      21      0       1    228    1024        0
kernel_get_rows_mxfp4                                            378      17      34      0       3    224    1024        0
kernel_get_rows_q                                              FALLA
kernel_get_rows_q1_0                                             288       2       3      0       3    206    1024        0
kernel_get_rows_q2_0                                             298       4       5      0       3    206    1024        0
kernel_get_rows_q2_K                                             330       4       5      0      19    252    1024        0
kernel_get_rows_q3_K                                             421       4       5      0      19    240    1024        0
kernel_get_rows_q4_0                                             295       9      10      0      18    206    1024        0
kernel_get_rows_q4_1                                             293      10      11      0      18    206    1024        0
kernel_get_rows_q4_K                                             312       0       3      0      19    206    1024        0
kernel_get_rows_q5_0                                             382       4       5      0      17    206    1024        0
kernel_get_rows_q5_1                                             371       4       6      0      17    206    1024        0
kernel_get_rows_q5_K                                             406       7      10      0      19    240    1024        0
kernel_get_rows_q6_K                                             383       1       2      0      15    206    1024        0
kernel_get_rows_q8_0                                             282      16      17      0       3    206    1024        0
kernel_group_norm_f32                                            552       0       3      6       4    238    1024        0
kernel_im2col                                                  FALLA
kernel_im2col_ext                                              FALLA
kernel_im2col_ext_f16                                            578       0       1      0       5    255    1024        0
kernel_im2col_ext_f32                                            578       0       1      0       4    255    1024        0
kernel_im2col_f16                                                398       0       1      0       1    204    1024        0
kernel_im2col_f32                                                396       0       1      0       0    204    1024        0
kernel_l2_norm_f32_f32                                           238       0       2      3       3    248    1024        0
kernel_l2_norm_f32_f32_4                                         246       0       2      3       2    230    1024        0
kernel_l2_norm_impl                                            FALLA
kernel_lightning_indexer                                       FALLA
kernel_lightning_indexer_bf16                                  FALLA
kernel_lightning_indexer_f16                                   FALLA
kernel_lightning_indexer_f32                                   FALLA
kernel_lightning_indexer_q4_0                                  FALLA
kernel_lightning_indexer_q4_1                                  FALLA
kernel_lightning_indexer_q5_0                                  FALLA
kernel_lightning_indexer_q5_1                                  FALLA
kernel_lightning_indexer_q8_0                                  FALLA
kernel_memset                                                  FALLA
kernel_memset_i64                                                 75       0       0      0       0    186    1024        0
kernel_mul_mm                                                  FALLA
kernel_mul_mm_bf16_f32                                         FALLA
kernel_mul_mm_f16_f16                                          FALLA
kernel_mul_mm_f16_f32                                          FALLA
kernel_mul_mm_f16_f32_w                                        FALLA
kernel_mul_mm_f32_f16                                          FALLA
kernel_mul_mm_f32_f32                                          FALLA
kernel_mul_mm_id                                               FALLA
kernel_mul_mm_id_bf16_f32                                      FALLA
kernel_mul_mm_id_f16_f16                                       FALLA
kernel_mul_mm_id_f16_f32                                       FALLA
kernel_mul_mm_id_f16_f32_w                                     FALLA
kernel_mul_mm_id_f32_f16                                       FALLA
kernel_mul_mm_id_f32_f32                                       FALLA
kernel_mul_mm_id_iq1_m_f16                                     FALLA
kernel_mul_mm_id_iq1_m_f32                                     FALLA
kernel_mul_mm_id_iq1_m_f32_w                                   FALLA
kernel_mul_mm_id_iq1_s_f16                                     FALLA
kernel_mul_mm_id_iq1_s_f32                                     FALLA
kernel_mul_mm_id_iq1_s_f32_w                                   FALLA
kernel_mul_mm_id_iq2_s_f16                                     FALLA
kernel_mul_mm_id_iq2_s_f32                                     FALLA
kernel_mul_mm_id_iq2_s_f32_w                                   FALLA
kernel_mul_mm_id_iq2_xs_f16                                    FALLA
kernel_mul_mm_id_iq2_xs_f32                                    FALLA
kernel_mul_mm_id_iq2_xs_f32_w                                  FALLA
kernel_mul_mm_id_iq2_xxs_f16                                   FALLA
kernel_mul_mm_id_iq2_xxs_f32                                   FALLA
kernel_mul_mm_id_iq2_xxs_f32_w                                 FALLA
kernel_mul_mm_id_iq3_s_f16                                     FALLA
kernel_mul_mm_id_iq3_s_f32                                     FALLA
kernel_mul_mm_id_iq3_s_f32_w                                   FALLA
kernel_mul_mm_id_iq3_xxs_f16                                   FALLA
kernel_mul_mm_id_iq3_xxs_f32                                   FALLA
kernel_mul_mm_id_iq3_xxs_f32_w                                 FALLA
kernel_mul_mm_id_iq4_nl_f16                                    FALLA
kernel_mul_mm_id_iq4_nl_f32                                    FALLA
kernel_mul_mm_id_iq4_nl_f32_w                                  FALLA
kernel_mul_mm_id_iq4_xs_f16                                    FALLA
kernel_mul_mm_id_iq4_xs_f32                                    FALLA
kernel_mul_mm_id_iq4_xs_f32_w                                  FALLA
kernel_mul_mm_id_map0                                          FALLA
kernel_mul_mm_id_map0_ne20_10                                    236      10      10     20       4    230    1024        0
kernel_mul_mm_id_map0_ne20_16                                    277      16      16     32       3    183    1024        0
kernel_mul_mm_id_map0_ne20_22                                    329      22      22     44       3    183    1024        0
kernel_mul_mm_id_mxfp4_f16                                     FALLA
kernel_mul_mm_id_mxfp4_f32                                     FALLA
kernel_mul_mm_id_mxfp4_f32_w                                   FALLA
kernel_mul_mm_id_q1_0_f16                                      FALLA
kernel_mul_mm_id_q1_0_f32                                      FALLA
kernel_mul_mm_id_q2_0_f16                                      FALLA
kernel_mul_mm_id_q2_0_f32                                      FALLA
kernel_mul_mm_id_q2_K_f16                                      FALLA
kernel_mul_mm_id_q2_K_f32                                      FALLA
kernel_mul_mm_id_q2_K_f32_w                                    FALLA
kernel_mul_mm_id_q3_K_f16                                      FALLA
kernel_mul_mm_id_q3_K_f32                                      FALLA
kernel_mul_mm_id_q3_K_f32_w                                    FALLA
kernel_mul_mm_id_q4_0_f16                                      FALLA
kernel_mul_mm_id_q4_0_f32                                      FALLA
kernel_mul_mm_id_q4_0_f32_w                                    FALLA
kernel_mul_mm_id_q4_1_f16                                      FALLA
kernel_mul_mm_id_q4_1_f32                                      FALLA
kernel_mul_mm_id_q4_1_f32_w                                    FALLA
kernel_mul_mm_id_q4_K_f16                                      FALLA
kernel_mul_mm_id_q4_K_f32                                      FALLA
kernel_mul_mm_id_q4_K_f32_w                                    FALLA
kernel_mul_mm_id_q5_0_f16                                      FALLA
kernel_mul_mm_id_q5_0_f32                                      FALLA
kernel_mul_mm_id_q5_0_f32_w                                    FALLA
kernel_mul_mm_id_q5_1_f16                                      FALLA
kernel_mul_mm_id_q5_1_f32                                      FALLA
kernel_mul_mm_id_q5_1_f32_w                                    FALLA
kernel_mul_mm_id_q5_K_f16                                      FALLA
kernel_mul_mm_id_q5_K_f32                                      FALLA
kernel_mul_mm_id_q5_K_f32_w                                    FALLA
kernel_mul_mm_id_q6_K_f16                                      FALLA
kernel_mul_mm_id_q6_K_f32                                      FALLA
kernel_mul_mm_id_q6_K_f32_w                                    FALLA
kernel_mul_mm_id_q8_0_f16                                      FALLA
kernel_mul_mm_id_q8_0_f32                                      FALLA
kernel_mul_mm_id_q8_0_f32_w                                    FALLA
kernel_mul_mm_iq1_m_f16                                        FALLA
kernel_mul_mm_iq1_m_f32                                        FALLA
kernel_mul_mm_iq1_m_f32_w                                      FALLA
kernel_mul_mm_iq1_s_f16                                        FALLA
kernel_mul_mm_iq1_s_f32                                        FALLA
kernel_mul_mm_iq1_s_f32_w                                      FALLA
kernel_mul_mm_iq2_s_f16                                        FALLA
kernel_mul_mm_iq2_s_f32                                        FALLA
kernel_mul_mm_iq2_s_f32_w                                      FALLA
kernel_mul_mm_iq2_xs_f16                                       FALLA
kernel_mul_mm_iq2_xs_f32                                       FALLA
kernel_mul_mm_iq2_xs_f32_w                                     FALLA
kernel_mul_mm_iq2_xxs_f16                                      FALLA
kernel_mul_mm_iq2_xxs_f32                                      FALLA
kernel_mul_mm_iq2_xxs_f32_w                                    FALLA
kernel_mul_mm_iq3_s_f16                                        FALLA
kernel_mul_mm_iq3_s_f32                                        FALLA
kernel_mul_mm_iq3_s_f32_w                                      FALLA
kernel_mul_mm_iq3_xxs_f16                                      FALLA
kernel_mul_mm_iq3_xxs_f32                                      FALLA
kernel_mul_mm_iq3_xxs_f32_w                                    FALLA
kernel_mul_mm_iq4_nl_f16                                       FALLA
kernel_mul_mm_iq4_nl_f32                                       FALLA
kernel_mul_mm_iq4_nl_f32_w                                     FALLA
kernel_mul_mm_iq4_xs_f16                                       FALLA
kernel_mul_mm_iq4_xs_f32                                       FALLA
kernel_mul_mm_iq4_xs_f32_w                                     FALLA
kernel_mul_mm_mxfp4_f16                                        FALLA
kernel_mul_mm_mxfp4_f32                                        FALLA
kernel_mul_mm_mxfp4_f32_w                                      FALLA
kernel_mul_mm_q1_0_f16                                         FALLA
kernel_mul_mm_q1_0_f32                                         FALLA
kernel_mul_mm_q2_0_f16                                         FALLA
kernel_mul_mm_q2_0_f32                                         FALLA
kernel_mul_mm_q2_K_f16                                         FALLA
kernel_mul_mm_q2_K_f32                                         FALLA
kernel_mul_mm_q2_K_f32_w                                       FALLA
kernel_mul_mm_q3_K_f16                                         FALLA
kernel_mul_mm_q3_K_f32                                         FALLA
kernel_mul_mm_q3_K_f32_w                                       FALLA
kernel_mul_mm_q4_0_f16                                         FALLA
kernel_mul_mm_q4_0_f32                                         FALLA
kernel_mul_mm_q4_0_f32_w                                       FALLA
kernel_mul_mm_q4_1_f16                                         FALLA
kernel_mul_mm_q4_1_f32                                         FALLA
kernel_mul_mm_q4_1_f32_w                                       FALLA
kernel_mul_mm_q4_K_f16                                         FALLA
kernel_mul_mm_q4_K_f32                                         FALLA
kernel_mul_mm_q4_K_f32_w                                       FALLA
kernel_mul_mm_q5_0_f16                                         FALLA
kernel_mul_mm_q5_0_f32                                         FALLA
kernel_mul_mm_q5_0_f32_w                                       FALLA
kernel_mul_mm_q5_1_f16                                         FALLA
kernel_mul_mm_q5_1_f32                                         FALLA
kernel_mul_mm_q5_1_f32_w                                       FALLA
kernel_mul_mm_q5_K_f16                                         FALLA
kernel_mul_mm_q5_K_f32                                         FALLA
kernel_mul_mm_q5_K_f32_w                                       FALLA
kernel_mul_mm_q6_K_f16                                         FALLA
kernel_mul_mm_q6_K_f32                                         FALLA
kernel_mul_mm_q6_K_f32_w                                       FALLA
kernel_mul_mm_q8_0_f16                                         FALLA
kernel_mul_mm_q8_0_f32                                         FALLA
kernel_mul_mm_q8_0_f32_w                                       FALLA
kernel_mul_mv_bf16_bf16                                        FALLA
kernel_mul_mv_bf16_bf16_4                                      FALLA
kernel_mul_mv_bf16_bf16_short                                  FALLA
kernel_mul_mv_bf16_f32                                         FALLA
kernel_mul_mv_bf16_f32_4                                       FALLA
kernel_mul_mv_bf16_f32_short                                   FALLA
kernel_mul_mv_ext_bf16_f32_r1_2                                FALLA
kernel_mul_mv_ext_bf16_f32_r1_3                                FALLA
kernel_mul_mv_ext_bf16_f32_r1_4                                FALLA
kernel_mul_mv_ext_bf16_f32_r1_5                                FALLA
kernel_mul_mv_ext_f16_f32_r1_2                                 FALLA
kernel_mul_mv_ext_f16_f32_r1_3                                 FALLA
kernel_mul_mv_ext_f16_f32_r1_4                                 FALLA
kernel_mul_mv_ext_f16_f32_r1_5                                 FALLA
kernel_mul_mv_ext_f32_f32_r1_2                                 FALLA
kernel_mul_mv_ext_f32_f32_r1_3                                 FALLA
kernel_mul_mv_ext_f32_f32_r1_4                                 FALLA
kernel_mul_mv_ext_f32_f32_r1_5                                 FALLA
kernel_mul_mv_ext_iq4_nl_f32_r1_2                              FALLA
kernel_mul_mv_ext_iq4_nl_f32_r1_3                              FALLA
kernel_mul_mv_ext_iq4_nl_f32_r1_4                              FALLA
kernel_mul_mv_ext_iq4_nl_f32_r1_5                              FALLA
kernel_mul_mv_ext_mxfp4_f32_r1_2                               FALLA
kernel_mul_mv_ext_mxfp4_f32_r1_3                               FALLA
kernel_mul_mv_ext_mxfp4_f32_r1_4                               FALLA
kernel_mul_mv_ext_mxfp4_f32_r1_5                               FALLA
kernel_mul_mv_ext_q                                            FALLA
kernel_mul_mv_ext_q1_0_f32_r1_2                                FALLA
kernel_mul_mv_ext_q1_0_f32_r1_3                                FALLA
kernel_mul_mv_ext_q1_0_f32_r1_4                                FALLA
kernel_mul_mv_ext_q1_0_f32_r1_5                                FALLA
kernel_mul_mv_ext_q2_0_f32_r1_2                                FALLA
kernel_mul_mv_ext_q2_0_f32_r1_3                                FALLA
kernel_mul_mv_ext_q2_0_f32_r1_4                                FALLA
kernel_mul_mv_ext_q2_0_f32_r1_5                                FALLA
kernel_mul_mv_ext_q2_K_f32_r1_2                                FALLA
kernel_mul_mv_ext_q2_K_f32_r1_2_nr0_2                          FALLA
kernel_mul_mv_ext_q2_K_f32_r1_3                                FALLA
kernel_mul_mv_ext_q2_K_f32_r1_3_nr0_2                          FALLA
kernel_mul_mv_ext_q2_K_f32_r1_4                                FALLA
kernel_mul_mv_ext_q2_K_f32_r1_4_nr0_2                          FALLA
kernel_mul_mv_ext_q2_K_f32_r1_5                                FALLA
kernel_mul_mv_ext_q2_K_f32_r1_5_nr0_2                          FALLA
kernel_mul_mv_ext_q2_K_hdr_f32_r1_2                            FALLA
kernel_mul_mv_ext_q2_K_hdr_f32_r1_2_nr0_2                      FALLA
kernel_mul_mv_ext_q2_K_hdr_f32_r1_3                            FALLA
kernel_mul_mv_ext_q2_K_hdr_f32_r1_3_nr0_2                      FALLA
kernel_mul_mv_ext_q2_K_hdr_f32_r1_4                            FALLA
kernel_mul_mv_ext_q2_K_hdr_f32_r1_4_nr0_2                      FALLA
kernel_mul_mv_ext_q2_K_hdr_f32_r1_5                            FALLA
kernel_mul_mv_ext_q2_K_hdr_f32_r1_5_nr0_2                      FALLA
kernel_mul_mv_ext_q3_K_f32_r1_2                                FALLA
kernel_mul_mv_ext_q3_K_f32_r1_2_nr0_2                          FALLA
kernel_mul_mv_ext_q3_K_f32_r1_3                                FALLA
kernel_mul_mv_ext_q3_K_f32_r1_3_nr0_2                          FALLA
kernel_mul_mv_ext_q3_K_f32_r1_4                                FALLA
kernel_mul_mv_ext_q3_K_f32_r1_4_nr0_2                          FALLA
kernel_mul_mv_ext_q3_K_f32_r1_5                                FALLA
kernel_mul_mv_ext_q3_K_f32_r1_5_nr0_2                          FALLA
kernel_mul_mv_ext_q3_K_hdr_f32_r1_2                            FALLA
kernel_mul_mv_ext_q3_K_hdr_f32_r1_2_nr0_2                      FALLA
kernel_mul_mv_ext_q3_K_hdr_f32_r1_3                            FALLA
kernel_mul_mv_ext_q3_K_hdr_f32_r1_3_nr0_2                      FALLA
kernel_mul_mv_ext_q3_K_hdr_f32_r1_4                            FALLA
kernel_mul_mv_ext_q3_K_hdr_f32_r1_4_nr0_2                      FALLA
kernel_mul_mv_ext_q3_K_hdr_f32_r1_5                            FALLA
kernel_mul_mv_ext_q3_K_hdr_f32_r1_5_nr0_2                      FALLA
kernel_mul_mv_ext_q4_                                          FALLA
kernel_mul_mv_ext_q4_0_f32_r1_2                                FALLA
kernel_mul_mv_ext_q4_0_f32_r1_3                                FALLA
kernel_mul_mv_ext_q4_0_f32_r1_4                                FALLA
kernel_mul_mv_ext_q4_0_f32_r1_5                                FALLA
kernel_mul_mv_ext_q4_1_f32_r1_2                                FALLA
kernel_mul_mv_ext_q4_1_f32_r1_3                                FALLA
kernel_mul_mv_ext_q4_1_f32_r1_4                                FALLA
kernel_mul_mv_ext_q4_1_f32_r1_5                                FALLA
kernel_mul_mv_ext_q4_K_f32_r1_2                                FALLA
kernel_mul_mv_ext_q4_K_f32_r1_2_nr0_2                          FALLA
kernel_mul_mv_ext_q4_K_f32_r1_3                                FALLA
kernel_mul_mv_ext_q4_K_f32_r1_3_nr0_2                          FALLA
kernel_mul_mv_ext_q4_K_f32_r1_4                                FALLA
kernel_mul_mv_ext_q4_K_f32_r1_4_nr0_2                          FALLA
kernel_mul_mv_ext_q4_K_f32_r1_5                                FALLA
kernel_mul_mv_ext_q4_K_f32_r1_5_nr0_2                          FALLA
kernel_mul_mv_ext_q4_K_raw_f32_r1_2                            FALLA
kernel_mul_mv_ext_q4_K_raw_f32_r1_2_nr0_2                      FALLA
kernel_mul_mv_ext_q4_K_raw_f32_r1_3                            FALLA
kernel_mul_mv_ext_q4_K_raw_f32_r1_3_nr0_2                      FALLA
kernel_mul_mv_ext_q4_K_raw_f32_r1_4                            FALLA
kernel_mul_mv_ext_q4_K_raw_f32_r1_4_nr0_2                      FALLA
kernel_mul_mv_ext_q4_K_raw_f32_r1_5                            FALLA
kernel_mul_mv_ext_q4_K_raw_f32_r1_5_nr0_2                      FALLA
kernel_mul_mv_ext_q4_f32_disp                                  FALLA
kernel_mul_mv_ext_q4x4_f32_disp                                FALLA
kernel_mul_mv_ext_q5_0_f32_r1_2                                FALLA
kernel_mul_mv_ext_q5_0_f32_r1_3                                FALLA
kernel_mul_mv_ext_q5_0_f32_r1_4                                FALLA
kernel_mul_mv_ext_q5_0_f32_r1_5                                FALLA
kernel_mul_mv_ext_q5_1_f32_r1_2                                FALLA
kernel_mul_mv_ext_q5_1_f32_r1_3                                FALLA
kernel_mul_mv_ext_q5_1_f32_r1_4                                FALLA
kernel_mul_mv_ext_q5_1_f32_r1_5                                FALLA
kernel_mul_mv_ext_q5_K_f32_r1_2                                FALLA
kernel_mul_mv_ext_q5_K_f32_r1_2_nr0_2                          FALLA
kernel_mul_mv_ext_q5_K_f32_r1_3                                FALLA
kernel_mul_mv_ext_q5_K_f32_r1_3_nr0_2                          FALLA
kernel_mul_mv_ext_q5_K_f32_r1_4                                FALLA
kernel_mul_mv_ext_q5_K_f32_r1_4_nr0_2                          FALLA
kernel_mul_mv_ext_q5_K_f32_r1_5                                FALLA
kernel_mul_mv_ext_q5_K_f32_r1_5_nr0_2                          FALLA
kernel_mul_mv_ext_q5_K_hdr_f32_r1_2                            FALLA
kernel_mul_mv_ext_q5_K_hdr_f32_r1_2_nr0_2                      FALLA
kernel_mul_mv_ext_q5_K_hdr_f32_r1_3                            FALLA
kernel_mul_mv_ext_q5_K_hdr_f32_r1_3_nr0_2                      FALLA
kernel_mul_mv_ext_q5_K_hdr_f32_r1_4                            FALLA
kernel_mul_mv_ext_q5_K_hdr_f32_r1_4_nr0_2                      FALLA
kernel_mul_mv_ext_q5_K_hdr_f32_r1_5                            FALLA
kernel_mul_mv_ext_q5_K_hdr_f32_r1_5_nr0_2                      FALLA
kernel_mul_mv_ext_q6_K_f32_r1_2                                FALLA
kernel_mul_mv_ext_q6_K_f32_r1_2_nr0_2                          FALLA
kernel_mul_mv_ext_q6_K_f32_r1_3                                FALLA
kernel_mul_mv_ext_q6_K_f32_r1_3_nr0_2                          FALLA
kernel_mul_mv_ext_q6_K_f32_r1_4                                FALLA
kernel_mul_mv_ext_q6_K_f32_r1_4_nr0_2                          FALLA
kernel_mul_mv_ext_q6_K_f32_r1_5                                FALLA
kernel_mul_mv_ext_q6_K_f32_r1_5_nr0_2                          FALLA
kernel_mul_mv_ext_q8_0_f32_r1_2                                FALLA
kernel_mul_mv_ext_q8_0_f32_r1_3                                FALLA
kernel_mul_mv_ext_q8_0_f32_r1_4                                FALLA
kernel_mul_mv_ext_q8_0_f32_r1_5                                FALLA
kernel_mul_mv_f16_f16                                          FALLA
kernel_mul_mv_f16_f16_4                                        FALLA
kernel_mul_mv_f16_f16_short                                    FALLA
kernel_mul_mv_f16_f32                                          FALLA
kernel_mul_mv_f16_f32_4                                        FALLA
kernel_mul_mv_f16_f32_short                                    FALLA
kernel_mul_mv_f32_f32                                          FALLA
kernel_mul_mv_f32_f32_4                                        FALLA
kernel_mul_mv_f32_f32_short                                    FALLA
kernel_mul_mv_id                                               FALLA
kernel_mul_mv_id_bf16_f32                                      FALLA
kernel_mul_mv_id_bf16_f32_4                                    FALLA
kernel_mul_mv_id_f16_f32                                       FALLA
kernel_mul_mv_id_f16_f32_4                                     FALLA
kernel_mul_mv_id_f32_f32                                       FALLA
kernel_mul_mv_id_f32_f32_4                                     FALLA
kernel_mul_mv_id_iq1_m_f32                                     FALLA
kernel_mul_mv_id_iq1_s_f32                                     FALLA
kernel_mul_mv_id_iq2_s_f32                                     FALLA
kernel_mul_mv_id_iq2_xs_f32                                    FALLA
kernel_mul_mv_id_iq2_xxs_f32                                   FALLA
kernel_mul_mv_id_iq3_s_f32                                     FALLA
kernel_mul_mv_id_iq3_xxs_f32                                   FALLA
kernel_mul_mv_id_iq4_nl_f32                                    FALLA
kernel_mul_mv_id_iq4_xs_f32                                    FALLA
kernel_mul_mv_id_mxfp4_f32                                     FALLA
kernel_mul_mv_id_q1_0_f32                                      FALLA
kernel_mul_mv_id_q2_0_f32                                      FALLA
kernel_mul_mv_id_q2_K_f32                                      FALLA
kernel_mul_mv_id_q3_K_f32                                      FALLA
kernel_mul_mv_id_q4_0_f32                                      FALLA
kernel_mul_mv_id_q4_1_f32                                      FALLA
kernel_mul_mv_id_q4_K_f32                                      FALLA
kernel_mul_mv_id_q5_0_f32                                      FALLA
kernel_mul_mv_id_q5_1_f32                                      FALLA
kernel_mul_mv_id_q5_K_f32                                      FALLA
kernel_mul_mv_id_q6_K_f32                                      FALLA
kernel_mul_mv_id_q8_0_f32                                      FALLA
kernel_mul_mv_iq1_m_f32                                        FALLA
kernel_mul_mv_iq1_s_f32                                        FALLA
kernel_mul_mv_iq2_s_f32                                        FALLA
kernel_mul_mv_iq2_xs_f32                                       FALLA
kernel_mul_mv_iq2_xxs_f32                                      FALLA
kernel_mul_mv_iq3_s_f32                                        FALLA
kernel_mul_mv_iq3_xxs_f32                                      FALLA
kernel_mul_mv_iq4_nl_f32                                       FALLA
kernel_mul_mv_iq4_xs_f32                                       FALLA
kernel_mul_mv_mxfp4_f32                                        FALLA
kernel_mul_mv_q1_0_f32                                         FALLA
kernel_mul_mv_q2_                                              FALLA
kernel_mul_mv_q2_0_f32                                         FALLA
kernel_mul_mv_q2_K_f32                                         FALLA
kernel_mul_mv_q3_                                              FALLA
kernel_mul_mv_q3_K_f32                                         FALLA
kernel_mul_mv_q4_                                              FALLA
kernel_mul_mv_q4_0_f32                                         FALLA
kernel_mul_mv_q4_1_f32                                         FALLA
kernel_mul_mv_q4_K_f32                                         FALLA
kernel_mul_mv_q4_K_f32_r4                                      FALLA
kernel_mul_mv_q5_                                              FALLA
kernel_mul_mv_q5_0_f32                                         FALLA
kernel_mul_mv_q5_1_f32                                         FALLA
kernel_mul_mv_q5_K_f32                                         FALLA
kernel_mul_mv_q6_                                              FALLA
kernel_mul_mv_q6_K_f32                                         FALLA
kernel_mul_mv_q8_0_f32                                         FALLA
kernel_mul_mv_qvk_q2_                                          FALLA
kernel_mul_mv_qvk_q2_K_q4_K_q2_K_f32                           FALLA
kernel_mul_mv_qvk_q4_                                          FALLA
kernel_mul_mv_qvk_q4_0_q4_0_q4_0_f32                           FALLA
kernel_mul_mv_qvk_q4_K_q6_K_q4_K_f32                           FALLA
kernel_mul_mv_t_t                                              FALLA
kernel_mul_mv_t_t_4                                            FALLA
kernel_mul_mv_t_t_short                                        FALLA
kernel_norm_f32                                                  303       0       3      5       2    185    1024        0
kernel_norm_f32_4                                                340       0       3      5       5    238    1024        0
kernel_norm_fuse_impl                                          FALLA
kernel_norm_mul_add_f32                                          658       0       5      5       2    239    1024        0
kernel_norm_mul_add_f32_4                                        684       0       5      5       9    205    1024        0
kernel_norm_mul_f32                                              490       0       4      5       2    185    1024        0
kernel_norm_mul_f32_4                                            529       0       4      5       5    228    1024        0
kernel_op_sum_f32                                                236       0       1      2       1    206    1024        0
kernel_opt_step_adamw_f32                                        140       0       4      0       4    212    1024        0
kernel_opt_step_sgd_f32                                          101       0       2      0       2    184    1024        0
kernel_pad_f32                                                   301       0       1      0       4    184    1024        0
kernel_pad_f32_4                                                 311       0       1      0       2    218    1024        0
kernel_pad_impl                                                FALLA
kernel_pad_reflect_1d_f32                                        208       0       3      0       1    204    1024        0
kernel_pool_1d_avg_f32                                           402       0       1      0       5    183    1024        0
kernel_pool_1d_max_f32                                           392       0       1      0       7    188    1024        0
kernel_pool_2d_avg_f32                                           493       0       1      0       6    183    1024        0
kernel_pool_2d_max_f32                                           484       0       1      0       6    188    1024        0
kernel_reglu                                                   FALLA
kernel_reglu_f16                                                 145       1       1      0       1    226    1024        0
kernel_reglu_f32                                                 141       0       2      0       0    218    1024        0
kernel_repeat                                                  FALLA
kernel_repeat_bf16                                               333       1       1      0       1    185    1024        0
kernel_repeat_f16                                                333       1       1      0       1    204    1024        0
kernel_repeat_f32                                                334       0       1      0       0    204    1024        0
kernel_repeat_i16                                                333       1       1      0       1    204    1024        0
kernel_repeat_i32                                                334       0       1      0       0    204    1024        0
kernel_rms_norm_f32                                              237       0       2      3       1    220    1024        0
kernel_rms_norm_f32_4                                            246       0       2      3       1    246    1024        0
kernel_rms_norm_fuse_impl                                      FALLA
kernel_rms_norm_mul_add_f32                                      585       0       4      3       1    185    1024        0
kernel_rms_norm_mul_add_f32_4                                    598       0       4      3       5    202    1024        0
kernel_rms_norm_mul_f32                                          416       0       3      3       1    185    1024        0
kernel_rms_norm_mul_f32_4                                        438       0       3      3       1    202    1024        0
kernel_roll_f32                                                  166       0       1      0       0    182    1024        0
kernel_rope_multi                                              FALLA
kernel_rope_multi_f16                                          FALLA
kernel_rope_multi_f32                                          FALLA
kernel_rope_neox                                               FALLA
kernel_rope_neox_f16                                           FALLA
kernel_rope_neox_f32                                           FALLA
kernel_rope_norm                                               FALLA
kernel_rope_norm_f16                                           FALLA
kernel_rope_norm_f32                                           FALLA
kernel_rope_vision                                             FALLA
kernel_rope_vision_f16                                         FALLA
kernel_rope_vision_f32                                         FALLA
kernel_rwkv_wkv6_f32                                            1144       0      69     67      92    187     512     1024
kernel_rwkv_wkv7_f32                                             863       0      70     83      69    234     512     1280
kernel_set_rows_bf16_i32_bf16                                    313       1       2      0       3    206    1024        0
kernel_set_rows_bf16_i64_bf16                                    313       1       2      0       3    206    1024        0
kernel_set_rows_f                                              FALLA
kernel_set_rows_f16_i32_f16                                      313       1       2      0       1    210    1024        0
kernel_set_rows_f16_i64_f16                                      313       1       2      0       1    210    1024        0
kernel_set_rows_f32_i32_bf16                                     322       0       2      0       2    206    1024        0
kernel_set_rows_f32_i32_f16                                      315       0       2      0       1    187    1024        0
kernel_set_rows_f32_i32_f32                                      313       0       2      0       1    210    1024        0
kernel_set_rows_f32_i32_iq4_nl                                  2647       0     175      0      11    249    1024        0
kernel_set_rows_f32_i32_q4_0                                     796       0      44      0       7    229    1024        0
kernel_set_rows_f32_i32_q4_1                                     833       0      41      0       7    249    1024        0
kernel_set_rows_f32_i32_q5_0                                     943       0      44      0       7    232    1024        0
kernel_set_rows_f32_i32_q5_1                                     876       0      41      0       7    206    1024        0
kernel_set_rows_f32_i32_q8_0                                     790       0      41      0       7    224    1024        0
kernel_set_rows_f32_i64_bf16                                     322       0       2      0       2    206    1024        0
kernel_set_rows_f32_i64_f16                                      315       0       2      0       1    187    1024        0
kernel_set_rows_f32_i64_f32                                      313       0       2      0       1    210    1024        0
kernel_set_rows_f32_i64_iq4_nl                                  2647       0     175      0      11    249    1024        0
kernel_set_rows_f32_i64_q4_0                                     796       0      44      0       7    229    1024        0
kernel_set_rows_f32_i64_q4_1                                     833       0      41      0       7    249    1024        0
kernel_set_rows_f32_i64_q5_0                                     943       0      44      0       7    232    1024        0
kernel_set_rows_f32_i64_q5_1                                     876       0      41      0       7    206    1024        0
kernel_set_rows_f32_i64_q8_0                                     790       0      41      0       7    224    1024        0
kernel_set_rows_q32                                            FALLA
kernel_set_rows_turbo                                          FALLA
kernel_set_rows_turbo2                                         FALLA
kernel_set_rows_turbo2_i32                                      3603       9     542      0       3    240     256        0
kernel_set_rows_turbo2_i64                                      3603       9     542      0       3    240     256        0
kernel_set_rows_turbo3_coop                                    FALLA
kernel_set_rows_turbo3_coop_w64                                FALLA
kernel_set_rows_turbo3_i32                                       830       0       8      0       8    248    1024        0
kernel_set_rows_turbo3_i32_w64                                   948       0       8      0       5    248    1024        0
kernel_set_rows_turbo3_i64                                       829       0       8      0       6    248    1024        0
kernel_set_rows_turbo3_i64_w64                                   945       0       8      0       5    248    1024        0
kernel_set_rows_turbo4                                         FALLA
kernel_set_rows_turbo4_coop                                    FALLA
kernel_set_rows_turbo4_coop_w64                                FALLA
kernel_set_rows_turbo4_i32                                      1033       0       8      0       6    252    1024        0
kernel_set_rows_turbo4_i32_w64                                  1032       0       8      0       4    252    1024        0
kernel_set_rows_turbo4_i64                                      1030       0       8      0       8    252    1024        0
kernel_set_rows_turbo4_i64_w64                                  1029       0       8      0       4    252    1024        0
kernel_silu_back_f32                                             107       0       2      0       1    186    1024        0
kernel_snake                                                   FALLA
kernel_snake_bf16                                                161       0       2      0       2    248    1024        0
kernel_snake_f16                                                 156       0       2      0       1    236    1024        0
kernel_snake_f32                                                 156       0       3      0       0    234    1024        0
kernel_soft_max                                                FALLA
kernel_soft_max_4                                              FALLA
kernel_soft_max_f16                                              593       0       3      6       3    188    1024        0
kernel_soft_max_f16_4                                            648       0       5      6       6    202    1024        0
kernel_soft_max_f32                                              577       0       6      6       3    252    1024        0
kernel_soft_max_f32_4                                            638       0       5      6       5    212    1024        0
kernel_solve_tri_f32                                           FALLA
kernel_ssm_conv_f32_f32                                          137       0       2      0       0    187    1024        0
kernel_ssm_conv_f32_f32_4                                        145       0       2      0       1    216    1024        0
kernel_ssm_conv_f32_f32_batched                                FALLA
kernel_ssm_conv_f32_f32_batched_4                              FALLA
kernel_ssm_conv_f32_f32_decode                                   153       0       2      0       1    187    1024        0
kernel_ssm_conv_f32_f32_decode_4                                 164       0       2      0       0    206    1024        0
kernel_ssm_conv_f32_f32_decode_silu                              160       0       2      0       0    187    1024        0
kernel_ssm_conv_f32_f32_decode_silu_4                            175       0       2      0       1    206    1024        0
kernel_ssm_scan_f32                                              745       0       7      7       4    188    1024        0
kernel_sum_rows_f32_f32                                        FALLA
kernel_sum_rows_f32_f32_4                                      FALLA
kernel_sum_rows_impl                                           FALLA
kernel_swiglu                                                  FALLA
kernel_swiglu_f16                                                151       0       0      0       2    230    1024        0
kernel_swiglu_f32                                                151       0       2      0       1    228    1024        0
kernel_swiglu_oai                                              FALLA
kernel_swiglu_oai_f16                                            157       0       0      0       2    242    1024        0
kernel_swiglu_oai_f32                                            156       0       2      0       1    240    1024        0
kernel_timestep_embedding_f32                                    147       0       0      0       0    206    1024        0
kernel_topk_moe_f32                                              725       0       8     51       1    237    1024        0
kernel_tri                                                     FALLA
kernel_tri_bf16_0                                                156       0       0      0       1    204    1024        0
kernel_tri_bf16_1                                                156       0       0      0       1    204    1024        0
kernel_tri_bf16_2                                                156       0       0      0       1    204    1024        0
kernel_tri_bf16_3                                                156       0       0      0       1    204    1024        0
kernel_tri_f16_0                                                 156       1       1      0       1    204    1024        0
kernel_tri_f16_1                                                 156       1       1      0       1    204    1024        0
kernel_tri_f16_2                                                 157       1       1      0       0    204    1024        0
kernel_tri_f16_3                                                 157       1       1      0       0    204    1024        0
kernel_tri_f32_0                                                 150       0       1      0       1    204    1024        0
kernel_tri_f32_1                                                 150       0       1      0       1    204    1024        0
kernel_tri_f32_2                                                 151       0       1      0       0    204    1024        0
kernel_tri_f32_3                                                 151       0       1      0       0    204    1024        0
kernel_turbo_wht                                                1579       0      32      0       5    186     768        0
kernel_unary_f16_f16                                           FALLA
kernel_unary_f16_f16_4                                         FALLA
kernel_unary_f32_f32                                           FALLA
kernel_unary_f32_f32_4                                         FALLA
kernel_unary_impl                                              FALLA
kernel_upscale_bicubic_f32                                       765       0      16      0      14    185    1024        0
kernel_upscale_bilinear_f32                                    FALLA
kernel_upscale_nearest_f32                                       349       0       1      0       2    186    1024        0
