ggml-cpu : split arch-specific implementations #13892
xctan
force pushed
194 days ago
xctan
force pushed
193 days ago
xctan
marked this pull request as ready for review 193 days ago
move ggml-cpu-aarch64 to repack
6814bd48
split quantize_row_q8_0/1
a07340a1
split helper functions
82d7410f
split ggml_vec_dot_q4_0_q8_0
ead57621
split ggml_vec_dot_q4_1_q8_1
627e1ecd
split ggml_vec_dot_q5_0_q8_0
9582518e
split ggml_vec_dot_q5_1_q8_1
beca2195
split ggml_vec_dot_q8_0_q8_0
a32715a8
split ggml_vec_dot_tq1_0_q8_K
a46eca7d
split ggml_vec_dot_tq2_0_q8_K
96a7f516
split ggml_vec_dot_q2_K_q8_K
5f881c9f
split ggml_vec_dot_q3_K_q8_K
91fbf27d
split ggml_vec_dot_q4_K_q8_K
58b6c627
split ggml_vec_dot_q5_K_q8_K
6272e0cc
split ggml_vec_dot_q6_K_q8_K
7c7223f2
split ggml_vec_dot_iq2_xxs_q8_K
9671c0e2
split ggml_vec_dot_iq2_xs_q8_K
e4e1cfc2
split ggml_vec_dot_iq2_s_q8_K
c9efc9ee
split ggml_vec_dot_iq3_xxs_q8_K
d1d2e24d
split ggml_vec_dot_iq3_s_q8_K
da6fcec8
split ggml_vec_dot_iq1_s_q8_K
3334b107
split ggml_vec_dot_iq1_m_q8_K
93f0c4f6
split ggml_vec_dot_iq4_nl_q8_0
3f4866f5
split ggml_vec_dot_iq4_xs_q8_K
740b3c9f
fix typos
9487b769
fix missing prototypes
88e7e42a
rename ggml-cpu-quants.c
2252aa2b
rename ggml-cpu-traits
6df3dd57
rename arm folder
3566ee8d
move cpu-feats-x86.cpp
f40ad8c9
rename ggml-cpu-hbm
1ac2d5ec
update arm detection macro in quants.c
321b3ac4
move iq quant tables
7b5bf50f
split ggml_quantize_mat_q8_0/K
bf3dbea0
split ggml_gemv_*
868c895c
split ggml_gemm_*
6a2ba77c
rename namespace aarch64 to repack
72ddf5ad
use weak aliases to replace test macros
ad523494
rename GGML_CPU_AARCH64 to GGML_CPU_REPACK
62dc3fd8
rename more aarch64 to repack
46b1e49e
clean up rebase leftover
5601df65
fix compilation errors
827aec0d
remove trailing spaces
58210b8c
try to fix clang compilation errors
2739f4c5
try to fix clang compilation errors again
8713f877
try to fix clang compilation errors, 3rd attempt
df278103
try to fix clang compilation errors, 4th attempt
553d8ca6
try to fix clang compilation errors, 5th attempt
9bfcd7e2
try to fix clang compilation errors, 6th attempt
08ebdd91
try to fix clang compilation errors, 7th attempt
67eceec5
try to fix clang compilation errors, 8th attempt
01a1c5c0
try to fix clang compilation errors, 9th attempt
bef5b8d1
more cleanup
47701d51
fix compilation errors
e5b6fdb7
fix apple targets
2573662f
fix a typo in arm version of ggml_vec_dot_q4_K_q8_K
93e6718b
xctan
force pushed
to
93e6718b
187 days ago
ggerganov
approved these changes
on 2025-06-06
slaren
approved these changes
on 2025-06-09
slaren
merged
f470bc36
into master 183 days ago
Assignees
No one assigned
Labels
Nvidia GPU
ggml
SYCL
Login to write a write a comment.
Login via GitHub