Skip to content

Armv8.1-M: Add native Keccak x4 XORBytes and ExtractBytes#972

Open
mkannwischer wants to merge 3 commits intomainfrom
mve-keccakx4-bitinterleaving
Open

Armv8.1-M: Add native Keccak x4 XORBytes and ExtractBytes#972
mkannwischer wants to merge 3 commits intomainfrom
mve-keccakx4-bitinterleaving

Conversation

@mkannwischer
Copy link
Copy Markdown
Contributor

@oqs-bot
Copy link
Copy Markdown
Contributor

oqs-bot commented Feb 19, 2026

CBMC Results (ML-DSA-65)

Full Results (203 proofs)
Proof Status Current Previous Change
**TOTAL** 2170s 2488s -12.8%
polyvecl_pointwise_acc_montgomery_c 480s 622s -23%
polyvec_matrix_expand 124s 137s -9%
rej_uniform_native 123s 136s -10%
sign_verify_internal 115s 125s -8%
poly_pointwise_montgomery_c 93s 109s -15%
mld_invntt_layer 92s 104s -12%
mld_ct_memcmp 74s 86s -14%
mld_attempt_signature_generation 48s 54s -11%
mld_ntt_layer 44s 48s -8%
sign_signature_internal 32s 33s -3%
sign_keypair_internal 29s 31s -6%
fqmul 26s 33s -21%
polyvec_matrix_expand_serial 25s 30s -17%
keccakf1600x4_permute_native 22s 22s +0%
polyvec_matrix_pointwise_montgomery 22s 23s -4%
polyvec_matrix_pointwise_montgomery_yvec 21s 21s +0%
sign_pk_from_sk 20s 20s +0%
rej_uniform_c 19s 19s +0%
mld_ntt_butterfly_block 16s 18s -11%
rej_uniform 16s 20s -20%
polyveck_decompose 15s 17s -12%
polyveck_power2round 15s 18s -17%
poly_chknorm_c 14s 17s -18%
polyt0_unpack 14s 16s -12%
poly_uniform_eta_4x 13s 14s -7%
poly_uniform_4x 12s 13s -8%
poly_add 11s 12s -8%
polyveck_add 10s 12s -17%
polyveck_use_hint 10s 10s +0%
keccak_absorb 9s 9s +0%
mld_check_pct 9s 10s -10%
poly_invntt_tomont_c 9s 10s -10%
poly_power2round 9s 10s -10%
polyveck_pointwise_poly_montgomery 9s 8s +12%
keccak_absorb_once_x4 8s 11s -27%
keccakf1600_permute 7s 8s -12%
mld_compute_pack_z 7s 9s -22%
polyveck_shiftl 7s 9s -22%
polyveck_sub 7s 9s -22%
polyvecl_ntt 7s 9s -22%
intt_native_x86_64 6s 3s +100%
keccakf1600_permute_native 6s 6s +0%
pack_pk 6s 2s +200%
pointwise_acc_native_x86_64 6s 7s -14%
polyveck_ntt 6s 6s +0%
polyveck_unpack_t0 6s 4s +50%
rej_eta_native 6s 3s +100%
sign_signature_extmu 6s 3s +100%
keccak_squeezeblocks_x4 5s 6s -17%
mld_ct_abs_i32 5s 1s +400%
mld_h 5s 2s +150%
mld_prepare_domain_separation_prefix 5s 6s -17%
mld_sample_s1_s2_serial 5s 7s -29%
pointwise_acc_native_aarch64 5s 7s -29%
poly_challenge 5s 3s +67%
poly_pointwise_montgomery_native 5s 4s +25%
poly_uniform 5s 4s +25%
poly_uniform_gamma1 5s 3s +67%
polyt0_pack 5s 5s +0%
polyvec_matrix_pointwise_montgomery_row 5s 5s +0%
polyveck_chknorm 5s 5s +0%
polyveck_invntt_tomont 5s 8s -38%
polyveck_reduce 5s 7s -29%
polyvecl_chknorm 5s 9s -44%
polyvecl_unpack_eta 5s 2s +150%
polyw1_pack 5s 3s +67%
polyz_unpack_c 5s 7s -29%
sign 5s 9s -44%
sign_open 5s 7s -29%
sign_verify 5s 3s +67%
unpack_hints 5s 7s -29%
decompose 4s 6s -33%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid 4s 3s +33%
keccak_squeeze 4s 2s +100%
keccakf1600x4_xor_bytes_native 4s - new
mld_polymat_expand_entry 4s 2s +100%
pack_sig_z 4s 5s -20%
poly_caddq_c 4s 4s +0%
poly_caddq_native_aarch64 4s 2s +100%
poly_decompose 4s 6s -33%
poly_invntt_tomont_native 4s 3s +33%
poly_ntt_c 4s 3s +33%
poly_reduce 4s 3s +33%
poly_shiftl 4s 3s +33%
poly_use_hint_native 4s 2s +100%
polyt1_pack 4s 3s +33%
polyveck_caddq 4s 7s -43%
polyveck_unpack_eta 4s 4s +0%
polyz_unpack 4s 2s +100%
polyz_unpack_17_native_aarch64 4s 3s +33%
polyz_unpack_19_native_aarch64 4s 2s +100%
polyz_unpack_native 4s 1s +300%
power2round 4s 2s +100%
rej_eta_c 4s 4s +0%
shake128_absorb 4s 2s +100%
shake256x4_squeezeblocks 4s 8s -50%
sign_verify_extmu 4s 6s -33%
sign_verify_pre_hash_internal 4s 7s -43%
unpack_sig 4s 4s +0%
unpack_sk 4s 3s +33%
use_hint 4s 3s +33%
yvec_get_poly 4s 3s +33%
yvec_init 4s 2s +100%
keccak_f1600_x1_native_aarch64 3s 3s +0%
keccak_f1600_x4_native_aarch64_v84a 3s 3s +0%
keccak_finalize 3s 2s +50%
keccakf1600x4_extract_bytes_native 3s - new
keccakf1600x4_permute 3s 1s +200%
keccakf1600x4_xor_bytes 3s 3s +0%
mld_ct_get_optblocker_u32 3s 3s +0%
mld_ct_get_optblocker_u8 3s 2s +50%
mld_sample_s1_s2 3s 4s -25%
mld_value_barrier_i64 3s 2s +50%
mld_value_barrier_u32 3s 1s +200%
mld_value_barrier_u8 3s 1s +200%
montgomery_reduce 3s 2s +50%
ntt_native_aarch64 3s 5s -40%
nttunpack_native_x86_64 3s 4s -25%
pack_sig_c 3s 4s -25%
pack_sig_h_poly 3s 3s +0%
pack_sk_s1 3s 2s +50%
pointwise_native_aarch64 3s 4s -25%
pointwise_native_x86_64 3s 2s +50%
poly_caddq 3s 3s +0%
poly_chknorm_native 3s 2s +50%
poly_chknorm_native_aarch64 3s 4s -25%
poly_decompose_c 3s 4s -25%
poly_permute_bitrev_to_custom_optional 3s 3s +0%
poly_uniform_gamma1_4x 3s 4s -25%
poly_use_hint_c 3s 5s -40%
poly_use_hint_native_aarch64 3s 3s +0%
polyt1_unpack 3s 5s -40%
polyveck_pack_eta 3s 5s -40%
polyvecl_pack_eta 3s 5s -40%
polyvecl_pointwise_acc_montgomery 3s 6s -50%
polyvecl_pointwise_acc_montgomery_native 3s 6s -50%
polyz_pack 3s 3s +0%
reduce32 3s 4s -25%
shake128_release 3s 2s +50%
shake128_squeeze 3s 2s +50%
shake128x4_absorb_once 3s 4s -25%
shake128x4_squeezeblocks 3s 4s -25%
shake256_init 3s 1s +200%
shake256x4_absorb_once 3s 2s +50%
sign_signature_pre_hash_internal 3s 4s -25%
sign_signature_pre_hash_shake256 3s 3s +0%
sk_s2hat_get_poly 3s 3s +0%
sk_t0hat_get_poly 3s 3s +0%
unpack_pk 3s 4s -25%
unpack_sk_s1hat 3s 3s +0%
unpack_sk_t0hat 3s 2s +50%
caddq 2s 3s -33%
fqscale 2s 2s +0%
keccak_f1600_x1_native_aarch64_v84a 2s 2s +0%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid 2s 1s +100%
keccak_f1600_x4_native_avx2 2s 2s +0%
keccak_init 2s 3s -33%
keccakf1600_extract_bytes (big endian) 2s 2s +0%
keccakf1600_xor_bytes 2s 3s -33%
keccakf1600_xor_bytes (big endian) 2s 2s +0%
keccakf1600x4_extract_bytes 2s 2s +0%
make_hint 2s 4s -50%
mld_ct_cmask_neg_i32 2s 2s +0%
mld_ct_cmask_nonzero_u8 2s 2s +0%
mld_ct_sel_int32 2s 2s +0%
mld_keccakf1600_extract_bytes 2s 4s -50%
pack_sk_rho_key_tr_s2_t0 2s 2s +0%
poly_decompose_native 2s 4s -50%
poly_make_hint 2s 4s -50%
poly_ntt 2s 4s -50%
poly_ntt_native 2s 1s +100%
poly_pointwise_montgomery 2s 4s -50%
poly_sub 2s 3s -33%
poly_uniform_eta 2s 3s -33%
poly_use_hint 2s 4s -50%
polyeta_pack 2s 4s -50%
polyveck_pack_t0 2s 4s -50%
polyvecl_uniform_gamma1 2s 4s -50%
polyvecl_unpack_z 2s 3s -33%
rej_eta 2s 3s -33%
shake128_finalize 2s 3s -33%
shake256 2s 3s -33%
shake256_absorb 2s 3s -33%
shake256_finalize 2s 2s +0%
shake256_release 2s 2s +0%
shake256_squeeze 2s 3s -33%
sign_keypair 2s 3s -33%
sign_signature 2s 3s -33%
sign_verify_pre_hash_shake256 2s 5s -60%
sk_s1hat_get_poly 2s 4s -50%
unpack_sk_s2hat 2s 3s -33%
mld_ct_cmask_nonzero_u32 1s 3s -67%
mld_ct_get_optblocker_i64 1s 3s -67%
ntt_native_x86_64 1s 3s -67%
poly_caddq_native 1s 3s -67%
poly_chknorm 1s 2s -50%
poly_invntt_tomont 1s 2s -50%
poly_permute_bitrev_to_custom_optional_native 1s 5s -80%
polyeta_unpack 1s 3s -67%
polyveck_pack_w1 1s 4s -75%
polyvecl_uniform_gamma1_serial 1s 3s -67%
shake128_init 1s 1s +0%
sys_check_capability 1s 3s -67%

@oqs-bot
Copy link
Copy Markdown
Contributor

oqs-bot commented Feb 19, 2026

CBMC Results (ML-DSA-44)

Full Results (203 proofs)
Proof Status Current Previous Change
**TOTAL** 1779s 1843s -3.5%
sign_verify_internal 205s 211s -3%
rej_uniform_native 126s 130s -3%
polyvecl_pointwise_acc_montgomery_c 122s 138s -12%
poly_pointwise_montgomery_c 99s 107s -7%
mld_invntt_layer 96s 100s -4%
mld_ct_memcmp 73s 85s -14%
mld_attempt_signature_generation 50s 48s +4%
mld_ntt_layer 44s 45s -2%
fqmul 28s 31s -10%
polyvec_matrix_expand 26s 27s -4%
keccakf1600x4_permute_native 23s 22s +5%
polyvec_matrix_pointwise_montgomery 23s 25s -8%
sign_keypair_internal 23s 22s +5%
sign_pk_from_sk 20s 19s +5%
poly_chknorm_c 17s 17s +0%
polyt0_unpack 17s 14s +21%
rej_uniform_c 17s 18s -6%
sign_signature_internal 17s 14s +21%
mld_ntt_butterfly_block 16s 16s +0%
polyz_unpack_c 16s 13s +23%
rej_uniform 16s 17s -6%
polyveck_chknorm 14s 20s -30%
poly_uniform_eta_4x 12s 15s -20%
polyeta_unpack 12s 12s +0%
polyvec_matrix_pointwise_montgomery_yvec 12s 12s +0%
poly_add 11s 11s +0%
poly_uniform_4x 11s 12s -8%
mld_check_pct 10s 9s +11%
mld_compute_pack_z 10s 6s +67%
keccak_absorb 9s 6s +50%
keccak_absorb_once_x4 9s 8s +12%
poly_power2round 9s 9s +0%
keccakf1600_permute_native 8s 6s +33%
polyveck_use_hint 8s 8s +0%
mld_sample_s1_s2_serial 7s 3s +133%
poly_invntt_tomont_c 7s 9s -22%
polyvec_matrix_expand_serial 7s 5s +40%
polyveck_add 7s 5s +40%
polyveck_invntt_tomont 7s 6s +17%
ntt_native_x86_64 6s 3s +100%
poly_uniform_eta 6s 2s +200%
polyveck_pointwise_poly_montgomery 6s 7s -14%
polyveck_power2round 6s 4s +50%
polyveck_reduce 6s 5s +20%
polyvecl_chknorm 6s 5s +20%
keccakf1600_permute 5s 7s -29%
mld_polymat_expand_entry 5s 4s +25%
mld_sample_s1_s2 5s 5s +0%
pack_sk_rho_key_tr_s2_t0 5s 3s +67%
pointwise_acc_native_aarch64 5s 4s +25%
pointwise_acc_native_x86_64 5s 5s +0%
poly_challenge 5s 3s +67%
poly_decompose_c 5s 4s +25%
poly_decompose_native 5s 1s +400%
poly_ntt 5s 3s +67%
poly_uniform_gamma1_4x 5s 4s +25%
polyt0_pack 5s 3s +67%
polyveck_caddq 5s 3s +67%
polyveck_decompose 5s 6s -17%
polyveck_unpack_t0 5s 4s +25%
polyvecl_unpack_z 5s 4s +25%
polyz_unpack_19_native_aarch64 5s 3s +67%
shake256_squeeze 5s 2s +150%
sign 5s 6s -17%
sign_keypair 5s 3s +67%
sign_signature 5s 6s -17%
sign_verify_extmu 5s 4s +25%
keccak_f1600_x1_native_aarch64 4s 2s +100%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid 4s 2s +100%
keccak_f1600_x4_native_avx2 4s 3s +33%
keccak_squeezeblocks_x4 4s 3s +33%
keccakf1600x4_xor_bytes_native 4s - new
mld_ct_cmask_neg_i32 4s 2s +100%
mld_prepare_domain_separation_prefix 4s 2s +100%
pack_sig_h_poly 4s 2s +100%
pointwise_native_aarch64 4s 2s +100%
poly_caddq 4s 3s +33%
poly_chknorm 4s 4s +0%
poly_decompose 4s 2s +100%
poly_shiftl 4s 5s -20%
poly_uniform 4s 4s +0%
polyt1_pack 4s 9s -56%
polyt1_unpack 4s 2s +100%
polyveck_pack_t0 4s 5s -20%
polyvecl_ntt 4s 5s -20%
polyz_unpack_native 4s 2s +100%
rej_eta_c 4s 3s +33%
rej_eta_native 4s 2s +100%
shake128_finalize 4s 4s +0%
shake128_init 4s 3s +33%
shake128_squeeze 4s 1s +300%
shake128x4_absorb_once 4s 3s +33%
shake256 4s 2s +100%
sign_signature_pre_hash_internal 4s 5s -20%
sign_verify_pre_hash_internal 4s 6s -33%
sign_verify_pre_hash_shake256 4s 3s +33%
unpack_hints 4s 5s -20%
unpack_pk 4s 5s -20%
unpack_sig 4s 4s +0%
unpack_sk 4s 4s +0%
yvec_get_poly 4s 3s +33%
decompose 3s 2s +50%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid 3s 2s +50%
keccak_finalize 3s 3s +0%
keccak_init 3s 1s +200%
keccak_squeeze 3s 4s -25%
keccakf1600x4_extract_bytes 3s 3s +0%
keccakf1600x4_xor_bytes 3s 6s -50%
mld_ct_abs_i32 3s 3s +0%
mld_ct_cmask_nonzero_u32 3s 3s +0%
mld_ct_cmask_nonzero_u8 3s 3s +0%
mld_ct_get_optblocker_i64 3s 2s +50%
mld_ct_get_optblocker_u32 3s 4s -25%
mld_ct_get_optblocker_u8 3s 1s +200%
mld_h 3s 5s -40%
mld_keccakf1600_extract_bytes 3s 3s +0%
mld_value_barrier_i64 3s 4s -25%
ntt_native_aarch64 3s 4s -25%
nttunpack_native_x86_64 3s 5s -40%
pack_pk 3s 4s -25%
pack_sig_z 3s 4s -25%
pack_sk_s1 3s 4s -25%
poly_caddq_c 3s 4s -25%
poly_chknorm_native 3s 3s +0%
poly_chknorm_native_aarch64 3s 3s +0%
poly_invntt_tomont 3s 4s -25%
poly_pointwise_montgomery 3s 2s +50%
poly_pointwise_montgomery_native 3s 3s +0%
poly_sub 3s 2s +50%
poly_uniform_gamma1 3s 4s -25%
poly_use_hint 3s 2s +50%
poly_use_hint_native 3s 3s +0%
poly_use_hint_native_aarch64 3s 2s +50%
polyeta_pack 3s 4s -25%
polyvec_matrix_pointwise_montgomery_row 3s 3s +0%
polyveck_ntt 3s 4s -25%
polyveck_sub 3s 5s -40%
polyveck_unpack_eta 3s 5s -40%
polyvecl_pointwise_acc_montgomery 3s 2s +50%
polyvecl_pointwise_acc_montgomery_native 3s 3s +0%
polyvecl_uniform_gamma1_serial 3s 2s +50%
polyvecl_unpack_eta 3s 3s +0%
polyz_pack 3s 3s +0%
polyz_unpack_17_native_aarch64 3s 4s -25%
power2round 3s 7s -57%
shake128_absorb 3s 3s +0%
shake128_release 3s 2s +50%
shake128x4_squeezeblocks 3s 1s +200%
shake256_finalize 3s 3s +0%
shake256x4_squeezeblocks 3s 3s +0%
sign_open 3s 6s -50%
sign_signature_extmu 3s 2s +50%
sign_signature_pre_hash_shake256 3s 6s -50%
sign_verify 3s 6s -50%
sk_s2hat_get_poly 3s 3s +0%
sk_t0hat_get_poly 3s 4s -25%
unpack_sk_t0hat 3s 5s -40%
caddq 2s 3s -33%
intt_native_x86_64 2s 4s -50%
keccak_f1600_x1_native_aarch64_v84a 2s 3s -33%
keccakf1600_extract_bytes (big endian) 2s 2s +0%
keccakf1600_xor_bytes 2s 2s +0%
keccakf1600x4_extract_bytes_native 2s - new
keccakf1600x4_permute 2s 3s -33%
make_hint 2s 3s -33%
mld_ct_sel_int32 2s 4s -50%
mld_value_barrier_u32 2s 2s +0%
mld_value_barrier_u8 2s 3s -33%
montgomery_reduce 2s 6s -67%
pack_sig_c 2s 3s -33%
pointwise_native_x86_64 2s 3s -33%
poly_caddq_native_aarch64 2s 2s +0%
poly_invntt_tomont_native 2s 2s +0%
poly_make_hint 2s 2s +0%
poly_ntt_c 2s 3s -33%
poly_ntt_native 2s 4s -50%
poly_permute_bitrev_to_custom_optional 2s 3s -33%
poly_permute_bitrev_to_custom_optional_native 2s 2s +0%
poly_reduce 2s 3s -33%
polyveck_pack_eta 2s 4s -50%
polyveck_pack_w1 2s 4s -50%
polyveck_shiftl 2s 3s -33%
polyvecl_pack_eta 2s 7s -71%
polyvecl_uniform_gamma1 2s 3s -33%
polyw1_pack 2s 3s -33%
polyz_unpack 2s 2s +0%
reduce32 2s 2s +0%
rej_eta 2s 2s +0%
shake256_absorb 2s 3s -33%
shake256_release 2s 2s +0%
shake256x4_absorb_once 2s 4s -50%
sk_s1hat_get_poly 2s 2s +0%
sys_check_capability 2s 6s -67%
unpack_sk_s1hat 2s 3s -33%
use_hint 2s 3s -33%
yvec_init 2s 5s -60%
fqscale 1s 5s -80%
keccak_f1600_x4_native_aarch64_v84a 1s 3s -67%
keccakf1600_xor_bytes (big endian) 1s 3s -67%
poly_caddq_native 1s 4s -75%
poly_use_hint_c 1s 4s -75%
shake256_init 1s 4s -75%
unpack_sk_s2hat 1s 3s -67%

@oqs-bot
Copy link
Copy Markdown
Contributor

oqs-bot commented Feb 19, 2026

CBMC Results (ML-DSA-87)

Full Results (203 proofs)
Proof Status Current Previous Change
**TOTAL** 2521s 2375s +6.1%
polyvecl_pointwise_acc_montgomery_c 384s 345s +11%
sign_verify_internal 270s 253s +7%
polyvec_matrix_expand 176s 169s +4%
rej_uniform_native 127s 126s +1%
poly_pointwise_montgomery_c 102s 92s +11%
mld_invntt_layer 99s 90s +10%
mld_ct_memcmp 84s 73s +15%
mld_attempt_signature_generation 63s 62s +2%
polyvec_matrix_expand_serial 58s 56s +4%
sign_keypair_internal 53s 52s +2%
mld_ntt_layer 45s 45s +0%
polyveck_power2round 36s 34s +6%
sign_signature_internal 35s 36s -3%
sign_pk_from_sk 34s 34s +0%
fqmul 29s 27s +7%
polyvec_matrix_pointwise_montgomery 29s 29s +0%
keccakf1600x4_permute_native 24s 23s +4%
rej_uniform_c 21s 17s +24%
polyvec_matrix_pointwise_montgomery_yvec 19s 19s +0%
mld_ntt_butterfly_block 17s 16s +6%
poly_chknorm_c 16s 16s +0%
poly_uniform_eta_4x 16s 13s +23%
rej_uniform 16s 19s -16%
polyt0_unpack 15s 17s -12%
polyeta_unpack 13s 12s +8%
poly_power2round 12s 9s +33%
poly_uniform_4x 12s 12s +0%
polyveck_decompose 12s 12s +0%
polyveck_ntt 12s 10s +20%
mld_check_pct 11s 9s +22%
polyveck_add 11s 11s +0%
keccak_absorb_once_x4 10s 9s +11%
poly_add 10s 12s -17%
polyveck_use_hint 10s 11s -9%
mld_sample_s1_s2_serial 9s 8s +12%
polyveck_pointwise_poly_montgomery 9s 9s +0%
polyveck_sub 9s 8s +12%
mld_compute_pack_z 8s 8s +0%
pointwise_acc_native_x86_64 8s 8s +0%
poly_decompose_c 8s 5s +60%
poly_invntt_tomont_c 8s 8s +0%
polyveck_shiftl 8s 8s +0%
polyvecl_ntt 8s 5s +60%
polyvecl_unpack_eta 8s 3s +167%
sign_signature 8s 2s +300%
keccakf1600_permute 7s 7s +0%
pointwise_acc_native_aarch64 7s 5s +40%
polyveck_invntt_tomont 7s 5s +40%
sign 7s 7s +0%
sign_signature_pre_hash_shake256 7s 8s -12%
keccak_absorb 6s 7s -14%
keccakf1600_permute_native 6s 7s -14%
mld_h 6s 4s +50%
mld_sample_s1_s2 6s 6s +0%
ntt_native_x86_64 6s 2s +200%
pack_pk 6s 6s +0%
poly_use_hint_native_aarch64 6s 2s +200%
polyeta_pack 6s 3s +100%
polyveck_unpack_t0 6s 4s +50%
polyz_unpack_c 6s 7s -14%
sign_verify_extmu 6s 4s +50%
unpack_pk 6s 3s +100%
mld_ct_cmask_nonzero_u32 5s 3s +67%
poly_challenge 5s 3s +67%
poly_decompose 5s 5s +0%
poly_permute_bitrev_to_custom_optional 5s 2s +150%
poly_uniform_eta 5s 4s +25%
poly_uniform_gamma1_4x 5s 5s +0%
poly_use_hint 5s 2s +150%
poly_use_hint_c 5s 6s -17%
polyveck_caddq 5s 4s +25%
polyveck_chknorm 5s 7s -29%
polyveck_reduce 5s 7s -29%
polyvecl_chknorm 5s 4s +25%
polyvecl_pack_eta 5s 2s +150%
polyvecl_uniform_gamma1_serial 5s 3s +67%
shake128_squeeze 5s 4s +25%
shake128x4_squeezeblocks 5s 2s +150%
sign_keypair 5s 4s +25%
sign_open 5s 4s +25%
sign_verify_pre_hash_shake256 5s 2s +150%
unpack_sk_t0hat 5s 4s +25%
decompose 4s 3s +33%
intt_native_x86_64 4s 2s +100%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid 4s 3s +33%
keccak_squeezeblocks_x4 4s 4s +0%
keccakf1600_xor_bytes 4s 1s +300%
keccakf1600x4_extract_bytes 4s 2s +100%
keccakf1600x4_xor_bytes_native 4s - new
make_hint 4s 4s +0%
mld_ct_abs_i32 4s 3s +33%
mld_ct_cmask_nonzero_u8 4s 3s +33%
mld_keccakf1600_extract_bytes 4s 1s +300%
mld_polymat_expand_entry 4s 5s -20%
nttunpack_native_x86_64 4s 3s +33%
pack_sig_z 4s 4s +0%
pointwise_native_aarch64 4s 2s +100%
poly_decompose_native 4s 2s +100%
poly_pointwise_montgomery_native 4s 3s +33%
poly_reduce 4s 2s +100%
polyt1_pack 4s 3s +33%
polyveck_unpack_eta 4s 3s +33%
polyvecl_pointwise_acc_montgomery_native 4s 3s +33%
polyvecl_unpack_z 4s 2s +100%
polyw1_pack 4s 2s +100%
polyz_unpack_17_native_aarch64 4s 4s +0%
polyz_unpack_19_native_aarch64 4s 3s +33%
reduce32 4s 4s +0%
shake256_absorb 4s 2s +100%
shake256x4_squeezeblocks 4s 3s +33%
sign_signature_extmu 4s 6s -33%
sk_s1hat_get_poly 4s 5s -20%
unpack_sk 4s 6s -33%
unpack_sk_s2hat 4s 3s +33%
use_hint 4s 4s +0%
caddq 3s 6s -50%
fqscale 3s 5s -40%
keccak_f1600_x4_native_aarch64_v84a 3s 2s +50%
keccak_init 3s 3s +0%
keccakf1600_extract_bytes (big endian) 3s 4s -25%
keccakf1600x4_permute 3s 2s +50%
mld_ct_cmask_neg_i32 3s 1s +200%
montgomery_reduce 3s 2s +50%
ntt_native_aarch64 3s 2s +50%
pack_sig_c 3s 3s +0%
pack_sk_s1 3s 5s -40%
poly_caddq 3s 2s +50%
poly_caddq_c 3s 4s -25%
poly_chknorm_native 3s 2s +50%
poly_invntt_tomont 3s 4s -25%
poly_invntt_tomont_native 3s 4s -25%
poly_ntt_c 3s 2s +50%
poly_ntt_native 3s 3s +0%
poly_shiftl 3s 3s +0%
poly_sub 3s 2s +50%
poly_uniform_gamma1 3s 6s -50%
poly_use_hint_native 3s 4s -25%
polyt0_pack 3s 1s +200%
polyt1_unpack 3s 2s +50%
polyveck_pack_eta 3s 2s +50%
polyveck_pack_t0 3s 3s +0%
polyveck_pack_w1 3s 4s -25%
polyvecl_pointwise_acc_montgomery 3s 3s +0%
polyz_pack 3s 4s -25%
polyz_unpack_native 3s 2s +50%
rej_eta 3s 3s +0%
rej_eta_c 3s 4s -25%
rej_eta_native 3s 6s -50%
shake128_absorb 3s 2s +50%
shake128_finalize 3s 2s +50%
shake128_init 3s 3s +0%
shake256 3s 4s -25%
shake256_finalize 3s 2s +50%
shake256_squeeze 3s 1s +200%
sign_signature_pre_hash_internal 3s 4s -25%
sign_verify 3s 5s -40%
sign_verify_pre_hash_internal 3s 5s -40%
sk_s2hat_get_poly 3s 3s +0%
sys_check_capability 3s 5s -40%
unpack_hints 3s 7s -57%
keccak_f1600_x1_native_aarch64_v84a 2s 3s -33%
keccak_squeeze 2s 3s -33%
keccakf1600_xor_bytes (big endian) 2s 2s +0%
keccakf1600x4_extract_bytes_native 2s - new
keccakf1600x4_xor_bytes 2s 2s +0%
mld_ct_get_optblocker_i64 2s 5s -60%
mld_ct_get_optblocker_u32 2s 2s +0%
mld_prepare_domain_separation_prefix 2s 3s -33%
mld_value_barrier_i64 2s 2s +0%
mld_value_barrier_u32 2s 1s +100%
mld_value_barrier_u8 2s 1s +100%
pack_sig_h_poly 2s 3s -33%
pack_sk_rho_key_tr_s2_t0 2s 5s -60%
pointwise_native_x86_64 2s 5s -60%
poly_caddq_native 2s 4s -50%
poly_caddq_native_aarch64 2s 4s -50%
poly_chknorm 2s 2s +0%
poly_chknorm_native_aarch64 2s 3s -33%
poly_make_hint 2s 2s +0%
poly_ntt 2s 3s -33%
poly_pointwise_montgomery 2s 3s -33%
poly_uniform 2s 4s -50%
polyvec_matrix_pointwise_montgomery_row 2s 1s +100%
polyvecl_uniform_gamma1 2s 3s -33%
polyz_unpack 2s 3s -33%
shake128_release 2s 2s +0%
shake128x4_absorb_once 2s 3s -33%
sk_t0hat_get_poly 2s 4s -50%
unpack_sig 2s 4s -50%
unpack_sk_s1hat 2s 5s -60%
yvec_get_poly 2s 2s +0%
yvec_init 2s 4s -50%
keccak_f1600_x1_native_aarch64 1s 2s -50%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid 1s 3s -67%
keccak_f1600_x4_native_avx2 1s 3s -67%
keccak_finalize 1s 2s -50%
mld_ct_get_optblocker_u8 1s 2s -50%
mld_ct_sel_int32 1s 3s -67%
poly_permute_bitrev_to_custom_optional_native 1s 2s -50%
power2round 1s 2s -50%
shake256_init 1s 3s -67%
shake256_release 1s 4s -75%
shake256x4_absorb_once 1s 2s -50%

@mkannwischer mkannwischer force-pushed the mve-keccakx4-bitinterleaving branch from 32b5350 to 3823f1f Compare February 24, 2026 08:44
@mkannwischer mkannwischer marked this pull request as ready for review February 24, 2026 08:45
@mkannwischer mkannwischer requested a review from a team as a code owner February 24, 2026 08:45
Replace test_keccakf1600x4_permute with test_keccakf1600x4_xor_permute_extract
that tests the full x4 Keccak flow (xor_bytes, permute, extract_bytes) against
the x1 C reference implementation.

Testing through the public interface rather than comparing internal state
directly allows verifying backends that use custom state representations
(e.g., bit-interleaved) without requiring state conversion functions.

The test uses random offsets and lengths for both xor_bytes and extract_bytes,
and verifies each of the 4 lanes independently against the x1 reference.

Also reduce functional test iterations for M55 baremetal platform.

Signed-off-by: Matthias J. Kannwischer <matthias@kannwischer.eu>
Extend the FIPS202 native backend API to support implementing XORBytes and
ExtractBytes steps in native code.

This is essential for backends using custom state representations (e.g.,
bit-interleaved state), where these functions handle conversion to/from
the internal format on-the-fly. In such cases, they also account for a
significant amount of processing time.

New flags:
- MLD_USE_FIPS202_X4_XOR_BYTES_NATIVE: Backend provides native XOR bytes
- MLD_USE_FIPS202_X4_EXTRACT_BYTES_NATIVE: Backend provides native extract bytes

When set, backends provide native implementations for:
- mld_keccakf1600_xor_bytes_x4_native: XOR input data into state
- mld_keccakf1600_extract_bytes_x4_native: Extract output from state

Signed-off-by: Matthias J. Kannwischer <matthias@kannwischer.eu>
@mkannwischer mkannwischer force-pushed the mve-keccakx4-bitinterleaving branch from 3823f1f to af01704 Compare May 3, 2026 03:59
Add native MVE implementations of XORBytes and ExtractBytes that perform
bit-interleaving/deinterleaving on-the-fly, enabling use of a bit-interleaved
state representation without temporary conversions in the permutation.

This improves performance by:
- Reducing the number of bit-interleaving operations
- Accelerating bit-interleaving using MVE vector instructions

The backend uses bit-interleaved state representation where each 64-bit
lane is split into even and odd 32-bit halves for efficient 32-bit
MVE processing.

Co-Authored-By: Brendan Moran <brendan.moran@arm.com>
Signed-off-by: Matthias J. Kannwischer <matthias@kannwischer.eu>
@mkannwischer mkannwischer force-pushed the mve-keccakx4-bitinterleaving branch from af01704 to 473cbfb Compare May 3, 2026 04:07
@oqs-bot
Copy link
Copy Markdown
Contributor

oqs-bot commented May 3, 2026

CBMC Results (ML-DSA-65, REDUCE-RAM)

Full Results (203 proofs)
Proof Status Current Previous Change
**TOTAL** 1618s 1737s -6.9%
poly_pointwise_montgomery_c 169s 201s -16%
sign_verify_internal 116s 125s -7%
mld_invntt_layer 106s 116s -9%
rej_uniform_native 103s 116s -11%
mld_ct_memcmp 73s 81s -10%
polyvec_matrix_pointwise_montgomery_yvec 73s 80s -9%
mld_ntt_layer 41s 45s -9%
fqmul 28s 30s -7%
mld_attempt_signature_generation 26s 30s -13%
keccakf1600x4_permute_native 21s 23s -9%
rej_uniform 19s 20s -5%
sign_keypair_internal 19s 22s -14%
polyveck_decompose 17s 18s -6%
rej_uniform_c 17s 19s -11%
mld_ntt_butterfly_block 16s 17s -6%
polyveck_power2round 16s 17s -6%
poly_chknorm_c 15s 14s +7%
poly_uniform_eta_4x 13s 11s +18%
polyvec_matrix_pointwise_montgomery_row 13s 15s -13%
sign_pk_from_sk 13s 14s -7%
polyt0_unpack 12s 10s +20%
polyveck_add 11s 12s -8%
mld_check_pct 10s 12s -17%
poly_add 10s 11s -9%
poly_power2round 10s 7s +43%
polyveck_chknorm 10s 7s +43%
keccak_absorb_once_x4 9s 8s +12%
polyveck_caddq 9s 8s +12%
keccakf1600_permute_native 8s 9s -11%
polyveck_ntt 8s 6s +33%
polyveck_pointwise_poly_montgomery 8s 8s +0%
polyveck_use_hint 8s 12s -33%
polyvecl_chknorm 8s 9s -11%
polyvecl_ntt 8s 9s -11%
sign 8s 9s -11%
keccakf1600_permute 7s 7s +0%
mld_compute_pack_z 7s 6s +17%
poly_caddq_c 7s 9s -22%
poly_chknorm_native_aarch64 7s 4s +75%
poly_invntt_tomont_c 7s 8s -12%
polyveck_shiftl 7s 5s +40%
polyveck_sub 7s 6s +17%
intt_native_x86_64 6s 5s +20%
keccak_absorb 6s 9s -33%
mld_h 6s 5s +20%
mld_sample_s1_s2 6s 5s +20%
pointwise_acc_native_aarch64 6s 7s -14%
polyveck_invntt_tomont 6s 8s -25%
sign_verify_pre_hash_shake256 6s 4s +50%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid 5s 3s +67%
keccak_squeeze 5s 4s +25%
keccak_squeezeblocks_x4 5s 4s +25%
mld_sample_s1_s2_serial 5s 3s +67%
nttunpack_native_x86_64 5s 5s +0%
poly_challenge 5s 6s -17%
poly_uniform 5s 6s -17%
polyt0_pack 5s 5s +0%
polyveck_reduce 5s 5s +0%
polyvecl_pointwise_acc_montgomery_c 5s 3s +67%
polyvecl_unpack_eta 5s 6s -17%
rej_eta_native 5s 4s +25%
sign_signature 5s 8s -38%
sign_signature_internal 5s 6s -17%
sign_verify 5s 4s +25%
unpack_hints 5s 4s +25%
unpack_sig 5s 2s +150%
unpack_sk_s1hat 5s 1s +400%
mld_ct_abs_i32 4s 2s +100%
mld_ct_cmask_nonzero_u8 4s 2s +100%
ntt_native_aarch64 4s 4s +0%
ntt_native_x86_64 4s 3s +33%
pack_sig_z 4s 1s +300%
pack_sk_rho_key_tr_s2_t0 4s 2s +100%
pointwise_acc_native_x86_64 4s 8s -50%
poly_caddq_native 4s 3s +33%
poly_decompose_c 4s 5s -20%
poly_decompose_native 4s 3s +33%
poly_invntt_tomont_native 4s 5s -20%
poly_permute_bitrev_to_custom_optional 4s 2s +100%
poly_shiftl 4s 5s -20%
poly_use_hint_native_aarch64 4s 2s +100%
polyeta_unpack 4s 3s +33%
polyveck_pack_t0 4s 3s +33%
polyveck_unpack_eta 4s 3s +33%
polyvecl_pack_eta 4s 3s +33%
polyvecl_pointwise_acc_montgomery 4s 2s +100%
polyvecl_uniform_gamma1_serial 4s 2s +100%
polyz_unpack_19_native_aarch64 4s 3s +33%
rej_eta_c 4s 4s +0%
shake128x4_squeezeblocks 4s 2s +100%
shake256_release 4s 2s +100%
sign_keypair 4s 3s +33%
sign_open 4s 3s +33%
sign_signature_extmu 4s 4s +0%
sign_signature_pre_hash_internal 4s 5s -20%
sign_signature_pre_hash_shake256 4s 4s +0%
sign_verify_extmu 4s 4s +0%
sign_verify_pre_hash_internal 4s 3s +33%
sk_t0hat_get_poly 4s 5s -20%
unpack_pk 4s 3s +33%
unpack_sk_t0hat 4s 2s +100%
caddq 3s 2s +50%
decompose 3s 2s +50%
keccak_f1600_x4_native_aarch64_v84a 3s 2s +50%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid 3s 3s +0%
keccak_finalize 3s 3s +0%
keccakf1600_xor_bytes (big endian) 3s 2s +50%
keccakf1600x4_extract_bytes_native 3s - new
keccakf1600x4_permute 3s 3s +0%
keccakf1600x4_xor_bytes 3s 2s +50%
keccakf1600x4_xor_bytes_native 3s - new
mld_ct_cmask_neg_i32 3s 2s +50%
mld_ct_cmask_nonzero_u32 3s 3s +0%
mld_ct_get_optblocker_u32 3s 3s +0%
mld_ct_get_optblocker_u8 3s 3s +0%
mld_ct_sel_int32 3s 2s +50%
mld_polymat_expand_entry 3s 3s +0%
pack_pk 3s 5s -40%
pack_sk_s1 3s 3s +0%
pointwise_native_aarch64 3s 4s -25%
pointwise_native_x86_64 3s 2s +50%
poly_chknorm_native 3s 4s -25%
poly_ntt_c 3s 4s -25%
poly_permute_bitrev_to_custom_optional_native 3s 3s +0%
poly_pointwise_montgomery 3s 5s -40%
poly_pointwise_montgomery_native 3s 3s +0%
poly_reduce 3s 3s +0%
poly_uniform_4x 3s 4s -25%
poly_uniform_eta 3s 4s -25%
poly_use_hint 3s 3s +0%
poly_use_hint_c 3s 4s -25%
poly_use_hint_native 3s 5s -40%
polyt1_unpack 3s 2s +50%
polyvec_matrix_pointwise_montgomery 3s 6s -50%
polyveck_pack_eta 3s 4s -25%
polyvecl_pointwise_acc_montgomery_native 3s 3s +0%
polyvecl_uniform_gamma1 3s 3s +0%
polyvecl_unpack_z 3s 4s -25%
polyz_pack 3s 3s +0%
polyz_unpack 3s 2s +50%
polyz_unpack_17_native_aarch64 3s 3s +0%
reduce32 3s 2s +50%
rej_eta 3s 3s +0%
shake128_absorb 3s 4s -25%
shake128_release 3s 4s -25%
shake128_squeeze 3s 3s +0%
shake256 3s 4s -25%
sk_s2hat_get_poly 3s 6s -50%
sys_check_capability 3s 3s +0%
unpack_sk 3s 3s +0%
unpack_sk_s2hat 3s 3s +0%
use_hint 3s 4s -25%
yvec_get_poly 3s 4s -25%
yvec_init 3s 2s +50%
fqscale 2s 4s -50%
keccak_f1600_x1_native_aarch64 2s 4s -50%
keccak_f1600_x1_native_aarch64_v84a 2s 2s +0%
keccak_f1600_x4_native_avx2 2s 3s -33%
keccak_init 2s 4s -50%
keccakf1600_extract_bytes (big endian) 2s 2s +0%
keccakf1600_xor_bytes 2s 2s +0%
keccakf1600x4_extract_bytes 2s 2s +0%
make_hint 2s 3s -33%
mld_ct_get_optblocker_i64 2s 3s -33%
mld_keccakf1600_extract_bytes 2s 2s +0%
mld_prepare_domain_separation_prefix 2s 6s -67%
mld_value_barrier_i64 2s 2s +0%
mld_value_barrier_u32 2s 2s +0%
montgomery_reduce 2s 3s -33%
pack_sig_c 2s 3s -33%
pack_sig_h_poly 2s 2s +0%
poly_caddq 2s 3s -33%
poly_caddq_native_aarch64 2s 3s -33%
poly_chknorm 2s 6s -67%
poly_invntt_tomont 2s 4s -50%
poly_make_hint 2s 3s -33%
poly_ntt 2s 2s +0%
poly_ntt_native 2s 3s -33%
poly_sub 2s 3s -33%
poly_uniform_gamma1 2s 4s -50%
poly_uniform_gamma1_4x 2s 7s -71%
polyeta_pack 2s 3s -33%
polyt1_pack 2s 2s +0%
polyvec_matrix_expand 2s 1s +100%
polyvec_matrix_expand_serial 2s 3s -33%
polyveck_pack_w1 2s 4s -50%
polyveck_unpack_t0 2s 3s -33%
polyw1_pack 2s 3s -33%
polyz_unpack_c 2s 4s -50%
polyz_unpack_native 2s 4s -50%
power2round 2s 3s -33%
shake128_init 2s 3s -33%
shake128x4_absorb_once 2s 2s +0%
shake256_squeeze 2s 1s +100%
shake256x4_absorb_once 2s 1s +100%
sk_s1hat_get_poly 2s 2s +0%
mld_value_barrier_u8 1s 2s -50%
poly_decompose 1s 2s -50%
shake128_finalize 1s 3s -67%
shake256_absorb 1s 3s -67%
shake256_finalize 1s 2s -50%
shake256_init 1s 3s -67%
shake256x4_squeezeblocks 1s 2s -50%

@oqs-bot
Copy link
Copy Markdown
Contributor

oqs-bot commented May 3, 2026

CBMC Results (ML-DSA-44, REDUCE-RAM)

Full Results (203 proofs)
Proof Status Current Previous Change
**TOTAL** 1624s 1562s +4.0%
poly_pointwise_montgomery_c 201s 185s +9%
rej_uniform_native 114s 112s +2%
mld_invntt_layer 112s 107s +5%
mld_ct_memcmp 82s 78s +5%
sign_verify_internal 67s 65s +3%
polyvec_matrix_pointwise_montgomery_yvec 66s 65s +2%
mld_ntt_layer 44s 45s -2%
fqmul 31s 32s -3%
mld_attempt_signature_generation 29s 24s +21%
keccakf1600x4_permute_native 22s 23s -4%
polyeta_unpack 21s 21s +0%
rej_uniform 21s 21s +0%
rej_uniform_c 21s 21s +0%
mld_ntt_butterfly_block 18s 15s +20%
poly_chknorm_c 16s 13s +23%
polyveck_use_hint 15s 15s +0%
sign_keypair_internal 14s 15s -7%
poly_uniform_eta_4x 13s 15s -13%
polyt0_unpack 13s 12s +8%
sign_pk_from_sk 13s 16s -19%
keccak_absorb_once_x4 11s 9s +22%
mld_check_pct 11s 14s -21%
poly_add 11s 12s -8%
poly_invntt_tomont_c 10s 10s +0%
keccak_absorb 8s 6s +33%
poly_caddq_c 8s 8s +0%
poly_power2round 8s 7s +14%
polyveck_power2round 8s 4s +100%
keccakf1600_permute 7s 7s +0%
keccakf1600_permute_native 7s 6s +17%
pointwise_acc_native_x86_64 7s 8s -12%
poly_decompose_c 7s 7s +0%
poly_shiftl 7s 5s +40%
poly_use_hint_native 7s 4s +75%
polyveck_chknorm 7s 6s +17%
polyveck_decompose 7s 7s +0%
polyveck_pointwise_poly_montgomery 7s 9s -22%
polyvecl_chknorm 7s 5s +40%
pack_sk_rho_key_tr_s2_t0 6s 4s +50%
poly_uniform 6s 4s +50%
poly_uniform_4x 6s 4s +50%
polyvec_matrix_pointwise_montgomery_row 6s 8s -25%
polyveck_add 6s 8s -25%
polyveck_unpack_eta 6s 2s +200%
polyvecl_ntt 6s 7s -14%
polyvecl_pointwise_acc_montgomery_native 6s 3s +100%
polyvecl_uniform_gamma1_serial 6s 3s +100%
reduce32 6s 3s +100%
sign_open 6s 5s +20%
sign_signature_extmu 6s 3s +100%
sign_signature_internal 6s 3s +100%
sk_s2hat_get_poly 6s 4s +50%
unpack_hints 6s 6s +0%
keccak_squeeze 5s 2s +150%
keccak_squeezeblocks_x4 5s 4s +25%
keccakf1600x4_xor_bytes_native 5s - new
mld_compute_pack_z 5s 6s -17%
mld_prepare_domain_separation_prefix 5s 5s +0%
pointwise_acc_native_aarch64 5s 5s +0%
poly_decompose_native 5s 3s +67%
poly_invntt_tomont 5s 1s +400%
polyveck_invntt_tomont 5s 6s -17%
polyveck_sub 5s 4s +25%
polyz_unpack 5s 2s +150%
rej_eta 5s 3s +67%
sign_signature_pre_hash_internal 5s 5s +0%
sign_signature_pre_hash_shake256 5s 3s +67%
sign_verify_pre_hash_shake256 5s 2s +150%
keccak_init 4s 2s +100%
keccakf1600x4_permute 4s 4s +0%
mld_ct_cmask_nonzero_u8 4s 2s +100%
montgomery_reduce 4s 4s +0%
ntt_native_aarch64 4s 3s +33%
pack_pk 4s 2s +100%
pack_sig_z 4s 3s +33%
poly_caddq_native 4s 1s +300%
poly_challenge 4s 3s +33%
poly_chknorm 4s 2s +100%
poly_chknorm_native_aarch64 4s 3s +33%
poly_ntt_c 4s 4s +0%
poly_uniform_gamma1 4s 3s +33%
polyt1_unpack 4s 3s +33%
polyvec_matrix_expand 4s 2s +100%
polyvec_matrix_expand_serial 4s 2s +100%
polyvec_matrix_pointwise_montgomery 4s 2s +100%
polyveck_caddq 4s 6s -33%
polyveck_shiftl 4s 4s +0%
polyvecl_pointwise_acc_montgomery_c 4s 2s +100%
polyvecl_unpack_z 4s 2s +100%
polyw1_pack 4s 3s +33%
polyz_unpack_17_native_aarch64 4s 4s +0%
power2round 4s 3s +33%
rej_eta_c 4s 6s -33%
rej_eta_native 4s 3s +33%
shake256_init 4s 4s +0%
sign 4s 5s -20%
sign_signature 4s 5s -20%
sign_verify 4s 4s +0%
sign_verify_pre_hash_internal 4s 4s +0%
fqscale 3s 1s +200%
keccak_f1600_x1_native_aarch64 3s 1s +200%
keccak_f1600_x1_native_aarch64_v84a 3s 3s +0%
keccak_f1600_x4_native_aarch64_v84a 3s 3s +0%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid 3s 2s +50%
keccakf1600_xor_bytes (big endian) 3s 3s +0%
keccakf1600x4_extract_bytes_native 3s - new
make_hint 3s 5s -40%
mld_ct_cmask_nonzero_u32 3s 3s +0%
mld_ct_get_optblocker_u32 3s 2s +50%
mld_ct_get_optblocker_u8 3s 3s +0%
mld_ct_sel_int32 3s 2s +50%
mld_h 3s 4s -25%
mld_keccakf1600_extract_bytes 3s 3s +0%
mld_sample_s1_s2_serial 3s 6s -50%
mld_value_barrier_u32 3s 4s -25%
mld_value_barrier_u8 3s 1s +200%
ntt_native_x86_64 3s 3s +0%
nttunpack_native_x86_64 3s 5s -40%
pack_sig_c 3s 3s +0%
pack_sig_h_poly 3s 3s +0%
pointwise_native_aarch64 3s 2s +50%
pointwise_native_x86_64 3s 4s -25%
poly_caddq_native_aarch64 3s 3s +0%
poly_chknorm_native 3s 3s +0%
poly_invntt_tomont_native 3s 4s -25%
poly_ntt 3s 2s +50%
poly_reduce 3s 5s -40%
poly_sub 3s 6s -50%
poly_uniform_eta 3s 3s +0%
poly_uniform_gamma1_4x 3s 2s +50%
polyt0_pack 3s 3s +0%
polyt1_pack 3s 2s +50%
polyveck_ntt 3s 5s -40%
polyveck_pack_eta 3s 5s -40%
polyveck_pack_t0 3s 4s -25%
polyveck_unpack_t0 3s 2s +50%
polyvecl_pointwise_acc_montgomery 3s 2s +50%
polyz_unpack_c 3s 2s +50%
shake128_release 3s 2s +50%
shake128x4_absorb_once 3s 2s +50%
shake128x4_squeezeblocks 3s 2s +50%
shake256 3s 4s -25%
shake256_finalize 3s 2s +50%
shake256_squeeze 3s 1s +200%
shake256x4_absorb_once 3s 3s +0%
sign_verify_extmu 3s 3s +0%
sk_t0hat_get_poly 3s 2s +50%
sys_check_capability 3s 2s +50%
unpack_pk 3s 2s +50%
unpack_sig 3s 3s +0%
unpack_sk_s1hat 3s 3s +0%
unpack_sk_t0hat 3s 3s +0%
yvec_init 3s 3s +0%
caddq 2s 5s -60%
decompose 2s 4s -50%
intt_native_x86_64 2s 3s -33%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid 2s 2s +0%
keccak_f1600_x4_native_avx2 2s 2s +0%
keccak_finalize 2s 2s +0%
keccakf1600_xor_bytes 2s 3s -33%
keccakf1600x4_extract_bytes 2s 2s +0%
keccakf1600x4_xor_bytes 2s 2s +0%
mld_ct_abs_i32 2s 4s -50%
mld_ct_cmask_neg_i32 2s 4s -50%
mld_ct_get_optblocker_i64 2s 2s +0%
mld_polymat_expand_entry 2s 3s -33%
mld_sample_s1_s2 2s 4s -50%
mld_value_barrier_i64 2s 3s -33%
pack_sk_s1 2s 3s -33%
poly_caddq 2s 2s +0%
poly_decompose 2s 3s -33%
poly_make_hint 2s 3s -33%
poly_ntt_native 2s 2s +0%
poly_permute_bitrev_to_custom_optional 2s 3s -33%
poly_permute_bitrev_to_custom_optional_native 2s 4s -50%
poly_pointwise_montgomery 2s 3s -33%
poly_pointwise_montgomery_native 2s 2s +0%
poly_use_hint 2s 3s -33%
poly_use_hint_c 2s 2s +0%
poly_use_hint_native_aarch64 2s 3s -33%
polyeta_pack 2s 3s -33%
polyveck_pack_w1 2s 3s -33%
polyveck_reduce 2s 5s -60%
polyvecl_pack_eta 2s 2s +0%
polyvecl_uniform_gamma1 2s 4s -50%
polyvecl_unpack_eta 2s 4s -50%
polyz_pack 2s 2s +0%
polyz_unpack_19_native_aarch64 2s 3s -33%
shake128_absorb 2s 3s -33%
shake128_finalize 2s 4s -50%
shake128_squeeze 2s 3s -33%
shake256_release 2s 4s -50%
shake256x4_squeezeblocks 2s 2s +0%
sign_keypair 2s 6s -67%
sk_s1hat_get_poly 2s 3s -33%
unpack_sk 2s 3s -33%
use_hint 2s 2s +0%
yvec_get_poly 2s 2s +0%
keccakf1600_extract_bytes (big endian) 1s 3s -67%
polyz_unpack_native 1s 2s -50%
shake128_init 1s 3s -67%
shake256_absorb 1s 5s -80%
unpack_sk_s2hat 1s 5s -80%

@oqs-bot
Copy link
Copy Markdown
Contributor

oqs-bot commented May 3, 2026

CBMC Results (ML-DSA-87, REDUCE-RAM)

Full Results (203 proofs)
Proof Status Current Previous Change
**TOTAL** 1910s 1808s +5.6%
sign_verify_internal 249s 238s +5%
poly_pointwise_montgomery_c 184s 169s +9%
polyvec_matrix_pointwise_montgomery_yvec 146s 140s +4%
rej_uniform_native 114s 105s +9%
mld_invntt_layer 106s 105s +1%
mld_ct_memcmp 77s 73s +5%
mld_ntt_layer 44s 43s +2%
sign_keypair_internal 33s 33s +0%
fqmul 28s 27s +4%
mld_attempt_signature_generation 26s 23s +13%
keccakf1600x4_permute_native 22s 22s +0%
rej_uniform 20s 24s -17%
sign_pk_from_sk 20s 18s +11%
polyeta_unpack 18s 20s -10%
rej_uniform_c 18s 17s +6%
mld_ntt_butterfly_block 16s 15s +7%
poly_uniform_eta_4x 15s 13s +15%
poly_chknorm_c 14s 14s +0%
mld_check_pct 12s 12s +0%
polyvec_matrix_pointwise_montgomery_row 12s 13s -8%
polyveck_decompose 12s 15s -20%
poly_add 10s 13s -23%
polyt0_unpack 10s 14s -29%
polyveck_add 10s 9s +11%
polyveck_ntt 10s 8s +25%
keccak_absorb_once_x4 9s 10s -10%
keccakf1600_permute_native 9s 7s +29%
polyveck_pointwise_poly_montgomery 9s 9s +0%
polyveck_use_hint 9s 8s +12%
mld_sample_s1_s2 8s 6s +33%
pointwise_acc_native_x86_64 8s 6s +33%
poly_invntt_tomont_c 8s 9s -11%
poly_power2round 8s 8s +0%
polyveck_invntt_tomont 8s 9s -11%
polyveck_power2round 8s 8s +0%
keccak_absorb 7s 7s +0%
keccakf1600_permute 7s 7s +0%
poly_caddq_c 7s 8s -12%
poly_shiftl 7s 7s +0%
polyt1_unpack 7s 3s +133%
polyveck_caddq 7s 5s +40%
polyveck_chknorm 7s 7s +0%
polyvecl_ntt 7s 3s +133%
pointwise_acc_native_aarch64 6s 8s -25%
poly_ntt_c 6s 5s +20%
poly_uniform_4x 6s 4s +50%
polyvec_matrix_pointwise_montgomery 6s 7s -14%
polyveck_reduce 6s 5s +20%
polyveck_shiftl 6s 7s -14%
polyveck_sub 6s 7s -14%
polyvecl_chknorm 6s 6s +0%
polyvecl_unpack_z 6s 2s +200%
sign_open 6s 2s +200%
sign_signature_internal 6s 4s +50%
sign_signature_pre_hash_shake256 6s 4s +50%
sign_verify 6s 5s +20%
sign_verify_pre_hash_internal 6s 3s +100%
unpack_hints 6s 5s +20%
keccak_squeeze 5s 2s +150%
keccak_squeezeblocks_x4 5s 5s +0%
mld_compute_pack_z 5s 6s -17%
mld_h 5s 4s +25%
mld_sample_s1_s2_serial 5s 6s -17%
poly_caddq_native_aarch64 5s 2s +150%
poly_challenge 5s 4s +25%
poly_permute_bitrev_to_custom_optional_native 5s 3s +67%
poly_use_hint_native_aarch64 5s 3s +67%
polyt1_pack 5s 1s +400%
polyvec_matrix_expand 5s 3s +67%
polyvecl_pack_eta 5s 3s +67%
polyz_unpack_native 5s 5s +0%
rej_eta_native 5s 4s +25%
sign 5s 7s -29%
sign_signature 5s 2s +150%
sign_signature_extmu 5s 2s +150%
sign_signature_pre_hash_internal 5s 4s +25%
sk_s2hat_get_poly 5s 3s +67%
unpack_pk 5s 5s +0%
keccakf1600_extract_bytes (big endian) 4s 3s +33%
keccakf1600_xor_bytes 4s 4s +0%
keccakf1600x4_extract_bytes_native 4s - new
keccakf1600x4_permute 4s 3s +33%
keccakf1600x4_xor_bytes_native 4s - new
make_hint 4s 4s +0%
mld_ct_cmask_nonzero_u32 4s 2s +100%
mld_keccakf1600_extract_bytes 4s 3s +33%
montgomery_reduce 4s 3s +33%
ntt_native_aarch64 4s 2s +100%
ntt_native_x86_64 4s 2s +100%
pack_sig_h_poly 4s 3s +33%
pointwise_native_x86_64 4s 3s +33%
poly_decompose_c 4s 5s -20%
poly_invntt_tomont 4s 1s +300%
poly_ntt_native 4s 3s +33%
poly_pointwise_montgomery 4s 3s +33%
poly_pointwise_montgomery_native 4s 2s +100%
poly_uniform 4s 4s +0%
poly_uniform_gamma1 4s 4s +0%
poly_use_hint 4s 2s +100%
polyeta_pack 4s 5s -20%
polyz_unpack_19_native_aarch64 4s 3s +33%
polyz_unpack_c 4s 3s +33%
rej_eta 4s 3s +33%
shake128_init 4s 4s +0%
shake256_init 4s 4s +0%
sign_keypair 4s 4s +0%
sk_t0hat_get_poly 4s 3s +33%
decompose 3s 2s +50%
fqscale 3s 1s +200%
intt_native_x86_64 3s 2s +50%
keccak_f1600_x4_native_aarch64_v84a 3s 4s -25%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid 3s 2s +50%
keccak_f1600_x4_native_avx2 3s 4s -25%
keccakf1600x4_xor_bytes 3s 4s -25%
mld_ct_cmask_neg_i32 3s 2s +50%
mld_ct_get_optblocker_i64 3s 1s +200%
mld_ct_sel_int32 3s 2s +50%
mld_prepare_domain_separation_prefix 3s 3s +0%
mld_value_barrier_i64 3s 2s +50%
nttunpack_native_x86_64 3s 3s +0%
pack_sig_c 3s 3s +0%
pack_sig_z 3s 4s -25%
pack_sk_rho_key_tr_s2_t0 3s 5s -40%
pack_sk_s1 3s 2s +50%
pointwise_native_aarch64 3s 2s +50%
poly_chknorm_native 3s 3s +0%
poly_chknorm_native_aarch64 3s 2s +50%
poly_decompose 3s 2s +50%
poly_decompose_native 3s 3s +0%
poly_invntt_tomont_native 3s 1s +200%
poly_ntt 3s 2s +50%
poly_permute_bitrev_to_custom_optional 3s 2s +50%
poly_sub 3s 2s +50%
poly_uniform_gamma1_4x 3s 2s +50%
poly_use_hint_c 3s 5s -40%
polyt0_pack 3s 5s -40%
polyvec_matrix_expand_serial 3s 2s +50%
polyveck_pack_t0 3s 3s +0%
polyveck_pack_w1 3s 3s +0%
polyveck_unpack_eta 3s 4s -25%
polyveck_unpack_t0 3s 3s +0%
polyvecl_pointwise_acc_montgomery 3s 3s +0%
polyvecl_pointwise_acc_montgomery_c 3s 3s +0%
polyvecl_uniform_gamma1 3s 2s +50%
polyz_unpack_17_native_aarch64 3s 4s -25%
power2round 3s 2s +50%
rej_eta_c 3s 4s -25%
shake128_absorb 3s 3s +0%
shake128_finalize 3s 3s +0%
shake256 3s 1s +200%
shake256_absorb 3s 3s +0%
shake256_finalize 3s 5s -40%
shake256_release 3s 1s +200%
shake256_squeeze 3s 3s +0%
shake256x4_absorb_once 3s 3s +0%
sk_s1hat_get_poly 3s 2s +50%
unpack_sk 3s 2s +50%
unpack_sk_s2hat 3s 3s +0%
unpack_sk_t0hat 3s 4s -25%
yvec_get_poly 3s 2s +50%
yvec_init 3s 4s -25%
caddq 2s 2s +0%
keccak_f1600_x1_native_aarch64 2s 3s -33%
keccak_f1600_x1_native_aarch64_v84a 2s 1s +100%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid 2s 1s +100%
keccak_init 2s 3s -33%
keccakf1600_xor_bytes (big endian) 2s 2s +0%
keccakf1600x4_extract_bytes 2s 4s -50%
mld_ct_abs_i32 2s 1s +100%
mld_ct_cmask_nonzero_u8 2s 2s +0%
mld_ct_get_optblocker_u8 2s 2s +0%
mld_polymat_expand_entry 2s 4s -50%
mld_value_barrier_u32 2s 1s +100%
mld_value_barrier_u8 2s 2s +0%
pack_pk 2s 2s +0%
poly_caddq 2s 3s -33%
poly_caddq_native 2s 5s -60%
poly_chknorm 2s 1s +100%
poly_make_hint 2s 3s -33%
poly_reduce 2s 4s -50%
poly_uniform_eta 2s 4s -50%
poly_use_hint_native 2s 4s -50%
polyvecl_pointwise_acc_montgomery_native 2s 3s -33%
polyvecl_uniform_gamma1_serial 2s 2s +0%
polyvecl_unpack_eta 2s 3s -33%
polyw1_pack 2s 1s +100%
polyz_unpack 2s 1s +100%
reduce32 2s 4s -50%
shake128_squeeze 2s 2s +0%
shake128x4_squeezeblocks 2s 4s -50%
sign_verify_extmu 2s 7s -71%
sign_verify_pre_hash_shake256 2s 6s -67%
sys_check_capability 2s 2s +0%
unpack_sig 2s 2s +0%
unpack_sk_s1hat 2s 4s -50%
keccak_finalize 1s 2s -50%
mld_ct_get_optblocker_u32 1s 3s -67%
polyveck_pack_eta 1s 2s -50%
polyz_pack 1s 5s -80%
shake128_release 1s 1s +0%
shake128x4_absorb_once 1s 2s -50%
shake256x4_squeezeblocks 1s 2s -50%
use_hint 1s 4s -75%

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants