Skip to content

Add HOL Light poly_use_hint proofs for AArch64#1037

Open
jakemas wants to merge 1 commit intomainfrom
hol-light-aarch64-use-hint
Open

Add HOL Light poly_use_hint proofs for AArch64#1037
jakemas wants to merge 1 commit intomainfrom
hol-light-aarch64-use-hint

Conversation

@jakemas
Copy link
Copy Markdown
Contributor

@jakemas jakemas commented Apr 13, 2026

Resolves:

Summary

@jakemas jakemas requested a review from a team as a code owner April 13, 2026 16:54
@jakemas jakemas force-pushed the hol-light-aarch64-use-hint branch from 2053888 to 0b42385 Compare April 13, 2026 17:02
@jakemas jakemas marked this pull request as draft April 13, 2026 17:12
@jakemas jakemas force-pushed the hol-light-aarch64-use-hint branch from 0b42385 to c5c76af Compare April 13, 2026 17:21
@jakemas jakemas marked this pull request as ready for review April 13, 2026 17:52
@oqs-bot
Copy link
Copy Markdown
Contributor

oqs-bot commented Apr 13, 2026

CBMC Results (ML-DSA-87)

Full Results (187 proofs)
Proof Status Current Previous Change
**TOTAL** 3519s 3549s -0.8%
polyvecl_pointwise_acc_montgomery_c 1347s 1354s -1%
sign_verify_internal 250s 250s +0%
polyvec_matrix_expand 182s 182s +0%
poly_pointwise_montgomery_c 163s 168s -3%
rej_uniform_native 147s 147s +0%
polyvec_matrix_expand_serial 125s 124s +1%
mld_attempt_signature_generation 89s 88s +1%
mld_ct_memcmp 81s 79s +3%
mld_invntt_layer 65s 68s -4%
mld_ntt_layer 53s 53s +0%
sign_keypair_internal 47s 46s +2%
sign_signature_internal 34s 34s +0%
polymat_permute_bitrev_to_custom 27s 27s +0%
polyveck_invntt_tomont 27s 27s +0%
sign_pk_from_sk 23s 24s -4%
poly_chknorm_c 21s 21s +0%
rej_uniform 21s 23s -9%
fqmul 20s 20s +0%
polyveck_decompose 18s 21s -14%
poly_uniform_eta_4x 17s 19s -11%
poly_uniform_4x 16s 14s +14%
polyeta_unpack 16s 16s +0%
polyveck_add 15s 12s +25%
rej_uniform_c 15s 15s +0%
keccakf1600x4_permute_native 14s 15s -7%
polyt0_unpack 14s 12s +17%
poly_add 13s 12s +8%
mld_check_pct 12s 9s +33%
mld_ntt_butterfly_block 12s 10s +20%
polyz_unpack_c 11s 9s +22%
keccakf1600_permute 10s 8s +25%
polyveck_caddq 10s 10s +0%
keccak_absorb_once_x4 9s 9s +0%
keccakf1600_permute_native 9s 11s -18%
polyveck_power2round 9s 9s +0%
rej_eta_native 9s 7s +29%
mld_polyvecl_permute_bitrev_to_custom_native 8s 8s +0%
poly_decompose_c 8s 6s +33%
polyvec_matrix_pointwise_montgomery 8s 9s -11%
polyvecl_ntt 8s 7s +14%
unpack_hints 8s 7s +14%
unpack_sk 8s 6s +33%
keccak_absorb 7s 6s +17%
mld_compute_pack_z 7s 8s -12%
mld_sample_s1_s2 7s 9s -22%
pointwise_acc_native_aarch64 7s 7s +0%
pointwise_acc_native_x86_64 7s 9s -22%
polyveck_ntt 7s 8s -12%
polyveck_pointwise_poly_montgomery 7s 8s -12%
polyveck_reduce 7s 6s +17%
polyveck_shiftl 7s 8s -12%
polyveck_use_hint 7s 9s -22%
polyvecl_chknorm 7s 6s +17%
sign_verify_pre_hash_internal 7s 6s +17%
poly_invntt_tomont_c 6s 6s +0%
poly_pointwise_montgomery 6s 1s +500%
polyt0_pack 6s 4s +50%
polyveck_chknorm 6s 6s +0%
polyvecl_uniform_gamma1_serial 6s 4s +50%
sign_keypair 6s 2s +200%
sign_verify 6s 3s +100%
unpack_pk 6s 5s +20%
keccak_squeezeblocks_x4 5s 4s +25%
mld_h 5s 6s -17%
mld_keccakf1600_extract_bytes 5s 1s +400%
pack_pk 5s 3s +67%
pack_sig_c 5s 3s +67%
pack_sig_h_poly 5s 4s +25%
poly_caddq_c 5s 6s -17%
poly_caddq_native_aarch64 5s 4s +25%
poly_power2round 5s 8s -38%
poly_use_hint_native_aarch64 5s - new
polyveck_sub 5s 6s -17%
polyvecl_pointwise_acc_montgomery_native 5s 3s +67%
polyvecl_uniform_gamma1 5s 5s +0%
polyz_unpack_native 5s 5s +0%
shake128_init 5s 2s +150%
shake256_release 5s 1s +400%
sign 5s 6s -17%
sign_signature_extmu 5s 4s +25%
sign_signature_pre_hash_internal 5s 5s +0%
sign_signature_pre_hash_shake256 5s 3s +67%
sign_verify_extmu 5s 3s +67%
sign_verify_pre_hash_shake256 5s 5s +0%
decompose 4s 5s -20%
mld_prepare_domain_separation_prefix 4s 4s +0%
mld_sample_s1_s2_serial 4s 6s -33%
pack_sig_z 4s 3s +33%
pack_sk_s1 4s 3s +33%
pointwise_native_aarch64 4s 3s +33%
poly_challenge 4s 7s -43%
poly_ntt_c 4s 5s -20%
poly_reduce 4s 3s +33%
poly_sub 4s 3s +33%
poly_uniform 4s 7s -43%
poly_uniform_eta 4s 6s -33%
poly_uniform_gamma1_4x 4s 6s -33%
polyeta_pack 4s 3s +33%
polyveck_pack_t0 4s 3s +33%
polyveck_unpack_eta 4s 6s -33%
polyvecl_permute_bitrev_to_custom 4s 1s +300%
polyz_pack 4s 4s +0%
polyz_unpack 4s 4s +0%
shake256 4s 3s +33%
shake256_init 4s 1s +300%
sign_open 4s 4s +0%
unpack_sig 4s 4s +0%
keccak_f1600_x1_native_aarch64 3s 3s +0%
keccak_squeeze 3s 2s +50%
keccakf1600_xor_bytes 3s 4s -25%
mld_ct_abs_i32 3s 2s +50%
mld_ct_cmask_nonzero_u8 3s 2s +50%
mld_value_barrier_u32 3s 4s -25%
ntt_native_x86_64 3s 5s -40%
pointwise_native_x86_64 3s 3s +0%
poly_chknorm_native_aarch64 3s 1s +200%
poly_decompose_native 3s 2s +50%
poly_invntt_tomont_native 3s 4s -25%
poly_make_hint 3s 4s -25%
poly_ntt 3s 4s -25%
poly_ntt_native 3s 3s +0%
poly_shiftl 3s 3s +0%
poly_use_hint 3s 2s +50%
poly_use_hint_native 3s 2s +50%
polyt1_pack 3s 2s +50%
polyt1_unpack 3s 3s +0%
polyveck_pack_eta 3s 4s -25%
polyveck_pack_w1 3s 4s -25%
polyveck_unpack_t0 3s 4s -25%
polyvecl_pack_eta 3s 2s +50%
polyvecl_pointwise_acc_montgomery 3s 3s +0%
polyvecl_unpack_eta 3s 8s -62%
polyvecl_unpack_z 3s 5s -40%
polyw1_pack 3s 3s +0%
reduce32 3s 5s -40%
rej_eta 3s 3s +0%
rej_eta_c 3s 5s -40%
shake128_release 3s 3s +0%
shake128_squeeze 3s 2s +50%
shake256_finalize 3s 1s +200%
sign_signature 3s 4s -25%
caddq 2s 2s +0%
fqscale 2s 2s +0%
intt_native_x86_64 2s 3s -33%
keccak_f1600_x1_native_aarch64_v84a 2s 3s -33%
keccak_f1600_x4_native_aarch64_v84a 2s 3s -33%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid 2s 1s +100%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid 2s 3s -33%
keccak_finalize 2s 5s -60%
keccak_init 2s 2s +0%
keccakf1600_extract_bytes (big endian) 2s 5s -60%
keccakf1600_xor_bytes (big endian) 2s 3s -33%
keccakf1600x4_extract_bytes 2s 2s +0%
keccakf1600x4_permute 2s 4s -50%
keccakf1600x4_xor_bytes 2s 2s +0%
make_hint 2s 2s +0%
mld_ct_cmask_nonzero_u32 2s 2s +0%
mld_ct_get_optblocker_u32 2s 4s -50%
mld_ct_get_optblocker_u8 2s 4s -50%
mld_value_barrier_i64 2s 3s -33%
montgomery_reduce 2s 5s -60%
ntt_native_aarch64 2s 4s -50%
pack_sk_rho_key_tr_s2_t0 2s 4s -50%
poly_caddq 2s 3s -33%
poly_caddq_native 2s 3s -33%
poly_chknorm 2s 3s -33%
poly_chknorm_native 2s 2s +0%
poly_decompose 2s 2s +0%
poly_invntt_tomont 2s 3s -33%
poly_pointwise_montgomery_native 2s 3s -33%
poly_uniform_gamma1 2s 6s -67%
power2round 2s 2s +0%
shake128_finalize 2s 3s -33%
shake128x4_absorb_once 2s 1s +100%
shake128x4_squeezeblocks 2s 2s +0%
shake256_absorb 2s 4s -50%
shake256_squeeze 2s 2s +0%
shake256x4_squeezeblocks 2s 3s -33%
use_hint 2s 4s -50%
mld_ct_cmask_neg_i32 1s 3s -67%
mld_ct_get_optblocker_i64 1s 3s -67%
mld_ct_sel_int32 1s 3s -67%
mld_value_barrier_u8 1s 3s -67%
poly_use_hint_c 1s 4s -75%
shake128_absorb 1s 3s -67%
shake256x4_absorb_once 1s 2s -50%
sys_check_capability 1s 5s -80%

@oqs-bot
Copy link
Copy Markdown
Contributor

oqs-bot commented Apr 13, 2026

CBMC Results (ML-DSA-65)

Full Results (187 proofs)
Proof Status Current Previous Change
**TOTAL** 2405s 2438s -1.4%
polyvecl_pointwise_acc_montgomery_c 599s 581s +3%
sign_verify_internal 220s 224s -2%
poly_pointwise_montgomery_c 159s 152s +5%
rej_uniform_native 137s 140s -2%
polyvec_matrix_expand 86s 90s -4%
mld_ct_memcmp 76s 75s +1%
mld_attempt_signature_generation 63s 64s -2%
mld_invntt_layer 63s 67s -6%
mld_ntt_layer 51s 56s -9%
polyvec_matrix_expand_serial 48s 49s -2%
polymat_permute_bitrev_to_custom 35s 35s +0%
sign_signature_internal 27s 24s +12%
sign_keypair_internal 26s 29s -10%
poly_chknorm_c 23s 19s +21%
rej_uniform 23s 23s +0%
fqmul 22s 19s +16%
sign_pk_from_sk 18s 20s -10%
polyveck_power2round 17s 17s +0%
polyt0_unpack 16s 14s +14%
poly_uniform_eta_4x 15s 18s -17%
poly_uniform_4x 14s 13s +8%
rej_uniform_c 14s 13s +8%
keccakf1600x4_permute_native 13s 14s -7%
polyveck_decompose 12s 15s -20%
keccak_absorb_once_x4 11s 12s -8%
poly_add 11s 11s +0%
polyvec_matrix_pointwise_montgomery 11s 12s -8%
polyveck_add 11s 10s +10%
polyveck_invntt_tomont 10s 5s +100%
mld_check_pct 9s 11s -18%
mld_ntt_butterfly_block 9s 13s -31%
polyveck_caddq 9s 8s +12%
polyveck_shiftl 9s 8s +12%
keccak_absorb 8s 7s +14%
polyvecl_ntt 8s 7s +14%
keccak_squeezeblocks_x4 7s 8s -12%
keccakf1600_permute 7s 8s -12%
keccakf1600_permute_native 7s 8s -12%
pointwise_acc_native_aarch64 7s 5s +40%
pointwise_acc_native_x86_64 7s 6s +17%
poly_decompose_c 7s 7s +0%
polyeta_unpack 7s 6s +17%
polyveck_chknorm 7s 6s +17%
polyveck_pointwise_poly_montgomery 7s 6s +17%
polyveck_reduce 7s 6s +17%
sign_verify_pre_hash_shake256 7s 4s +75%
poly_challenge 6s 4s +50%
poly_uniform_gamma1 6s 4s +50%
polyveck_ntt 6s 6s +0%
polyveck_sub 6s 8s -25%
polyveck_use_hint 6s 5s +20%
sign_signature 6s 4s +50%
unpack_sk 6s 10s -40%
intt_native_x86_64 5s 4s +25%
mld_compute_pack_z 5s 9s -44%
mld_ct_get_optblocker_i64 5s 1s +400%
mld_polyvecl_permute_bitrev_to_custom_native 5s 6s -17%
mld_sample_s1_s2_serial 5s 5s +0%
mld_value_barrier_i64 5s 3s +67%
pack_sig_c 5s 2s +150%
poly_invntt_tomont_c 5s 8s -38%
poly_power2round 5s 8s -38%
polyvecl_chknorm 5s 5s +0%
rej_eta_c 5s 3s +67%
shake256_init 5s 3s +67%
sign 5s 7s -29%
sys_check_capability 5s 2s +150%
caddq 4s 2s +100%
decompose 4s 2s +100%
mld_ct_cmask_nonzero_u32 4s 3s +33%
mld_h 4s 8s -50%
mld_prepare_domain_separation_prefix 4s 4s +0%
mld_sample_s1_s2 4s 9s -56%
ntt_native_aarch64 4s 3s +33%
ntt_native_x86_64 4s 4s +0%
pack_sig_h_poly 4s 3s +33%
poly_caddq_native_aarch64 4s 3s +33%
poly_chknorm_native_aarch64 4s 5s -20%
poly_decompose_native 4s 1s +300%
poly_invntt_tomont_native 4s 4s +0%
poly_ntt_native 4s 4s +0%
poly_sub 4s 4s +0%
poly_uniform 4s 6s -33%
poly_use_hint_c 4s 7s -43%
poly_use_hint_native 4s 4s +0%
poly_use_hint_native_aarch64 4s - new
polyeta_pack 4s 1s +300%
polyt0_pack 4s 4s +0%
polyveck_pack_eta 4s 5s -20%
polyvecl_permute_bitrev_to_custom 4s 3s +33%
polyvecl_pointwise_acc_montgomery_native 4s 3s +33%
polyvecl_unpack_z 4s 2s +100%
polyz_unpack 4s 3s +33%
shake128_squeeze 4s 4s +0%
shake128x4_squeezeblocks 4s 2s +100%
sign_keypair 4s 5s -20%
sign_open 4s 6s -33%
sign_verify 4s 5s -20%
unpack_hints 4s 4s +0%
unpack_sig 4s 2s +100%
keccak_f1600_x1_native_aarch64_v84a 3s 4s -25%
keccak_finalize 3s 3s +0%
keccak_squeeze 3s 4s -25%
keccakf1600_xor_bytes 3s 4s -25%
mld_ct_cmask_nonzero_u8 3s 2s +50%
mld_ct_sel_int32 3s 2s +50%
mld_value_barrier_u8 3s 2s +50%
pack_pk 3s 1s +200%
pack_sig_z 3s 2s +50%
pointwise_native_x86_64 3s 3s +0%
poly_caddq 3s 3s +0%
poly_caddq_c 3s 6s -50%
poly_caddq_native 3s 3s +0%
poly_chknorm_native 3s 3s +0%
poly_make_hint 3s 3s +0%
poly_pointwise_montgomery_native 3s 5s -40%
poly_shiftl 3s 1s +200%
poly_uniform_eta 3s 5s -40%
poly_uniform_gamma1_4x 3s 2s +50%
poly_use_hint 3s 4s -25%
polyt1_pack 3s 4s -25%
polyt1_unpack 3s 2s +50%
polyveck_unpack_eta 3s 5s -40%
polyveck_unpack_t0 3s 5s -40%
polyvecl_uniform_gamma1_serial 3s 5s -40%
polyw1_pack 3s 4s -25%
polyz_unpack_c 3s 3s +0%
polyz_unpack_native 3s 2s +50%
rej_eta 3s 2s +50%
rej_eta_native 3s 6s -50%
shake128_finalize 3s 3s +0%
shake256_finalize 3s 1s +200%
shake256_squeeze 3s 2s +50%
shake256x4_absorb_once 3s 3s +0%
shake256x4_squeezeblocks 3s 1s +200%
sign_signature_pre_hash_internal 3s 5s -40%
sign_signature_pre_hash_shake256 3s 5s -40%
sign_verify_extmu 3s 4s -25%
sign_verify_pre_hash_internal 3s 3s +0%
unpack_pk 3s 3s +0%
fqscale 2s 3s -33%
keccak_f1600_x4_native_aarch64_v84a 2s 2s +0%
keccakf1600_extract_bytes (big endian) 2s 4s -50%
keccakf1600_xor_bytes (big endian) 2s 4s -50%
keccakf1600x4_extract_bytes 2s 2s +0%
keccakf1600x4_permute 2s 1s +100%
keccakf1600x4_xor_bytes 2s 2s +0%
make_hint 2s 3s -33%
mld_ct_abs_i32 2s 3s -33%
mld_ct_cmask_neg_i32 2s 2s +0%
mld_ct_get_optblocker_u32 2s 4s -50%
mld_value_barrier_u32 2s 4s -50%
montgomery_reduce 2s 4s -50%
pack_sk_s1 2s 2s +0%
poly_chknorm 2s 3s -33%
poly_decompose 2s 4s -50%
poly_invntt_tomont 2s 1s +100%
poly_ntt 2s 2s +0%
poly_pointwise_montgomery 2s 3s -33%
poly_reduce 2s 3s -33%
polyveck_pack_t0 2s 4s -50%
polyveck_pack_w1 2s 2s +0%
polyvecl_pack_eta 2s 6s -67%
polyvecl_pointwise_acc_montgomery 2s 4s -50%
polyvecl_uniform_gamma1 2s 3s -33%
polyvecl_unpack_eta 2s 3s -33%
power2round 2s 1s +100%
shake128_init 2s 3s -33%
shake128_release 2s 2s +0%
shake128x4_absorb_once 2s 2s +0%
shake256 2s 1s +100%
shake256_absorb 2s 3s -33%
shake256_release 2s 2s +0%
sign_signature_extmu 2s 3s -33%
keccak_f1600_x1_native_aarch64 1s 2s -50%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid 1s 2s -50%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid 1s 3s -67%
keccak_init 1s 3s -67%
mld_ct_get_optblocker_u8 1s 2s -50%
mld_keccakf1600_extract_bytes 1s 1s +0%
pack_sk_rho_key_tr_s2_t0 1s 2s -50%
pointwise_native_aarch64 1s 4s -75%
poly_ntt_c 1s 1s +0%
polyz_pack 1s 2s -50%
reduce32 1s 4s -75%
shake128_absorb 1s 5s -80%
use_hint 1s 5s -80%

@oqs-bot
Copy link
Copy Markdown
Contributor

oqs-bot commented Apr 13, 2026

CBMC Results (ML-DSA-44)

Full Results (187 proofs)
Proof Status Current Previous Change
**TOTAL** 1817s 1968s -7.7%
polyvecl_pointwise_acc_montgomery_c 169s 199s -15%
poly_pointwise_montgomery_c 163s 189s -14%
sign_verify_internal 156s 173s -10%
rej_uniform_native 142s 153s -7%
mld_ct_memcmp 79s 90s -12%
mld_invntt_layer 66s 72s -8%
mld_attempt_signature_generation 55s 57s -4%
mld_ntt_layer 54s 57s -5%
polymat_permute_bitrev_to_custom 30s 28s +7%
polyvec_matrix_expand 27s 27s +0%
rej_uniform 21s 25s -16%
sign_keypair_internal 21s 21s +0%
fqmul 20s 23s -13%
poly_chknorm_c 20s 22s -9%
polyeta_unpack 19s 16s +19%
poly_uniform_eta_4x 18s 18s +0%
sign_pk_from_sk 17s 19s -11%
sign_signature_internal 17s 16s +6%
poly_uniform_4x 16s 17s -6%
keccakf1600x4_permute_native 15s 14s +7%
polyt0_unpack 14s 16s -12%
mld_check_pct 12s 10s +20%
mld_ntt_butterfly_block 12s 13s -8%
polyz_unpack_c 12s 12s +0%
rej_uniform_c 12s 18s -33%
keccak_absorb_once_x4 10s 13s -23%
poly_add 10s 12s -17%
mld_compute_pack_z 9s 8s +12%
poly_decompose_c 9s 9s +0%
polyvec_matrix_pointwise_montgomery 9s 8s +12%
polyveck_add 9s 8s +12%
keccak_absorb 8s 7s +14%
ntt_native_x86_64 8s 3s +167%
poly_power2round 8s 7s +14%
polyveck_pointwise_poly_montgomery 8s 7s +14%
polyveck_power2round 8s 12s -33%
polyveck_use_hint 8s 8s +0%
unpack_sk 8s 5s +60%
keccakf1600_permute 7s 10s -30%
poly_invntt_tomont_c 7s 7s +0%
polyveck_sub 7s 6s +17%
rej_eta_native 7s 8s -12%
sign_signature_pre_hash_shake256 7s 2s +250%
keccakf1600_permute_native 6s 7s -14%
mld_polyvecl_permute_bitrev_to_custom_native 6s 3s +100%
pointwise_acc_native_aarch64 6s 5s +20%
pointwise_acc_native_x86_64 6s 7s -14%
poly_caddq_c 6s 6s +0%
poly_chknorm_native 6s 4s +50%
polyvec_matrix_expand_serial 6s 11s -45%
polyveck_shiftl 6s 7s -14%
sign 6s 6s +0%
sign_signature 6s 4s +50%
sign_signature_pre_hash_internal 6s 4s +50%
intt_native_x86_64 5s 3s +67%
make_hint 5s 4s +25%
pack_sk_rho_key_tr_s2_t0 5s 3s +67%
pointwise_native_aarch64 5s 2s +150%
poly_challenge 5s 5s +0%
poly_decompose_native 5s 3s +67%
poly_reduce 5s 2s +150%
poly_uniform 5s 6s -17%
poly_uniform_gamma1 5s 2s +150%
poly_use_hint_c 5s 4s +25%
polyt0_pack 5s 4s +25%
polyt1_unpack 5s 3s +67%
polyveck_decompose 5s 5s +0%
polyveck_ntt 5s 4s +25%
polyvecl_permute_bitrev_to_custom 5s 3s +67%
polyvecl_uniform_gamma1 5s 2s +150%
polyvecl_unpack_z 5s 4s +25%
shake128x4_absorb_once 5s 3s +67%
shake256_finalize 5s 4s +25%
shake256_squeeze 5s 3s +67%
sign_open 5s 8s -38%
sign_verify 5s 5s +0%
sign_verify_extmu 5s 7s -29%
sign_verify_pre_hash_internal 5s 7s -29%
unpack_hints 5s 7s -29%
keccak_squeezeblocks_x4 4s 7s -43%
mld_ct_abs_i32 4s 3s +33%
mld_ct_get_optblocker_u32 4s 1s +300%
mld_ct_get_optblocker_u8 4s 3s +33%
mld_h 4s 4s +0%
mld_prepare_domain_separation_prefix 4s 8s -50%
mld_sample_s1_s2 4s 4s +0%
mld_sample_s1_s2_serial 4s 4s +0%
pack_pk 4s 6s -33%
pack_sk_s1 4s 4s +0%
poly_chknorm 4s 4s +0%
poly_make_hint 4s 2s +100%
poly_ntt_c 4s 4s +0%
poly_shiftl 4s 4s +0%
poly_use_hint_native_aarch64 4s - new
polyveck_caddq 4s 7s -43%
polyveck_chknorm 4s 6s -33%
polyveck_invntt_tomont 4s 6s -33%
polyveck_pack_t0 4s 4s +0%
polyvecl_ntt 4s 5s -20%
polyz_unpack 4s 3s +33%
rej_eta_c 4s 4s +0%
shake128x4_squeezeblocks 4s 2s +100%
fqscale 3s 2s +50%
keccak_f1600_x1_native_aarch64_v84a 3s 3s +0%
keccak_f1600_x4_native_aarch64_v84a 3s 3s +0%
keccak_finalize 3s 5s -40%
keccak_init 3s 3s +0%
keccakf1600_extract_bytes (big endian) 3s 1s +200%
keccakf1600_xor_bytes 3s 2s +50%
mld_ct_cmask_nonzero_u8 3s 2s +50%
mld_value_barrier_i64 3s 3s +0%
ntt_native_aarch64 3s 2s +50%
pack_sig_h_poly 3s 3s +0%
pointwise_native_x86_64 3s 2s +50%
poly_caddq 3s 3s +0%
poly_caddq_native 3s 4s -25%
poly_decompose 3s 3s +0%
poly_invntt_tomont 3s 1s +200%
poly_ntt 3s 3s +0%
poly_pointwise_montgomery_native 3s 2s +50%
poly_use_hint 3s 5s -40%
poly_use_hint_native 3s 2s +50%
polyeta_pack 3s 2s +50%
polyt1_pack 3s 2s +50%
polyveck_pack_w1 3s 4s -25%
polyveck_reduce 3s 5s -40%
polyveck_unpack_eta 3s 2s +50%
polyveck_unpack_t0 3s 5s -40%
polyvecl_chknorm 3s 7s -57%
polyvecl_pack_eta 3s 3s +0%
polyvecl_pointwise_acc_montgomery 3s 4s -25%
polyvecl_pointwise_acc_montgomery_native 3s 5s -40%
polyvecl_uniform_gamma1_serial 3s 4s -25%
polyz_pack 3s 4s -25%
polyz_unpack_native 3s 4s -25%
power2round 3s 4s -25%
rej_eta 3s 2s +50%
shake128_finalize 3s 1s +200%
shake128_release 3s 4s -25%
shake256 3s 3s +0%
sign_signature_extmu 3s 5s -40%
sign_verify_pre_hash_shake256 3s 3s +0%
sys_check_capability 3s 3s +0%
unpack_pk 3s 4s -25%
caddq 2s 3s -33%
decompose 2s 4s -50%
keccak_f1600_x1_native_aarch64 2s 4s -50%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid 2s 2s +0%
keccakf1600_xor_bytes (big endian) 2s 2s +0%
keccakf1600x4_permute 2s 5s -60%
mld_ct_cmask_neg_i32 2s 1s +100%
mld_ct_cmask_nonzero_u32 2s 2s +0%
mld_ct_get_optblocker_i64 2s 2s +0%
mld_ct_sel_int32 2s 1s +100%
mld_keccakf1600_extract_bytes 2s 2s +0%
montgomery_reduce 2s 1s +100%
pack_sig_z 2s 3s -33%
poly_caddq_native_aarch64 2s 4s -50%
poly_chknorm_native_aarch64 2s 6s -67%
poly_invntt_tomont_native 2s 3s -33%
poly_pointwise_montgomery 2s 5s -60%
poly_uniform_eta 2s 5s -60%
poly_uniform_gamma1_4x 2s 4s -50%
polyveck_pack_eta 2s 2s +0%
polyvecl_unpack_eta 2s 6s -67%
polyw1_pack 2s 2s +0%
reduce32 2s 2s +0%
shake128_absorb 2s 2s +0%
shake256_absorb 2s 2s +0%
shake256_init 2s 3s -33%
shake256_release 2s 4s -50%
shake256x4_absorb_once 2s 1s +100%
sign_keypair 2s 4s -50%
unpack_sig 2s 4s -50%
use_hint 2s 3s -33%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid 1s 1s +0%
keccak_squeeze 1s 1s +0%
keccakf1600x4_extract_bytes 1s 2s -50%
keccakf1600x4_xor_bytes 1s 2s -50%
mld_value_barrier_u32 1s 2s -50%
mld_value_barrier_u8 1s 3s -67%
pack_sig_c 1s 3s -67%
poly_ntt_native 1s 5s -80%
poly_sub 1s 4s -75%
shake128_init 1s 4s -75%
shake128_squeeze 1s 2s -50%
shake256x4_squeezeblocks 1s 4s -75%

Copy link
Copy Markdown
Contributor

@mkannwischer mkannwischer left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks @jakemas. Please add the CBMC proofs.

@jakemas jakemas force-pushed the hol-light-aarch64-use-hint branch 5 times, most recently from e7ca229 to e7a3a45 Compare April 14, 2026 07:44
@jakemas
Copy link
Copy Markdown
Contributor Author

jakemas commented Apr 14, 2026

Thanks @jakemas. Please add the CBMC proofs.

Thank you, added the harness as a single file with split functionality on GAMMA2

@jakemas jakemas force-pushed the hol-light-aarch64-use-hint branch 4 times, most recently from 1ea7797 to 817aa3d Compare April 17, 2026 10:22
Comment on lines +131 to +133
(\s. read PC s = word(pc + 0x110) /\
(!i. i < 256 ==>
val(read(memory :> bytes32(word_add b (word(4 * i)))) s) <= 15))
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This does not meet the bar of a functional correctness proof -- it merely provides bounds for the output?

Comment on lines +139 to +141
(\s. read PC s = word(pc + 0x130) /\
(!i. i < 256 ==>
val(read(memory :> bytes32(word_add b (word(4 * i)))) s) <= 43))
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Same here

Copy link
Copy Markdown
Contributor

@hanno-becker hanno-becker left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you @jakemas for working on this!

The specification seems far too weak to claim functional correctness? We seem to only say something about the bounds of the output.

@jakemas jakemas force-pushed the hol-light-aarch64-use-hint branch 7 times, most recently from 40ec688 to de080de Compare April 17, 2026 15:39
@jakemas jakemas force-pushed the hol-light-aarch64-use-hint branch 5 times, most recently from 8f24a11 to f491734 Compare April 17, 2026 17:58
Port the ML-DSA poly_use_hint HOL Light proofs from s2n-bignum to
mldsa-native for AArch64, covering both GAMMA2=(Q-1)/32 (l=5,7)
and GAMMA2=(Q-1)/88 (l=4) parameter sets.

Each proof includes correctness, subroutine form, and constant-time
with memory safety. Adds CBMC contracts for the assembly functions.

The correctness proof now establishes full functional equivalence
with the FIPS 204 UseHint specification (mldsa_use_hint_32_spec),
not just output bounds. Includes Barrett reduction equivalence proof
and shared infrastructure lemmas in common/mldsa_specs.ml.

Resolves: #486

Ported from s2n-bignum PRs #372 and #375.

Signed-off-by: Jake Massimo <jakemas@amazon.com>
@jakemas jakemas force-pushed the hol-light-aarch64-use-hint branch from f491734 to 2d5c44a Compare April 17, 2026 18:17
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants