C: Switch to [inv]NTT with 2+2+2+1 structure

[hanno-becker]

Rewrite mlk_poly_ntt_c / mlk_poly_invntt_tomont_c to process two layers at a time, with three 2-layer passes plus the leftover layer 7 as a single layer.

Introduces shared mlk_ct_butterfly and mlk_gs_butterfly helpers; the inverse 2-layer block applies four GS butterflies and then Barrett-reduces the additive outputs explicitly.

mlk_fqmul now takes a precomputed b_twisted = b * MLKEM_Q^{-1} mod 2^16 and uses a hi-mul / lo-mul-and-correct sequence in place of an inline mlk_montgomery_reduce, dropping the QINV multiply. The mlk_zetas table is regenerated as int16_t[128][2] of (zeta_mont, zeta_twisted) pairs.

[oqs-bot]

CBMC Results (ML-KEM-512)

⚠️ Attention Required

Proof	Status	Current	Previous	Change
**TOTAL**	⚠️	1689s	1202s	+40.5%

Full Results (195 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	⚠️	1689s	1202s	+40.5%
mlk_ntt_2_layers_block	✅	402s	-	new
mlk_indcpa_keypair_derand	✅	244s	232s	+5%
mlk_indcpa_enc	✅	161s	158s	+2%
mlk_rej_uniform_c	✅	108s	106s	+2%
mlk_polyvec_basemul_acc_montgomery_cached_c	✅	53s	49s	+8%
mlk_poly_rej_uniform	✅	29s	28s	+4%
mlk_keccak_squeezeblocks_x4	✅	26s	24s	+8%
mlk_ntt_2_layers	✅	24s	-	new
poly_ntt_native	✅	24s	23s	+4%
mlk_ntt_layer	✅	22s	25s	-12%
mlk_poly_reduce_native	✅	19s	17s	+12%
keccakf1600x4_permute_native_x4	✅	16s	19s	-16%
mlk_poly_decompress_d10_native	✅	15s	12s	+25%
mlk_indcpa_dec	✅	14s	15s	-7%
mlk_poly_decompress_d4_native	✅	13s	13s	+0%
mlk_invntt_2_layers_block	✅	11s	-	new
mlk_polyvec_add	✅	11s	12s	-8%
mlk_poly_frommsg	✅	9s	10s	-10%
mlk_poly_rej_uniform_x4	✅	9s	5s	+80%
mlk_keccak_squeezeblocks	✅	8s	6s	+33%
polyvec_basemul_acc_montgomery_cached_native	✅	8s	6s	+33%
mlk_invntt_layer	✅	7s	4s	+75%
mlk_keccak_squeeze_once	✅	7s	7s	+0%
mlk_ntt_butterfly_block	✅	7s	7s	+0%
mlk_poly_frombytes_native	✅	7s	8s	-12%
mlk_poly_invntt_tomont_c	✅	6s	2s	+200%
rej_uniform_native_x86_64	✅	6s	6s	+0%
kem_check_pk	✅	5s	2s	+150%
mlk_fqmul	✅	5s	13s	-62%
mlk_keccakf1600_permute_c	✅	5s	4s	+25%
mlk_poly_cbd_eta2	✅	5s	7s	-29%
mlk_poly_decompress_d11_c	✅	5s	3s	+67%
mlk_poly_ntt	✅	5s	6s	-17%
nttunpack_native_x86_64	✅	5s	3s	+67%
poly_decompress_d10_native_x86_64	✅	5s	2s	+150%
poly_decompress_d4_native_x86_64	✅	5s	7s	-29%
sys_check_capability	✅	5s	3s	+67%
keccakf1600_permute_native	✅	4s	2s	+100%
kem_enc	✅	4s	2s	+100%
mlk_ct_cmov_zero	✅	4s	2s	+100%
mlk_keccak_absorb_once_x4	✅	4s	4s	+0%
mlk_montgomery_reduce	✅	4s	2s	+100%
mlk_poly_compress_d11_c	✅	4s	2s	+100%
mlk_poly_compress_du	✅	4s	2s	+100%
mlk_polyvec_reduce	✅	4s	2s	+100%
mlk_polyvec_tomont	✅	4s	1s	+300%
mlk_scalar_decompress_d4	✅	4s	2s	+100%
poly_frombytes_native_x86_64	✅	4s	5s	-20%
poly_invntt_tomont_native	✅	4s	2s	+100%
poly_mulcache_compute_native_aarch64	✅	4s	3s	+33%
polyvec_basemul_acc_montgomery_cached_k3_native_x86_64	✅	4s	3s	+33%
intt_native_x86_64	✅	3s	2s	+50%
keccak_f1600_x1_native_aarch64	✅	3s	3s	+0%
keccak_f1600_x1_native_aarch64_v84a	✅	3s	3s	+0%
keccak_f1600_x4_native_aarch64_v84a	✅	3s	5s	-40%
keccakf1600x4_extract_bytes_native	✅	3s	2s	+50%
keccakf1600x4_xor_bytes_native	✅	3s	3s	+0%
kem_check_sk	✅	3s	1s	+200%
kem_dec	✅	3s	6s	-50%
mlk_barrett_reduce	✅	3s	2s	+50%
mlk_check_pct	✅	3s	1s	+200%
mlk_ct_get_optblocker_i32	✅	3s	2s	+50%
mlk_ct_memcmp	✅	3s	2s	+50%
mlk_gen_matrix	✅	3s	1s	+200%
mlk_invntt_2_layers	✅	3s	-	new
mlk_keccak_absorb_once	✅	3s	2s	+50%
mlk_keccakf1600_extract_bytes	✅	3s	3s	+0%
mlk_keccakf1600x4_extract_bytes_c	✅	3s	2s	+50%
mlk_keccakf1600x4_xor_bytes	✅	3s	1s	+200%
mlk_poly_add	✅	3s	2s	+50%
mlk_poly_cbd_eta1	✅	3s	2s	+50%
mlk_poly_compress_d5_c	✅	3s	1s	+200%
mlk_poly_compress_d5_native	✅	3s	5s	-40%
mlk_poly_decompress_d4_c	✅	3s	1s	+200%
mlk_poly_decompress_du	✅	3s	2s	+50%
mlk_poly_frombytes	✅	3s	3s	+0%
mlk_poly_mulcache_compute	✅	3s	3s	+0%
mlk_poly_reduce_c	✅	3s	1s	+200%
mlk_poly_tobytes	✅	3s	1s	+200%
mlk_poly_tobytes_c	✅	3s	4s	-25%
mlk_poly_tomont_native	✅	3s	3s	+0%
mlk_polyvec_compress_du	✅	3s	2s	+50%
mlk_polyvec_invntt_tomont	✅	3s	2s	+50%
mlk_polyvec_permute_bitrev_to_custom_native	✅	3s	2s	+50%
mlk_scalar_compress_d1	✅	3s	1s	+200%
mlk_scalar_compress_d4	✅	3s	2s	+50%
mlk_scalar_decompress_d10	✅	3s	2s	+50%
mlk_scalar_decompress_d5	✅	3s	3s	+0%
mlk_scalar_signed_to_unsigned_q	✅	3s	2s	+50%
mlk_sha3_512	✅	3s	3s	+0%
mlk_shake128x4_absorb_once	✅	3s	1s	+200%
mlk_shake256x4	✅	3s	2s	+50%
mlk_value_barrier_u8	✅	3s	2s	+50%
poly_compress_d10_native_x86_64	✅	3s	2s	+50%
poly_compress_d5_native_x86_64	✅	3s	1s	+200%
poly_decompress_d11_native_x86_64	✅	3s	4s	-25%
poly_reduce_native_aarch64	✅	3s	2s	+50%
poly_tobytes_native_x86_64	✅	3s	1s	+200%
polyvec_basemul_acc_montgomery_cached_k2_native_aarch64	✅	3s	2s	+50%
polyvec_basemul_acc_montgomery_cached_k4_native_aarch64	✅	3s	4s	-25%
rej_uniform_native	✅	3s	2s	+50%
intt_native_aarch64	✅	2s	2s	+0%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	2s	2s	+0%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	2s	4s	-50%
kem_enc_derand	✅	2s	2s	+0%
kem_keypair	✅	2s	3s	-33%
kem_keypair_derand	✅	2s	2s	+0%
mlk_ct_cmask_neg_i16	✅	2s	2s	+0%
mlk_ct_cmask_nonzero_u16	✅	2s	2s	+0%
mlk_ct_get_optblocker_u32	✅	2s	1s	+100%
mlk_ct_get_optblocker_u8	✅	2s	2s	+0%
mlk_ct_sel_uint8	✅	2s	1s	+100%
mlk_enc_getnoise_eta1_eta2	✅	2s	3s	-33%
mlk_gen_matrix_serial	✅	2s	3s	-33%
mlk_keccakf1600_xor_bytes (big endian)	✅	2s	5s	-60%
mlk_keccakf1600x4_permute	✅	2s	1s	+100%
mlk_keccakf1600x4_xor_bytes_c	✅	2s	2s	+0%
mlk_keypair_getnoise_eta1	✅	2s	2s	+0%
mlk_matvec_mul	✅	2s	1s	+100%
mlk_poly_compress_d10	✅	2s	2s	+0%
mlk_poly_compress_d10_c	✅	2s	3s	-33%
mlk_poly_compress_d11	✅	2s	2s	+0%
mlk_poly_compress_d11_native	✅	2s	2s	+0%
mlk_poly_compress_d4	✅	2s	1s	+100%
mlk_poly_compress_d4_c	✅	2s	2s	+0%
mlk_poly_compress_d4_native	✅	2s	2s	+0%
mlk_poly_compress_d5	✅	2s	3s	-33%
mlk_poly_compress_dv	✅	2s	2s	+0%
mlk_poly_decompress_d11	✅	2s	4s	-50%
mlk_poly_decompress_d11_native	✅	2s	1s	+100%
mlk_poly_decompress_d5	✅	2s	2s	+0%
mlk_poly_decompress_d5_native	✅	2s	3s	-33%
mlk_poly_decompress_dv	✅	2s	5s	-60%
mlk_poly_frombytes_c	✅	2s	3s	-33%
mlk_poly_getnoise_eta1122_4x	✅	2s	2s	+0%
mlk_poly_getnoise_eta1_4x	✅	2s	3s	-33%
mlk_poly_getnoise_eta1_4x_native	✅	2s	2s	+0%
mlk_poly_getnoise_eta2	✅	2s	4s	-50%
mlk_poly_invntt_tomont	✅	2s	1s	+100%
mlk_poly_mulcache_compute_c	✅	2s	3s	-33%
mlk_poly_mulcache_compute_native	✅	2s	3s	-33%
mlk_poly_ntt_c	✅	2s	4s	-50%
mlk_poly_sub	✅	2s	3s	-33%
mlk_poly_tobytes_native	✅	2s	2s	+0%
mlk_poly_tomont	✅	2s	2s	+0%
mlk_poly_tomont_c	✅	2s	2s	+0%
mlk_poly_tomsg	✅	2s	3s	-33%
mlk_polymat_permute_bitrev_to_custom	✅	2s	2s	+0%
mlk_polyvec_basemul_acc_montgomery_cached	✅	2s	1s	+100%
mlk_polyvec_decompress_du	✅	2s	2s	+0%
mlk_polyvec_frombytes	✅	2s	2s	+0%
mlk_polyvec_mulcache_compute	✅	2s	3s	-33%
mlk_polyvec_permute_bitrev_to_custom	✅	2s	2s	+0%
mlk_polyvec_tobytes	✅	2s	3s	-33%
mlk_rej_uniform	✅	2s	3s	-33%
mlk_scalar_compress_d10	✅	2s	1s	+100%
mlk_scalar_compress_d11	✅	2s	3s	-33%
mlk_scalar_compress_d5	✅	2s	2s	+0%
mlk_scalar_decompress_d11	✅	2s	1s	+100%
mlk_sha3_256	✅	2s	1s	+100%
mlk_shake128_absorb_once	✅	2s	2s	+0%
mlk_shake128_squeezeblocks	✅	2s	1s	+100%
mlk_shake256	✅	2s	1s	+100%
mlk_value_barrier_u32	✅	2s	2s	+0%
ntt_native_aarch64	✅	2s	2s	+0%
poly_compress_d11_native_x86_64	✅	2s	3s	-33%
poly_decompress_d5_native_x86_64	✅	2s	1s	+100%
poly_mulcache_compute_native_x86_64	✅	2s	5s	-60%
poly_reduce_native_x86_64	✅	2s	3s	-33%
poly_tomont_native_aarch64	✅	2s	3s	-33%
poly_tomont_native_x86_64	✅	2s	1s	+100%
polyvec_basemul_acc_montgomery_cached_k2_native_x86_64	✅	2s	2s	+0%
polyvec_basemul_acc_montgomery_cached_k3_native_aarch64	✅	2s	2s	+0%
rej_uniform_native_aarch64	✅	2s	2s	+0%
keccak_f1600_x4_native_avx2	✅	1s	2s	-50%
mlk_ct_cmask_nonzero_u8	✅	1s	1s	+0%
mlk_ct_sel_int16	✅	1s	2s	-50%
mlk_keccakf1600_extract_bytes (big endian)	✅	1s	2s	-50%
mlk_keccakf1600_permute	✅	1s	4s	-75%
mlk_keccakf1600_xor_bytes	✅	1s	2s	-50%
mlk_keccakf1600x4_extract_bytes	✅	1s	2s	-50%
mlk_poly_compress_d10_native	✅	1s	1s	+0%
mlk_poly_decompress_d10	✅	1s	3s	-67%
mlk_poly_decompress_d10_c	✅	1s	1s	+0%
mlk_poly_decompress_d4	✅	1s	1s	+0%
mlk_poly_decompress_d5_c	✅	1s	1s	+0%
mlk_poly_reduce	✅	1s	3s	-67%
mlk_polyvec_ntt	✅	1s	2s	-50%
mlk_shake128x4_squeezeblocks	✅	1s	3s	-67%
mlk_value_barrier_i32	✅	1s	3s	-67%
ntt_native_x86_64	✅	1s	2s	-50%
poly_compress_d4_native_x86_64	✅	1s	3s	-67%
poly_getnoise_eta1122_4x_native	✅	1s	2s	-50%
poly_tobytes_native_aarch64	✅	1s	1s	+0%
polyvec_basemul_acc_montgomery_cached_k4_native_x86_64	✅	1s	5s	-80%

[oqs-bot]

CBMC Results (ML-KEM-768)

⚠️ Attention Required

Proof	Status	Current	Previous	Change
**TOTAL**	⚠️	1727s	1303s	+32.5%

Full Results (195 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	⚠️	1727s	1303s	+32.5%
mlk_ntt_2_layers_block	✅	436s	-	new
mlk_indcpa_keypair_derand	✅	196s	202s	-3%
mlk_indcpa_enc	✅	172s	177s	-3%
mlk_rej_uniform_c	✅	122s	136s	-10%
mlk_polyvec_basemul_acc_montgomery_cached_c	✅	43s	46s	-7%
mlk_poly_rej_uniform	✅	32s	35s	-9%
mlk_keccak_squeezeblocks_x4	✅	27s	25s	+8%
mlk_ntt_2_layers	✅	27s	-	new
poly_ntt_native	✅	27s	28s	-4%
mlk_ntt_layer	✅	25s	35s	-29%
mlk_poly_reduce_native	✅	19s	21s	-10%
keccakf1600x4_permute_native_x4	✅	17s	17s	+0%
polyvec_basemul_acc_montgomery_cached_native	✅	16s	19s	-16%
mlk_indcpa_dec	✅	15s	14s	+7%
mlk_poly_decompress_d4_native	✅	14s	13s	+8%
mlk_poly_decompress_d10_native	✅	13s	16s	-19%
mlk_poly_frommsg	✅	12s	8s	+50%
mlk_invntt_2_layers_block	✅	11s	-	new
mlk_poly_ntt	✅	10s	7s	+43%
mlk_ntt_butterfly_block	✅	9s	10s	-10%
mlk_poly_frombytes_native	✅	9s	12s	-25%
mlk_polyvec_add	✅	9s	9s	+0%
mlk_invntt_layer	✅	8s	5s	+60%
mlk_keccak_squeeze_once	✅	8s	7s	+14%
mlk_keccak_squeezeblocks	✅	8s	10s	-20%
mlk_poly_compress_d4_c	✅	8s	2s	+300%
kem_dec	✅	7s	5s	+40%
mlk_poly_rej_uniform_x4	✅	7s	6s	+17%
rej_uniform_native_x86_64	✅	7s	7s	+0%
poly_frombytes_native_x86_64	✅	6s	5s	+20%
mlk_keccak_absorb_once_x4	✅	5s	7s	-29%
mlk_keccakf1600_permute_c	✅	5s	6s	-17%
mlk_poly_cbd_eta1	✅	5s	2s	+150%
mlk_scalar_signed_to_unsigned_q	✅	5s	2s	+150%
rej_uniform_native_aarch64	✅	5s	3s	+67%
intt_native_aarch64	✅	4s	1s	+300%
intt_native_x86_64	✅	4s	1s	+300%
kem_check_pk	✅	4s	3s	+33%
mlk_ct_cmask_nonzero_u16	✅	4s	1s	+300%
mlk_enc_getnoise_eta1_eta2	✅	4s	3s	+33%
mlk_fqmul	✅	4s	14s	-71%
mlk_gen_matrix_serial	✅	4s	3s	+33%
mlk_keccak_absorb_once	✅	4s	3s	+33%
mlk_keccakf1600x4_permute	✅	4s	1s	+300%
mlk_keypair_getnoise_eta1	✅	4s	3s	+33%
mlk_poly_compress_d10_c	✅	4s	3s	+33%
mlk_poly_compress_du	✅	4s	2s	+100%
mlk_poly_decompress_d11	✅	4s	2s	+100%
mlk_poly_getnoise_eta1_4x_native	✅	4s	6s	-33%
mlk_poly_getnoise_eta2	✅	4s	3s	+33%
mlk_poly_tomont_native	✅	4s	3s	+33%
mlk_polyvec_reduce	✅	4s	2s	+100%
mlk_scalar_decompress_d4	✅	4s	4s	+0%
mlk_shake256x4	✅	4s	5s	-20%
mlk_value_barrier_u8	✅	4s	1s	+300%
poly_compress_d5_native_x86_64	✅	4s	6s	-33%
poly_decompress_d10_native_x86_64	✅	4s	3s	+33%
poly_decompress_d5_native_x86_64	✅	4s	3s	+33%
polyvec_basemul_acc_montgomery_cached_k2_native_aarch64	✅	4s	4s	+0%
polyvec_basemul_acc_montgomery_cached_k3_native_x86_64	✅	4s	5s	-20%
mlk_check_pct	✅	3s	2s	+50%
mlk_ct_cmask_nonzero_u8	✅	3s	1s	+200%
mlk_ct_get_optblocker_u8	✅	3s	2s	+50%
mlk_ct_sel_uint8	✅	3s	3s	+0%
mlk_gen_matrix	✅	3s	3s	+0%
mlk_keccakf1600x4_extract_bytes_c	✅	3s	2s	+50%
mlk_matvec_mul	✅	3s	1s	+200%
mlk_poly_cbd_eta2	✅	3s	3s	+0%
mlk_poly_compress_d10_native	✅	3s	3s	+0%
mlk_poly_compress_d4	✅	3s	2s	+50%
mlk_poly_compress_d5_native	✅	3s	2s	+50%
mlk_poly_decompress_d4_c	✅	3s	1s	+200%
mlk_poly_decompress_d5_c	✅	3s	2s	+50%
mlk_poly_getnoise_eta1_4x	✅	3s	3s	+0%
mlk_poly_invntt_tomont_c	✅	3s	3s	+0%
mlk_poly_mulcache_compute	✅	3s	2s	+50%
mlk_poly_tomont	✅	3s	3s	+0%
mlk_polyvec_frombytes	✅	3s	3s	+0%
mlk_polyvec_invntt_tomont	✅	3s	3s	+0%
mlk_polyvec_ntt	✅	3s	3s	+0%
mlk_polyvec_permute_bitrev_to_custom_native	✅	3s	4s	-25%
mlk_scalar_compress_d1	✅	3s	2s	+50%
mlk_scalar_compress_d4	✅	3s	3s	+0%
mlk_scalar_compress_d5	✅	3s	3s	+0%
mlk_scalar_decompress_d10	✅	3s	3s	+0%
mlk_shake128x4_squeezeblocks	✅	3s	2s	+50%
mlk_shake256	✅	3s	1s	+200%
mlk_value_barrier_i32	✅	3s	3s	+0%
ntt_native_aarch64	✅	3s	4s	-25%
ntt_native_x86_64	✅	3s	2s	+50%
nttunpack_native_x86_64	✅	3s	4s	-25%
poly_compress_d10_native_x86_64	✅	3s	3s	+0%
poly_compress_d11_native_x86_64	✅	3s	3s	+0%
poly_decompress_d11_native_x86_64	✅	3s	2s	+50%
poly_decompress_d4_native_x86_64	✅	3s	4s	-25%
poly_getnoise_eta1122_4x_native	✅	3s	5s	-40%
poly_mulcache_compute_native_aarch64	✅	3s	4s	-25%
poly_reduce_native_aarch64	✅	3s	3s	+0%
poly_tobytes_native_aarch64	✅	3s	3s	+0%
polyvec_basemul_acc_montgomery_cached_k4_native_aarch64	✅	3s	4s	-25%
keccak_f1600_x1_native_aarch64	✅	2s	5s	-60%
keccak_f1600_x1_native_aarch64_v84a	✅	2s	1s	+100%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	2s	2s	+0%
keccak_f1600_x4_native_avx2	✅	2s	2s	+0%
keccakf1600_permute_native	✅	2s	1s	+100%
keccakf1600x4_extract_bytes_native	✅	2s	3s	-33%
keccakf1600x4_xor_bytes_native	✅	2s	2s	+0%
kem_check_sk	✅	2s	3s	-33%
kem_enc	✅	2s	3s	-33%
kem_enc_derand	✅	2s	3s	-33%
kem_keypair	✅	2s	2s	+0%
mlk_ct_cmask_neg_i16	✅	2s	3s	-33%
mlk_ct_cmov_zero	✅	2s	1s	+100%
mlk_ct_get_optblocker_i32	✅	2s	2s	+0%
mlk_ct_sel_int16	✅	2s	3s	-33%
mlk_invntt_2_layers	✅	2s	-	new
mlk_keccakf1600_extract_bytes	✅	2s	3s	-33%
mlk_keccakf1600_extract_bytes (big endian)	✅	2s	5s	-60%
mlk_keccakf1600_permute	✅	2s	2s	+0%
mlk_keccakf1600_xor_bytes	✅	2s	3s	-33%
mlk_keccakf1600_xor_bytes (big endian)	✅	2s	2s	+0%
mlk_keccakf1600x4_xor_bytes	✅	2s	3s	-33%
mlk_keccakf1600x4_xor_bytes_c	✅	2s	1s	+100%
mlk_poly_add	✅	2s	3s	-33%
mlk_poly_compress_d10	✅	2s	3s	-33%
mlk_poly_compress_d11	✅	2s	2s	+0%
mlk_poly_compress_d11_c	✅	2s	2s	+0%
mlk_poly_compress_d4_native	✅	2s	3s	-33%
mlk_poly_compress_d5_c	✅	2s	4s	-50%
mlk_poly_decompress_d10	✅	2s	4s	-50%
mlk_poly_decompress_d10_c	✅	2s	2s	+0%
mlk_poly_decompress_d11_c	✅	2s	3s	-33%
mlk_poly_decompress_d11_native	✅	2s	2s	+0%
mlk_poly_decompress_d4	✅	2s	1s	+100%
mlk_poly_decompress_d5	✅	2s	3s	-33%
mlk_poly_decompress_du	✅	2s	2s	+0%
mlk_poly_decompress_dv	✅	2s	3s	-33%
mlk_poly_frombytes	✅	2s	2s	+0%
mlk_poly_frombytes_c	✅	2s	1s	+100%
mlk_poly_getnoise_eta1122_4x	✅	2s	1s	+100%
mlk_poly_invntt_tomont	✅	2s	3s	-33%
mlk_poly_mulcache_compute_native	✅	2s	2s	+0%
mlk_poly_reduce	✅	2s	1s	+100%
mlk_poly_sub	✅	2s	1s	+100%
mlk_poly_tobytes_native	✅	2s	2s	+0%
mlk_poly_tomsg	✅	2s	3s	-33%
mlk_polymat_permute_bitrev_to_custom	✅	2s	3s	-33%
mlk_polyvec_basemul_acc_montgomery_cached	✅	2s	3s	-33%
mlk_polyvec_decompress_du	✅	2s	2s	+0%
mlk_polyvec_mulcache_compute	✅	2s	4s	-50%
mlk_polyvec_tomont	✅	2s	1s	+100%
mlk_rej_uniform	✅	2s	3s	-33%
mlk_sha3_256	✅	2s	2s	+0%
mlk_sha3_512	✅	2s	1s	+100%
mlk_shake128_squeezeblocks	✅	2s	2s	+0%
mlk_shake128x4_absorb_once	✅	2s	2s	+0%
mlk_value_barrier_u32	✅	2s	1s	+100%
poly_invntt_tomont_native	✅	2s	2s	+0%
poly_mulcache_compute_native_x86_64	✅	2s	2s	+0%
poly_tobytes_native_x86_64	✅	2s	1s	+100%
poly_tomont_native_aarch64	✅	2s	3s	-33%
poly_tomont_native_x86_64	✅	2s	1s	+100%
polyvec_basemul_acc_montgomery_cached_k2_native_x86_64	✅	2s	2s	+0%
polyvec_basemul_acc_montgomery_cached_k3_native_aarch64	✅	2s	2s	+0%
polyvec_basemul_acc_montgomery_cached_k4_native_x86_64	✅	2s	1s	+100%
rej_uniform_native	✅	2s	4s	-50%
sys_check_capability	✅	2s	4s	-50%
keccak_f1600_x4_native_aarch64_v84a	✅	1s	4s	-75%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	1s	1s	+0%
kem_keypair_derand	✅	1s	3s	-67%
mlk_barrett_reduce	✅	1s	2s	-50%
mlk_ct_get_optblocker_u32	✅	1s	3s	-67%
mlk_ct_memcmp	✅	1s	3s	-67%
mlk_keccakf1600x4_extract_bytes	✅	1s	1s	+0%
mlk_montgomery_reduce	✅	1s	1s	+0%
mlk_poly_compress_d11_native	✅	1s	2s	-50%
mlk_poly_compress_d5	✅	1s	4s	-75%
mlk_poly_compress_dv	✅	1s	1s	+0%
mlk_poly_decompress_d5_native	✅	1s	1s	+0%
mlk_poly_mulcache_compute_c	✅	1s	3s	-67%
mlk_poly_ntt_c	✅	1s	2s	-50%
mlk_poly_reduce_c	✅	1s	6s	-83%
mlk_poly_tobytes	✅	1s	2s	-50%
mlk_poly_tobytes_c	✅	1s	2s	-50%
mlk_poly_tomont_c	✅	1s	1s	+0%
mlk_polyvec_compress_du	✅	1s	1s	+0%
mlk_polyvec_permute_bitrev_to_custom	✅	1s	1s	+0%
mlk_polyvec_tobytes	✅	1s	1s	+0%
mlk_scalar_compress_d10	✅	1s	1s	+0%
mlk_scalar_compress_d11	✅	1s	2s	-50%
mlk_scalar_decompress_d11	✅	1s	1s	+0%
mlk_scalar_decompress_d5	✅	1s	2s	-50%
mlk_shake128_absorb_once	✅	1s	2s	-50%
poly_compress_d4_native_x86_64	✅	1s	2s	-50%
poly_reduce_native_x86_64	✅	1s	3s	-67%

[oqs-bot]

CBMC Results (ML-KEM-1024)

⚠️ Attention Required

Proof	Status	Current	Previous	Change
**TOTAL**	⚠️	1562s	1207s	+29.4%

Full Results (195 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	⚠️	1562s	1207s	+29.4%
mlk_ntt_2_layers_block	✅	376s	-	new
mlk_indcpa_enc	✅	135s	141s	-4%
mlk_indcpa_keypair_derand	✅	123s	118s	+4%
mlk_rej_uniform_c	✅	113s	117s	-3%
mlk_polyvec_basemul_acc_montgomery_cached_c	✅	70s	75s	-7%
polyvec_basemul_acc_montgomery_cached_native	✅	33s	34s	-3%
mlk_poly_rej_uniform	✅	29s	30s	-3%
mlk_keccak_squeezeblocks_x4	✅	24s	25s	-4%
mlk_ntt_2_layers	✅	23s	-	new
mlk_ntt_layer	✅	23s	29s	-21%
poly_ntt_native	✅	23s	25s	-8%
mlk_poly_reduce_native	✅	18s	24s	-25%
keccakf1600x4_permute_native_x4	✅	15s	17s	-12%
mlk_poly_decompress_d11_native	✅	13s	14s	-7%
mlk_poly_decompress_d5_native	✅	12s	14s	-14%
mlk_invntt_2_layers_block	✅	11s	-	new
mlk_polyvec_add	✅	11s	11s	+0%
kem_dec	✅	8s	5s	+60%
mlk_indcpa_dec	✅	8s	10s	-20%
mlk_invntt_layer	✅	8s	9s	-11%
mlk_keccak_squeezeblocks	✅	8s	8s	+0%
mlk_poly_frombytes_native	✅	8s	7s	+14%
mlk_poly_ntt	✅	8s	8s	+0%
mlk_keccak_absorb_once_x4	✅	7s	5s	+40%
mlk_keccak_squeeze_once	✅	7s	10s	-30%
mlk_poly_frommsg	✅	7s	6s	+17%
mlk_ntt_butterfly_block	✅	6s	8s	-25%
mlk_poly_compress_d10_native	✅	6s	2s	+200%
mlk_polymat_permute_bitrev_to_custom	✅	6s	7s	-14%
rej_uniform_native_x86_64	✅	6s	5s	+20%
mlk_fqmul	✅	5s	16s	-69%
mlk_gen_matrix_serial	✅	5s	6s	-17%
mlk_keccakf1600_permute_c	✅	5s	7s	-29%
mlk_poly_tomont_c	✅	5s	1s	+400%
mlk_shake128x4_absorb_once	✅	5s	4s	+25%
poly_getnoise_eta1122_4x_native	✅	5s	5s	+0%
kem_check_pk	✅	4s	3s	+33%
kem_keypair_derand	✅	4s	3s	+33%
mlk_gen_matrix	✅	4s	6s	-33%
mlk_invntt_2_layers	✅	4s	-	new
mlk_keccak_absorb_once	✅	4s	4s	+0%
mlk_matvec_mul	✅	4s	3s	+33%
mlk_montgomery_reduce	✅	4s	2s	+100%
mlk_poly_compress_d11	✅	4s	1s	+300%
mlk_poly_compress_d11_c	✅	4s	4s	+0%
mlk_poly_compress_du	✅	4s	2s	+100%
mlk_poly_decompress_d10	✅	4s	1s	+300%
mlk_poly_getnoise_eta1_4x	✅	4s	3s	+33%
mlk_poly_invntt_tomont_c	✅	4s	3s	+33%
mlk_poly_rej_uniform_x4	✅	4s	7s	-43%
mlk_polyvec_mulcache_compute	✅	4s	3s	+33%
mlk_polyvec_permute_bitrev_to_custom_native	✅	4s	3s	+33%
mlk_polyvec_tobytes	✅	4s	2s	+100%
ntt_native_aarch64	✅	4s	2s	+100%
nttunpack_native_x86_64	✅	4s	3s	+33%
poly_decompress_d11_native_x86_64	✅	4s	5s	-20%
poly_decompress_d5_native_x86_64	✅	4s	4s	+0%
poly_mulcache_compute_native_aarch64	✅	4s	1s	+300%
keccak_f1600_x1_native_aarch64_v84a	✅	3s	3s	+0%
keccak_f1600_x4_native_aarch64_v84a	✅	3s	2s	+50%
keccak_f1600_x4_native_avx2	✅	3s	1s	+200%
keccakf1600_permute_native	✅	3s	2s	+50%
kem_check_sk	✅	3s	5s	-40%
kem_enc	✅	3s	2s	+50%
kem_enc_derand	✅	3s	3s	+0%
mlk_check_pct	✅	3s	1s	+200%
mlk_ct_cmask_nonzero_u16	✅	3s	4s	-25%
mlk_ct_cmov_zero	✅	3s	4s	-25%
mlk_ct_memcmp	✅	3s	2s	+50%
mlk_enc_getnoise_eta1_eta2	✅	3s	4s	-25%
mlk_keccakf1600_extract_bytes	✅	3s	2s	+50%
mlk_keccakf1600_extract_bytes (big endian)	✅	3s	3s	+0%
mlk_keccakf1600_xor_bytes	✅	3s	2s	+50%
mlk_keccakf1600_xor_bytes (big endian)	✅	3s	2s	+50%
mlk_keccakf1600x4_permute	✅	3s	1s	+200%
mlk_poly_add	✅	3s	2s	+50%
mlk_poly_cbd_eta1	✅	3s	3s	+0%
mlk_poly_cbd_eta2	✅	3s	2s	+50%
mlk_poly_compress_d5	✅	3s	3s	+0%
mlk_poly_decompress_d11	✅	3s	4s	-25%
mlk_poly_decompress_d11_c	✅	3s	4s	-25%
mlk_poly_decompress_d5	✅	3s	3s	+0%
mlk_poly_decompress_d5_c	✅	3s	2s	+50%
mlk_poly_decompress_dv	✅	3s	3s	+0%
mlk_poly_reduce	✅	3s	1s	+200%
mlk_poly_reduce_c	✅	3s	4s	-25%
mlk_poly_tobytes_c	✅	3s	2s	+50%
mlk_poly_tomont_native	✅	3s	1s	+200%
mlk_polyvec_ntt	✅	3s	3s	+0%
mlk_polyvec_permute_bitrev_to_custom	✅	3s	2s	+50%
mlk_polyvec_reduce	✅	3s	2s	+50%
mlk_scalar_compress_d10	✅	3s	1s	+200%
mlk_scalar_compress_d4	✅	3s	1s	+200%
mlk_scalar_decompress_d10	✅	3s	2s	+50%
mlk_scalar_decompress_d11	✅	3s	1s	+200%
mlk_sha3_256	✅	3s	2s	+50%
mlk_sha3_512	✅	3s	1s	+200%
mlk_shake128_absorb_once	✅	3s	2s	+50%
mlk_shake256x4	✅	3s	4s	-25%
mlk_value_barrier_u32	✅	3s	1s	+200%
poly_compress_d10_native_x86_64	✅	3s	1s	+200%
poly_compress_d5_native_x86_64	✅	3s	2s	+50%
poly_decompress_d10_native_x86_64	✅	3s	4s	-25%
poly_frombytes_native_x86_64	✅	3s	4s	-25%
poly_invntt_tomont_native	✅	3s	1s	+200%
poly_tobytes_native_x86_64	✅	3s	3s	+0%
poly_tomont_native_aarch64	✅	3s	3s	+0%
poly_tomont_native_x86_64	✅	3s	2s	+50%
polyvec_basemul_acc_montgomery_cached_k2_native_x86_64	✅	3s	4s	-25%
polyvec_basemul_acc_montgomery_cached_k3_native_aarch64	✅	3s	4s	-25%
polyvec_basemul_acc_montgomery_cached_k3_native_x86_64	✅	3s	2s	+50%
rej_uniform_native_aarch64	✅	3s	3s	+0%
intt_native_aarch64	✅	2s	1s	+100%
intt_native_x86_64	✅	2s	2s	+0%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	2s	1s	+100%
keccakf1600x4_extract_bytes_native	✅	2s	3s	-33%
keccakf1600x4_xor_bytes_native	✅	2s	1s	+100%
kem_keypair	✅	2s	4s	-50%
mlk_barrett_reduce	✅	2s	3s	-33%
mlk_ct_cmask_neg_i16	✅	2s	1s	+100%
mlk_ct_get_optblocker_i32	✅	2s	1s	+100%
mlk_ct_get_optblocker_u32	✅	2s	3s	-33%
mlk_ct_sel_int16	✅	2s	2s	+0%
mlk_ct_sel_uint8	✅	2s	1s	+100%
mlk_keccakf1600x4_extract_bytes	✅	2s	3s	-33%
mlk_keccakf1600x4_extract_bytes_c	✅	2s	4s	-50%
mlk_keccakf1600x4_xor_bytes	✅	2s	2s	+0%
mlk_keccakf1600x4_xor_bytes_c	✅	2s	4s	-50%
mlk_poly_compress_d11_native	✅	2s	1s	+100%
mlk_poly_compress_d4	✅	2s	1s	+100%
mlk_poly_compress_d4_c	✅	2s	5s	-60%
mlk_poly_compress_d4_native	✅	2s	1s	+100%
mlk_poly_compress_dv	✅	2s	2s	+0%
mlk_poly_decompress_d10_native	✅	2s	3s	-33%
mlk_poly_decompress_d4	✅	2s	2s	+0%
mlk_poly_decompress_d4_c	✅	2s	4s	-50%
mlk_poly_getnoise_eta1122_4x	✅	2s	1s	+100%
mlk_poly_getnoise_eta1_4x_native	✅	2s	2s	+0%
mlk_poly_getnoise_eta2	✅	2s	3s	-33%
mlk_poly_mulcache_compute	✅	2s	4s	-50%
mlk_poly_ntt_c	✅	2s	4s	-50%
mlk_poly_sub	✅	2s	2s	+0%
mlk_poly_tobytes	✅	2s	2s	+0%
mlk_poly_tobytes_native	✅	2s	1s	+100%
mlk_poly_tomont	✅	2s	4s	-50%
mlk_polyvec_basemul_acc_montgomery_cached	✅	2s	1s	+100%
mlk_polyvec_compress_du	✅	2s	2s	+0%
mlk_polyvec_frombytes	✅	2s	4s	-50%
mlk_polyvec_invntt_tomont	✅	2s	2s	+0%
mlk_polyvec_tomont	✅	2s	3s	-33%
mlk_rej_uniform	✅	2s	1s	+100%
mlk_scalar_compress_d1	✅	2s	2s	+0%
mlk_scalar_compress_d5	✅	2s	1s	+100%
mlk_scalar_decompress_d4	✅	2s	4s	-50%
mlk_scalar_decompress_d5	✅	2s	2s	+0%
mlk_scalar_signed_to_unsigned_q	✅	2s	4s	-50%
mlk_shake128_squeezeblocks	✅	2s	2s	+0%
mlk_shake128x4_squeezeblocks	✅	2s	1s	+100%
mlk_shake256	✅	2s	2s	+0%
mlk_value_barrier_u8	✅	2s	2s	+0%
poly_mulcache_compute_native_x86_64	✅	2s	6s	-67%
poly_reduce_native_aarch64	✅	2s	2s	+0%
poly_reduce_native_x86_64	✅	2s	3s	-33%
polyvec_basemul_acc_montgomery_cached_k4_native_aarch64	✅	2s	2s	+0%
polyvec_basemul_acc_montgomery_cached_k4_native_x86_64	✅	2s	4s	-50%
rej_uniform_native	✅	2s	3s	-33%
keccak_f1600_x1_native_aarch64	✅	1s	1s	+0%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	1s	2s	-50%
mlk_ct_cmask_nonzero_u8	✅	1s	1s	+0%
mlk_ct_get_optblocker_u8	✅	1s	2s	-50%
mlk_keccakf1600_permute	✅	1s	2s	-50%
mlk_keypair_getnoise_eta1	✅	1s	5s	-80%
mlk_poly_compress_d10	✅	1s	2s	-50%
mlk_poly_compress_d10_c	✅	1s	2s	-50%
mlk_poly_compress_d5_c	✅	1s	4s	-75%
mlk_poly_compress_d5_native	✅	1s	1s	+0%
mlk_poly_decompress_d10_c	✅	1s	3s	-67%
mlk_poly_decompress_d4_native	✅	1s	1s	+0%
mlk_poly_decompress_du	✅	1s	1s	+0%
mlk_poly_frombytes	✅	1s	4s	-75%
mlk_poly_frombytes_c	✅	1s	5s	-80%
mlk_poly_invntt_tomont	✅	1s	2s	-50%
mlk_poly_mulcache_compute_c	✅	1s	4s	-75%
mlk_poly_mulcache_compute_native	✅	1s	3s	-67%
mlk_poly_tomsg	✅	1s	2s	-50%
mlk_polyvec_decompress_du	✅	1s	4s	-75%
mlk_scalar_compress_d11	✅	1s	2s	-50%
mlk_value_barrier_i32	✅	1s	3s	-67%
ntt_native_x86_64	✅	1s	2s	-50%
poly_compress_d11_native_x86_64	✅	1s	4s	-75%
poly_compress_d4_native_x86_64	✅	1s	3s	-67%
poly_decompress_d4_native_x86_64	✅	1s	4s	-75%
poly_tobytes_native_aarch64	✅	1s	4s	-75%
polyvec_basemul_acc_montgomery_cached_k2_native_aarch64	✅	1s	2s	-50%
sys_check_capability	✅	1s	1s	+0%

[oqs-bot]

Mac Mini (M1, 2020) benchmarks

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	12320 cycles	12319 cycles	1.00
ML-KEM-512 encaps	14999 cycles	14997 cycles	1.00
ML-KEM-512 decaps	19554 cycles	19549 cycles	1.00
ML-KEM-768 keypair	21264 cycles	21264 cycles	1
ML-KEM-768 encaps	23873 cycles	23871 cycles	1.00
ML-KEM-768 decaps	30416 cycles	30423 cycles	1.00
ML-KEM-1024 keypair	30328 cycles	30327 cycles	1.00
ML-KEM-1024 encaps	34574 cycles	34573 cycles	1.00
ML-KEM-1024 decaps	44191 cycles	44191 cycles	1

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Intel Xeon 4th gen (c7i)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	12042 cycles	12031 cycles	1.00
ML-KEM-512 encaps	13614 cycles	13792 cycles	0.99
ML-KEM-512 decaps	17818 cycles	17802 cycles	1.00
ML-KEM-768 keypair	21294 cycles	21035 cycles	1.01
ML-KEM-768 encaps	22008 cycles	22107 cycles	1.00
ML-KEM-768 decaps	28034 cycles	28330 cycles	0.99
ML-KEM-1024 keypair	29563 cycles	29964 cycles	0.99
ML-KEM-1024 encaps	31689 cycles	31704 cycles	1.00
ML-KEM-1024 decaps	39367 cycles	39312 cycles	1.00

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Intel Xeon 4th gen (c7i) (no-opt)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	24431 cycles	28199 cycles	0.87
ML-KEM-512 encaps	29933 cycles	36622 cycles	0.82
ML-KEM-512 decaps	37694 cycles	45214 cycles	0.83
ML-KEM-768 keypair	40053 cycles	46304 cycles	0.87
ML-KEM-768 encaps	50467 cycles	55843 cycles	0.90
ML-KEM-768 decaps	59728 cycles	69876 cycles	0.85
ML-KEM-1024 keypair	64223 cycles	70436 cycles	0.91
ML-KEM-1024 encaps	76545 cycles	82480 cycles	0.93
ML-KEM-1024 decaps	86915 cycles	99348 cycles	0.87

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

AMD EPYC 3rd gen (c6a)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	14215 cycles	14239 cycles	1.00
ML-KEM-512 encaps	15990 cycles	15964 cycles	1.00
ML-KEM-512 decaps	21534 cycles	21528 cycles	1.00
ML-KEM-768 keypair	25122 cycles	24710 cycles	1.02
ML-KEM-768 encaps	25669 cycles	25470 cycles	1.01
ML-KEM-768 decaps	33537 cycles	33335 cycles	1.01
ML-KEM-1024 keypair	34894 cycles	37146 cycles	0.94
ML-KEM-1024 encaps	36116 cycles	36786 cycles	0.98
ML-KEM-1024 decaps	47225 cycles	46716 cycles	1.01

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

AMD EPYC 4th gen (c7a)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	12789 cycles	12790 cycles	1.00
ML-KEM-512 encaps	14279 cycles	14273 cycles	1.00
ML-KEM-512 decaps	19139 cycles	19129 cycles	1.00
ML-KEM-768 keypair	22564 cycles	22413 cycles	1.01
ML-KEM-768 encaps	23063 cycles	23072 cycles	1.00
ML-KEM-768 decaps	30067 cycles	30061 cycles	1.00
ML-KEM-1024 keypair	34215 cycles	33027 cycles	1.04
ML-KEM-1024 encaps	33003 cycles	33126 cycles	1.00
ML-KEM-1024 decaps	42408 cycles	42412 cycles	1.00

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'AMD EPYC 4th gen (c7a)'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.03.

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-1024 keypair	34215 cycles	33027 cycles	1.04

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

ppc64le (POWER10) benchmarks

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	52364 cycles	59434 cycles	0.88
ML-KEM-512 encaps	62294 cycles	72134 cycles	0.86
ML-KEM-512 decaps	76687 cycles	92082 cycles	0.83
ML-KEM-768 keypair	89778 cycles	99316 cycles	0.90
ML-KEM-768 encaps	103607 cycles	115930 cycles	0.89
ML-KEM-768 decaps	122375 cycles	141912 cycles	0.86
ML-KEM-1024 keypair	139005 cycles	150195 cycles	0.93
ML-KEM-1024 encaps	155445 cycles	169079 cycles	0.92
ML-KEM-1024 decaps	177695 cycles	200415 cycles	0.89

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Graviton4

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	17690 cycles	17644 cycles	1.00
ML-KEM-512 encaps	20636 cycles	20596 cycles	1.00
ML-KEM-512 decaps	27083 cycles	27048 cycles	1.00
ML-KEM-768 keypair	29976 cycles	29903 cycles	1.00
ML-KEM-768 encaps	32752 cycles	32771 cycles	1.00
ML-KEM-768 decaps	42010 cycles	41962 cycles	1.00
ML-KEM-1024 keypair	43720 cycles	43743 cycles	1.00
ML-KEM-1024 encaps	48775 cycles	48657 cycles	1.00
ML-KEM-1024 decaps	61383 cycles	61383 cycles	1

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Intel Xeon 3rd gen (c6i)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	17600 cycles	17540 cycles	1.00
ML-KEM-512 encaps	19907 cycles	19937 cycles	1.00
ML-KEM-512 decaps	26420 cycles	26445 cycles	1.00
ML-KEM-768 keypair	31206 cycles	31159 cycles	1.00
ML-KEM-768 encaps	31864 cycles	32046 cycles	0.99
ML-KEM-768 decaps	41472 cycles	41536 cycles	1.00
ML-KEM-1024 keypair	43815 cycles	43957 cycles	1.00
ML-KEM-1024 encaps	45880 cycles	45616 cycles	1.01
ML-KEM-1024 decaps	58050 cycles	58219 cycles	1.00

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

AMD EPYC 3rd gen (c6a) (no-opt)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	33786 cycles	40258 cycles	0.84
ML-KEM-512 encaps	41402 cycles	48385 cycles	0.86
ML-KEM-512 decaps	51313 cycles	62592 cycles	0.82
ML-KEM-768 keypair	54176 cycles	63729 cycles	0.85
ML-KEM-768 encaps	65367 cycles	74928 cycles	0.87
ML-KEM-768 decaps	78299 cycles	93722 cycles	0.84
ML-KEM-1024 keypair	84344 cycles	95285 cycles	0.89
ML-KEM-1024 encaps	98201 cycles	109505 cycles	0.90
ML-KEM-1024 decaps	114088 cycles	132331 cycles	0.86

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

AMD EPYC 4th gen (c7a) (no-opt)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	30684 cycles	36614 cycles	0.84
ML-KEM-512 encaps	36951 cycles	43076 cycles	0.86
ML-KEM-512 decaps	45696 cycles	55713 cycles	0.82
ML-KEM-768 keypair	49371 cycles	58664 cycles	0.84
ML-KEM-768 encaps	58602 cycles	67519 cycles	0.87
ML-KEM-768 decaps	69939 cycles	84462 cycles	0.83
ML-KEM-1024 keypair	76257 cycles	88980 cycles	0.86
ML-KEM-1024 encaps	87602 cycles	99212 cycles	0.88
ML-KEM-1024 decaps	101779 cycles	120642 cycles	0.84

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Graviton4 (no-opt)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	34213 cycles	35412 cycles	0.97
ML-KEM-512 encaps	38932 cycles	40111 cycles	0.97
ML-KEM-512 decaps	48961 cycles	51138 cycles	0.96
ML-KEM-768 keypair	54993 cycles	56668 cycles	0.97
ML-KEM-768 encaps	63100 cycles	65152 cycles	0.97
ML-KEM-768 decaps	75917 cycles	79299 cycles	0.96
ML-KEM-1024 keypair	85413 cycles	87866 cycles	0.97
ML-KEM-1024 encaps	94656 cycles	96875 cycles	0.98
ML-KEM-1024 decaps	111743 cycles	115827 cycles	0.96

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Graviton3

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	18675 cycles	18637 cycles	1.00
ML-KEM-512 encaps	21886 cycles	21874 cycles	1.00
ML-KEM-512 decaps	28890 cycles	28863 cycles	1.00
ML-KEM-768 keypair	31630 cycles	31540 cycles	1.00
ML-KEM-768 encaps	34788 cycles	34773 cycles	1.00
ML-KEM-768 decaps	44835 cycles	44778 cycles	1.00
ML-KEM-1024 keypair	46068 cycles	46080 cycles	1.00
ML-KEM-1024 encaps	51494 cycles	51490 cycles	1.00
ML-KEM-1024 decaps	65004 cycles	65028 cycles	1.00

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Intel Xeon 3rd gen (c6i) (no-opt)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	39358 cycles	45700 cycles	0.86
ML-KEM-512 encaps	46964 cycles	54451 cycles	0.86
ML-KEM-512 decaps	57447 cycles	69774 cycles	0.82
ML-KEM-768 keypair	62990 cycles	74220 cycles	0.85
ML-KEM-768 encaps	74954 cycles	86044 cycles	0.87
ML-KEM-768 decaps	88809 cycles	106669 cycles	0.83
ML-KEM-1024 keypair	100417 cycles	112098 cycles	0.90
ML-KEM-1024 encaps	110546 cycles	124743 cycles	0.89
ML-KEM-1024 decaps	128093 cycles	150712 cycles	0.85

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Arm Cortex-A55 (Snapdragon 888) benchmarks

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	59837 cycles	59732 cycles	1.00
ML-KEM-512 encaps	67447 cycles	67418 cycles	1.00
ML-KEM-512 decaps	86186 cycles	86116 cycles	1.00
ML-KEM-768 keypair	97444 cycles	97471 cycles	1.00
ML-KEM-768 encaps	110872 cycles	111029 cycles	1.00
ML-KEM-768 decaps	137941 cycles	137995 cycles	1.00
ML-KEM-1024 keypair	154689 cycles	154794 cycles	1.00
ML-KEM-1024 encaps	171850 cycles	171103 cycles	1.00
ML-KEM-1024 decaps	209734 cycles	208406 cycles	1.01

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Arm Cortex-A76 (Raspberry Pi 5) benchmarks

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	28264 cycles	28220 cycles	1.00
ML-KEM-512 encaps	34156 cycles	34107 cycles	1.00
ML-KEM-512 decaps	44374 cycles	44335 cycles	1.00
ML-KEM-768 keypair	47618 cycles	47615 cycles	1.00
ML-KEM-768 encaps	53933 cycles	53937 cycles	1.00
ML-KEM-768 decaps	68339 cycles	68365 cycles	1.00
ML-KEM-1024 keypair	70245 cycles	70246 cycles	1.00
ML-KEM-1024 encaps	78734 cycles	78726 cycles	1.00
ML-KEM-1024 decaps	98418 cycles	98445 cycles	1.00

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Graviton3 (no-opt)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	37514 cycles	38880 cycles	0.96
ML-KEM-512 encaps	42772 cycles	44586 cycles	0.96
ML-KEM-512 decaps	53687 cycles	56659 cycles	0.95
ML-KEM-768 keypair	60007 cycles	62298 cycles	0.96
ML-KEM-768 encaps	68948 cycles	72317 cycles	0.95
ML-KEM-768 decaps	82493 cycles	87701 cycles	0.94
ML-KEM-1024 keypair	93052 cycles	96154 cycles	0.97
ML-KEM-1024 encaps	103167 cycles	106126 cycles	0.97
ML-KEM-1024 decaps	121291 cycles	126570 cycles	0.96

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Graviton2

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	28238 cycles	28269 cycles	1.00
ML-KEM-512 encaps	34162 cycles	34122 cycles	1.00
ML-KEM-512 decaps	44342 cycles	44378 cycles	1.00
ML-KEM-768 keypair	47642 cycles	47674 cycles	1.00
ML-KEM-768 encaps	53923 cycles	53908 cycles	1.00
ML-KEM-768 decaps	68400 cycles	68363 cycles	1.00
ML-KEM-1024 keypair	70382 cycles	70273 cycles	1.00
ML-KEM-1024 encaps	78782 cycles	78768 cycles	1.00
ML-KEM-1024 decaps	98584 cycles	98473 cycles	1.00

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Graviton2 (no-opt)

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	55880 cycles	59124 cycles	0.95
ML-KEM-512 encaps	65578 cycles	68626 cycles	0.96
ML-KEM-512 decaps	82286 cycles	87341 cycles	0.94
ML-KEM-768 keypair	90360 cycles	95326 cycles	0.95
ML-KEM-768 encaps	104886 cycles	109860 cycles	0.95
ML-KEM-768 decaps	126772 cycles	134332 cycles	0.94
ML-KEM-1024 keypair	140425 cycles	147915 cycles	0.95
ML-KEM-1024 encaps	157358 cycles	163791 cycles	0.96
ML-KEM-1024 decaps	185033 cycles	195404 cycles	0.95

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

Arm Cortex-A72 (Raspberry Pi 4) benchmarks

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	51033 cycles	50865 cycles	1.00
ML-KEM-512 encaps	58913 cycles	58841 cycles	1.00
ML-KEM-512 decaps	74849 cycles	74794 cycles	1.00
ML-KEM-768 keypair	86918 cycles	86024 cycles	1.01
ML-KEM-768 encaps	95132 cycles	94487 cycles	1.01
ML-KEM-768 decaps	118067 cycles	119530 cycles	0.99
ML-KEM-1024 keypair	130121 cycles	130071 cycles	1.00
ML-KEM-1024 encaps	142355 cycles	142892 cycles	1.00
ML-KEM-1024 decaps	174938 cycles	173373 cycles	1.01

This comment was automatically generated by workflow using github-action-benchmark.

[oqs-bot]

SpacemiT K1 8 (Banana Pi F3) benchmarks

Details

Benchmark suite	Current: ee59089	Previous: 070028c	Ratio
ML-KEM-512 keypair	155485 cycles	155504 cycles	1.00
ML-KEM-512 encaps	163394 cycles	163399 cycles	1.00
ML-KEM-512 decaps	206591 cycles	206667 cycles	1.00
ML-KEM-768 keypair	249903 cycles	249893 cycles	1.00
ML-KEM-768 encaps	270434 cycles	270406 cycles	1.00
ML-KEM-768 decaps	332188 cycles	332823 cycles	1.00
ML-KEM-1024 keypair	395688 cycles	395922 cycles	1.00
ML-KEM-1024 encaps	422596 cycles	423034 cycles	1.00
ML-KEM-1024 decaps	506524 cycles	506558 cycles	1.00

This comment was automatically generated by workflow using github-action-benchmark.