Skip to content

HOL-Light: Add AArch64 poly_decompose_{32,88} correctness proofs#977

Draft
mkannwischer wants to merge 8 commits intomainfrom
hol-arm-decompose
Draft

HOL-Light: Add AArch64 poly_decompose_{32,88} correctness proofs#977
mkannwischer wants to merge 8 commits intomainfrom
hol-arm-decompose

Conversation

@mkannwischer
Copy link
Contributor

mkannwischer and others added 8 commits February 27, 2026 19:40
Add a HOL Light functional correctness proof for the aarch64 ML-DSA
function poly_chknorm, which checks whether any polynomial coefficient
has absolute value >= a given bound.

This commit includes:
- Functional correctness proof showing the assembly computes
  `bitval(?i. i < 256 /\ abs(ival(x i)) >= ival bound)`
- autogen support for generating aarch64 HOL Light assembly
- Update of s2n-bignum

Signed-off-by: Matthias J. Kannwischer <matthias@kannwischer.eu>
This also fixes the contract of poly_chknorm_native - it was incorrectly
modelled as only returning -1 or 0, never 1.
This was actually a proof gap - CBMC was not happy with 0U - ret. This commit
changes it to use mld_ct_cmask_nonzero_u32 - which has exactly the behavior
we want here.

Signed-off-by: Matthias J. Kannwischer <matthias@kannwischer.eu>
- Rewrite expressions during symbolic execution to keep
  system states readable
- Keep quantified propositions folded to the point where
  case-by-case analysis is needed
- Hoist all helper lemmas out of the main proof for
  better readability

Signed-off-by: Hanno Becker <beckphan@amazon.co.uk>
Add functional correctness proofs for polyz_unpack_17 and polyz_unpack_19.
Closely following the decompress proofs from mlkem-native:
pq-code-package/mlkem-native#1543

Signed-off-by: Matthias J. Kannwischer <matthias@kannwischer.eu>
Add CBMC contracts for native AArch64 polyz_unpack_17 and polyz_unpack_19
following corresponding HOL-Light specs.

Signed-off-by: Matthias J. Kannwischer <matthias@kannwischer.eu>
Move ival bounds out of CORRECT postcondition and derive them
as a corollary in SUBROUTINE_CORRECT.

Signed-off-by: Matthias J. Kannwischer <matthias@kannwischer.eu>
Add functional correctness proofs for the AArch64 assembly implementations
of poly_decompose for both GAMMA2 variants:
- poly_decompose_32 (GAMMA2 = (Q-1)/32 = 261888, used in ML-DSA-65/87)
- poly_decompose_88 (GAMMA2 = (Q-1)/88 = 95232, used in ML-DSA-44)

Signed-off-by: Matthias J. Kannwischer <matthias@kannwischer.eu>
Signed-off-by: Matthias J. Kannwischer <matthias@kannwischer.eu>
@oqs-bot
Copy link
Contributor

oqs-bot commented Feb 27, 2026

CBMC Results (ML-DSA-44)

Full Results (180 proofs)
Proof Status Current Previous Change
**TOTAL** 1996s 2036s -2.0%
sign_verify_internal 243s 247s -2%
mld_attempt_signature_generation 210s 220s -5%
polyvecl_pointwise_acc_montgomery_c 197s 206s -4%
rej_uniform_native 139s 138s +1%
poly_pointwise_montgomery_c 127s 133s -5%
mld_ct_memcmp 80s 80s +0%
mld_invntt_layer 51s 52s -2%
mld_ntt_layer 44s 45s -2%
keccak_squeezeblocks_x4 43s 42s +2%
sign_signature_internal 42s 45s -7%
poly_invntt_tomont_c 36s 39s -8%
rej_uniform 20s 21s -5%
fqmul 19s 17s +12%
poly_uniform_eta_4x 19s 15s +27%
rej_uniform_c 19s 17s +12%
polymat_permute_bitrev_to_custom 17s 19s -11%
mld_polyvecl_permute_bitrev_to_custom_native 15s 12s +25%
polyvec_matrix_expand 15s 15s +0%
mld_compute_t0_t1_tr_from_sk_components 14s 16s -12%
mld_ntt_butterfly_block 14s 11s +27%
poly_uniform_4x 14s 17s -18%
polyt0_unpack 14s 15s -7%
poly_chknorm_c 13s 12s +8%
keccakf1600x4_permute_native 12s 15s -20%
polyeta_unpack 12s 12s +0%
keccak_absorb_once_x4 10s 10s +0%
polyz_unpack_c 9s 12s -25%
keccakf1600_permute 8s 8s +0%
mld_check_pct 8s 6s +33%
polyvec_matrix_pointwise_montgomery 8s 5s +60%
polyveck_add 8s 7s +14%
polyveck_decompose 8s 8s +0%
keccakf1600_permute_native 7s 7s +0%
polyveck_pointwise_poly_montgomery 7s 6s +17%
polyveck_reduce 7s 6s +17%
polyvecl_chknorm 7s 5s +40%
polyvecl_ntt 7s 5s +40%
mld_h 6s 5s +20%
mld_sample_s1_s2_serial 6s 4s +50%
poly_uniform 6s 7s -14%
polyvec_matrix_expand_serial 6s 10s -40%
polyvecl_pointwise_acc_montgomery_native 6s 3s +100%
rej_eta 6s 3s +100%
sign 6s 3s +100%
sign_pk_from_sk 6s 8s -25%
decompose 5s 4s +25%
mld_compute_pack_z 5s 8s -38%
mld_ct_sel_int32 5s 1s +400%
poly_caddq_native_aarch64 5s 4s +25%
poly_make_hint 5s 4s +25%
poly_uniform_gamma1_4x 5s 3s +67%
polyt1_unpack 5s 5s +0%
polyveck_caddq 5s 5s +0%
polyveck_make_hint 5s 4s +25%
polyveck_use_hint 5s 6s -17%
polyvecl_uniform_gamma1 5s 3s +67%
polyw1_pack 5s 2s +150%
rej_eta_native 5s 4s +25%
sign_signature_pre_hash_internal 5s 3s +67%
sign_signature_pre_hash_shake256 5s 5s +0%
keccak_absorb 4s 6s -33%
keccakf1600_xor_bytes 4s 2s +100%
keccakf1600x4_permute 4s 2s +100%
mld_ct_cmask_nonzero_u8 4s 4s +0%
mld_keccakf1600_extract_bytes 4s 2s +100%
mld_value_barrier_u8 4s 2s +100%
pack_sk 4s 5s -20%
poly_challenge 4s 6s -33%
poly_chknorm 4s 3s +33%
poly_chknorm_native 4s 4s +0%
poly_decompose 4s 3s +33%
poly_decompose_c 4s 5s -20%
poly_decompose_native 4s 6s -33%
poly_invntt_tomont 4s 3s +33%
poly_ntt_c 4s 3s +33%
poly_ntt_native 4s 3s +33%
poly_pointwise_montgomery_native 4s 3s +33%
poly_reduce 4s 3s +33%
poly_uniform_eta 4s 4s +0%
polyveck_invntt_tomont 4s 5s -20%
polyveck_ntt 4s 6s -33%
polyveck_pack_eta 4s 3s +33%
polyveck_power2round 4s 6s -33%
polyveck_shiftl 4s 5s -20%
polyveck_unpack_eta 4s 3s +33%
polyveck_unpack_t0 4s 3s +33%
polyvecl_pack_eta 4s 4s +0%
polyvecl_permute_bitrev_to_custom 4s 6s -33%
polyvecl_unpack_eta 4s 4s +0%
polyz_pack 4s 3s +33%
polyz_unpack 4s 3s +33%
polyz_unpack_17_native_aarch64 4s - new
polyz_unpack_native 4s 2s +100%
reduce32 4s 5s -20%
shake128x4_absorb_once 4s 2s +100%
shake256x4_absorb_once 4s 1s +300%
sign_keypair 4s 2s +100%
sign_keypair_internal 4s 4s +0%
sign_open 4s 3s +33%
unpack_hints 4s 8s -50%
unpack_sig 4s 4s +0%
fqscale 3s 2s +50%
intt_native_x86_64 3s 4s -25%
keccak_squeeze 3s 3s +0%
keccakf1600x4_extract_bytes 3s 3s +0%
mld_ct_abs_i32 3s 2s +50%
mld_ct_cmask_neg_i32 3s 2s +50%
mld_ct_get_optblocker_i64 3s 3s +0%
mld_sample_s1_s2 3s 6s -50%
mld_value_barrier_i64 3s 3s +0%
ntt_native_x86_64 3s 5s -40%
poly_add 3s 2s +50%
poly_caddq_c 3s 2s +50%
poly_caddq_native 3s 3s +0%
poly_chknorm_native_aarch64 3s - new
poly_decompose_88_native_aarch64 3s - new
poly_invntt_tomont_native 3s 3s +0%
poly_ntt 3s 3s +0%
poly_pointwise_montgomery 3s 4s -25%
poly_power2round 3s 2s +50%
polyeta_pack 3s 2s +50%
polyt0_pack 3s 6s -50%
polyt1_pack 3s 6s -50%
polyveck_chknorm 3s 3s +0%
polyveck_pack_t0 3s 5s -40%
polyveck_pack_w1 3s 3s +0%
polyveck_sub 3s 4s -25%
polyvecl_uniform_gamma1_serial 3s 2s +50%
polyz_unpack_19_native_aarch64 3s - new
rej_eta_c 3s 2s +50%
shake128_finalize 3s 4s -25%
shake128_squeeze 3s 2s +50%
shake256_init 3s 3s +0%
shake256_release 3s 3s +0%
sign_signature_extmu 3s 3s +0%
sign_verify_pre_hash_internal 3s 4s -25%
sign_verify_pre_hash_shake256 3s 4s -25%
unpack_sk 3s 5s -40%
use_hint 3s 5s -40%
caddq 2s 2s +0%
keccakf1600_extract_bytes (big endian) 2s 2s +0%
keccakf1600_xor_bytes (big endian) 2s 2s +0%
make_hint 2s 1s +100%
mld_ct_get_optblocker_u32 2s 3s -33%
mld_ct_get_optblocker_u8 2s 3s -33%
mld_prepare_domain_separation_prefix 2s 4s -50%
mld_value_barrier_u32 2s 2s +0%
montgomery_reduce 2s 4s -50%
pack_pk 2s 2s +0%
pack_sig_c_h 2s 3s -33%
pack_sig_z 2s 2s +0%
poly_caddq 2s 2s +0%
poly_decompose_32_native_aarch64 2s - new
poly_uniform_gamma1 2s 2s +0%
poly_use_hint 2s 2s +0%
poly_use_hint_c 2s 6s -67%
poly_use_hint_native 2s 4s -50%
polyvecl_pointwise_acc_montgomery 2s 5s -60%
polyvecl_unpack_z 2s 5s -60%
power2round 2s 3s -33%
shake128_init 2s 3s -33%
shake256 2s 3s -33%
shake256_absorb 2s 2s +0%
shake256_finalize 2s 2s +0%
shake256_squeeze 2s 3s -33%
shake256x4_squeezeblocks 2s 3s -33%
sign_signature 2s 2s +0%
sign_verify 2s 3s -33%
sign_verify_extmu 2s 5s -60%
sys_check_capability 2s 2s +0%
keccak_finalize 1s 3s -67%
keccak_init 1s 3s -67%
keccakf1600x4_xor_bytes 1s 2s -50%
mld_ct_cmask_nonzero_u32 1s 2s -50%
poly_shiftl 1s 1s +0%
poly_sub 1s 6s -83%
shake128_absorb 1s 2s -50%
shake128_release 1s 4s -75%
shake128x4_squeezeblocks 1s 4s -75%
unpack_pk 1s 3s -67%

@oqs-bot
Copy link
Contributor

oqs-bot commented Feb 27, 2026

CBMC Results (ML-DSA-65)

Full Results (180 proofs)
Proof Status Current Previous Change
**TOTAL** 2257s 2495s -9.5%
polyvecl_pointwise_acc_montgomery_c 215s 270s -20%
mld_attempt_signature_generation 184s 219s -16%
sign_verify_internal 175s 193s -9%
polyvec_matrix_expand 145s 158s -8%
rej_uniform_native 139s 161s -14%
poly_pointwise_montgomery_c 129s 166s -22%
mld_invntt_layer 116s 132s -12%
mld_ct_memcmp 81s 87s -7%
polyvec_matrix_expand_serial 62s 70s -11%
sign_signature_internal 52s 53s -2%
mld_ntt_layer 44s 49s -10%
keccak_squeezeblocks_x4 42s 46s -9%
mld_compute_t0_t1_tr_from_sk_components 27s 26s +4%
fqmul 20s 22s -9%
rej_uniform 20s 22s -9%
polymat_permute_bitrev_to_custom 19s 23s -17%
rej_uniform_c 18s 20s -10%
poly_chknorm_c 16s 19s -16%
polyveck_decompose 16s 19s -16%
poly_uniform_eta_4x 15s 17s -12%
polyt0_unpack 14s 16s -12%
polyvec_matrix_pointwise_montgomery 14s 20s -30%
polyveck_use_hint 14s 15s -7%
keccakf1600x4_permute_native 13s 15s -13%
mld_ntt_butterfly_block 13s 15s -13%
mld_polyvecl_permute_bitrev_to_custom_native 13s 14s -7%
poly_uniform_4x 13s 16s -19%
polyveck_add 12s 11s +9%
keccak_absorb_once_x4 10s 12s -17%
polyveck_caddq 10s 10s +0%
polyveck_invntt_tomont 10s 9s +11%
sign 10s 10s +0%
mld_check_pct 9s 12s -25%
polyveck_power2round 9s 11s -18%
keccakf1600_permute_native 8s 10s -20%
poly_invntt_tomont_c 8s 11s -27%
polyveck_ntt 8s 10s -20%
polyveck_reduce 8s 8s +0%
keccakf1600_permute 7s 10s -30%
mld_prepare_domain_separation_prefix 7s 6s +17%
poly_challenge 7s 5s +40%
poly_decompose_c 7s 7s +0%
polyveck_shiftl 7s 8s -12%
polyveck_sub 7s 7s +0%
polyvecl_ntt 7s 8s -12%
sign_pk_from_sk 7s 9s -22%
unpack_sk 7s 4s +75%
mld_compute_pack_z 6s 6s +0%
mld_sample_s1_s2 6s 3s +100%
pack_sk 6s 2s +200%
poly_add 6s 5s +20%
poly_decompose_native 6s 2s +200%
poly_invntt_tomont 6s 3s +100%
poly_use_hint_c 6s 4s +50%
polyveck_chknorm 6s 6s +0%
polyvecl_permute_bitrev_to_custom 6s 4s +50%
power2round 6s 4s +50%
sign_keypair_internal 6s 3s +100%
sign_signature_extmu 6s 3s +100%
unpack_hints 6s 7s -14%
keccak_absorb 5s 5s +0%
keccakf1600x4_permute 5s 3s +67%
mld_sample_s1_s2_serial 5s 4s +25%
poly_caddq_native 5s 3s +67%
poly_ntt_native 5s 3s +67%
poly_sub 5s 4s +25%
poly_uniform_gamma1 5s 5s +0%
polyeta_unpack 5s 5s +0%
polyveck_make_hint 5s 6s -17%
polyveck_pack_t0 5s 2s +150%
polyveck_pointwise_poly_montgomery 5s 6s -17%
polyvecl_pointwise_acc_montgomery 5s 1s +400%
polyvecl_pointwise_acc_montgomery_native 5s 4s +25%
polyvecl_uniform_gamma1 5s 4s +25%
polyz_pack 5s 3s +67%
shake128_init 5s 3s +67%
sign_verify_pre_hash_shake256 5s 2s +150%
caddq 4s 3s +33%
keccak_squeeze 4s 3s +33%
keccakf1600_extract_bytes (big endian) 4s 3s +33%
make_hint 4s 3s +33%
mld_ct_cmask_nonzero_u32 4s 3s +33%
mld_ct_cmask_nonzero_u8 4s 3s +33%
mld_h 4s 5s -20%
pack_pk 4s 4s +0%
poly_caddq_c 4s 3s +33%
poly_caddq_native_aarch64 4s 2s +100%
poly_chknorm 4s 3s +33%
poly_chknorm_native_aarch64 4s - new
poly_invntt_tomont_native 4s 4s +0%
poly_make_hint 4s 4s +0%
poly_pointwise_montgomery 4s 2s +100%
poly_pointwise_montgomery_native 4s 4s +0%
poly_uniform 4s 5s -20%
poly_use_hint_native 4s 4s +0%
polyveck_pack_eta 4s 4s +0%
polyveck_pack_w1 4s 4s +0%
polyvecl_uniform_gamma1_serial 4s 2s +100%
rej_eta_c 4s 3s +33%
shake128_finalize 4s 3s +33%
shake256x4_absorb_once 4s 2s +100%
sign_keypair 4s 4s +0%
sign_open 4s 3s +33%
sign_signature_pre_hash_shake256 4s 5s -20%
sign_verify_extmu 4s 2s +100%
sign_verify_pre_hash_internal 4s 2s +100%
sys_check_capability 4s 5s -20%
use_hint 4s 4s +0%
decompose 3s 4s -25%
intt_native_x86_64 3s 4s -25%
keccakf1600x4_xor_bytes 3s 3s +0%
mld_ct_cmask_neg_i32 3s 3s +0%
mld_ct_get_optblocker_i64 3s 4s -25%
mld_ct_get_optblocker_u8 3s 3s +0%
mld_ct_sel_int32 3s 2s +50%
mld_keccakf1600_extract_bytes 3s 2s +50%
ntt_native_x86_64 3s 3s +0%
poly_caddq 3s 2s +50%
poly_chknorm_native 3s 5s -40%
poly_decompose 3s 6s -50%
poly_decompose_32_native_aarch64 3s - new
poly_ntt_c 3s 2s +50%
poly_power2round 3s 4s -25%
poly_uniform_eta 3s 5s -40%
poly_uniform_gamma1_4x 3s 4s -25%
poly_use_hint 3s 3s +0%
polyeta_pack 3s 3s +0%
polyt0_pack 3s 5s -40%
polyveck_unpack_eta 3s 5s -40%
polyvecl_chknorm 3s 6s -50%
polyvecl_pack_eta 3s 6s -50%
polyw1_pack 3s 3s +0%
polyz_unpack 3s 2s +50%
polyz_unpack_17_native_aarch64 3s - new
polyz_unpack_c 3s 6s -50%
reduce32 3s 3s +0%
shake128_squeeze 3s 1s +200%
shake128x4_absorb_once 3s 4s -25%
shake128x4_squeezeblocks 3s 3s +0%
shake256_finalize 3s 1s +200%
shake256_squeeze 3s 4s -25%
shake256x4_squeezeblocks 3s 1s +200%
sign_signature 3s 4s -25%
sign_signature_pre_hash_internal 3s 4s -25%
sign_verify 3s 3s +0%
keccak_finalize 2s 3s -33%
keccak_init 2s 2s +0%
keccakf1600_xor_bytes (big endian) 2s 4s -50%
keccakf1600x4_extract_bytes 2s 3s -33%
mld_ct_abs_i32 2s 3s -33%
mld_ct_get_optblocker_u32 2s 1s +100%
mld_value_barrier_i64 2s 1s +100%
mld_value_barrier_u32 2s 4s -50%
mld_value_barrier_u8 2s 1s +100%
montgomery_reduce 2s 3s -33%
pack_sig_c_h 2s 4s -50%
pack_sig_z 2s 4s -50%
poly_decompose_88_native_aarch64 2s - new
poly_ntt 2s 4s -50%
poly_reduce 2s 5s -60%
polyt1_pack 2s 3s -33%
polyt1_unpack 2s 5s -60%
polyveck_unpack_t0 2s 3s -33%
polyvecl_unpack_eta 2s 3s -33%
polyvecl_unpack_z 2s 3s -33%
polyz_unpack_19_native_aarch64 2s - new
rej_eta 2s 3s -33%
rej_eta_native 2s 4s -50%
shake128_absorb 2s 2s +0%
shake128_release 2s 5s -60%
shake256 2s 1s +100%
shake256_absorb 2s 2s +0%
shake256_init 2s 2s +0%
shake256_release 2s 2s +0%
unpack_pk 2s 4s -50%
fqscale 1s 3s -67%
keccakf1600_xor_bytes 1s 2s -50%
poly_shiftl 1s 1s +0%
polyz_unpack_native 1s 3s -67%
unpack_sig 1s 2s -50%

@oqs-bot
Copy link
Contributor

oqs-bot commented Feb 27, 2026

CBMC Results (ML-DSA-87)

Full Results (180 proofs)
Proof Status Current Previous Change
**TOTAL** 2639s 2686s -1.7%
sign_verify_internal 381s 383s -1%
mld_attempt_signature_generation 248s 250s -1%
polyvecl_pointwise_acc_montgomery_c 193s 204s -5%
polyvec_matrix_expand 162s 161s +1%
rej_uniform_native 151s 152s -1%
poly_pointwise_montgomery_c 150s 160s -6%
mld_invntt_layer 121s 125s -3%
polyvec_matrix_expand_serial 113s 111s +2%
mld_ct_memcmp 85s 92s -8%
sign_signature_internal 48s 49s -2%
mld_ntt_layer 46s 50s -8%
keccak_squeezeblocks_x4 43s 45s -4%
polymat_permute_bitrev_to_custom 27s 25s +8%
mld_compute_t0_t1_tr_from_sk_components 26s 28s -7%
rej_uniform 24s 20s +20%
rej_uniform_c 22s 20s +10%
fqmul 19s 21s -10%
poly_chknorm_c 17s 20s -15%
poly_uniform_eta_4x 17s 16s +6%
polyt0_unpack 17s 18s -6%
poly_uniform_4x 16s 15s +7%
polyvec_matrix_pointwise_montgomery 15s 17s -12%
polyeta_unpack 14s 14s +0%
polyveck_add 14s 16s -12%
polyveck_power2round 14s 12s +17%
mld_ntt_butterfly_block 13s 12s +8%
keccakf1600x4_permute_native 12s 13s -8%
polyvecl_ntt 11s 8s +38%
keccak_absorb_once_x4 10s 11s -9%
keccakf1600_permute_native 10s 7s +43%
poly_decompose_c 10s 9s +11%
poly_invntt_tomont_c 10s 9s +11%
polyveck_reduce 10s 10s +0%
polyveck_use_hint 10s 12s -17%
polyveck_caddq 9s 10s -10%
polyveck_chknorm 9s 10s -10%
polyveck_shiftl 9s 7s +29%
sign 9s 8s +12%
keccak_absorb 8s 6s +33%
mld_check_pct 8s 8s +0%
mld_h 8s 6s +33%
mld_polyvecl_permute_bitrev_to_custom_native 8s 8s +0%
sign_pk_from_sk 8s 9s -11%
keccakf1600_permute 7s 8s -12%
mld_sample_s1_s2 7s 6s +17%
polyveck_decompose 7s 7s +0%
polyveck_invntt_tomont 7s 8s -12%
polyveck_ntt 7s 8s -12%
intt_native_x86_64 6s 4s +50%
mld_compute_pack_z 6s 8s -25%
mld_sample_s1_s2_serial 6s 6s +0%
poly_caddq_native 6s 3s +100%
poly_challenge 6s 3s +100%
poly_decompose_32_native_aarch64 6s - new
polyt1_pack 6s 6s +0%
polyveck_pointwise_poly_montgomery 6s 7s -14%
polyveck_sub 6s 6s +0%
polyvecl_chknorm 6s 3s +100%
polyvecl_unpack_eta 6s 6s +0%
sign_signature_extmu 6s 3s +100%
sign_verify_pre_hash_internal 6s 2s +200%
unpack_sk 6s 5s +20%
caddq 5s 3s +67%
keccakf1600x4_permute 5s 3s +67%
montgomery_reduce 5s 3s +67%
poly_invntt_tomont_native 5s 5s +0%
poly_uniform_eta 5s 5s +0%
polyveck_pack_eta 5s 3s +67%
polyveck_unpack_t0 5s 5s +0%
polyvecl_pack_eta 5s 4s +25%
polyvecl_unpack_z 5s 5s +0%
sign_open 5s 2s +150%
sign_signature 5s 5s +0%
sign_signature_pre_hash_internal 5s 3s +67%
sign_signature_pre_hash_shake256 5s 3s +67%
unpack_hints 5s 8s -38%
keccakf1600_xor_bytes (big endian) 4s 3s +33%
mld_ct_cmask_nonzero_u32 4s 6s -33%
mld_prepare_domain_separation_prefix 4s 4s +0%
ntt_native_x86_64 4s 3s +33%
pack_pk 4s 3s +33%
pack_sig_c_h 4s 3s +33%
pack_sig_z 4s 3s +33%
poly_chknorm 4s 4s +0%
poly_chknorm_native 4s 4s +0%
poly_decompose_88_native_aarch64 4s - new
poly_ntt_c 4s 3s +33%
poly_ntt_native 4s 4s +0%
poly_power2round 4s 4s +0%
poly_uniform_gamma1_4x 4s 4s +0%
polyt0_pack 4s 4s +0%
polyveck_pack_w1 4s 4s +0%
polyveck_unpack_eta 4s 3s +33%
polyvecl_pointwise_acc_montgomery 4s 4s +0%
polyvecl_pointwise_acc_montgomery_native 4s 5s -20%
polyvecl_uniform_gamma1 4s 5s -20%
polyvecl_uniform_gamma1_serial 4s 7s -43%
polyz_unpack_c 4s 6s -33%
polyz_unpack_native 4s 5s -20%
rej_eta 4s 4s +0%
rej_eta_native 4s 4s +0%
sign_keypair 4s 2s +100%
sign_keypair_internal 4s 7s -43%
sign_verify 4s 5s -20%
sign_verify_extmu 4s 5s -20%
keccak_finalize 3s 2s +50%
keccak_init 3s 2s +50%
keccakf1600_extract_bytes (big endian) 3s 3s +0%
keccakf1600_xor_bytes 3s 2s +50%
make_hint 3s 2s +50%
mld_ct_abs_i32 3s 4s -25%
mld_ct_cmask_nonzero_u8 3s 3s +0%
mld_ct_sel_int32 3s 3s +0%
mld_keccakf1600_extract_bytes 3s 3s +0%
mld_value_barrier_i64 3s 4s -25%
mld_value_barrier_u8 3s 2s +50%
poly_add 3s 5s -40%
poly_caddq 3s 3s +0%
poly_caddq_native_aarch64 3s 4s -25%
poly_invntt_tomont 3s 4s -25%
poly_make_hint 3s 2s +50%
poly_ntt 3s 3s +0%
poly_reduce 3s 2s +50%
poly_shiftl 3s 4s -25%
poly_sub 3s 3s +0%
poly_uniform 3s 4s -25%
poly_uniform_gamma1 3s 4s -25%
poly_use_hint_native 3s 4s -25%
polyeta_pack 3s 6s -50%
polyveck_make_hint 3s 5s -40%
polyveck_pack_t0 3s 2s +50%
polyvecl_permute_bitrev_to_custom 3s 1s +200%
polyw1_pack 3s 3s +0%
power2round 3s 3s +0%
rej_eta_c 3s 5s -40%
shake128_init 3s 1s +200%
shake128_release 3s 3s +0%
shake128x4_absorb_once 3s 2s +50%
shake256_release 3s 2s +50%
shake256x4_squeezeblocks 3s 5s -40%
unpack_pk 3s 3s +0%
decompose 2s 3s -33%
fqscale 2s 3s -33%
keccak_squeeze 2s 5s -60%
keccakf1600x4_xor_bytes 2s 1s +100%
mld_ct_cmask_neg_i32 2s 4s -50%
mld_ct_get_optblocker_i64 2s 1s +100%
mld_ct_get_optblocker_u32 2s 2s +0%
mld_ct_get_optblocker_u8 2s 3s -33%
pack_sk 2s 4s -50%
poly_chknorm_native_aarch64 2s - new
poly_decompose 2s 4s -50%
poly_decompose_native 2s 4s -50%
poly_pointwise_montgomery 2s 3s -33%
poly_pointwise_montgomery_native 2s 3s -33%
poly_use_hint 2s 3s -33%
polyt1_unpack 2s 4s -50%
polyz_pack 2s 4s -50%
polyz_unpack 2s 3s -33%
polyz_unpack_17_native_aarch64 2s - new
polyz_unpack_19_native_aarch64 2s - new
reduce32 2s 2s +0%
shake128_finalize 2s 3s -33%
shake128_squeeze 2s 3s -33%
shake128x4_squeezeblocks 2s 4s -50%
shake256 2s 3s -33%
shake256_finalize 2s 4s -50%
shake256_squeeze 2s 4s -50%
shake256x4_absorb_once 2s 2s +0%
sign_verify_pre_hash_shake256 2s 4s -50%
sys_check_capability 2s 4s -50%
unpack_sig 2s 2s +0%
use_hint 2s 3s -33%
keccakf1600x4_extract_bytes 1s 2s -50%
mld_value_barrier_u32 1s 1s +0%
poly_caddq_c 1s 4s -75%
poly_use_hint_c 1s 3s -67%
shake128_absorb 1s 4s -75%
shake256_absorb 1s 3s -67%
shake256_init 1s 3s -67%

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.

HOL-Light: Prove AArch64 poly_decompose

3 participants