x86_64 + HOL-Light: Replace poly_use_hint AVX2 intrinsics with hand-written assembly and HOL-Light proofs

[jakemas]

Resolves #484
Resolves #485

This is the only remaining intrinsic to be picked up, and proved. I know @willieyz has a PR in flight, I've taken it, brought it up to main, fixed conflicts, adjusted asm, fixed #1074/#1076, added hol-light CBMC proofs and supporting infra.

Replaces the poly_use_hint_32 (ML-DSA-65/87) and poly_use_hint_88 (ML-DSA-44) AVX2 intrinsics with hand-written assembly, and adds HOL-Light functional-correctness and constant-time / memory-safety proofs for both x86_64 routines.

Builds on @willieyz's original assembly (#940). Changes from that code:

• Reworked to the in-place 2-arg signature (int32_t *a, const int32_t *h) instead of the 3-arg out-of-place form.
• Renamed to the _asm convention and wired the native backend to the assembly (not the .c intrinsic).
• Fixed the poly_use_hint_32 compare bound 0x15BA900 (87·GAMMA2) → 0x7BE100 (31·GAMMA2), per Potentially wrong value for q_bound in poly_use_hint_32_avx2.c #1074/x86_64: Fix wrong qbound 87*q -> 31*q in use_hint_32 #1076.

Performance

The performance degradation we previously saw in willieyz's PR is fixed.
poly_use_hint component benchmark, median cycles on Intel Xeon Platinum 8175M, OPT=1 CYCLES=PMU:

Variant	AVX2 intrinsics (baseline, main)	@willieyz asm (#940)	Hand-written AVX2 asm (this PR)
use_hint_32 (ML-DSA-65/87)	~314	~300	~303
use_hint_88 (ML-DSA-44)	~377	~351	~356

[oqs-bot]

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

Full Results (206 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	✅	1661s	1495s	+11.1%
mld_invntt_layer	✅	180s	163s	+10%
poly_pointwise_montgomery_c	✅	149s	133s	+12%
rej_uniform_native	✅	130s	119s	+9%
polyvec_matrix_pointwise_montgomery_yvec	✅	88s	81s	+9%
mld_ct_memcmp	✅	73s	65s	+12%
mld_ntt_layer	✅	48s	41s	+17%
fqmul	✅	44s	39s	+13%
polyveck_chknorm	✅	41s	36s	+14%
mld_attempt_signature_generation	✅	24s	21s	+14%
keccakf1600x4_permute_native	✅	23s	22s	+5%
mld_ntt_butterfly_block	✅	23s	21s	+10%
poly_chknorm_c	✅	21s	17s	+24%
polyt0_unpack	✅	17s	17s	+0%
rej_uniform_c	✅	17s	16s	+6%
sign_verify_internal	✅	17s	16s	+6%
polyveck_decompose	✅	16s	16s	+0%
polyvecl_chknorm	✅	16s	14s	+14%
mld_check_pct	✅	15s	14s	+7%
poly_add	✅	12s	11s	+9%
poly_uniform_eta_4x	✅	12s	13s	-8%
keccak_absorb_once_x4	✅	10s	8s	+25%
polyvecl_ntt	✅	10s	6s	+67%
rej_uniform	✅	10s	7s	+43%
poly_invntt_tomont_c	✅	9s	10s	-10%
polyvec_matrix_pointwise_montgomery_row	✅	9s	11s	-18%
poly_power2round	✅	8s	7s	+14%
polyveck_caddq	✅	8s	7s	+14%
sign	✅	8s	6s	+33%
compute_pack_t0_t1	✅	7s	7s	+0%
mld_keccakf1600_permute_c	✅	7s	6s	+17%
mld_sample_s1_s2_serial	✅	7s	3s	+133%
pointwise_acc_native_aarch64	✅	7s	6s	+17%
polyveck_invntt_tomont	✅	7s	5s	+40%
sign_verify_extmu	✅	7s	4s	+75%
keccak_absorb	✅	6s	5s	+20%
keccakf1600_permute_native	✅	6s	5s	+20%
mld_compute_pack_z	✅	6s	6s	+0%
mld_h	✅	6s	3s	+100%
poly_caddq_c	✅	6s	3s	+100%
poly_invntt_tomont_native	✅	6s	2s	+200%
poly_shiftl	✅	6s	4s	+50%
polyveck_unpack_eta	✅	6s	2s	+200%
power2round	✅	6s	2s	+200%
rej_eta_native	✅	6s	4s	+50%
sign_pk_from_sk	✅	6s	7s	-14%
sign_verify	✅	6s	2s	+200%
decompose	✅	5s	2s	+150%
intt_native_x86_64	✅	5s	5s	+0%
keccak_squeeze	✅	5s	4s	+25%
keccak_squeezeblocks_x4	✅	5s	5s	+0%
keccakf1600x4_permute	✅	5s	3s	+67%
mld_polymat_expand_entry	✅	5s	3s	+67%
pointwise_acc_native_x86_64	✅	5s	6s	-17%
poly_chknorm_native_x86_64	✅	5s	2s	+150%
poly_permute_bitrev_to_custom_optional	✅	5s	4s	+25%
poly_pointwise_montgomery	✅	5s	2s	+150%
poly_reduce	✅	5s	3s	+67%
poly_uniform	✅	5s	3s	+67%
poly_uniform_gamma1_4x	✅	5s	2s	+150%
polyeta_unpack	✅	5s	7s	-29%
polyveck_pack_eta	✅	5s	2s	+150%
polyvecl_uniform_gamma1_serial	✅	5s	3s	+67%
polyz_unpack_c	✅	5s	6s	-17%
rej_eta_c	✅	5s	2s	+150%
rej_uniform_eta_native_aarch64	✅	5s	3s	+67%
shake256x4_absorb_once	✅	5s	1s	+400%
sign_keypair	✅	5s	6s	-17%
sign_open	✅	5s	5s	+0%
sign_signature_extmu	✅	5s	3s	+67%
sign_signature_pre_hash_shake256	✅	5s	6s	-17%
sign_verify_pre_hash_internal	✅	5s	2s	+150%
intt_native_aarch64	✅	4s	4s	+0%
keccakf1600_extract_bytes (big endian)	✅	4s	1s	+300%
keccakf1600x4_xor_bytes_native	✅	4s	1s	+300%
mld_ct_abs_i32	✅	4s	3s	+33%
mld_ct_cmask_neg_i32	✅	4s	4s	+0%
mld_keccakf1600x4_xor_bytes_c	✅	4s	1s	+300%
mld_sample_s1_s2	✅	4s	4s	+0%
ntt_native_aarch64	✅	4s	3s	+33%
ntt_native_x86_64	✅	4s	4s	+0%
pack_sig_c	✅	4s	3s	+33%
pack_sk_rho_key_tr_s2	✅	4s	2s	+100%
pointwise_native_aarch64	✅	4s	2s	+100%
poly_caddq_native	✅	4s	3s	+33%
poly_caddq_native_aarch64	✅	4s	4s	+0%
poly_challenge	✅	4s	5s	-20%
poly_decompose	✅	4s	2s	+100%
poly_decompose_88_native_aarch64	✅	4s	2s	+100%
poly_decompose_native	✅	4s	3s	+33%
poly_use_hint_native_x86_64	✅	4s	-	new
polyeta_pack	✅	4s	2s	+100%
polyt0_pack	✅	4s	6s	-33%
polyt1_unpack	✅	4s	3s	+33%
polyveck_ntt	✅	4s	4s	+0%
polyveck_reduce	✅	4s	4s	+0%
polyvecl_pointwise_acc_montgomery_native	✅	4s	3s	+33%
polyvecl_unpack_eta	✅	4s	3s	+33%
polyvecl_unpack_z	✅	4s	3s	+33%
shake256_finalize	✅	4s	2s	+100%
sign_keypair_internal	✅	4s	4s	+0%
sign_signature	✅	4s	3s	+33%
sign_signature_internal	✅	4s	4s	+0%
sign_signature_pre_hash_internal	✅	4s	4s	+0%
unpack_pk_t1	✅	4s	1s	+300%
unpack_sk_s2hat	✅	4s	2s	+100%
caddq	✅	3s	3s	+0%
fqscale	✅	3s	4s	-25%
keccak_f1600_x1_native_aarch64	✅	3s	2s	+50%
keccakf1600_xor_bytes	✅	3s	2s	+50%
keccakf1600x4_extract_bytes	✅	3s	2s	+50%
keccakf1600x4_extract_bytes_native	✅	3s	3s	+0%
mld_ct_get_optblocker_i64	✅	3s	2s	+50%
mld_ct_get_optblocker_u8	✅	3s	3s	+0%
mld_keccakf1600_extract_bytes	✅	3s	3s	+0%
mld_prepare_domain_separation_prefix	✅	3s	3s	+0%
mld_value_barrier_i64	✅	3s	3s	+0%
mld_value_barrier_u32	✅	3s	2s	+50%
mld_value_barrier_u8	✅	3s	2s	+50%
montgomery_reduce	✅	3s	5s	-40%
nttunpack_native_x86_64	✅	3s	3s	+0%
pack_sig_h	✅	3s	4s	-25%
pack_sig_z	✅	3s	3s	+0%
pack_sk_s1	✅	3s	4s	-25%
pointwise_native_x86_64	✅	3s	3s	+0%
poly_caddq	✅	3s	3s	+0%
poly_caddq_native_x86_64	✅	3s	2s	+50%
poly_chknorm_native	✅	3s	4s	-25%
poly_decompose_32_native_aarch64	✅	3s	1s	+200%
poly_decompose_c	✅	3s	5s	-40%
poly_ntt	✅	3s	3s	+0%
poly_ntt_c	✅	3s	2s	+50%
poly_sub	✅	3s	3s	+0%
poly_uniform_4x	✅	3s	3s	+0%
poly_uniform_eta	✅	3s	2s	+50%
poly_uniform_gamma1	✅	3s	3s	+0%
poly_use_hint_c	✅	3s	3s	+0%
poly_use_hint_native	✅	3s	4s	-25%
poly_use_hint_native_aarch64	✅	3s	3s	+0%
polyt1_pack	✅	3s	4s	-25%
polyvec_matrix_expand	✅	3s	1s	+200%
polyvec_matrix_expand_serial	✅	3s	3s	+0%
polyvecl_pointwise_acc_montgomery_c	✅	3s	3s	+0%
polyvecl_uniform_gamma1	✅	3s	2s	+50%
polyw1_pack_88	✅	3s	3s	+0%
polyz_unpack	✅	3s	3s	+0%
polyz_unpack_17_native_aarch64	✅	3s	4s	-25%
polyz_unpack_native	✅	3s	2s	+50%
polyz_unpack_native_x86_64	✅	3s	3s	+0%
reduce32	✅	3s	3s	+0%
rej_uniform_native_aarch64	✅	3s	3s	+0%
shake128_absorb	✅	3s	3s	+0%
shake128_init	✅	3s	2s	+50%
shake128_squeeze	✅	3s	2s	+50%
shake256_absorb	✅	3s	3s	+0%
shake256_init	✅	3s	2s	+50%
shake256_squeeze	✅	3s	2s	+50%
sk_s1hat_get_poly	✅	3s	3s	+0%
sk_s2hat_get_poly	✅	3s	1s	+200%
unpack_sk	✅	3s	3s	+0%
unpack_sk_s1hat	✅	3s	4s	-25%
yvec_get_poly	✅	3s	3s	+0%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	2s	2s	+0%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	2s	3s	-33%
keccak_f1600_x4_native_avx2	✅	2s	2s	+0%
keccak_finalize	✅	2s	4s	-50%
keccak_init	✅	2s	3s	-33%
keccakf1600_xor_bytes (big endian)	✅	2s	3s	-33%
keccakf1600x4_xor_bytes	✅	2s	3s	-33%
make_hint	✅	2s	4s	-50%
mld_ct_cmask_nonzero_u32	✅	2s	4s	-50%
mld_ct_cmask_nonzero_u8	✅	2s	4s	-50%
mld_ct_get_optblocker_u32	✅	2s	4s	-50%
mld_keccakf1600x4_extract_bytes_c	✅	2s	4s	-50%
poly_chknorm	✅	2s	5s	-60%
poly_chknorm_native_aarch64	✅	2s	4s	-50%
poly_invntt_tomont	✅	2s	4s	-50%
poly_ntt_native	✅	2s	4s	-50%
poly_permute_bitrev_to_custom_optional_native	✅	2s	5s	-60%
poly_use_hint	✅	2s	4s	-50%
polyvecl_pack_eta	✅	2s	3s	-33%
polyvecl_pointwise_acc_montgomery	✅	2s	2s	+0%
polyw1_pack	✅	2s	4s	-50%
polyw1_pack_32	✅	2s	2s	+0%
polyz_pack	✅	2s	3s	-33%
polyz_unpack_19_native_aarch64	✅	2s	4s	-50%
rej_eta	✅	2s	2s	+0%
shake128_finalize	✅	2s	4s	-50%
shake128x4_absorb_once	✅	2s	1s	+100%
shake128x4_squeezeblocks	✅	2s	2s	+0%
shake256	✅	2s	2s	+0%
shake256_release	✅	2s	2s	+0%
sig_unpack_hints	✅	2s	3s	-33%
sign_verify_pre_hash_shake256	✅	2s	5s	-60%
sk_t0hat_get_poly	✅	2s	2s	+0%
unpack_sk_t0hat	✅	2s	1s	+100%
yvec_init	✅	2s	5s	-60%
keccak_f1600_x1_native_aarch64_v84a	✅	1s	2s	-50%
keccak_f1600_x4_native_aarch64_v84a	✅	1s	1s	+0%
keccakf1600_permute	✅	1s	2s	-50%
mld_ct_sel_int32	✅	1s	4s	-75%
poly_pointwise_montgomery_native	✅	1s	2s	-50%
polyveck_pack_w1	✅	1s	2s	-50%
shake128_release	✅	1s	1s	+0%
shake256x4_squeezeblocks	✅	1s	2s	-50%
sys_check_capability	✅	1s	2s	-50%
use_hint	✅	1s	3s	-67%

[oqs-bot]

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

Full Results (206 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	✅	1474s	1486s	-0.8%
mld_invntt_layer	✅	156s	160s	-3%
poly_pointwise_montgomery_c	✅	121s	131s	-8%
rej_uniform_native	✅	119s	115s	+3%
polyvec_matrix_pointwise_montgomery_yvec	✅	109s	111s	-2%
mld_ct_memcmp	✅	63s	63s	+0%
mld_ntt_layer	✅	44s	40s	+10%
fqmul	✅	40s	42s	-5%
keccakf1600x4_permute_native	✅	24s	22s	+9%
mld_attempt_signature_generation	✅	24s	24s	+0%
sign_verify_internal	✅	22s	23s	-4%
mld_ntt_butterfly_block	✅	21s	19s	+11%
poly_chknorm_c	✅	18s	17s	+6%
polyeta_unpack	✅	17s	14s	+21%
mld_check_pct	✅	15s	17s	-12%
polyt0_unpack	✅	14s	18s	-22%
rej_uniform_c	✅	14s	16s	-12%
poly_uniform_eta_4x	✅	13s	12s	+8%
polyz_unpack_c	✅	11s	11s	+0%
compute_pack_t0_t1	✅	9s	8s	+12%
poly_add	✅	9s	11s	-18%
polyvec_matrix_pointwise_montgomery_row	✅	9s	9s	+0%
keccak_absorb_once_x4	✅	8s	9s	-11%
sign_keypair	✅	8s	3s	+167%
poly_decompose_c	✅	7s	6s	+17%
poly_invntt_tomont_c	✅	7s	11s	-36%
polyveck_chknorm	✅	7s	9s	-22%
rej_uniform	✅	7s	8s	-12%
sign	✅	7s	7s	+0%
mld_compute_pack_z	✅	6s	6s	+0%
poly_uniform_eta	✅	6s	3s	+100%
polyvecl_chknorm	✅	6s	5s	+20%
sign_pk_from_sk	✅	6s	6s	+0%
sign_signature_extmu	✅	6s	5s	+20%
sign_signature_pre_hash_internal	✅	6s	5s	+20%
caddq	✅	5s	4s	+25%
keccak_absorb	✅	5s	6s	-17%
mld_keccakf1600_permute_c	✅	5s	5s	+0%
ntt_native_x86_64	✅	5s	4s	+25%
nttunpack_native_x86_64	✅	5s	3s	+67%
pointwise_acc_native_x86_64	✅	5s	6s	-17%
pointwise_native_aarch64	✅	5s	2s	+150%
poly_caddq_native	✅	5s	3s	+67%
poly_challenge	✅	5s	5s	+0%
poly_chknorm_native	✅	5s	3s	+67%
poly_pointwise_montgomery	✅	5s	3s	+67%
poly_power2round	✅	5s	8s	-38%
poly_reduce	✅	5s	5s	+0%
poly_shiftl	✅	5s	4s	+25%
poly_sub	✅	5s	3s	+67%
poly_use_hint	✅	5s	4s	+25%
polyveck_decompose	✅	5s	5s	+0%
polyvecl_ntt	✅	5s	3s	+67%
sign_signature_internal	✅	5s	3s	+67%
sign_verify	✅	5s	6s	-17%
sign_verify_extmu	✅	5s	5s	+0%
intt_native_aarch64	✅	4s	2s	+100%
keccak_squeezeblocks_x4	✅	4s	3s	+33%
keccakf1600_permute	✅	4s	3s	+33%
keccakf1600_xor_bytes (big endian)	✅	4s	3s	+33%
keccakf1600x4_extract_bytes_native	✅	4s	3s	+33%
mld_ct_cmask_nonzero_u32	✅	4s	4s	+0%
mld_h	✅	4s	3s	+33%
mld_keccakf1600_extract_bytes	✅	4s	4s	+0%
mld_polymat_expand_entry	✅	4s	5s	-20%
mld_sample_s1_s2	✅	4s	5s	-20%
mld_value_barrier_i64	✅	4s	1s	+300%
pack_sk_rho_key_tr_s2	✅	4s	2s	+100%
pointwise_acc_native_aarch64	✅	4s	6s	-33%
poly_caddq_c	✅	4s	4s	+0%
poly_caddq_native_aarch64	✅	4s	2s	+100%
poly_decompose_88_native_aarch64	✅	4s	2s	+100%
poly_invntt_tomont_native	✅	4s	3s	+33%
poly_pointwise_montgomery_native	✅	4s	3s	+33%
poly_uniform	✅	4s	3s	+33%
poly_uniform_4x	✅	4s	3s	+33%
poly_use_hint_c	✅	4s	3s	+33%
poly_use_hint_native_aarch64	✅	4s	3s	+33%
polyt1_unpack	✅	4s	3s	+33%
polyvec_matrix_expand_serial	✅	4s	2s	+100%
polyveck_invntt_tomont	✅	4s	7s	-43%
polyveck_reduce	✅	4s	4s	+0%
polyw1_pack_88	✅	4s	2s	+100%
polyz_unpack	✅	4s	3s	+33%
rej_uniform_eta_native_aarch64	✅	4s	5s	-20%
shake128_absorb	✅	4s	1s	+300%
shake128_release	✅	4s	2s	+100%
shake256	✅	4s	2s	+100%
sign_keypair_internal	✅	4s	4s	+0%
sign_open	✅	4s	5s	-20%
sign_signature	✅	4s	4s	+0%
sign_signature_pre_hash_shake256	✅	4s	2s	+100%
yvec_get_poly	✅	4s	1s	+300%
decompose	✅	3s	3s	+0%
fqscale	✅	3s	2s	+50%
intt_native_x86_64	✅	3s	3s	+0%
keccak_finalize	✅	3s	1s	+200%
keccak_init	✅	3s	2s	+50%
keccak_squeeze	✅	3s	3s	+0%
keccakf1600_extract_bytes (big endian)	✅	3s	1s	+200%
keccakf1600_xor_bytes	✅	3s	2s	+50%
keccakf1600x4_xor_bytes	✅	3s	2s	+50%
mld_ct_cmask_neg_i32	✅	3s	3s	+0%
mld_ct_get_optblocker_i64	✅	3s	2s	+50%
mld_prepare_domain_separation_prefix	✅	3s	5s	-40%
montgomery_reduce	✅	3s	2s	+50%
ntt_native_aarch64	✅	3s	2s	+50%
pack_sig_c	✅	3s	2s	+50%
pointwise_native_x86_64	✅	3s	2s	+50%
poly_caddq	✅	3s	2s	+50%
poly_caddq_native_x86_64	✅	3s	2s	+50%
poly_chknorm_native_x86_64	✅	3s	2s	+50%
poly_decompose_native	✅	3s	4s	-25%
poly_invntt_tomont	✅	3s	2s	+50%
poly_ntt	✅	3s	3s	+0%
poly_permute_bitrev_to_custom_optional_native	✅	3s	2s	+50%
poly_uniform_gamma1	✅	3s	3s	+0%
poly_uniform_gamma1_4x	✅	3s	3s	+0%
poly_use_hint_native_x86_64	✅	3s	-	new
polyt0_pack	✅	3s	5s	-40%
polyveck_caddq	✅	3s	4s	-25%
polyveck_ntt	✅	3s	2s	+50%
polyveck_pack_w1	✅	3s	3s	+0%
polyvecl_pointwise_acc_montgomery_native	✅	3s	3s	+0%
polyvecl_uniform_gamma1	✅	3s	2s	+50%
polyvecl_unpack_eta	✅	3s	4s	-25%
polyvecl_unpack_z	✅	3s	4s	-25%
polyw1_pack_32	✅	3s	4s	-25%
polyz_unpack_19_native_aarch64	✅	3s	3s	+0%
polyz_unpack_native	✅	3s	3s	+0%
reduce32	✅	3s	3s	+0%
rej_eta	✅	3s	2s	+50%
rej_eta_native	✅	3s	4s	-25%
rej_uniform_native_aarch64	✅	3s	3s	+0%
shake128_finalize	✅	3s	1s	+200%
shake128_squeeze	✅	3s	1s	+200%
shake128x4_squeezeblocks	✅	3s	2s	+50%
shake256_init	✅	3s	2s	+50%
shake256_squeeze	✅	3s	2s	+50%
shake256x4_squeezeblocks	✅	3s	2s	+50%
sign_verify_pre_hash_internal	✅	3s	5s	-40%
sign_verify_pre_hash_shake256	✅	3s	4s	-25%
sys_check_capability	✅	3s	4s	-25%
unpack_pk_t1	✅	3s	3s	+0%
unpack_sk_s1hat	✅	3s	3s	+0%
unpack_sk_s2hat	✅	3s	4s	-25%
keccak_f1600_x1_native_aarch64	✅	2s	3s	-33%
keccak_f1600_x1_native_aarch64_v84a	✅	2s	2s	+0%
keccak_f1600_x4_native_avx2	✅	2s	3s	-33%
keccakf1600_permute_native	✅	2s	2s	+0%
keccakf1600x4_permute	✅	2s	1s	+100%
mld_ct_get_optblocker_u32	✅	2s	2s	+0%
mld_ct_get_optblocker_u8	✅	2s	2s	+0%
mld_keccakf1600x4_extract_bytes_c	✅	2s	1s	+100%
mld_keccakf1600x4_xor_bytes_c	✅	2s	3s	-33%
mld_sample_s1_s2_serial	✅	2s	4s	-50%
mld_value_barrier_u32	✅	2s	2s	+0%
pack_sig_h	✅	2s	3s	-33%
pack_sig_z	✅	2s	3s	-33%
pack_sk_s1	✅	2s	4s	-50%
poly_chknorm	✅	2s	5s	-60%
poly_decompose_32_native_aarch64	✅	2s	2s	+0%
poly_ntt_c	✅	2s	2s	+0%
poly_ntt_native	✅	2s	5s	-60%
poly_permute_bitrev_to_custom_optional	✅	2s	2s	+0%
poly_use_hint_native	✅	2s	6s	-67%
polyeta_pack	✅	2s	2s	+0%
polyt1_pack	✅	2s	4s	-50%
polyveck_unpack_eta	✅	2s	2s	+0%
polyvecl_pack_eta	✅	2s	3s	-33%
polyvecl_pointwise_acc_montgomery	✅	2s	4s	-50%
polyvecl_pointwise_acc_montgomery_c	✅	2s	3s	-33%
polyvecl_uniform_gamma1_serial	✅	2s	5s	-60%
polyw1_pack	✅	2s	2s	+0%
polyz_unpack_17_native_aarch64	✅	2s	7s	-71%
polyz_unpack_native_x86_64	✅	2s	3s	-33%
power2round	✅	2s	1s	+100%
rej_eta_c	✅	2s	3s	-33%
shake128_init	✅	2s	3s	-33%
shake256_absorb	✅	2s	2s	+0%
shake256_finalize	✅	2s	3s	-33%
sk_s1hat_get_poly	✅	2s	4s	-50%
sk_s2hat_get_poly	✅	2s	2s	+0%
sk_t0hat_get_poly	✅	2s	3s	-33%
unpack_sk	✅	2s	1s	+100%
unpack_sk_t0hat	✅	2s	1s	+100%
use_hint	✅	2s	3s	-33%
yvec_init	✅	2s	4s	-50%
keccak_f1600_x4_native_aarch64_v84a	✅	1s	4s	-75%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	1s	3s	-67%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	1s	2s	-50%
keccakf1600x4_extract_bytes	✅	1s	3s	-67%
keccakf1600x4_xor_bytes_native	✅	1s	3s	-67%
make_hint	✅	1s	2s	-50%
mld_ct_abs_i32	✅	1s	2s	-50%
mld_ct_cmask_nonzero_u8	✅	1s	3s	-67%
mld_ct_sel_int32	✅	1s	3s	-67%
mld_value_barrier_u8	✅	1s	1s	+0%
poly_chknorm_native_aarch64	✅	1s	3s	-67%
poly_decompose	✅	1s	4s	-75%
polyvec_matrix_expand	✅	1s	2s	-50%
polyveck_pack_eta	✅	1s	2s	-50%
polyz_pack	✅	1s	2s	-50%
shake128x4_absorb_once	✅	1s	3s	-67%
shake256_release	✅	1s	3s	-67%
shake256x4_absorb_once	✅	1s	1s	+0%
sig_unpack_hints	✅	1s	3s	-67%

[oqs-bot]

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

Full Results (206 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	✅	1653s	1664s	-0.7%
mld_invntt_layer	✅	175s	172s	+2%
polyvec_matrix_pointwise_montgomery_yvec	✅	163s	162s	+1%
poly_pointwise_montgomery_c	✅	138s	135s	+2%
rej_uniform_native	✅	127s	128s	-1%
mld_ct_memcmp	✅	72s	69s	+4%
mld_ntt_layer	✅	46s	44s	+5%
fqmul	✅	43s	44s	-2%
mld_attempt_signature_generation	✅	35s	35s	+0%
keccakf1600x4_permute_native	✅	25s	22s	+14%
mld_ntt_butterfly_block	✅	24s	23s	+4%
sign_verify_internal	✅	23s	22s	+5%
poly_chknorm_c	✅	19s	19s	+0%
polyt0_unpack	✅	17s	15s	+13%
polyveck_decompose	✅	17s	16s	+6%
rej_uniform_c	✅	16s	16s	+0%
polyeta_unpack	✅	15s	15s	+0%
mld_check_pct	✅	13s	11s	+18%
poly_add	✅	13s	12s	+8%
poly_uniform_eta_4x	✅	12s	14s	-14%
compute_pack_t0_t1	✅	11s	11s	+0%
polyvecl_chknorm	✅	10s	11s	-9%
polyvecl_ntt	✅	10s	9s	+11%
keccak_absorb_once_x4	✅	8s	8s	+0%
pointwise_acc_native_x86_64	✅	8s	7s	+14%
poly_invntt_tomont_c	✅	8s	7s	+14%
rej_uniform	✅	8s	9s	-11%
sign_pk_from_sk	✅	8s	6s	+33%
mld_sample_s1_s2_serial	✅	7s	6s	+17%
poly_power2round	✅	7s	8s	-12%
polyveck_invntt_tomont	✅	7s	7s	+0%
sign	✅	7s	7s	+0%
sign_keypair_internal	✅	7s	8s	-12%
keccak_absorb	✅	6s	6s	+0%
mld_sample_s1_s2	✅	6s	5s	+20%
pointwise_acc_native_aarch64	✅	6s	8s	-25%
poly_caddq_c	✅	6s	5s	+20%
poly_decompose_c	✅	6s	10s	-40%
poly_ntt	✅	6s	2s	+200%
poly_reduce	✅	6s	4s	+50%
polyveck_caddq	✅	6s	6s	+0%
polyveck_reduce	✅	6s	6s	+0%
polyz_unpack_c	✅	6s	7s	-14%
rej_eta_native	✅	6s	5s	+20%
rej_uniform_native_aarch64	✅	6s	5s	+20%
sign_signature_pre_hash_internal	✅	6s	3s	+100%
sign_verify_pre_hash_shake256	✅	6s	4s	+50%
intt_native_x86_64	✅	5s	1s	+400%
keccakf1600x4_permute	✅	5s	2s	+150%
mld_compute_pack_z	✅	5s	4s	+25%
mld_keccakf1600_permute_c	✅	5s	5s	+0%
pointwise_native_aarch64	✅	5s	4s	+25%
poly_permute_bitrev_to_custom_optional_native	✅	5s	3s	+67%
poly_uniform	✅	5s	3s	+67%
poly_use_hint_native	✅	5s	3s	+67%
poly_use_hint_native_x86_64	✅	5s	-	new
polyvec_matrix_expand_serial	✅	5s	4s	+25%
polyvec_matrix_pointwise_montgomery_row	✅	5s	9s	-44%
sign_signature	✅	5s	7s	-29%
intt_native_aarch64	✅	4s	2s	+100%
keccak_squeezeblocks_x4	✅	4s	4s	+0%
keccakf1600_permute	✅	4s	1s	+300%
keccakf1600x4_xor_bytes_native	✅	4s	2s	+100%
mld_ct_cmask_neg_i32	✅	4s	3s	+33%
mld_polymat_expand_entry	✅	4s	3s	+33%
montgomery_reduce	✅	4s	3s	+33%
nttunpack_native_x86_64	✅	4s	5s	-20%
poly_caddq	✅	4s	4s	+0%
poly_caddq_native	✅	4s	5s	-20%
poly_caddq_native_aarch64	✅	4s	3s	+33%
poly_challenge	✅	4s	6s	-33%
poly_chknorm_native_aarch64	✅	4s	3s	+33%
poly_ntt_native	✅	4s	2s	+100%
poly_permute_bitrev_to_custom_optional	✅	4s	2s	+100%
poly_pointwise_montgomery_native	✅	4s	3s	+33%
poly_shiftl	✅	4s	5s	-20%
poly_uniform_eta	✅	4s	2s	+100%
poly_uniform_gamma1	✅	4s	2s	+100%
poly_use_hint_c	✅	4s	4s	+0%
polyeta_pack	✅	4s	2s	+100%
polyveck_unpack_eta	✅	4s	3s	+33%
polyvecl_pack_eta	✅	4s	2s	+100%
polyvecl_pointwise_acc_montgomery_native	✅	4s	3s	+33%
polyvecl_uniform_gamma1	✅	4s	2s	+100%
polyvecl_uniform_gamma1_serial	✅	4s	4s	+0%
polyvecl_unpack_z	✅	4s	2s	+100%
polyz_unpack	✅	4s	3s	+33%
sign_keypair	✅	4s	5s	-20%
sign_open	✅	4s	6s	-33%
sign_signature_extmu	✅	4s	5s	-20%
sign_signature_pre_hash_shake256	✅	4s	6s	-33%
sign_verify	✅	4s	5s	-20%
sign_verify_extmu	✅	4s	4s	+0%
sign_verify_pre_hash_internal	✅	4s	2s	+100%
sk_s2hat_get_poly	✅	4s	2s	+100%
unpack_pk_t1	✅	4s	2s	+100%
unpack_sk_s1hat	✅	4s	2s	+100%
unpack_sk_s2hat	✅	4s	2s	+100%
caddq	✅	3s	2s	+50%
fqscale	✅	3s	3s	+0%
keccak_f1600_x1_native_aarch64	✅	3s	1s	+200%
keccak_f1600_x1_native_aarch64_v84a	✅	3s	2s	+50%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	3s	2s	+50%
keccak_f1600_x4_native_avx2	✅	3s	1s	+200%
keccak_squeeze	✅	3s	3s	+0%
keccakf1600_extract_bytes (big endian)	✅	3s	1s	+200%
keccakf1600_xor_bytes	✅	3s	4s	-25%
mld_ct_abs_i32	✅	3s	1s	+200%
mld_ct_cmask_nonzero_u8	✅	3s	6s	-50%
mld_ct_sel_int32	✅	3s	2s	+50%
mld_h	✅	3s	3s	+0%
mld_keccakf1600x4_extract_bytes_c	✅	3s	2s	+50%
mld_value_barrier_u32	✅	3s	5s	-40%
pack_sig_c	✅	3s	4s	-25%
pack_sig_z	✅	3s	3s	+0%
pointwise_native_x86_64	✅	3s	5s	-40%
poly_caddq_native_x86_64	✅	3s	3s	+0%
poly_chknorm	✅	3s	3s	+0%
poly_chknorm_native	✅	3s	3s	+0%
poly_decompose_88_native_aarch64	✅	3s	3s	+0%
poly_decompose_native	✅	3s	2s	+50%
poly_invntt_tomont_native	✅	3s	4s	-25%
poly_pointwise_montgomery	✅	3s	3s	+0%
poly_sub	✅	3s	2s	+50%
poly_uniform_gamma1_4x	✅	3s	5s	-40%
polyveck_ntt	✅	3s	2s	+50%
polyvecl_pointwise_acc_montgomery_c	✅	3s	3s	+0%
polyw1_pack	✅	3s	3s	+0%
polyw1_pack_32	✅	3s	2s	+50%
polyz_unpack_17_native_aarch64	✅	3s	6s	-50%
polyz_unpack_19_native_aarch64	✅	3s	3s	+0%
power2round	✅	3s	3s	+0%
reduce32	✅	3s	3s	+0%
rej_eta_c	✅	3s	4s	-25%
rej_uniform_eta_native_aarch64	✅	3s	3s	+0%
shake128_release	✅	3s	3s	+0%
shake256	✅	3s	2s	+50%
shake256_absorb	✅	3s	3s	+0%
shake256_squeeze	✅	3s	2s	+50%
shake256x4_squeezeblocks	✅	3s	5s	-40%
sig_unpack_hints	✅	3s	5s	-40%
sign_signature_internal	✅	3s	4s	-25%
sk_s1hat_get_poly	✅	3s	3s	+0%
sk_t0hat_get_poly	✅	3s	3s	+0%
sys_check_capability	✅	3s	2s	+50%
use_hint	✅	3s	2s	+50%
keccak_f1600_x4_native_aarch64_v84a	✅	2s	5s	-60%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	2s	4s	-50%
keccak_finalize	✅	2s	2s	+0%
keccak_init	✅	2s	3s	-33%
keccakf1600_permute_native	✅	2s	3s	-33%
keccakf1600x4_extract_bytes	✅	2s	5s	-60%
keccakf1600x4_extract_bytes_native	✅	2s	4s	-50%
keccakf1600x4_xor_bytes	✅	2s	5s	-60%
make_hint	✅	2s	3s	-33%
mld_ct_cmask_nonzero_u32	✅	2s	3s	-33%
mld_ct_get_optblocker_i64	✅	2s	1s	+100%
mld_ct_get_optblocker_u32	✅	2s	3s	-33%
mld_keccakf1600_extract_bytes	✅	2s	4s	-50%
mld_keccakf1600x4_xor_bytes_c	✅	2s	3s	-33%
mld_prepare_domain_separation_prefix	✅	2s	4s	-50%
mld_value_barrier_i64	✅	2s	4s	-50%
mld_value_barrier_u8	✅	2s	2s	+0%
ntt_native_aarch64	✅	2s	2s	+0%
ntt_native_x86_64	✅	2s	4s	-50%
pack_sk_rho_key_tr_s2	✅	2s	2s	+0%
pack_sk_s1	✅	2s	3s	-33%
poly_chknorm_native_x86_64	✅	2s	5s	-60%
poly_decompose	✅	2s	4s	-50%
poly_decompose_32_native_aarch64	✅	2s	2s	+0%
poly_invntt_tomont	✅	2s	3s	-33%
poly_ntt_c	✅	2s	3s	-33%
poly_uniform_4x	✅	2s	4s	-50%
poly_use_hint	✅	2s	4s	-50%
poly_use_hint_native_aarch64	✅	2s	4s	-50%
polyt0_pack	✅	2s	2s	+0%
polyt1_pack	✅	2s	4s	-50%
polyt1_unpack	✅	2s	3s	-33%
polyveck_chknorm	✅	2s	2s	+0%
polyveck_pack_eta	✅	2s	5s	-60%
polyveck_pack_w1	✅	2s	3s	-33%
polyvecl_pointwise_acc_montgomery	✅	2s	3s	-33%
polyvecl_unpack_eta	✅	2s	4s	-50%
polyw1_pack_88	✅	2s	5s	-60%
polyz_unpack_native	✅	2s	7s	-71%
polyz_unpack_native_x86_64	✅	2s	4s	-50%
shake128_absorb	✅	2s	3s	-33%
shake128_finalize	✅	2s	2s	+0%
shake128x4_squeezeblocks	✅	2s	3s	-33%
shake256_release	✅	2s	3s	-33%
shake256x4_absorb_once	✅	2s	1s	+100%
unpack_sk	✅	2s	5s	-60%
unpack_sk_t0hat	✅	2s	3s	-33%
yvec_get_poly	✅	2s	4s	-50%
yvec_init	✅	2s	4s	-50%
decompose	✅	1s	2s	-50%
keccakf1600_xor_bytes (big endian)	✅	1s	2s	-50%
mld_ct_get_optblocker_u8	✅	1s	1s	+0%
pack_sig_h	✅	1s	3s	-67%
polyvec_matrix_expand	✅	1s	2s	-50%
polyz_pack	✅	1s	1s	+0%
rej_eta	✅	1s	3s	-67%
shake128_init	✅	1s	4s	-75%
shake128_squeeze	✅	1s	2s	-50%
shake128x4_absorb_once	✅	1s	2s	-50%
shake256_finalize	✅	1s	3s	-67%
shake256_init	✅	1s	2s	-50%

[oqs-bot]

CBMC Results (ML-DSA-65)

Full Results (206 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	✅	2062s	2203s	-6.4%
mld_invntt_layer	✅	281s	319s	-12%
polyvecl_pointwise_acc_montgomery_c	✅	205s	238s	-14%
rej_uniform_native	✅	148s	159s	-7%
polyvec_matrix_expand	✅	129s	140s	-8%
poly_pointwise_montgomery_c	✅	93s	112s	-17%
mld_attempt_signature_generation	✅	67s	68s	-1%
sign_verify_internal	✅	66s	67s	-1%
mld_ct_memcmp	✅	65s	75s	-13%
sign_signature_internal	✅	47s	48s	-2%
mld_ntt_layer	✅	44s	46s	-4%
fqmul	✅	41s	47s	-13%
keccakf1600x4_permute_native	✅	25s	23s	+9%
polyvec_matrix_expand_serial	✅	24s	27s	-11%
rej_uniform	✅	22s	22s	+0%
compute_pack_t0_t1	✅	21s	17s	+24%
mld_ntt_butterfly_block	✅	20s	23s	-13%
poly_chknorm_c	✅	18s	22s	-18%
polyt0_unpack	✅	17s	17s	+0%
mld_check_pct	✅	16s	18s	-11%
polyvecl_chknorm	✅	16s	15s	+7%
poly_uniform_eta_4x	✅	13s	12s	+8%
polyveck_decompose	✅	13s	15s	-13%
rej_uniform_c	✅	13s	15s	-13%
poly_uniform_4x	✅	12s	10s	+20%
keccak_absorb_once_x4	✅	10s	9s	+11%
poly_add	✅	10s	11s	-9%
polyveck_chknorm	✅	10s	12s	-17%
mld_compute_pack_z	✅	9s	8s	+12%
poly_invntt_tomont_c	✅	9s	9s	+0%
polyveck_invntt_tomont	✅	9s	8s	+12%
sign_pk_from_sk	✅	9s	6s	+50%
polyveck_ntt	✅	8s	7s	+14%
mld_keccakf1600_permute_c	✅	7s	7s	+0%
mld_polymat_expand_entry	✅	7s	3s	+133%
poly_pointwise_montgomery	✅	7s	3s	+133%
polyvec_matrix_pointwise_montgomery_yvec	✅	7s	8s	-12%
polyvecl_ntt	✅	7s	7s	+0%
polyz_unpack_c	✅	7s	7s	+0%
sign	✅	7s	8s	-12%
pointwise_acc_native_x86_64	✅	6s	7s	-14%
poly_challenge	✅	6s	4s	+50%
poly_invntt_tomont	✅	6s	1s	+500%
poly_ntt_native	✅	6s	1s	+500%
polyz_unpack_native_x86_64	✅	6s	3s	+100%
sign_keypair	✅	6s	4s	+50%
keccak_finalize	✅	5s	2s	+150%
keccakf1600_xor_bytes (big endian)	✅	5s	2s	+150%
ntt_native_x86_64	✅	5s	4s	+25%
pointwise_acc_native_aarch64	✅	5s	6s	-17%
poly_caddq_c	✅	5s	5s	+0%
poly_decompose	✅	5s	2s	+150%
poly_permute_bitrev_to_custom_optional_native	✅	5s	4s	+25%
poly_power2round	✅	5s	5s	+0%
poly_use_hint	✅	5s	2s	+150%
poly_use_hint_native_aarch64	✅	5s	3s	+67%
polyveck_caddq	✅	5s	7s	-29%
polyw1_pack_88	✅	5s	2s	+150%
power2round	✅	5s	2s	+150%
rej_uniform_eta_native_aarch64	✅	5s	5s	+0%
sign_open	✅	5s	5s	+0%
sign_signature_pre_hash_internal	✅	5s	5s	+0%
unpack_sk_t0hat	✅	5s	5s	+0%
decompose	✅	4s	2s	+100%
intt_native_aarch64	✅	4s	5s	-20%
keccak_absorb	✅	4s	7s	-43%
keccak_f1600_x1_native_aarch64_v84a	✅	4s	3s	+33%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	4s	2s	+100%
keccak_squeezeblocks_x4	✅	4s	5s	-20%
keccakf1600_permute	✅	4s	2s	+100%
keccakf1600x4_extract_bytes	✅	4s	2s	+100%
mld_ct_cmask_nonzero_u32	✅	4s	2s	+100%
mld_keccakf1600x4_extract_bytes_c	✅	4s	2s	+100%
mld_sample_s1_s2	✅	4s	6s	-33%
mld_value_barrier_u8	✅	4s	3s	+33%
ntt_native_aarch64	✅	4s	3s	+33%
poly_decompose_32_native_aarch64	✅	4s	4s	+0%
poly_permute_bitrev_to_custom_optional	✅	4s	3s	+33%
poly_pointwise_montgomery_native	✅	4s	3s	+33%
poly_uniform_gamma1_4x	✅	4s	4s	+0%
poly_use_hint_native	✅	4s	3s	+33%
polyeta_unpack	✅	4s	3s	+33%
polyt0_pack	✅	4s	4s	+0%
polyw1_pack	✅	4s	4s	+0%
polyz_unpack_native	✅	4s	5s	-20%
rej_eta_c	✅	4s	4s	+0%
rej_uniform_native_aarch64	✅	4s	2s	+100%
shake128_absorb	✅	4s	3s	+33%
sig_unpack_hints	✅	4s	2s	+100%
sign_keypair_internal	✅	4s	4s	+0%
sign_signature_pre_hash_shake256	✅	4s	4s	+0%
sign_verify	✅	4s	5s	-20%
sign_verify_pre_hash_internal	✅	4s	4s	+0%
unpack_pk_t1	✅	4s	2s	+100%
use_hint	✅	4s	2s	+100%
caddq	✅	3s	4s	-25%
intt_native_x86_64	✅	3s	3s	+0%
keccak_f1600_x1_native_aarch64	✅	3s	4s	-25%
keccak_f1600_x4_native_aarch64_v84a	✅	3s	2s	+50%
keccak_squeeze	✅	3s	2s	+50%
keccakf1600x4_extract_bytes_native	✅	3s	1s	+200%
keccakf1600x4_permute	✅	3s	1s	+200%
mld_ct_get_optblocker_i64	✅	3s	3s	+0%
mld_ct_get_optblocker_u32	✅	3s	2s	+50%
mld_keccakf1600x4_xor_bytes_c	✅	3s	4s	-25%
mld_sample_s1_s2_serial	✅	3s	8s	-62%
montgomery_reduce	✅	3s	3s	+0%
nttunpack_native_x86_64	✅	3s	2s	+50%
pack_sk_rho_key_tr_s2	✅	3s	4s	-25%
pack_sk_s1	✅	3s	2s	+50%
pointwise_native_aarch64	✅	3s	3s	+0%
poly_caddq_native_aarch64	✅	3s	3s	+0%
poly_caddq_native_x86_64	✅	3s	3s	+0%
poly_chknorm	✅	3s	4s	-25%
poly_chknorm_native_aarch64	✅	3s	3s	+0%
poly_decompose_c	✅	3s	3s	+0%
poly_decompose_native	✅	3s	1s	+200%
poly_invntt_tomont_native	✅	3s	4s	-25%
poly_ntt	✅	3s	3s	+0%
poly_ntt_c	✅	3s	3s	+0%
poly_uniform	✅	3s	4s	-25%
poly_uniform_eta	✅	3s	5s	-40%
poly_use_hint_native_x86_64	✅	3s	-	new
polyt1_pack	✅	3s	3s	+0%
polyveck_reduce	✅	3s	4s	-25%
polyveck_unpack_eta	✅	3s	3s	+0%
polyvecl_pack_eta	✅	3s	2s	+50%
polyvecl_uniform_gamma1_serial	✅	3s	3s	+0%
polyw1_pack_32	✅	3s	7s	-57%
polyz_pack	✅	3s	2s	+50%
polyz_unpack_17_native_aarch64	✅	3s	2s	+50%
reduce32	✅	3s	3s	+0%
rej_eta	✅	3s	6s	-50%
shake128_init	✅	3s	3s	+0%
shake128x4_squeezeblocks	✅	3s	3s	+0%
shake256	✅	3s	2s	+50%
shake256_absorb	✅	3s	2s	+50%
shake256_release	✅	3s	2s	+50%
shake256_squeeze	✅	3s	3s	+0%
sign_signature	✅	3s	5s	-40%
sign_signature_extmu	✅	3s	5s	-40%
sign_verify_extmu	✅	3s	4s	-25%
sign_verify_pre_hash_shake256	✅	3s	2s	+50%
sk_s1hat_get_poly	✅	3s	3s	+0%
sk_s2hat_get_poly	✅	3s	4s	-25%
unpack_sk	✅	3s	3s	+0%
unpack_sk_s1hat	✅	3s	3s	+0%
unpack_sk_s2hat	✅	3s	1s	+200%
yvec_get_poly	✅	3s	2s	+50%
fqscale	✅	2s	2s	+0%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	2s	5s	-60%
keccak_f1600_x4_native_avx2	✅	2s	2s	+0%
keccak_init	✅	2s	2s	+0%
keccakf1600_permute_native	✅	2s	1s	+100%
keccakf1600_xor_bytes	✅	2s	4s	-50%
keccakf1600x4_xor_bytes	✅	2s	3s	-33%
keccakf1600x4_xor_bytes_native	✅	2s	1s	+100%
make_hint	✅	2s	4s	-50%
mld_ct_abs_i32	✅	2s	2s	+0%
mld_ct_cmask_nonzero_u8	✅	2s	2s	+0%
mld_ct_get_optblocker_u8	✅	2s	1s	+100%
mld_ct_sel_int32	✅	2s	2s	+0%
mld_h	✅	2s	5s	-60%
mld_keccakf1600_extract_bytes	✅	2s	2s	+0%
mld_prepare_domain_separation_prefix	✅	2s	3s	-33%
mld_value_barrier_u32	✅	2s	2s	+0%
pack_sig_c	✅	2s	2s	+0%
pack_sig_h	✅	2s	4s	-50%
pack_sig_z	✅	2s	4s	-50%
pointwise_native_x86_64	✅	2s	5s	-60%
poly_caddq	✅	2s	2s	+0%
poly_caddq_native	✅	2s	2s	+0%
poly_chknorm_native	✅	2s	6s	-67%
poly_chknorm_native_x86_64	✅	2s	2s	+0%
poly_decompose_88_native_aarch64	✅	2s	4s	-50%
poly_reduce	✅	2s	2s	+0%
poly_shiftl	✅	2s	4s	-50%
poly_sub	✅	2s	3s	-33%
poly_uniform_gamma1	✅	2s	4s	-50%
poly_use_hint_c	✅	2s	4s	-50%
polyeta_pack	✅	2s	3s	-33%
polyt1_unpack	✅	2s	3s	-33%
polyvec_matrix_pointwise_montgomery_row	✅	2s	2s	+0%
polyveck_pack_eta	✅	2s	2s	+0%
polyveck_pack_w1	✅	2s	2s	+0%
polyvecl_pointwise_acc_montgomery	✅	2s	4s	-50%
polyvecl_pointwise_acc_montgomery_native	✅	2s	3s	-33%
polyvecl_uniform_gamma1	✅	2s	6s	-67%
polyvecl_unpack_eta	✅	2s	5s	-60%
polyvecl_unpack_z	✅	2s	3s	-33%
polyz_unpack	✅	2s	4s	-50%
polyz_unpack_19_native_aarch64	✅	2s	2s	+0%
rej_eta_native	✅	2s	4s	-50%
shake128_finalize	✅	2s	4s	-50%
shake128_release	✅	2s	3s	-33%
shake128x4_absorb_once	✅	2s	2s	+0%
shake256_finalize	✅	2s	3s	-33%
shake256_init	✅	2s	2s	+0%
shake256x4_absorb_once	✅	2s	2s	+0%
shake256x4_squeezeblocks	✅	2s	2s	+0%
sk_t0hat_get_poly	✅	2s	3s	-33%
sys_check_capability	✅	2s	4s	-50%
yvec_init	✅	2s	5s	-60%
keccakf1600_extract_bytes (big endian)	✅	1s	2s	-50%
mld_ct_cmask_neg_i32	✅	1s	1s	+0%
mld_value_barrier_i64	✅	1s	5s	-80%
shake128_squeeze	✅	1s	2s	-50%

[oqs-bot]

CBMC Results (ML-DSA-44)

Full Results (206 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	✅	1820s	1841s	-1.1%
mld_invntt_layer	✅	288s	302s	-5%
rej_uniform_native	✅	148s	152s	-3%
polyvecl_pointwise_acc_montgomery_c	✅	120s	130s	-8%
poly_pointwise_montgomery_c	✅	102s	105s	-3%
mld_ct_memcmp	✅	68s	72s	-6%
mld_attempt_signature_generation	✅	62s	64s	-3%
mld_ntt_layer	✅	46s	43s	+7%
fqmul	✅	44s	42s	+5%
keccakf1600x4_permute_native	✅	26s	25s	+4%
polyvec_matrix_expand	✅	26s	27s	-4%
sign_verify_internal	✅	26s	27s	-4%
sign_signature_internal	✅	23s	21s	+10%
mld_ntt_butterfly_block	✅	22s	24s	-8%
rej_uniform	✅	21s	21s	+0%
poly_chknorm_c	✅	19s	22s	-14%
polyt0_unpack	✅	17s	17s	+0%
compute_pack_t0_t1	✅	16s	16s	+0%
mld_check_pct	✅	16s	14s	+14%
polyeta_unpack	✅	15s	18s	-17%
polyveck_chknorm	✅	14s	12s	+17%
poly_uniform_eta_4x	✅	13s	11s	+18%
polyz_unpack_c	✅	12s	10s	+20%
rej_uniform_c	✅	12s	12s	+0%
poly_add	✅	11s	9s	+22%
polyvec_matrix_pointwise_montgomery_yvec	✅	11s	9s	+22%
keccak_absorb_once_x4	✅	10s	10s	+0%
poly_invntt_tomont_c	✅	10s	10s	+0%
poly_uniform_4x	✅	10s	10s	+0%
polyvec_matrix_expand_serial	✅	9s	9s	+0%
pointwise_acc_native_x86_64	✅	8s	7s	+14%
sign_pk_from_sk	✅	8s	7s	+14%
poly_caddq_c	✅	7s	5s	+40%
polyveck_decompose	✅	7s	8s	-12%
sign	✅	7s	8s	-12%
sign_verify_pre_hash_internal	✅	7s	4s	+75%
keccak_absorb	✅	6s	7s	-14%
mld_compute_pack_z	✅	6s	9s	-33%
mld_keccakf1600_permute_c	✅	6s	7s	-14%
mld_polymat_expand_entry	✅	6s	2s	+200%
poly_shiftl	✅	6s	2s	+200%
poly_use_hint_c	✅	6s	8s	-25%
polyz_unpack_native_x86_64	✅	6s	2s	+200%
sign_open	✅	6s	6s	+0%
sign_signature	✅	6s	3s	+100%
unpack_sk	✅	6s	3s	+100%
fqscale	✅	5s	3s	+67%
mld_ct_cmask_nonzero_u32	✅	5s	5s	+0%
mld_ct_get_optblocker_u8	✅	5s	3s	+67%
montgomery_reduce	✅	5s	3s	+67%
pack_sig_c	✅	5s	2s	+150%
pointwise_acc_native_aarch64	✅	5s	5s	+0%
poly_caddq	✅	5s	3s	+67%
poly_challenge	✅	5s	4s	+25%
poly_permute_bitrev_to_custom_optional	✅	5s	2s	+150%
poly_uniform_gamma1_4x	✅	5s	6s	-17%
poly_use_hint_native_aarch64	✅	5s	5s	+0%
polyt0_pack	✅	5s	4s	+25%
polyt1_unpack	✅	5s	3s	+67%
polyveck_caddq	✅	5s	4s	+25%
polyveck_ntt	✅	5s	4s	+25%
polyvecl_chknorm	✅	5s	3s	+67%
polyvecl_uniform_gamma1	✅	5s	3s	+67%
polyvecl_unpack_eta	✅	5s	2s	+150%
rej_eta	✅	5s	2s	+150%
rej_eta_native	✅	5s	4s	+25%
sig_unpack_hints	✅	5s	2s	+150%
sign_keypair_internal	✅	5s	2s	+150%
sign_signature_pre_hash_shake256	✅	5s	4s	+25%
sign_verify	✅	5s	4s	+25%
sk_t0hat_get_poly	✅	5s	5s	+0%
decompose	✅	4s	2s	+100%
keccakf1600x4_extract_bytes	✅	4s	2s	+100%
mld_prepare_domain_separation_prefix	✅	4s	3s	+33%
mld_sample_s1_s2_serial	✅	4s	4s	+0%
mld_value_barrier_i64	✅	4s	3s	+33%
mld_value_barrier_u32	✅	4s	3s	+33%
ntt_native_x86_64	✅	4s	4s	+0%
pack_sig_h	✅	4s	6s	-33%
pointwise_native_aarch64	✅	4s	5s	-20%
poly_caddq_native	✅	4s	5s	-20%
poly_chknorm_native	✅	4s	3s	+33%
poly_decompose_88_native_aarch64	✅	4s	3s	+33%
poly_pointwise_montgomery	✅	4s	2s	+100%
poly_reduce	✅	4s	5s	-20%
poly_sub	✅	4s	3s	+33%
poly_uniform	✅	4s	3s	+33%
polyvecl_pack_eta	✅	4s	4s	+0%
polyvecl_uniform_gamma1_serial	✅	4s	2s	+100%
polyvecl_unpack_z	✅	4s	3s	+33%
polyz_pack	✅	4s	2s	+100%
polyz_unpack_19_native_aarch64	✅	4s	3s	+33%
polyz_unpack_native	✅	4s	5s	-20%
power2round	✅	4s	1s	+300%
rej_eta_c	✅	4s	5s	-20%
shake128_squeeze	✅	4s	1s	+300%
sign_signature_extmu	✅	4s	4s	+0%
sign_verify_pre_hash_shake256	✅	4s	4s	+0%
sk_s1hat_get_poly	✅	4s	2s	+100%
unpack_sk_t0hat	✅	4s	3s	+33%
yvec_init	✅	4s	3s	+33%
caddq	✅	3s	2s	+50%
intt_native_aarch64	✅	3s	6s	-50%
intt_native_x86_64	✅	3s	4s	-25%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	3s	4s	-25%
keccak_squeezeblocks_x4	✅	3s	4s	-25%
keccakf1600_permute_native	✅	3s	2s	+50%
keccakf1600_xor_bytes	✅	3s	4s	-25%
keccakf1600x4_permute	✅	3s	4s	-25%
keccakf1600x4_xor_bytes	✅	3s	1s	+200%
make_hint	✅	3s	5s	-40%
mld_ct_abs_i32	✅	3s	1s	+200%
mld_ct_cmask_nonzero_u8	✅	3s	2s	+50%
mld_ct_get_optblocker_i64	✅	3s	2s	+50%
mld_h	✅	3s	4s	-25%
mld_keccakf1600_extract_bytes	✅	3s	3s	+0%
mld_keccakf1600x4_extract_bytes_c	✅	3s	2s	+50%
mld_keccakf1600x4_xor_bytes_c	✅	3s	1s	+200%
mld_sample_s1_s2	✅	3s	3s	+0%
mld_value_barrier_u8	✅	3s	1s	+200%
ntt_native_aarch64	✅	3s	5s	-40%
nttunpack_native_x86_64	✅	3s	5s	-40%
pack_sig_z	✅	3s	2s	+50%
pack_sk_s1	✅	3s	3s	+0%
poly_caddq_native_x86_64	✅	3s	4s	-25%
poly_chknorm	✅	3s	4s	-25%
poly_chknorm_native_aarch64	✅	3s	3s	+0%
poly_chknorm_native_x86_64	✅	3s	4s	-25%
poly_decompose	✅	3s	3s	+0%
poly_decompose_32_native_aarch64	✅	3s	5s	-40%
poly_decompose_c	✅	3s	3s	+0%
poly_decompose_native	✅	3s	3s	+0%
poly_ntt_c	✅	3s	5s	-40%
poly_ntt_native	✅	3s	3s	+0%
poly_permute_bitrev_to_custom_optional_native	✅	3s	2s	+50%
poly_pointwise_montgomery_native	✅	3s	4s	-25%
poly_uniform_eta	✅	3s	7s	-57%
poly_use_hint	✅	3s	3s	+0%
poly_use_hint_native_x86_64	✅	3s	-	new
polyt1_pack	✅	3s	4s	-25%
polyveck_invntt_tomont	✅	3s	4s	-25%
polyveck_unpack_eta	✅	3s	4s	-25%
polyvecl_ntt	✅	3s	5s	-40%
polyvecl_pointwise_acc_montgomery_native	✅	3s	6s	-50%
polyz_unpack	✅	3s	3s	+0%
rej_uniform_eta_native_aarch64	✅	3s	3s	+0%
shake128_release	✅	3s	3s	+0%
shake256	✅	3s	2s	+50%
shake256_absorb	✅	3s	2s	+50%
shake256_init	✅	3s	3s	+0%
shake256_release	✅	3s	1s	+200%
shake256_squeeze	✅	3s	1s	+200%
shake256x4_absorb_once	✅	3s	1s	+200%
sign_keypair	✅	3s	4s	-25%
sys_check_capability	✅	3s	2s	+50%
unpack_sk_s1hat	✅	3s	3s	+0%
unpack_sk_s2hat	✅	3s	3s	+0%
yvec_get_poly	✅	3s	3s	+0%
keccak_f1600_x1_native_aarch64	✅	2s	3s	-33%
keccak_f1600_x1_native_aarch64_v84a	✅	2s	4s	-50%
keccak_f1600_x4_native_aarch64_v84a	✅	2s	1s	+100%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	2s	3s	-33%
keccak_f1600_x4_native_avx2	✅	2s	3s	-33%
keccakf1600_extract_bytes (big endian)	✅	2s	4s	-50%
keccakf1600_xor_bytes (big endian)	✅	2s	2s	+0%
keccakf1600x4_extract_bytes_native	✅	2s	3s	-33%
pack_sk_rho_key_tr_s2	✅	2s	4s	-50%
pointwise_native_x86_64	✅	2s	3s	-33%
poly_caddq_native_aarch64	✅	2s	5s	-60%
poly_invntt_tomont	✅	2s	4s	-50%
poly_invntt_tomont_native	✅	2s	4s	-50%
poly_power2round	✅	2s	5s	-60%
poly_uniform_gamma1	✅	2s	2s	+0%
poly_use_hint_native	✅	2s	2s	+0%
polyeta_pack	✅	2s	3s	-33%
polyvec_matrix_pointwise_montgomery_row	✅	2s	2s	+0%
polyveck_pack_eta	✅	2s	3s	-33%
polyveck_pack_w1	✅	2s	2s	+0%
polyveck_reduce	✅	2s	2s	+0%
polyvecl_pointwise_acc_montgomery	✅	2s	3s	-33%
polyw1_pack	✅	2s	3s	-33%
polyw1_pack_32	✅	2s	3s	-33%
polyw1_pack_88	✅	2s	8s	-75%
polyz_unpack_17_native_aarch64	✅	2s	3s	-33%
rej_uniform_native_aarch64	✅	2s	3s	-33%
shake128_absorb	✅	2s	1s	+100%
shake128_init	✅	2s	1s	+100%
sign_signature_pre_hash_internal	✅	2s	3s	-33%
sign_verify_extmu	✅	2s	3s	-33%
sk_s2hat_get_poly	✅	2s	4s	-50%
unpack_pk_t1	✅	2s	3s	-33%
use_hint	✅	2s	2s	+0%
keccak_finalize	✅	1s	3s	-67%
keccak_init	✅	1s	3s	-67%
keccak_squeeze	✅	1s	5s	-80%
keccakf1600_permute	✅	1s	2s	-50%
keccakf1600x4_xor_bytes_native	✅	1s	2s	-50%
mld_ct_cmask_neg_i32	✅	1s	2s	-50%
mld_ct_get_optblocker_u32	✅	1s	1s	+0%
mld_ct_sel_int32	✅	1s	3s	-67%
poly_ntt	✅	1s	2s	-50%
reduce32	✅	1s	3s	-67%
shake128_finalize	✅	1s	3s	-67%
shake128x4_absorb_once	✅	1s	3s	-67%
shake128x4_squeezeblocks	✅	1s	2s	-50%
shake256_finalize	✅	1s	4s	-75%
shake256x4_squeezeblocks	✅	1s	2s	-50%

[oqs-bot]

CBMC Results (ML-DSA-87)

Full Results (206 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	✅	2486s	2584s	-3.8%
polyvecl_pointwise_acc_montgomery_c	✅	327s	367s	-11%
mld_invntt_layer	✅	303s	321s	-6%
polyvec_matrix_expand	✅	226s	244s	-7%
rej_uniform_native	✅	155s	160s	-3%
mld_attempt_signature_generation	✅	108s	107s	+1%
poly_pointwise_montgomery_c	✅	104s	112s	-7%
sign_signature_internal	✅	69s	66s	+5%
mld_ct_memcmp	✅	68s	73s	-7%
sign_verify_internal	✅	59s	64s	-8%
polyvec_matrix_expand_serial	✅	49s	50s	-2%
mld_ntt_layer	✅	44s	45s	-2%
fqmul	✅	42s	45s	-7%
compute_pack_t0_t1	✅	33s	32s	+3%
polyvec_matrix_pointwise_montgomery_yvec	✅	30s	32s	-6%
keccakf1600x4_permute_native	✅	23s	22s	+5%
mld_ntt_butterfly_block	✅	23s	22s	+5%
rej_uniform	✅	23s	23s	+0%
poly_chknorm_c	✅	22s	19s	+16%
mld_check_pct	✅	18s	16s	+12%
polyt0_unpack	✅	17s	17s	+0%
rej_uniform_c	✅	16s	15s	+7%
polyeta_unpack	✅	15s	17s	-12%
poly_uniform_eta_4x	✅	14s	13s	+8%
poly_add	✅	10s	13s	-23%
poly_uniform_4x	✅	10s	13s	-23%
polyvecl_ntt	✅	10s	9s	+11%
keccak_absorb_once_x4	✅	9s	12s	-25%
poly_invntt_tomont_c	✅	9s	9s	+0%
polyveck_decompose	✅	9s	9s	+0%
polyveck_invntt_tomont	✅	9s	10s	-10%
sign	✅	9s	6s	+50%
mld_compute_pack_z	✅	8s	8s	+0%
pointwise_acc_native_x86_64	✅	8s	8s	+0%
polyveck_ntt	✅	8s	7s	+14%
unpack_sk_t0hat	✅	8s	8s	+0%
mld_keccakf1600_permute_c	✅	7s	7s	+0%
mld_sample_s1_s2	✅	7s	6s	+17%
pointwise_acc_native_aarch64	✅	7s	8s	-12%
poly_uniform_gamma1_4x	✅	7s	3s	+133%
polyveck_caddq	✅	7s	6s	+17%
caddq	✅	6s	3s	+100%
poly_caddq_native	✅	6s	2s	+200%
poly_decompose_c	✅	6s	7s	-14%
polyvecl_chknorm	✅	6s	5s	+20%
sign_open	✅	6s	3s	+100%
sign_pk_from_sk	✅	6s	7s	-14%
sign_verify_pre_hash_internal	✅	6s	6s	+0%
decompose	✅	5s	2s	+150%
keccak_absorb	✅	5s	6s	-17%
keccak_init	✅	5s	3s	+67%
mld_prepare_domain_separation_prefix	✅	5s	5s	+0%
mld_sample_s1_s2_serial	✅	5s	7s	-29%
pack_sig_c	✅	5s	2s	+150%
poly_caddq_c	✅	5s	6s	-17%
poly_caddq_native_aarch64	✅	5s	4s	+25%
poly_decompose	✅	5s	4s	+25%
poly_power2round	✅	5s	6s	-17%
poly_reduce	✅	5s	1s	+400%
polyveck_chknorm	✅	5s	4s	+25%
polyz_unpack_c	✅	5s	5s	+0%
shake128x4_squeezeblocks	✅	5s	1s	+400%
sign_signature_extmu	✅	5s	5s	+0%
sign_signature_pre_hash_shake256	✅	5s	4s	+25%
sign_verify	✅	5s	5s	+0%
sign_verify_pre_hash_shake256	✅	5s	3s	+67%
sk_s2hat_get_poly	✅	5s	3s	+67%
unpack_sk_s1hat	✅	5s	3s	+67%
keccak_squeeze	✅	4s	2s	+100%
keccak_squeezeblocks_x4	✅	4s	4s	+0%
keccakf1600x4_xor_bytes	✅	4s	4s	+0%
mld_ct_get_optblocker_u32	✅	4s	3s	+33%
mld_h	✅	4s	4s	+0%
mld_value_barrier_u32	✅	4s	4s	+0%
montgomery_reduce	✅	4s	4s	+0%
ntt_native_aarch64	✅	4s	2s	+100%
ntt_native_x86_64	✅	4s	2s	+100%
nttunpack_native_x86_64	✅	4s	3s	+33%
pack_sig_z	✅	4s	1s	+300%
pack_sk_s1	✅	4s	3s	+33%
pointwise_native_x86_64	✅	4s	3s	+33%
poly_challenge	✅	4s	6s	-33%
poly_chknorm_native_aarch64	✅	4s	7s	-43%
poly_permute_bitrev_to_custom_optional	✅	4s	2s	+100%
poly_pointwise_montgomery_native	✅	4s	4s	+0%
poly_shiftl	✅	4s	3s	+33%
poly_sub	✅	4s	3s	+33%
poly_uniform_eta	✅	4s	5s	-20%
poly_uniform_gamma1	✅	4s	3s	+33%
poly_use_hint_native_aarch64	✅	4s	3s	+33%
poly_use_hint_native_x86_64	✅	4s	-	new
polyt0_pack	✅	4s	5s	-20%
polyt1_unpack	✅	4s	2s	+100%
polyveck_pack_eta	✅	4s	4s	+0%
polyveck_pack_w1	✅	4s	5s	-20%
polyveck_unpack_eta	✅	4s	3s	+33%
polyvecl_pack_eta	✅	4s	4s	+0%
polyvecl_pointwise_acc_montgomery_native	✅	4s	2s	+100%
polyvecl_uniform_gamma1_serial	✅	4s	2s	+100%
polyw1_pack_88	✅	4s	3s	+33%
polyz_unpack_19_native_aarch64	✅	4s	4s	+0%
polyz_unpack_native_x86_64	✅	4s	5s	-20%
power2round	✅	4s	4s	+0%
shake128_absorb	✅	4s	1s	+300%
sign_keypair	✅	4s	4s	+0%
sign_keypair_internal	✅	4s	8s	-50%
sign_signature	✅	4s	3s	+33%
sys_check_capability	✅	4s	3s	+33%
unpack_pk_t1	✅	4s	3s	+33%
use_hint	✅	4s	3s	+33%
intt_native_aarch64	✅	3s	3s	+0%
intt_native_x86_64	✅	3s	5s	-40%
keccak_f1600_x4_native_aarch64_v84a	✅	3s	1s	+200%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	3s	4s	-25%
keccakf1600_permute	✅	3s	3s	+0%
keccakf1600x4_extract_bytes	✅	3s	2s	+50%
keccakf1600x4_permute	✅	3s	3s	+0%
make_hint	✅	3s	4s	-25%
mld_ct_cmask_nonzero_u32	✅	3s	3s	+0%
mld_ct_get_optblocker_u8	✅	3s	1s	+200%
mld_ct_sel_int32	✅	3s	2s	+50%
mld_keccakf1600_extract_bytes	✅	3s	3s	+0%
mld_keccakf1600x4_extract_bytes_c	✅	3s	2s	+50%
mld_keccakf1600x4_xor_bytes_c	✅	3s	1s	+200%
mld_polymat_expand_entry	✅	3s	4s	-25%
mld_value_barrier_u8	✅	3s	2s	+50%
pack_sig_h	✅	3s	3s	+0%
pointwise_native_aarch64	✅	3s	3s	+0%
poly_caddq_native_x86_64	✅	3s	3s	+0%
poly_chknorm	✅	3s	3s	+0%
poly_chknorm_native	✅	3s	4s	-25%
poly_chknorm_native_x86_64	✅	3s	4s	-25%
poly_decompose_32_native_aarch64	✅	3s	2s	+50%
poly_decompose_native	✅	3s	3s	+0%
poly_invntt_tomont	✅	3s	2s	+50%
poly_invntt_tomont_native	✅	3s	5s	-40%
poly_ntt	✅	3s	6s	-50%
poly_ntt_native	✅	3s	4s	-25%
poly_pointwise_montgomery	✅	3s	5s	-40%
poly_uniform	✅	3s	3s	+0%
poly_use_hint	✅	3s	2s	+50%
poly_use_hint_native	✅	3s	4s	-25%
polyeta_pack	✅	3s	2s	+50%
polyt1_pack	✅	3s	2s	+50%
polyveck_reduce	✅	3s	6s	-50%
polyvecl_unpack_eta	✅	3s	4s	-25%
polyvecl_unpack_z	✅	3s	4s	-25%
polyw1_pack	✅	3s	3s	+0%
polyz_unpack	✅	3s	3s	+0%
polyz_unpack_17_native_aarch64	✅	3s	4s	-25%
polyz_unpack_native	✅	3s	4s	-25%
rej_eta	✅	3s	5s	-40%
rej_eta_native	✅	3s	5s	-40%
rej_uniform_eta_native_aarch64	✅	3s	5s	-40%
rej_uniform_native_aarch64	✅	3s	3s	+0%
shake128_release	✅	3s	3s	+0%
shake128x4_absorb_once	✅	3s	4s	-25%
shake256_absorb	✅	3s	3s	+0%
shake256_finalize	✅	3s	2s	+50%
shake256_squeeze	✅	3s	2s	+50%
shake256x4_absorb_once	✅	3s	2s	+50%
sig_unpack_hints	✅	3s	6s	-50%
sign_signature_pre_hash_internal	✅	3s	4s	-25%
sign_verify_extmu	✅	3s	3s	+0%
sk_t0hat_get_poly	✅	3s	2s	+50%
unpack_sk	✅	3s	4s	-25%
yvec_init	✅	3s	3s	+0%
fqscale	✅	2s	5s	-60%
keccak_f1600_x1_native_aarch64	✅	2s	2s	+0%
keccak_f1600_x1_native_aarch64_v84a	✅	2s	4s	-50%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	2s	1s	+100%
keccak_f1600_x4_native_avx2	✅	2s	4s	-50%
keccakf1600_extract_bytes (big endian)	✅	2s	1s	+100%
keccakf1600_permute_native	✅	2s	2s	+0%
keccakf1600_xor_bytes	✅	2s	2s	+0%
keccakf1600_xor_bytes (big endian)	✅	2s	2s	+0%
keccakf1600x4_extract_bytes_native	✅	2s	3s	-33%
keccakf1600x4_xor_bytes_native	✅	2s	3s	-33%
mld_ct_abs_i32	✅	2s	2s	+0%
mld_ct_cmask_neg_i32	✅	2s	4s	-50%
mld_ct_get_optblocker_i64	✅	2s	4s	-50%
mld_value_barrier_i64	✅	2s	2s	+0%
pack_sk_rho_key_tr_s2	✅	2s	5s	-60%
poly_caddq	✅	2s	3s	-33%
poly_decompose_88_native_aarch64	✅	2s	2s	+0%
poly_ntt_c	✅	2s	3s	-33%
poly_permute_bitrev_to_custom_optional_native	✅	2s	4s	-50%
poly_use_hint_c	✅	2s	2s	+0%
polyvec_matrix_pointwise_montgomery_row	✅	2s	4s	-50%
polyvecl_pointwise_acc_montgomery	✅	2s	4s	-50%
polyvecl_uniform_gamma1	✅	2s	4s	-50%
polyw1_pack_32	✅	2s	3s	-33%
polyz_pack	✅	2s	1s	+100%
reduce32	✅	2s	3s	-33%
rej_eta_c	✅	2s	3s	-33%
shake128_finalize	✅	2s	2s	+0%
shake128_init	✅	2s	2s	+0%
shake128_squeeze	✅	2s	4s	-50%
shake256_release	✅	2s	6s	-67%
shake256x4_squeezeblocks	✅	2s	2s	+0%
sk_s1hat_get_poly	✅	2s	2s	+0%
yvec_get_poly	✅	2s	1s	+100%
keccak_finalize	✅	1s	2s	-50%
mld_ct_cmask_nonzero_u8	✅	1s	4s	-75%
shake256	✅	1s	2s	-50%
shake256_init	✅	1s	4s	-75%
unpack_sk_s2hat	✅	1s	5s	-80%

[github-actions]

Mac Mini (M1, 2020) benchmarks (opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	46543 cycles	46505 cycles	1.00
ML-DSA-44 sign	131110 cycles	131069 cycles	1.00
ML-DSA-44 verify	47346 cycles	47311 cycles	1.00
ML-DSA-65 keypair	81683 cycles	81686 cycles	1.00
ML-DSA-65 sign	215314 cycles	215297 cycles	1.00
ML-DSA-65 verify	79298 cycles	79302 cycles	1.00
ML-DSA-87 keypair	132406 cycles	132401 cycles	1.00
ML-DSA-87 sign	277266 cycles	277556 cycles	1.00
ML-DSA-87 verify	134048 cycles	134051 cycles	1.00

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

[github-actions]

Mac Mini (M1, 2020) benchmarks (no-opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	112979 cycles	112792 cycles	1.00
ML-DSA-44 sign	401852 cycles	401073 cycles	1.00
ML-DSA-44 verify	119681 cycles	119472 cycles	1.00
ML-DSA-65 keypair	193030 cycles	192928 cycles	1.00
ML-DSA-65 sign	650200 cycles	649950 cycles	1.00
ML-DSA-65 verify	192928 cycles	192868 cycles	1.00
ML-DSA-87 keypair	318757 cycles	318801 cycles	1.00
ML-DSA-87 sign	828689 cycles	828900 cycles	1.00
ML-DSA-87 verify	326680 cycles	326715 cycles	1.00

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

[github-actions]

Arm Cortex-A76 (Raspberry Pi 5) benchmarks (opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	112165 cycles	112135 cycles	1.00
ML-DSA-44 sign	353502 cycles	353883 cycles	1.00
ML-DSA-44 verify	117018 cycles	117213 cycles	1.00
ML-DSA-65 keypair	194781 cycles	194346 cycles	1.00
ML-DSA-65 sign	583902 cycles	583663 cycles	1.00
ML-DSA-65 verify	192714 cycles	193077 cycles	1.00
ML-DSA-87 keypair	320922 cycles	320080 cycles	1.00
ML-DSA-87 sign	747323 cycles	747211 cycles	1.00
ML-DSA-87 verify	318745 cycles	317932 cycles	1.00

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

[github-actions]

Arm Cortex-A76 (Raspberry Pi 5) benchmarks (no-opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	211556 cycles	211551 cycles	1.00
ML-DSA-44 sign	758823 cycles	759909 cycles	1.00
ML-DSA-44 verify	229004 cycles	229380 cycles	1.00
ML-DSA-65 keypair	376999 cycles	378142 cycles	1.00
ML-DSA-65 sign	1247052 cycles	1247244 cycles	1.00
ML-DSA-65 verify	371175 cycles	371972 cycles	1.00
ML-DSA-87 keypair	600407 cycles	601369 cycles	1.00
ML-DSA-87 sign	1582648 cycles	1581866 cycles	1.00
ML-DSA-87 verify	616058 cycles	617247 cycles	1.00

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

[oqs-bot]

Intel Xeon 4th gen (c7i)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	43539 cycles	43355 cycles	1.00
ML-DSA-44 sign	130496 cycles	130544 cycles	1.00
ML-DSA-44 verify	45200 cycles	45221 cycles	1.00
ML-DSA-65 keypair	75740 cycles	75384 cycles	1.00
ML-DSA-65 sign	214734 cycles	214301 cycles	1.00
ML-DSA-65 verify	74318 cycles	74395 cycles	1.00
ML-DSA-87 keypair	123030 cycles	123229 cycles	1.00
ML-DSA-87 sign	271641 cycles	270971 cycles	1.00
ML-DSA-87 verify	120752 cycles	120534 cycles	1.00

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

[oqs-bot]

Intel Xeon 3rd gen (c6i)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	61570 cycles	61347 cycles	1.00
ML-DSA-44 sign	188839 cycles	189018 cycles	1.00
ML-DSA-44 verify	66442 cycles	66490 cycles	1.00
ML-DSA-65 keypair	110091 cycles	107823 cycles	1.02
ML-DSA-65 sign	314575 cycles	311956 cycles	1.01
ML-DSA-65 verify	110040 cycles	108642 cycles	1.01
ML-DSA-87 keypair	171189 cycles	171470 cycles	1.00
ML-DSA-87 sign	379547 cycles	378524 cycles	1.00
ML-DSA-87 verify	171219 cycles	169439 cycles	1.01

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

[oqs-bot]

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

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	91607 cycles	91483 cycles	1.00
ML-DSA-44 sign	351599 cycles	351948 cycles	1.00
ML-DSA-44 verify	99770 cycles	99772 cycles	1.00
ML-DSA-65 keypair	153995 cycles	153840 cycles	1.00
ML-DSA-65 sign	572344 cycles	570802 cycles	1.00
ML-DSA-65 verify	159837 cycles	159638 cycles	1.00
ML-DSA-87 keypair	255277 cycles	255330 cycles	1.00
ML-DSA-87 sign	726258 cycles	726314 cycles	1.00
ML-DSA-87 verify	263867 cycles	264290 cycles	1.00

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

[oqs-bot]

AMD EPYC 3rd gen (c6a)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	55426 cycles	55817 cycles	0.99
ML-DSA-44 sign	159150 cycles	160142 cycles	0.99
ML-DSA-44 verify	57595 cycles	58345 cycles	0.99
ML-DSA-65 keypair	95738 cycles	96668 cycles	0.99
ML-DSA-65 sign	260961 cycles	264062 cycles	0.99
ML-DSA-65 verify	96292 cycles	97394 cycles	0.99
ML-DSA-87 keypair	154498 cycles	156370 cycles	0.99
ML-DSA-87 sign	323133 cycles	326034 cycles	0.99
ML-DSA-87 verify	151292 cycles	152600 cycles	0.99

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

[oqs-bot]

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

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	154549 cycles	154341 cycles	1.00
ML-DSA-44 sign	590341 cycles	588680 cycles	1.00
ML-DSA-44 verify	169731 cycles	169051 cycles	1.00
ML-DSA-65 keypair	262129 cycles	262010 cycles	1.00
ML-DSA-65 sign	961211 cycles	962118 cycles	1.00
ML-DSA-65 verify	271442 cycles	271267 cycles	1.00
ML-DSA-87 keypair	431136 cycles	431402 cycles	1.00
ML-DSA-87 sign	1212224 cycles	1210399 cycles	1.00
ML-DSA-87 verify	447543 cycles	447214 cycles	1.00

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

[oqs-bot]

Graviton4

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	67229 cycles	67424 cycles	1.00
ML-DSA-44 sign	198371 cycles	198161 cycles	1.00
ML-DSA-44 verify	70305 cycles	70119 cycles	1.00
ML-DSA-65 keypair	119276 cycles	119402 cycles	1.00
ML-DSA-65 sign	326136 cycles	326062 cycles	1.00
ML-DSA-65 verify	116872 cycles	116765 cycles	1.00
ML-DSA-87 keypair	196540 cycles	196679 cycles	1.00
ML-DSA-87 sign	421405 cycles	421900 cycles	1.00
ML-DSA-87 verify	193296 cycles	193403 cycles	1.00

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

[oqs-bot]

Graviton3

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	71403 cycles	71551 cycles	1.00
ML-DSA-44 sign	208982 cycles	208983 cycles	1.00
ML-DSA-44 verify	74836 cycles	74733 cycles	1.00
ML-DSA-65 keypair	125933 cycles	125927 cycles	1.00
ML-DSA-65 sign	345627 cycles	345490 cycles	1.00
ML-DSA-65 verify	124118 cycles	124185 cycles	1.00
ML-DSA-87 keypair	207065 cycles	206521 cycles	1.00
ML-DSA-87 sign	444070 cycles	439805 cycles	1.01
ML-DSA-87 verify	204100 cycles	204446 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: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	133180 cycles	134773 cycles	0.99
ML-DSA-44 sign	518749 cycles	524514 cycles	0.99
ML-DSA-44 verify	146562 cycles	147960 cycles	0.99
ML-DSA-65 keypair	225259 cycles	228152 cycles	0.99
ML-DSA-65 sign	846103 cycles	854970 cycles	0.99
ML-DSA-65 verify	234865 cycles	238032 cycles	0.99
ML-DSA-87 keypair	367122 cycles	371227 cycles	0.99
ML-DSA-87 sign	1060070 cycles	1069626 cycles	0.99
ML-DSA-87 verify	381059 cycles	384666 cycles	0.99

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

[oqs-bot]

AMD EPYC 4th gen (c7a)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	46429 cycles	46780 cycles	0.99
ML-DSA-44 sign	139202 cycles	138851 cycles	1.00
ML-DSA-44 verify	49535 cycles	49217 cycles	1.01
ML-DSA-65 keypair	82794 cycles	82672 cycles	1.00
ML-DSA-65 sign	227727 cycles	227571 cycles	1.00
ML-DSA-65 verify	82058 cycles	82025 cycles	1.00
ML-DSA-87 keypair	129470 cycles	130121 cycles	0.99
ML-DSA-87 sign	280908 cycles	280059 cycles	1.00
ML-DSA-87 verify	128521 cycles	128980 cycles	1.00

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

[oqs-bot]

Graviton4 (no-opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	127570 cycles	127652 cycles	1.00
ML-DSA-44 sign	441188 cycles	441196 cycles	1.00
ML-DSA-44 verify	136371 cycles	136381 cycles	1.00
ML-DSA-65 keypair	220497 cycles	220706 cycles	1.00
ML-DSA-65 sign	714325 cycles	713750 cycles	1.00
ML-DSA-65 verify	221016 cycles	220752 cycles	1.00
ML-DSA-87 keypair	364546 cycles	365114 cycles	1.00
ML-DSA-87 sign	915587 cycles	921254 cycles	0.99
ML-DSA-87 verify	370845 cycles	370795 cycles	1.00

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

[oqs-bot]

Graviton3 (no-opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	137943 cycles	138043 cycles	1.00
ML-DSA-44 sign	486093 cycles	486132 cycles	1.00
ML-DSA-44 verify	149025 cycles	149080 cycles	1.00
ML-DSA-65 keypair	241723 cycles	241858 cycles	1.00
ML-DSA-65 sign	792077 cycles	791595 cycles	1.00
ML-DSA-65 verify	242156 cycles	241299 cycles	1.00
ML-DSA-87 keypair	395769 cycles	396308 cycles	1.00
ML-DSA-87 sign	1013650 cycles	1019320 cycles	0.99
ML-DSA-87 verify	403627 cycles	403778 cycles	1.00

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

[github-actions]

Arm Cortex-A72 (Raspberry Pi 4) benchmarks (opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	227585 cycles	229314 cycles	0.99
ML-DSA-44 sign	617247 cycles	633574 cycles	0.97
ML-DSA-44 verify	223087 cycles	223313 cycles	1.00
ML-DSA-65 keypair	383083 cycles	387745 cycles	0.99
ML-DSA-65 sign	991379 cycles	1000126 cycles	0.99
ML-DSA-65 verify	365888 cycles	368820 cycles	0.99
ML-DSA-87 keypair	631971 cycles	640246 cycles	0.99
ML-DSA-87 sign	1311893 cycles	1340002 cycles	0.98
ML-DSA-87 verify	618225 cycles	625960 cycles	0.99

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

[oqs-bot]

Graviton2

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	112409 cycles	112383 cycles	1.00
ML-DSA-44 sign	354351 cycles	353857 cycles	1.00
ML-DSA-44 verify	117504 cycles	117230 cycles	1.00
ML-DSA-65 keypair	194514 cycles	194621 cycles	1.00
ML-DSA-65 sign	584390 cycles	584356 cycles	1.00
ML-DSA-65 verify	193521 cycles	193140 cycles	1.00
ML-DSA-87 keypair	320989 cycles	320887 cycles	1.00
ML-DSA-87 sign	747858 cycles	746586 cycles	1.00
ML-DSA-87 verify	318073 cycles	318597 cycles	1.00

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

[oqs-bot]

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

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	121906 cycles	118559 cycles	1.03
ML-DSA-44 sign	466366 cycles	458188 cycles	1.02
ML-DSA-44 verify	134415 cycles	130934 cycles	1.03
ML-DSA-65 keypair	200533 cycles	200695 cycles	1.00
ML-DSA-65 sign	745242 cycles	742901 cycles	1.00
ML-DSA-65 verify	208898 cycles	209938 cycles	1.00
ML-DSA-87 keypair	330451 cycles	330620 cycles	1.00
ML-DSA-87 sign	937675 cycles	935710 cycles	1.00
ML-DSA-87 verify	344116 cycles	343441 cycles	1.00

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

[oqs-bot]

Graviton2 (no-opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	212331 cycles	211560 cycles	1.00
ML-DSA-44 sign	761162 cycles	759550 cycles	1.00
ML-DSA-44 verify	229987 cycles	229101 cycles	1.00
ML-DSA-65 keypair	378737 cycles	377267 cycles	1.00
ML-DSA-65 sign	1248155 cycles	1247098 cycles	1.00
ML-DSA-65 verify	373444 cycles	371496 cycles	1.01
ML-DSA-87 keypair	602568 cycles	600625 cycles	1.00
ML-DSA-87 sign	1584683 cycles	1584726 cycles	1.00
ML-DSA-87 verify	618500 cycles	616267 cycles	1.00

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

[github-actions]

Arm Cortex-A72 (Raspberry Pi 4) benchmarks (no-opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	314774 cycles	316204 cycles	1.00
ML-DSA-44 sign	1195084 cycles	1200034 cycles	1.00
ML-DSA-44 verify	344766 cycles	346175 cycles	1.00
ML-DSA-65 keypair	586747 cycles	573340 cycles	1.02
ML-DSA-65 sign	1934847 cycles	1950745 cycles	0.99
ML-DSA-65 verify	562256 cycles	540488 cycles	1.04
ML-DSA-87 keypair	872532 cycles	851822 cycles	1.02
ML-DSA-87 sign	2477111 cycles	2431803 cycles	1.02
ML-DSA-87 verify	898628 cycles	891771 cycles	1.01

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

[github-actions]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Arm Cortex-A72 (Raspberry Pi 4) benchmarks (no-opt)'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.03.

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-65 verify	562256 cycles	540488 cycles	1.04

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

[github-actions]

Arm Cortex-A55 (Snapdragon 888) benchmarks (opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	271763 cycles	271174 cycles	1.00
ML-DSA-44 sign	821008 cycles	824814 cycles	1.00
ML-DSA-44 verify	276025 cycles	275199 cycles	1.00
ML-DSA-65 keypair	465651 cycles	464395 cycles	1.00
ML-DSA-65 sign	1327127 cycles	1344742 cycles	0.99
ML-DSA-65 verify	451152 cycles	453340 cycles	1.00
ML-DSA-87 keypair	803172 cycles	800977 cycles	1.00
ML-DSA-87 sign	1864325 cycles	1859208 cycles	1.00
ML-DSA-87 verify	781387 cycles	774585 cycles	1.01

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

[github-actions]

Arm Cortex-A55 (Snapdragon 888) benchmarks (no-opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	465923 cycles	463353 cycles	1.01
ML-DSA-44 sign	2142098 cycles	2132217 cycles	1.00
ML-DSA-44 verify	558746 cycles	555186 cycles	1.01
ML-DSA-65 keypair	784684 cycles	783632 cycles	1.00
ML-DSA-65 sign	3495012 cycles	3480344 cycles	1.00
ML-DSA-65 verify	868294 cycles	864386 cycles	1.00
ML-DSA-87 keypair	1269194 cycles	1259586 cycles	1.01
ML-DSA-87 sign	4369517 cycles	4283557 cycles	1.02
ML-DSA-87 verify	1391355 cycles	1383524 cycles	1.01

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

[github-actions]

SpacemiT K1 8 (Banana Pi F3) benchmarks (no-opt)

Details

Benchmark suite	Current: 2b012e5	Previous: e5884bb	Ratio
ML-DSA-44 keypair	760366 cycles	759827 cycles	1.00
ML-DSA-44 sign	3140093 cycles	3139576 cycles	1.00
ML-DSA-44 verify	859520 cycles	859029 cycles	1.00
ML-DSA-65 keypair	1284878 cycles	1285683 cycles	1.00
ML-DSA-65 sign	5074016 cycles	5080630 cycles	1.00
ML-DSA-65 verify	1363408 cycles	1364316 cycles	1.00
ML-DSA-87 keypair	2112289 cycles	2112201 cycles	1.00
ML-DSA-87 sign	6350994 cycles	6353149 cycles	1.00
ML-DSA-87 verify	2228031 cycles	2227965 cycles	1.00

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

[mkannwischer]

Thanks @jakemas for moving this forward!

[mkannwischer]

If they are shared, can we move them to a shared file, please?

[mkannwischer]

These comments seem to describe the AArch64 code, not the x86_64 code. Can you please adjust them to describe what this code is doing?

[mkannwischer]

This is no longer true since 3 weeks ago: pq-crystals/dilithium@bba9534. We should reference that commit here

[mkannwischer]

I find this rather confusing - this seems to mix the AArch64 branches (k * 262144 <= (2 * lo * 1074791425) DIV 4294967296 + 131072) and the x86 branches (k * 4194304 <= (lo + 127) DIV 128 * 1025 + 2097152). But the AArch64 branches are dead in this proof.
It seems easiest to remove the dead branches and not claim that this theorem is shared with AArch64.