Avx2 keccakx4 trivial loop bench

[manastasova]

This commit aims to compare performance with #1521. This commit differs from #1521 by:

• Changed loop-based ROUNDS_X2 from processing two rounds per iteration to a trivial loop MLK_ROUNDS processing one round per iteration.

[oqs-bot]

Mac Mini (M1, 2020) benchmarks

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	12327 cycles	12325 cycles	1.00
ML-KEM-512 encaps	15030 cycles	15029 cycles	1.00
ML-KEM-512 decaps	19606 cycles	19606 cycles	1
ML-KEM-768 keypair	21089 cycles	21089 cycles	1
ML-KEM-768 encaps	23851 cycles	23852 cycles	1.00
ML-KEM-768 decaps	30431 cycles	30427 cycles	1.00
ML-KEM-1024 keypair	30389 cycles	30388 cycles	1.00
ML-KEM-1024 encaps	34647 cycles	34647 cycles	1
ML-KEM-1024 decaps	44274 cycles	44273 cycles	1.00

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

[oqs-bot]

ppc64le (POWER10) benchmarks

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	59545 cycles	59172 cycles	1.01
ML-KEM-512 encaps	72248 cycles	71904 cycles	1.00
ML-KEM-512 decaps	91935 cycles	91485 cycles	1.00
ML-KEM-768 keypair	99436 cycles	98898 cycles	1.01
ML-KEM-768 encaps	115919 cycles	115162 cycles	1.01
ML-KEM-768 decaps	141731 cycles	140953 cycles	1.01
ML-KEM-1024 keypair	150862 cycles	148300 cycles	1.02
ML-KEM-1024 encaps	170096 cycles	167223 cycles	1.02
ML-KEM-1024 decaps	201671 cycles	197970 cycles	1.02

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

[oqs-bot]

Arm Cortex-A55 (Snapdragon 888) benchmarks

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	59523 cycles	59521 cycles	1.00
ML-KEM-512 encaps	67226 cycles	67229 cycles	1.00
ML-KEM-512 decaps	85809 cycles	85804 cycles	1.00
ML-KEM-768 keypair	97056 cycles	96994 cycles	1.00
ML-KEM-768 encaps	110877 cycles	110345 cycles	1.00
ML-KEM-768 decaps	137492 cycles	137732 cycles	1.00
ML-KEM-1024 keypair	154014 cycles	153929 cycles	1.00
ML-KEM-1024 encaps	171068 cycles	172288 cycles	0.99
ML-KEM-1024 decaps	206658 cycles	207746 cycles	0.99

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

[oqs-bot]

Arm Cortex-A76 (Raspberry Pi 5) benchmarks

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	28352 cycles	28397 cycles	1.00
ML-KEM-512 encaps	34316 cycles	34356 cycles	1.00
ML-KEM-512 decaps	44535 cycles	44625 cycles	1.00
ML-KEM-768 keypair	47860 cycles	47827 cycles	1.00
ML-KEM-768 encaps	54218 cycles	54220 cycles	1.00
ML-KEM-768 decaps	68712 cycles	68666 cycles	1.00
ML-KEM-1024 keypair	70593 cycles	70592 cycles	1.00
ML-KEM-1024 encaps	79130 cycles	79148 cycles	1.00
ML-KEM-1024 decaps	98827 cycles	98830 cycles	1.00

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

[oqs-bot]

SpacemiT K1 8 (Banana Pi F3) benchmarks

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	155135 cycles	155095 cycles	1.00
ML-KEM-512 encaps	163327 cycles	163265 cycles	1.00
ML-KEM-512 decaps	206580 cycles	206458 cycles	1.00
ML-KEM-768 keypair	249525 cycles	249494 cycles	1.00
ML-KEM-768 encaps	270302 cycles	270253 cycles	1.00
ML-KEM-768 decaps	332113 cycles	332056 cycles	1.00
ML-KEM-1024 keypair	395104 cycles	395171 cycles	1.00
ML-KEM-1024 encaps	423669 cycles	423760 cycles	1.00
ML-KEM-1024 decaps	505496 cycles	505500 cycles	1.00

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

[oqs-bot]

Arm Cortex-A72 (Raspberry Pi 4) benchmarks

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	50814 cycles	50728 cycles	1.00
ML-KEM-512 encaps	59134 cycles	58464 cycles	1.01
ML-KEM-512 decaps	75596 cycles	74142 cycles	1.02
ML-KEM-768 keypair	87232 cycles	86364 cycles	1.01
ML-KEM-768 encaps	95628 cycles	94317 cycles	1.01
ML-KEM-768 decaps	118116 cycles	117130 cycles	1.01
ML-KEM-1024 keypair	131476 cycles	130149 cycles	1.01
ML-KEM-1024 encaps	143198 cycles	142791 cycles	1.00
ML-KEM-1024 decaps	172979 cycles	174398 cycles	0.99

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

[oqs-bot]

CBMC Results (ML-KEM-512)

Full Results (151 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	✅	1244s	1152s	+8.0%
mlk_indcpa_keypair_derand	✅	177s	163s	+9%
mlk_indcpa_enc	✅	173s	161s	+7%
mlk_keccak_squeezeblocks_x4	✅	144s	131s	+10%
mlk_rej_uniform_c	✅	80s	67s	+19%
mlk_polyvec_basemul_acc_montgomery_cached_c	✅	51s	46s	+11%
mlk_poly_rej_uniform	✅	35s	34s	+3%
poly_ntt_native	✅	33s	28s	+18%
mlk_polyvec_add	✅	26s	23s	+13%
mlk_ntt_layer	✅	23s	19s	+21%
polyvec_basemul_acc_montgomery_cached_native	✅	22s	19s	+16%
keccakf1600x4_permute_native_x4	✅	20s	20s	+0%
mlk_poly_reduce_native	✅	17s	15s	+13%
mlk_indcpa_dec	✅	12s	11s	+9%
mlk_poly_frommsg	✅	10s	6s	+67%
mlk_poly_sub	✅	10s	10s	+0%
mlk_keccak_absorb_once_x4	✅	9s	9s	+0%
mlk_ntt_butterfly_block	✅	9s	10s	-10%
mlk_keccak_squeeze_once	✅	8s	7s	+14%
mlk_poly_frombytes_native	✅	8s	9s	-11%
mlk_invntt_layer	✅	7s	6s	+17%
mlk_keccak_squeezeblocks	✅	7s	7s	+0%
poly_frombytes_native_x86_64	✅	7s	5s	+40%
keccakf1600_permute_native	✅	6s	4s	+50%
mlk_fqmul	✅	6s	7s	-14%
mlk_poly_rej_uniform_x4	✅	6s	6s	+0%
mlk_check_pct	✅	5s	3s	+67%
mlk_poly_tomont	✅	5s	2s	+150%
mlk_poly_tomsg	✅	5s	3s	+67%
mlk_scalar_compress_d10	✅	5s	2s	+150%
poly_invntt_tomont_native	✅	5s	3s	+67%
kem_check_pk	✅	4s	6s	-33%
kem_check_sk	✅	4s	1s	+300%
kem_enc	✅	4s	2s	+100%
kem_keypair	✅	4s	3s	+33%
mlk_keccakf1600_permute	✅	4s	4s	+0%
mlk_keccakf1600x4_permute	✅	4s	2s	+100%
mlk_poly_cbd_eta1	✅	4s	5s	-20%
mlk_poly_frombytes_c	✅	4s	2s	+100%
mlk_poly_getnoise_eta1_4x	✅	4s	3s	+33%
mlk_poly_mulcache_compute_c	✅	4s	4s	+0%
mlk_poly_reduce	✅	4s	3s	+33%
mlk_polyvec_basemul_acc_montgomery_cached	✅	4s	2s	+100%
mlk_polyvec_mulcache_compute	✅	4s	3s	+33%
mlk_polyvec_permute_bitrev_to_custom_native	✅	4s	3s	+33%
mlk_shake256x4	✅	4s	7s	-43%
poly_mulcache_compute_native_aarch64	✅	4s	4s	+0%
poly_tomont_native_x86_64	✅	4s	2s	+100%
rej_uniform_native_aarch64	✅	4s	3s	+33%
intt_native_aarch64	✅	3s	3s	+0%
intt_native_x86_64	✅	3s	3s	+0%
keccak_f1600_x1_native_aarch64	✅	3s	3s	+0%
keccak_f1600_x1_native_aarch64_v84a	✅	3s	1s	+200%
kem_dec	✅	3s	4s	-25%
kem_enc_derand	✅	3s	2s	+50%
kem_keypair_derand	✅	3s	2s	+50%
mlk_barrett_reduce	✅	3s	4s	-25%
mlk_ct_cmask_nonzero_u8	✅	3s	3s	+0%
mlk_ct_memcmp	✅	3s	1s	+200%
mlk_ct_sel_uint8	✅	3s	2s	+50%
mlk_keccak_absorb_once	✅	3s	5s	-40%
mlk_keccakf1600_extract_bytes	✅	3s	2s	+50%
mlk_keccakf1600_xor_bytes	✅	3s	3s	+0%
mlk_keccakf1600x4_xor_bytes	✅	3s	1s	+200%
mlk_poly_add	✅	3s	2s	+50%
mlk_poly_compress_du	✅	3s	3s	+0%
mlk_poly_compress_dv	✅	3s	2s	+50%
mlk_poly_decompress_dv	✅	3s	4s	-25%
mlk_poly_invntt_tomont_c	✅	3s	1s	+200%
mlk_poly_mulcache_compute	✅	3s	2s	+50%
mlk_poly_ntt_c	✅	3s	1s	+200%
mlk_poly_tobytes_native	✅	3s	3s	+0%
mlk_poly_tomont_c	✅	3s	2s	+50%
mlk_polymat_permute_bitrev_to_custom	✅	3s	3s	+0%
mlk_polyvec_compress_du	✅	3s	3s	+0%
mlk_polyvec_decompress_du	✅	3s	2s	+50%
mlk_polyvec_frombytes	✅	3s	2s	+50%
mlk_polyvec_permute_bitrev_to_custom	✅	3s	3s	+0%
mlk_polyvec_tomont	✅	3s	2s	+50%
mlk_scalar_compress_d5	✅	3s	3s	+0%
mlk_scalar_decompress_d11	✅	3s	3s	+0%
mlk_scalar_decompress_d5	✅	3s	1s	+200%
mlk_scalar_signed_to_unsigned_q	✅	3s	3s	+0%
mlk_sha3_512	✅	3s	2s	+50%
mlk_shake128x4_squeezeblocks	✅	3s	3s	+0%
mlk_value_barrier_u8	✅	3s	3s	+0%
nttunpack_native_x86_64	✅	3s	4s	-25%
poly_getnoise_eta1122_4x_native	✅	3s	3s	+0%
poly_mulcache_compute_native_x86_64	✅	3s	2s	+50%
poly_tomont_native_aarch64	✅	3s	3s	+0%
polyvec_basemul_acc_montgomery_cached_k2_native_aarch64	✅	3s	5s	-40%
polyvec_basemul_acc_montgomery_cached_k2_native_x86_64	✅	3s	3s	+0%
rej_uniform_native	✅	3s	3s	+0%
rej_uniform_native_x86_64	✅	3s	1s	+200%
keccak_f1600_x4_native_aarch64_v84a	✅	2s	1s	+100%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	2s	2s	+0%
mlk_ct_cmask_nonzero_u16	✅	2s	2s	+0%
mlk_ct_get_optblocker_u32	✅	2s	2s	+0%
mlk_ct_get_optblocker_u8	✅	2s	4s	-50%
mlk_ct_sel_int16	✅	2s	2s	+0%
mlk_gen_matrix	✅	2s	4s	-50%
mlk_gen_matrix_serial	✅	2s	3s	-33%
mlk_keccakf1600_extract_bytes (big endian)	✅	2s	3s	-33%
mlk_keccakf1600x4_extract_bytes	✅	2s	3s	-33%
mlk_matvec_mul	✅	2s	2s	+0%
mlk_montgomery_reduce	✅	2s	1s	+100%
mlk_poly_cbd_eta2	✅	2s	2s	+0%
mlk_poly_decompress_du	✅	2s	3s	-33%
mlk_poly_frombytes	✅	2s	3s	-33%
mlk_poly_getnoise_eta1122_4x	✅	2s	2s	+0%
mlk_poly_getnoise_eta1_4x_native	✅	2s	1s	+100%
mlk_poly_getnoise_eta2	✅	2s	5s	-60%
mlk_poly_invntt_tomont	✅	2s	2s	+0%
mlk_poly_mulcache_compute_native	✅	2s	1s	+100%
mlk_poly_ntt	✅	2s	2s	+0%
mlk_poly_reduce_c	✅	2s	2s	+0%
mlk_poly_tobytes	✅	2s	2s	+0%
mlk_poly_tobytes_c	✅	2s	3s	-33%
mlk_poly_tomont_native	✅	2s	4s	-50%
mlk_polyvec_invntt_tomont	✅	2s	2s	+0%
mlk_polyvec_ntt	✅	2s	5s	-60%
mlk_polyvec_reduce	✅	2s	2s	+0%
mlk_polyvec_tobytes	✅	2s	5s	-60%
mlk_scalar_compress_d1	✅	2s	1s	+100%
mlk_scalar_compress_d11	✅	2s	4s	-50%
mlk_scalar_compress_d4	✅	2s	3s	-33%
mlk_sha3_256	✅	2s	1s	+100%
mlk_shake128_absorb_once	✅	2s	1s	+100%
mlk_shake128_squeezeblocks	✅	2s	3s	-33%
mlk_shake256	✅	2s	2s	+0%
mlk_value_barrier_i32	✅	2s	3s	-33%
mlk_value_barrier_u32	✅	2s	2s	+0%
ntt_native_aarch64	✅	2s	2s	+0%
ntt_native_x86_64	✅	2s	3s	-33%
poly_reduce_native_aarch64	✅	2s	1s	+100%
poly_tobytes_native_aarch64	✅	2s	2s	+0%
poly_tobytes_native_x86_64	✅	2s	2s	+0%
polyvec_basemul_acc_montgomery_cached_k3_native_aarch64	✅	2s	3s	-33%
polyvec_basemul_acc_montgomery_cached_k3_native_x86_64	✅	2s	2s	+0%
polyvec_basemul_acc_montgomery_cached_k4_native_aarch64	✅	2s	1s	+100%
polyvec_basemul_acc_montgomery_cached_k4_native_x86_64	✅	2s	3s	-33%
sys_check_capability	✅	2s	1s	+100%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	1s	2s	-50%
mlk_ct_cmask_neg_i16	✅	1s	1s	+0%
mlk_ct_cmov_zero	✅	1s	1s	+0%
mlk_ct_get_optblocker_i32	✅	1s	3s	-67%
mlk_keccakf1600_xor_bytes (big endian)	✅	1s	4s	-75%
mlk_rej_uniform	✅	1s	3s	-67%
mlk_scalar_decompress_d10	✅	1s	3s	-67%
mlk_scalar_decompress_d4	✅	1s	2s	-50%
mlk_shake128x4_absorb_once	✅	1s	3s	-67%
poly_reduce_native_x86_64	✅	1s	4s	-75%

[oqs-bot]

CBMC Results (ML-KEM-768)

Full Results (151 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	✅	1296s	1244s	+4.2%
mlk_indcpa_enc	✅	175s	174s	+1%
mlk_indcpa_keypair_derand	✅	154s	155s	-1%
mlk_keccak_squeezeblocks_x4	✅	144s	135s	+7%
mlk_rej_uniform_c	✅	83s	75s	+11%
mlk_polyvec_basemul_acc_montgomery_cached_c	✅	82s	75s	+9%
polyvec_basemul_acc_montgomery_cached_native	✅	58s	61s	-5%
mlk_poly_rej_uniform	✅	44s	36s	+22%
mlk_ntt_layer	✅	38s	29s	+31%
mlk_polyvec_add	✅	29s	28s	+4%
poly_ntt_native	✅	28s	27s	+4%
keccakf1600x4_permute_native_x4	✅	20s	19s	+5%
mlk_poly_reduce_native	✅	18s	15s	+20%
mlk_indcpa_dec	✅	14s	11s	+27%
mlk_ntt_butterfly_block	✅	13s	10s	+30%
mlk_poly_frombytes_native	✅	10s	7s	+43%
mlk_poly_sub	✅	10s	9s	+11%
mlk_keccak_absorb_once_x4	✅	9s	9s	+0%
mlk_poly_rej_uniform_x4	✅	9s	8s	+12%
mlk_keccak_squeeze_once	✅	8s	7s	+14%
mlk_keccak_squeezeblocks	✅	8s	8s	+0%
mlk_poly_frommsg	✅	8s	7s	+14%
mlk_fqmul	✅	6s	5s	+20%
mlk_polymat_permute_bitrev_to_custom	✅	6s	6s	+0%
poly_frombytes_native_x86_64	✅	6s	4s	+50%
keccakf1600_permute_native	✅	5s	6s	-17%
kem_dec	✅	5s	4s	+25%
kem_keypair_derand	✅	5s	3s	+67%
mlk_invntt_layer	✅	5s	4s	+25%
mlk_poly_ntt	✅	5s	2s	+150%
mlk_scalar_signed_to_unsigned_q	✅	5s	2s	+150%
poly_getnoise_eta1122_4x_native	✅	5s	2s	+150%
mlk_gen_matrix	✅	4s	4s	+0%
mlk_gen_matrix_serial	✅	4s	3s	+33%
mlk_keccak_absorb_once	✅	4s	5s	-20%
mlk_keccakf1600_permute	✅	4s	4s	+0%
mlk_poly_cbd_eta2	✅	4s	1s	+300%
mlk_poly_compress_dv	✅	4s	3s	+33%
mlk_poly_tobytes	✅	4s	1s	+300%
mlk_poly_tobytes_native	✅	4s	3s	+33%
mlk_shake256x4	✅	4s	3s	+33%
poly_mulcache_compute_native_x86_64	✅	4s	2s	+100%
poly_tomont_native_aarch64	✅	4s	1s	+300%
polyvec_basemul_acc_montgomery_cached_k3_native_aarch64	✅	4s	4s	+0%
intt_native_x86_64	✅	3s	2s	+50%
kem_check_sk	✅	3s	3s	+0%
kem_enc	✅	3s	2s	+50%
mlk_ct_cmask_neg_i16	✅	3s	3s	+0%
mlk_ct_get_optblocker_i32	✅	3s	1s	+200%
mlk_ct_sel_int16	✅	3s	2s	+50%
mlk_ct_sel_uint8	✅	3s	3s	+0%
mlk_poly_cbd_eta1	✅	3s	4s	-25%
mlk_poly_compress_du	✅	3s	4s	-25%
mlk_poly_getnoise_eta1_4x	✅	3s	3s	+0%
mlk_poly_getnoise_eta1_4x_native	✅	3s	3s	+0%
mlk_poly_getnoise_eta2	✅	3s	2s	+50%
mlk_poly_mulcache_compute_c	✅	3s	3s	+0%
mlk_poly_reduce	✅	3s	1s	+200%
mlk_poly_reduce_c	✅	3s	3s	+0%
mlk_poly_tobytes_c	✅	3s	2s	+50%
mlk_poly_tomont	✅	3s	1s	+200%
mlk_poly_tomont_native	✅	3s	2s	+50%
mlk_poly_tomsg	✅	3s	2s	+50%
mlk_polyvec_tobytes	✅	3s	2s	+50%
mlk_scalar_compress_d10	✅	3s	2s	+50%
mlk_scalar_compress_d11	✅	3s	2s	+50%
mlk_scalar_decompress_d10	✅	3s	1s	+200%
mlk_scalar_decompress_d11	✅	3s	4s	-25%
mlk_scalar_decompress_d4	✅	3s	3s	+0%
mlk_shake128_absorb_once	✅	3s	1s	+200%
mlk_shake128x4_squeezeblocks	✅	3s	1s	+200%
mlk_value_barrier_u8	✅	3s	2s	+50%
ntt_native_aarch64	✅	3s	3s	+0%
poly_mulcache_compute_native_aarch64	✅	3s	2s	+50%
poly_tobytes_native_x86_64	✅	3s	2s	+50%
poly_tomont_native_x86_64	✅	3s	2s	+50%
polyvec_basemul_acc_montgomery_cached_k3_native_x86_64	✅	3s	2s	+50%
rej_uniform_native	✅	3s	2s	+50%
intt_native_aarch64	✅	2s	2s	+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	2s	+0%
kem_check_pk	✅	2s	1s	+100%
kem_enc_derand	✅	2s	3s	-33%
kem_keypair	✅	2s	3s	-33%
mlk_barrett_reduce	✅	2s	3s	-33%
mlk_check_pct	✅	2s	5s	-60%
mlk_ct_get_optblocker_u8	✅	2s	2s	+0%
mlk_ct_memcmp	✅	2s	3s	-33%
mlk_keccakf1600_extract_bytes	✅	2s	3s	-33%
mlk_keccakf1600_extract_bytes (big endian)	✅	2s	4s	-50%
mlk_keccakf1600_xor_bytes	✅	2s	2s	+0%
mlk_keccakf1600_xor_bytes (big endian)	✅	2s	3s	-33%
mlk_keccakf1600x4_extract_bytes	✅	2s	2s	+0%
mlk_keccakf1600x4_permute	✅	2s	3s	-33%
mlk_keccakf1600x4_xor_bytes	✅	2s	3s	-33%
mlk_matvec_mul	✅	2s	1s	+100%
mlk_poly_add	✅	2s	1s	+100%
mlk_poly_decompress_du	✅	2s	1s	+100%
mlk_poly_decompress_dv	✅	2s	4s	-50%
mlk_poly_frombytes	✅	2s	2s	+0%
mlk_poly_getnoise_eta1122_4x	✅	2s	3s	-33%
mlk_poly_invntt_tomont	✅	2s	3s	-33%
mlk_poly_invntt_tomont_c	✅	2s	1s	+100%
mlk_poly_mulcache_compute_native	✅	2s	3s	-33%
mlk_poly_ntt_c	✅	2s	2s	+0%
mlk_poly_tomont_c	✅	2s	2s	+0%
mlk_polyvec_basemul_acc_montgomery_cached	✅	2s	4s	-50%
mlk_polyvec_compress_du	✅	2s	2s	+0%
mlk_polyvec_frombytes	✅	2s	3s	-33%
mlk_polyvec_mulcache_compute	✅	2s	4s	-50%
mlk_polyvec_ntt	✅	2s	3s	-33%
mlk_polyvec_permute_bitrev_to_custom	✅	2s	2s	+0%
mlk_polyvec_permute_bitrev_to_custom_native	✅	2s	4s	-50%
mlk_polyvec_reduce	✅	2s	3s	-33%
mlk_polyvec_tomont	✅	2s	3s	-33%
mlk_scalar_compress_d1	✅	2s	3s	-33%
mlk_scalar_compress_d4	✅	2s	2s	+0%
mlk_scalar_compress_d5	✅	2s	3s	-33%
mlk_shake128_squeezeblocks	✅	2s	2s	+0%
mlk_shake128x4_absorb_once	✅	2s	3s	-33%
mlk_shake256	✅	2s	4s	-50%
mlk_value_barrier_i32	✅	2s	2s	+0%
mlk_value_barrier_u32	✅	2s	5s	-60%
ntt_native_x86_64	✅	2s	3s	-33%
nttunpack_native_x86_64	✅	2s	3s	-33%
poly_invntt_tomont_native	✅	2s	3s	-33%
poly_reduce_native_aarch64	✅	2s	2s	+0%
poly_tobytes_native_aarch64	✅	2s	2s	+0%
polyvec_basemul_acc_montgomery_cached_k2_native_aarch64	✅	2s	4s	-50%
polyvec_basemul_acc_montgomery_cached_k2_native_x86_64	✅	2s	3s	-33%
polyvec_basemul_acc_montgomery_cached_k4_native_aarch64	✅	2s	2s	+0%
polyvec_basemul_acc_montgomery_cached_k4_native_x86_64	✅	2s	3s	-33%
rej_uniform_native_aarch64	✅	2s	2s	+0%
rej_uniform_native_x86_64	✅	2s	2s	+0%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	1s	1s	+0%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	1s	3s	-67%
mlk_ct_cmask_nonzero_u16	✅	1s	1s	+0%
mlk_ct_cmask_nonzero_u8	✅	1s	3s	-67%
mlk_ct_cmov_zero	✅	1s	2s	-50%
mlk_ct_get_optblocker_u32	✅	1s	1s	+0%
mlk_montgomery_reduce	✅	1s	1s	+0%
mlk_poly_frombytes_c	✅	1s	2s	-50%
mlk_poly_mulcache_compute	✅	1s	2s	-50%
mlk_polyvec_decompress_du	✅	1s	1s	+0%
mlk_polyvec_invntt_tomont	✅	1s	4s	-75%
mlk_rej_uniform	✅	1s	2s	-50%
mlk_scalar_decompress_d5	✅	1s	2s	-50%
mlk_sha3_256	✅	1s	2s	-50%
mlk_sha3_512	✅	1s	3s	-67%
poly_reduce_native_x86_64	✅	1s	3s	-67%
sys_check_capability	✅	1s	2s	-50%

[oqs-bot]

CBMC Results (ML-KEM-1024)

Full Results (151 proofs)

Proof	Status	Current	Previous	Change
**TOTAL**	✅	1536s	1494s	+2.8%
mlk_indcpa_keypair_derand	✅	311s	300s	+4%
mlk_indcpa_enc	✅	217s	212s	+2%
mlk_keccak_squeezeblocks_x4	✅	134s	134s	+0%
polyvec_basemul_acc_montgomery_cached_native	✅	118s	117s	+1%
mlk_rej_uniform_c	✅	75s	68s	+10%
mlk_polyvec_basemul_acc_montgomery_cached_c	✅	59s	54s	+9%
mlk_poly_rej_uniform	✅	38s	36s	+6%
mlk_ntt_layer	✅	35s	28s	+25%
poly_ntt_native	✅	27s	23s	+17%
mlk_poly_decompress_dv	✅	26s	29s	-10%
mlk_indcpa_dec	✅	24s	27s	-11%
keccakf1600x4_permute_native_x4	✅	19s	19s	+0%
mlk_poly_reduce_native	✅	14s	12s	+17%
mlk_polyvec_ntt	✅	12s	13s	-8%
mlk_keccak_absorb_once_x4	✅	11s	8s	+38%
mlk_poly_frommsg	✅	9s	7s	+29%
mlk_poly_rej_uniform_x4	✅	9s	7s	+29%
mlk_poly_sub	✅	9s	9s	+0%
mlk_polyvec_add	✅	9s	9s	+0%
keccakf1600_permute_native	✅	8s	5s	+60%
mlk_ntt_butterfly_block	✅	8s	8s	+0%
mlk_poly_compress_du	✅	8s	6s	+33%
mlk_poly_frombytes_native	✅	8s	9s	-11%
kem_dec	✅	7s	5s	+40%
mlk_gen_matrix	✅	7s	8s	-12%
mlk_keccak_squeeze_once	✅	7s	7s	+0%
mlk_keccak_squeezeblocks	✅	7s	6s	+17%
mlk_gen_matrix_serial	✅	6s	5s	+20%
keccak_f1600_x4_native_aarch64_v84a	✅	5s	3s	+67%
kem_enc_derand	✅	5s	3s	+67%
kem_keypair	✅	5s	3s	+67%
mlk_fqmul	✅	5s	6s	-17%
mlk_invntt_layer	✅	5s	4s	+25%
mlk_value_barrier_u8	✅	5s	4s	+25%
kem_check_pk	✅	4s	5s	-20%
kem_check_sk	✅	4s	4s	+0%
mlk_ct_sel_int16	✅	4s	5s	-20%
mlk_keccak_absorb_once	✅	4s	3s	+33%
mlk_keccakf1600_permute	✅	4s	3s	+33%
mlk_poly_frombytes	✅	4s	3s	+33%
mlk_poly_invntt_tomont	✅	4s	4s	+0%
mlk_poly_ntt_c	✅	4s	1s	+300%
mlk_poly_tomsg	✅	4s	4s	+0%
mlk_polymat_permute_bitrev_to_custom	✅	4s	4s	+0%
mlk_polyvec_decompress_du	✅	4s	1s	+300%
mlk_polyvec_invntt_tomont	✅	4s	1s	+300%
mlk_polyvec_permute_bitrev_to_custom_native	✅	4s	4s	+0%
poly_frombytes_native_x86_64	✅	4s	5s	-20%
poly_invntt_tomont_native	✅	4s	3s	+33%
poly_reduce_native_x86_64	✅	4s	3s	+33%
intt_native_x86_64	✅	3s	2s	+50%
kem_keypair_derand	✅	3s	2s	+50%
mlk_check_pct	✅	3s	2s	+50%
mlk_ct_sel_uint8	✅	3s	4s	-25%
mlk_keccakf1600_extract_bytes (big endian)	✅	3s	3s	+0%
mlk_keccakf1600x4_xor_bytes	✅	3s	2s	+50%
mlk_matvec_mul	✅	3s	2s	+50%
mlk_poly_cbd_eta1	✅	3s	4s	-25%
mlk_poly_cbd_eta2	✅	3s	2s	+50%
mlk_poly_decompress_du	✅	3s	3s	+0%
mlk_poly_getnoise_eta1_4x	✅	3s	2s	+50%
mlk_poly_tobytes_c	✅	3s	2s	+50%
mlk_poly_tobytes_native	✅	3s	1s	+200%
mlk_polyvec_basemul_acc_montgomery_cached	✅	3s	2s	+50%
mlk_polyvec_frombytes	✅	3s	3s	+0%
mlk_polyvec_mulcache_compute	✅	3s	2s	+50%
mlk_polyvec_reduce	✅	3s	2s	+50%
mlk_scalar_compress_d1	✅	3s	4s	-25%
mlk_scalar_decompress_d10	✅	3s	3s	+0%
mlk_scalar_signed_to_unsigned_q	✅	3s	2s	+50%
mlk_sha3_256	✅	3s	4s	-25%
mlk_shake128_squeezeblocks	✅	3s	1s	+200%
mlk_shake128x4_absorb_once	✅	3s	2s	+50%
mlk_shake128x4_squeezeblocks	✅	3s	4s	-25%
mlk_shake256	✅	3s	1s	+200%
mlk_shake256x4	✅	3s	4s	-25%
nttunpack_native_x86_64	✅	3s	5s	-40%
poly_getnoise_eta1122_4x_native	✅	3s	3s	+0%
poly_mulcache_compute_native_aarch64	✅	3s	2s	+50%
poly_mulcache_compute_native_x86_64	✅	3s	3s	+0%
poly_reduce_native_aarch64	✅	3s	2s	+50%
poly_tobytes_native_aarch64	✅	3s	3s	+0%
poly_tomont_native_x86_64	✅	3s	2s	+50%
polyvec_basemul_acc_montgomery_cached_k2_native_aarch64	✅	3s	2s	+50%
polyvec_basemul_acc_montgomery_cached_k2_native_x86_64	✅	3s	2s	+50%
polyvec_basemul_acc_montgomery_cached_k3_native_aarch64	✅	3s	3s	+0%
polyvec_basemul_acc_montgomery_cached_k3_native_x86_64	✅	3s	3s	+0%
polyvec_basemul_acc_montgomery_cached_k4_native_x86_64	✅	3s	3s	+0%
rej_uniform_native_aarch64	✅	3s	5s	-40%
rej_uniform_native_x86_64	✅	3s	2s	+50%
intt_native_aarch64	✅	2s	3s	-33%
keccak_f1600_x1_native_aarch64_v84a	✅	2s	3s	-33%
keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid	✅	2s	2s	+0%
keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid	✅	2s	3s	-33%
kem_enc	✅	2s	1s	+100%
mlk_ct_cmask_nonzero_u16	✅	2s	2s	+0%
mlk_ct_cmask_nonzero_u8	✅	2s	2s	+0%
mlk_ct_cmov_zero	✅	2s	4s	-50%
mlk_ct_get_optblocker_i32	✅	2s	2s	+0%
mlk_ct_get_optblocker_u32	✅	2s	2s	+0%
mlk_ct_get_optblocker_u8	✅	2s	2s	+0%
mlk_keccakf1600_extract_bytes	✅	2s	1s	+100%
mlk_keccakf1600_xor_bytes (big endian)	✅	2s	1s	+100%
mlk_keccakf1600x4_extract_bytes	✅	2s	2s	+0%
mlk_keccakf1600x4_permute	✅	2s	4s	-50%
mlk_montgomery_reduce	✅	2s	1s	+100%
mlk_poly_compress_dv	✅	2s	3s	-33%
mlk_poly_getnoise_eta1122_4x	✅	2s	3s	-33%
mlk_poly_getnoise_eta1_4x_native	✅	2s	2s	+0%
mlk_poly_getnoise_eta2	✅	2s	2s	+0%
mlk_poly_invntt_tomont_c	✅	2s	2s	+0%
mlk_poly_mulcache_compute	✅	2s	1s	+100%
mlk_poly_mulcache_compute_native	✅	2s	2s	+0%
mlk_poly_ntt	✅	2s	2s	+0%
mlk_poly_reduce	✅	2s	3s	-33%
mlk_poly_tobytes	✅	2s	4s	-50%
mlk_poly_tomont_c	✅	2s	2s	+0%
mlk_poly_tomont_native	✅	2s	2s	+0%
mlk_rej_uniform	✅	2s	2s	+0%
mlk_scalar_compress_d4	✅	2s	3s	-33%
mlk_scalar_compress_d5	✅	2s	3s	-33%
mlk_scalar_decompress_d4	✅	2s	2s	+0%
mlk_scalar_decompress_d5	✅	2s	3s	-33%
mlk_shake128_absorb_once	✅	2s	2s	+0%
mlk_value_barrier_i32	✅	2s	2s	+0%
mlk_value_barrier_u32	✅	2s	4s	-50%
ntt_native_aarch64	✅	2s	3s	-33%
poly_tobytes_native_x86_64	✅	2s	2s	+0%
polyvec_basemul_acc_montgomery_cached_k4_native_aarch64	✅	2s	3s	-33%
keccak_f1600_x1_native_aarch64	✅	1s	2s	-50%
mlk_barrett_reduce	✅	1s	2s	-50%
mlk_ct_cmask_neg_i16	✅	1s	1s	+0%
mlk_ct_memcmp	✅	1s	3s	-67%
mlk_keccakf1600_xor_bytes	✅	1s	3s	-67%
mlk_poly_add	✅	1s	1s	+0%
mlk_poly_frombytes_c	✅	1s	1s	+0%
mlk_poly_mulcache_compute_c	✅	1s	3s	-67%
mlk_poly_reduce_c	✅	1s	2s	-50%
mlk_poly_tomont	✅	1s	3s	-67%
mlk_polyvec_compress_du	✅	1s	2s	-50%
mlk_polyvec_permute_bitrev_to_custom	✅	1s	3s	-67%
mlk_polyvec_tobytes	✅	1s	1s	+0%
mlk_polyvec_tomont	✅	1s	2s	-50%
mlk_scalar_compress_d10	✅	1s	3s	-67%
mlk_scalar_compress_d11	✅	1s	2s	-50%
mlk_scalar_decompress_d11	✅	1s	4s	-75%
mlk_sha3_512	✅	1s	2s	-50%
ntt_native_x86_64	✅	1s	3s	-67%
poly_tomont_native_aarch64	✅	1s	2s	-50%
rej_uniform_native	✅	1s	2s	-50%
sys_check_capability	✅	1s	3s	-67%

[oqs-bot]

Intel Xeon 4th gen (c7i)

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	9556 cycles	9673 cycles	0.99
ML-KEM-512 encaps	10932 cycles	11477 cycles	0.95
ML-KEM-512 decaps	15058 cycles	15381 cycles	0.98
ML-KEM-768 keypair	16392 cycles	16417 cycles	1.00
ML-KEM-768 encaps	17697 cycles	17862 cycles	0.99
ML-KEM-768 decaps	23875 cycles	23516 cycles	1.02
ML-KEM-1024 keypair	22305 cycles	22324 cycles	1.00
ML-KEM-1024 encaps	24559 cycles	24598 cycles	1.00
ML-KEM-1024 decaps	32295 cycles	32492 cycles	0.99

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

[oqs-bot]

AMD EPYC 3rd gen (c6a)

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	14065 cycles	16694 cycles	0.84
ML-KEM-512 encaps	15769 cycles	18347 cycles	0.86
ML-KEM-512 decaps	21130 cycles	23698 cycles	0.89
ML-KEM-768 keypair	23751 cycles	28491 cycles	0.83
ML-KEM-768 encaps	25032 cycles	29745 cycles	0.84
ML-KEM-768 decaps	32982 cycles	37597 cycles	0.88
ML-KEM-1024 keypair	33466 cycles	41238 cycles	0.81
ML-KEM-1024 encaps	35796 cycles	43448 cycles	0.82
ML-KEM-1024 decaps	46212 cycles	53882 cycles	0.86

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

[oqs-bot]

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

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	28458 cycles	28486 cycles	1.00
ML-KEM-512 encaps	35805 cycles	35832 cycles	1.00
ML-KEM-512 decaps	45572 cycles	45388 cycles	1.00
ML-KEM-768 keypair	45946 cycles	45961 cycles	1.00
ML-KEM-768 encaps	55956 cycles	56259 cycles	0.99
ML-KEM-768 decaps	69443 cycles	69350 cycles	1.00
ML-KEM-1024 keypair	71747 cycles	71758 cycles	1.00
ML-KEM-1024 encaps	84456 cycles	84551 cycles	1.00
ML-KEM-1024 decaps	101536 cycles	101569 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: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	14466 cycles	16360 cycles	0.88
ML-KEM-512 encaps	16857 cycles	18696 cycles	0.90
ML-KEM-512 decaps	23388 cycles	25288 cycles	0.92
ML-KEM-768 keypair	25398 cycles	29401 cycles	0.86
ML-KEM-768 encaps	26818 cycles	29789 cycles	0.90
ML-KEM-768 decaps	36512 cycles	39271 cycles	0.93
ML-KEM-1024 keypair	35940 cycles	37691 cycles	0.95
ML-KEM-1024 encaps	36379 cycles	40668 cycles	0.89
ML-KEM-1024 decaps	49505 cycles	54398 cycles	0.91

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

[oqs-bot]

AMD EPYC 4th gen (c7a)

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	10566 cycles	11999 cycles	0.88
ML-KEM-512 encaps	12186 cycles	13180 cycles	0.92
ML-KEM-512 decaps	17026 cycles	18037 cycles	0.94
ML-KEM-768 keypair	18422 cycles	20544 cycles	0.90
ML-KEM-768 encaps	19458 cycles	21544 cycles	0.90
ML-KEM-768 decaps	26625 cycles	28655 cycles	0.93
ML-KEM-1024 keypair	24696 cycles	27796 cycles	0.89
ML-KEM-1024 encaps	26983 cycles	29927 cycles	0.90
ML-KEM-1024 decaps	36303 cycles	39285 cycles	0.92

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

[oqs-bot]

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

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	40291 cycles	40267 cycles	1.00
ML-KEM-512 encaps	48317 cycles	48316 cycles	1.00
ML-KEM-512 decaps	62449 cycles	62424 cycles	1.00
ML-KEM-768 keypair	63569 cycles	63564 cycles	1.00
ML-KEM-768 encaps	74735 cycles	74898 cycles	1.00
ML-KEM-768 decaps	93150 cycles	93130 cycles	1.00
ML-KEM-1024 keypair	94966 cycles	94896 cycles	1.00
ML-KEM-1024 encaps	109013 cycles	108876 cycles	1.00
ML-KEM-1024 decaps	131917 cycles	131809 cycles	1.00

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

[oqs-bot]

Graviton3

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	18745 cycles	18742 cycles	1.00
ML-KEM-512 encaps	22006 cycles	22003 cycles	1.00
ML-KEM-512 decaps	29029 cycles	29023 cycles	1.00
ML-KEM-768 keypair	31777 cycles	31761 cycles	1.00
ML-KEM-768 encaps	34923 cycles	34920 cycles	1.00
ML-KEM-768 decaps	45031 cycles	45029 cycles	1.00
ML-KEM-1024 keypair	46378 cycles	46373 cycles	1.00
ML-KEM-1024 encaps	51720 cycles	51719 cycles	1.00
ML-KEM-1024 decaps	65247 cycles	65250 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: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	36516 cycles	36497 cycles	1.00
ML-KEM-512 encaps	43033 cycles	42982 cycles	1.00
ML-KEM-512 decaps	55702 cycles	55646 cycles	1.00
ML-KEM-768 keypair	58639 cycles	58420 cycles	1.00
ML-KEM-768 encaps	67395 cycles	67359 cycles	1.00
ML-KEM-768 decaps	84468 cycles	84372 cycles	1.00
ML-KEM-1024 keypair	88706 cycles	88675 cycles	1.00
ML-KEM-1024 encaps	98904 cycles	98961 cycles	1.00
ML-KEM-1024 decaps	120395 cycles	120433 cycles	1.00

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

[oqs-bot]

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

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	45808 cycles	45849 cycles	1.00
ML-KEM-512 encaps	54744 cycles	54779 cycles	1.00
ML-KEM-512 decaps	70267 cycles	70327 cycles	1.00
ML-KEM-768 keypair	73772 cycles	73822 cycles	1.00
ML-KEM-768 encaps	85228 cycles	85296 cycles	1.00
ML-KEM-768 decaps	106252 cycles	106294 cycles	1.00
ML-KEM-1024 keypair	111786 cycles	111742 cycles	1.00
ML-KEM-1024 encaps	125899 cycles	125872 cycles	1.00
ML-KEM-1024 decaps	151997 cycles	151691 cycles	1.00

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

[oqs-bot]

Graviton3 (no-opt)

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	39060 cycles	39071 cycles	1.00
ML-KEM-512 encaps	44638 cycles	44643 cycles	1.00
ML-KEM-512 decaps	56702 cycles	56716 cycles	1.00
ML-KEM-768 keypair	62443 cycles	62450 cycles	1.00
ML-KEM-768 encaps	70937 cycles	70948 cycles	1.00
ML-KEM-768 decaps	86874 cycles	86891 cycles	1.00
ML-KEM-1024 keypair	96266 cycles	96265 cycles	1.00
ML-KEM-1024 encaps	106329 cycles	106329 cycles	1
ML-KEM-1024 decaps	126776 cycles	126794 cycles	1.00

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

[oqs-bot]

Graviton2

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	28353 cycles	28351 cycles	1.00
ML-KEM-512 encaps	34324 cycles	34327 cycles	1.00
ML-KEM-512 decaps	44558 cycles	44552 cycles	1.00
ML-KEM-768 keypair	47850 cycles	47859 cycles	1.00
ML-KEM-768 encaps	54200 cycles	54212 cycles	1.00
ML-KEM-768 decaps	68726 cycles	68737 cycles	1.00
ML-KEM-1024 keypair	70659 cycles	70597 cycles	1.00
ML-KEM-1024 encaps	79239 cycles	79162 cycles	1.00
ML-KEM-1024 decaps	99009 cycles	98902 cycles	1.00

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

[oqs-bot]

Graviton4

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	17765 cycles	17725 cycles	1.00
ML-KEM-512 encaps	20770 cycles	20732 cycles	1.00
ML-KEM-512 decaps	27205 cycles	27157 cycles	1.00
ML-KEM-768 keypair	30095 cycles	30056 cycles	1.00
ML-KEM-768 encaps	32881 cycles	32832 cycles	1.00
ML-KEM-768 decaps	42235 cycles	42114 cycles	1.00
ML-KEM-1024 keypair	44036 cycles	43983 cycles	1.00
ML-KEM-1024 encaps	49079 cycles	49010 cycles	1.00
ML-KEM-1024 decaps	61941 cycles	61573 cycles	1.01

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

[oqs-bot]

Graviton4 (no-opt)

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	35500 cycles	35503 cycles	1.00
ML-KEM-512 encaps	40226 cycles	40237 cycles	1.00
ML-KEM-512 decaps	51217 cycles	51238 cycles	1.00
ML-KEM-768 keypair	56803 cycles	56812 cycles	1.00
ML-KEM-768 encaps	64724 cycles	64720 cycles	1.00
ML-KEM-768 decaps	79038 cycles	79042 cycles	1.00
ML-KEM-1024 keypair	88036 cycles	88032 cycles	1.00
ML-KEM-1024 encaps	97177 cycles	97178 cycles	1.00
ML-KEM-1024 decaps	116101 cycles	116101 cycles	1

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

[oqs-bot]

Graviton2 (no-opt)

Details

Benchmark suite	Current: af21fb3	Previous: 09c31cb	Ratio
ML-KEM-512 keypair	59177 cycles	59157 cycles	1.00
ML-KEM-512 encaps	68651 cycles	68643 cycles	1.00
ML-KEM-512 decaps	87408 cycles	87390 cycles	1.00
ML-KEM-768 keypair	95658 cycles	95754 cycles	1.00
ML-KEM-768 encaps	109514 cycles	109601 cycles	1.00
ML-KEM-768 decaps	134406 cycles	134498 cycles	1.00
ML-KEM-1024 keypair	146936 cycles	148574 cycles	0.99
ML-KEM-1024 encaps	163016 cycles	164464 cycles	0.99
ML-KEM-1024 decaps	194266 cycles	195790 cycles	0.99

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

[hanno-becker]

@manastasova Is this still relevant? Closing for now. Please re-open and comment on your plans for it otherwise.