Skip to content

Commit 90a3ce6

Browse files
committed
native: Unify asm backend symbol naming
Every aarch64 asm symbol now ends in `_aarch64_asm`; every x86_64 avx2 symbol now ends in `_avx2_asm`. Signed-off-by: Matthias J. Kannwischer <matthias@zerorisc.com>
1 parent d4eb4e9 commit 90a3ce6

193 files changed

Lines changed: 1225 additions & 1203 deletions

File tree

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

BIBLIOGRAPHY.md

Lines changed: 59 additions & 59 deletions
Original file line numberDiff line numberDiff line change
@@ -23,10 +23,10 @@ source code and documentation.
2323
- Gregor Seiler
2424
* URL: https://eprint.iacr.org/2018/039
2525
* Referenced from:
26-
- [dev/x86_64/src/intt.S](dev/x86_64/src/intt.S)
27-
- [dev/x86_64/src/ntt.S](dev/x86_64/src/ntt.S)
28-
- [mlkem/src/native/x86_64/src/intt.S](mlkem/src/native/x86_64/src/intt.S)
29-
- [mlkem/src/native/x86_64/src/ntt.S](mlkem/src/native/x86_64/src/ntt.S)
26+
- [dev/x86_64/src/intt_avx2_asm.S](dev/x86_64/src/intt_avx2_asm.S)
27+
- [dev/x86_64/src/ntt_avx2_asm.S](dev/x86_64/src/ntt_avx2_asm.S)
28+
- [mlkem/src/native/x86_64/src/intt_avx2_asm.S](mlkem/src/native/x86_64/src/intt_avx2_asm.S)
29+
- [mlkem/src/native/x86_64/src/ntt_avx2_asm.S](mlkem/src/native/x86_64/src/ntt_avx2_asm.S)
3030
- [proofs/hol_light/x86_64/mlkem/mlkem_intt.S](proofs/hol_light/x86_64/mlkem/mlkem_intt.S)
3131
- [proofs/hol_light/x86_64/mlkem/mlkem_ntt.S](proofs/hol_light/x86_64/mlkem/mlkem_ntt.S)
3232

@@ -182,11 +182,11 @@ source code and documentation.
182182
* Referenced from:
183183
- [README.md](README.md)
184184
- [dev/fips202/aarch64/auto.h](dev/fips202/aarch64/auto.h)
185-
- [dev/fips202/aarch64/src/keccak_f1600_x1_v84a_asm.S](dev/fips202/aarch64/src/keccak_f1600_x1_v84a_asm.S)
186-
- [dev/fips202/aarch64/src/keccak_f1600_x2_v84a_asm.S](dev/fips202/aarch64/src/keccak_f1600_x2_v84a_asm.S)
185+
- [dev/fips202/aarch64/src/keccak_f1600_x1_v84a_aarch64_asm.S](dev/fips202/aarch64/src/keccak_f1600_x1_v84a_aarch64_asm.S)
186+
- [dev/fips202/aarch64/src/keccak_f1600_x2_v84a_aarch64_asm.S](dev/fips202/aarch64/src/keccak_f1600_x2_v84a_aarch64_asm.S)
187187
- [mlkem/src/fips202/native/aarch64/auto.h](mlkem/src/fips202/native/aarch64/auto.h)
188-
- [mlkem/src/fips202/native/aarch64/src/keccak_f1600_x1_v84a_asm.S](mlkem/src/fips202/native/aarch64/src/keccak_f1600_x1_v84a_asm.S)
189-
- [mlkem/src/fips202/native/aarch64/src/keccak_f1600_x2_v84a_asm.S](mlkem/src/fips202/native/aarch64/src/keccak_f1600_x2_v84a_asm.S)
188+
- [mlkem/src/fips202/native/aarch64/src/keccak_f1600_x1_v84a_aarch64_asm.S](mlkem/src/fips202/native/aarch64/src/keccak_f1600_x1_v84a_aarch64_asm.S)
189+
- [mlkem/src/fips202/native/aarch64/src/keccak_f1600_x2_v84a_aarch64_asm.S](mlkem/src/fips202/native/aarch64/src/keccak_f1600_x2_v84a_aarch64_asm.S)
190190
- [proofs/hol_light/README.md](proofs/hol_light/README.md)
191191
- [proofs/hol_light/aarch64/mlkem/keccak_f1600_x1_v84a.S](proofs/hol_light/aarch64/mlkem/keccak_f1600_x1_v84a.S)
192192
- [proofs/hol_light/aarch64/mlkem/keccak_f1600_x2_v84a.S](proofs/hol_light/aarch64/mlkem/keccak_f1600_x2_v84a.S)
@@ -222,23 +222,23 @@ source code and documentation.
222222
* URL: https://eprint.iacr.org/2021/986
223223
* Referenced from:
224224
- [dev/aarch64_clean/README.md](dev/aarch64_clean/README.md)
225-
- [dev/aarch64_clean/src/intt.S](dev/aarch64_clean/src/intt.S)
226-
- [dev/aarch64_clean/src/ntt.S](dev/aarch64_clean/src/ntt.S)
227-
- [dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_asm_k2.S](dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_asm_k2.S)
228-
- [dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_asm_k3.S](dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_asm_k3.S)
229-
- [dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_asm_k4.S](dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_asm_k4.S)
225+
- [dev/aarch64_clean/src/intt_aarch64_asm.S](dev/aarch64_clean/src/intt_aarch64_asm.S)
226+
- [dev/aarch64_clean/src/ntt_aarch64_asm.S](dev/aarch64_clean/src/ntt_aarch64_asm.S)
227+
- [dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_k2_aarch64_asm.S](dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_k2_aarch64_asm.S)
228+
- [dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_k3_aarch64_asm.S](dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_k3_aarch64_asm.S)
229+
- [dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_k4_aarch64_asm.S](dev/aarch64_clean/src/polyvec_basemul_acc_montgomery_cached_k4_aarch64_asm.S)
230230
- [dev/aarch64_opt/README.md](dev/aarch64_opt/README.md)
231-
- [dev/aarch64_opt/src/intt.S](dev/aarch64_opt/src/intt.S)
232-
- [dev/aarch64_opt/src/ntt.S](dev/aarch64_opt/src/ntt.S)
233-
- [dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_asm_k2.S](dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_asm_k2.S)
234-
- [dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_asm_k3.S](dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_asm_k3.S)
235-
- [dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_asm_k4.S](dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_asm_k4.S)
231+
- [dev/aarch64_opt/src/intt_aarch64_asm.S](dev/aarch64_opt/src/intt_aarch64_asm.S)
232+
- [dev/aarch64_opt/src/ntt_aarch64_asm.S](dev/aarch64_opt/src/ntt_aarch64_asm.S)
233+
- [dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_k2_aarch64_asm.S](dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_k2_aarch64_asm.S)
234+
- [dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_k3_aarch64_asm.S](dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_k3_aarch64_asm.S)
235+
- [dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_k4_aarch64_asm.S](dev/aarch64_opt/src/polyvec_basemul_acc_montgomery_cached_k4_aarch64_asm.S)
236236
- [mlkem/src/native/aarch64/README.md](mlkem/src/native/aarch64/README.md)
237-
- [mlkem/src/native/aarch64/src/intt.S](mlkem/src/native/aarch64/src/intt.S)
238-
- [mlkem/src/native/aarch64/src/ntt.S](mlkem/src/native/aarch64/src/ntt.S)
239-
- [mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_asm_k2.S](mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_asm_k2.S)
240-
- [mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_asm_k3.S](mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_asm_k3.S)
241-
- [mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_asm_k4.S](mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_asm_k4.S)
237+
- [mlkem/src/native/aarch64/src/intt_aarch64_asm.S](mlkem/src/native/aarch64/src/intt_aarch64_asm.S)
238+
- [mlkem/src/native/aarch64/src/ntt_aarch64_asm.S](mlkem/src/native/aarch64/src/ntt_aarch64_asm.S)
239+
- [mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_k2_aarch64_asm.S](mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_k2_aarch64_asm.S)
240+
- [mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_k3_aarch64_asm.S](mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_k3_aarch64_asm.S)
241+
- [mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_k4_aarch64_asm.S](mlkem/src/native/aarch64/src/polyvec_basemul_acc_montgomery_cached_k4_aarch64_asm.S)
242242
- [mlkem/src/poly.c](mlkem/src/poly.c)
243243
- [mlkem/src/poly_k.c](mlkem/src/poly_k.c)
244244
- [proofs/hol_light/aarch64/mlkem/mlkem_intt.S](proofs/hol_light/aarch64/mlkem/mlkem_intt.S)
@@ -290,36 +290,36 @@ source code and documentation.
290290
* URL: https://github.com/pq-crystals/kyber/tree/main/avx2
291291
* Referenced from:
292292
- [dev/x86_64/README.md](dev/x86_64/README.md)
293-
- [dev/x86_64/src/intt.S](dev/x86_64/src/intt.S)
294-
- [dev/x86_64/src/ntt.S](dev/x86_64/src/ntt.S)
295-
- [dev/x86_64/src/nttfrombytes.S](dev/x86_64/src/nttfrombytes.S)
296-
- [dev/x86_64/src/ntttobytes.S](dev/x86_64/src/ntttobytes.S)
297-
- [dev/x86_64/src/nttunpack.S](dev/x86_64/src/nttunpack.S)
298-
- [dev/x86_64/src/poly_compress_d10.S](dev/x86_64/src/poly_compress_d10.S)
299-
- [dev/x86_64/src/poly_compress_d11.S](dev/x86_64/src/poly_compress_d11.S)
300-
- [dev/x86_64/src/poly_compress_d4.S](dev/x86_64/src/poly_compress_d4.S)
301-
- [dev/x86_64/src/poly_compress_d5.S](dev/x86_64/src/poly_compress_d5.S)
302-
- [dev/x86_64/src/poly_decompress_d10.S](dev/x86_64/src/poly_decompress_d10.S)
303-
- [dev/x86_64/src/poly_decompress_d11.S](dev/x86_64/src/poly_decompress_d11.S)
304-
- [dev/x86_64/src/poly_decompress_d4.S](dev/x86_64/src/poly_decompress_d4.S)
305-
- [dev/x86_64/src/poly_decompress_d5.S](dev/x86_64/src/poly_decompress_d5.S)
306-
- [dev/x86_64/src/reduce.S](dev/x86_64/src/reduce.S)
307-
- [dev/x86_64/src/tomont.S](dev/x86_64/src/tomont.S)
308-
- [mlkem/src/native/x86_64/src/intt.S](mlkem/src/native/x86_64/src/intt.S)
309-
- [mlkem/src/native/x86_64/src/ntt.S](mlkem/src/native/x86_64/src/ntt.S)
310-
- [mlkem/src/native/x86_64/src/nttfrombytes.S](mlkem/src/native/x86_64/src/nttfrombytes.S)
311-
- [mlkem/src/native/x86_64/src/ntttobytes.S](mlkem/src/native/x86_64/src/ntttobytes.S)
312-
- [mlkem/src/native/x86_64/src/nttunpack.S](mlkem/src/native/x86_64/src/nttunpack.S)
313-
- [mlkem/src/native/x86_64/src/poly_compress_d10.S](mlkem/src/native/x86_64/src/poly_compress_d10.S)
314-
- [mlkem/src/native/x86_64/src/poly_compress_d11.S](mlkem/src/native/x86_64/src/poly_compress_d11.S)
315-
- [mlkem/src/native/x86_64/src/poly_compress_d4.S](mlkem/src/native/x86_64/src/poly_compress_d4.S)
316-
- [mlkem/src/native/x86_64/src/poly_compress_d5.S](mlkem/src/native/x86_64/src/poly_compress_d5.S)
317-
- [mlkem/src/native/x86_64/src/poly_decompress_d10.S](mlkem/src/native/x86_64/src/poly_decompress_d10.S)
318-
- [mlkem/src/native/x86_64/src/poly_decompress_d11.S](mlkem/src/native/x86_64/src/poly_decompress_d11.S)
319-
- [mlkem/src/native/x86_64/src/poly_decompress_d4.S](mlkem/src/native/x86_64/src/poly_decompress_d4.S)
320-
- [mlkem/src/native/x86_64/src/poly_decompress_d5.S](mlkem/src/native/x86_64/src/poly_decompress_d5.S)
321-
- [mlkem/src/native/x86_64/src/reduce.S](mlkem/src/native/x86_64/src/reduce.S)
322-
- [mlkem/src/native/x86_64/src/tomont.S](mlkem/src/native/x86_64/src/tomont.S)
293+
- [dev/x86_64/src/intt_avx2_asm.S](dev/x86_64/src/intt_avx2_asm.S)
294+
- [dev/x86_64/src/ntt_avx2_asm.S](dev/x86_64/src/ntt_avx2_asm.S)
295+
- [dev/x86_64/src/nttfrombytes_avx2_asm.S](dev/x86_64/src/nttfrombytes_avx2_asm.S)
296+
- [dev/x86_64/src/ntttobytes_avx2_asm.S](dev/x86_64/src/ntttobytes_avx2_asm.S)
297+
- [dev/x86_64/src/nttunpack_avx2_asm.S](dev/x86_64/src/nttunpack_avx2_asm.S)
298+
- [dev/x86_64/src/poly_compress_d10_avx2_asm.S](dev/x86_64/src/poly_compress_d10_avx2_asm.S)
299+
- [dev/x86_64/src/poly_compress_d11_avx2_asm.S](dev/x86_64/src/poly_compress_d11_avx2_asm.S)
300+
- [dev/x86_64/src/poly_compress_d4_avx2_asm.S](dev/x86_64/src/poly_compress_d4_avx2_asm.S)
301+
- [dev/x86_64/src/poly_compress_d5_avx2_asm.S](dev/x86_64/src/poly_compress_d5_avx2_asm.S)
302+
- [dev/x86_64/src/poly_decompress_d10_avx2_asm.S](dev/x86_64/src/poly_decompress_d10_avx2_asm.S)
303+
- [dev/x86_64/src/poly_decompress_d11_avx2_asm.S](dev/x86_64/src/poly_decompress_d11_avx2_asm.S)
304+
- [dev/x86_64/src/poly_decompress_d4_avx2_asm.S](dev/x86_64/src/poly_decompress_d4_avx2_asm.S)
305+
- [dev/x86_64/src/poly_decompress_d5_avx2_asm.S](dev/x86_64/src/poly_decompress_d5_avx2_asm.S)
306+
- [dev/x86_64/src/reduce_avx2_asm.S](dev/x86_64/src/reduce_avx2_asm.S)
307+
- [dev/x86_64/src/tomont_avx2_asm.S](dev/x86_64/src/tomont_avx2_asm.S)
308+
- [mlkem/src/native/x86_64/src/intt_avx2_asm.S](mlkem/src/native/x86_64/src/intt_avx2_asm.S)
309+
- [mlkem/src/native/x86_64/src/ntt_avx2_asm.S](mlkem/src/native/x86_64/src/ntt_avx2_asm.S)
310+
- [mlkem/src/native/x86_64/src/nttfrombytes_avx2_asm.S](mlkem/src/native/x86_64/src/nttfrombytes_avx2_asm.S)
311+
- [mlkem/src/native/x86_64/src/ntttobytes_avx2_asm.S](mlkem/src/native/x86_64/src/ntttobytes_avx2_asm.S)
312+
- [mlkem/src/native/x86_64/src/nttunpack_avx2_asm.S](mlkem/src/native/x86_64/src/nttunpack_avx2_asm.S)
313+
- [mlkem/src/native/x86_64/src/poly_compress_d10_avx2_asm.S](mlkem/src/native/x86_64/src/poly_compress_d10_avx2_asm.S)
314+
- [mlkem/src/native/x86_64/src/poly_compress_d11_avx2_asm.S](mlkem/src/native/x86_64/src/poly_compress_d11_avx2_asm.S)
315+
- [mlkem/src/native/x86_64/src/poly_compress_d4_avx2_asm.S](mlkem/src/native/x86_64/src/poly_compress_d4_avx2_asm.S)
316+
- [mlkem/src/native/x86_64/src/poly_compress_d5_avx2_asm.S](mlkem/src/native/x86_64/src/poly_compress_d5_avx2_asm.S)
317+
- [mlkem/src/native/x86_64/src/poly_decompress_d10_avx2_asm.S](mlkem/src/native/x86_64/src/poly_decompress_d10_avx2_asm.S)
318+
- [mlkem/src/native/x86_64/src/poly_decompress_d11_avx2_asm.S](mlkem/src/native/x86_64/src/poly_decompress_d11_avx2_asm.S)
319+
- [mlkem/src/native/x86_64/src/poly_decompress_d4_avx2_asm.S](mlkem/src/native/x86_64/src/poly_decompress_d4_avx2_asm.S)
320+
- [mlkem/src/native/x86_64/src/poly_decompress_d5_avx2_asm.S](mlkem/src/native/x86_64/src/poly_decompress_d5_avx2_asm.S)
321+
- [mlkem/src/native/x86_64/src/reduce_avx2_asm.S](mlkem/src/native/x86_64/src/reduce_avx2_asm.S)
322+
- [mlkem/src/native/x86_64/src/tomont_avx2_asm.S](mlkem/src/native/x86_64/src/tomont_avx2_asm.S)
323323
- [proofs/hol_light/x86_64/mlkem/mlkem_frombytes.S](proofs/hol_light/x86_64/mlkem/mlkem_frombytes.S)
324324
- [proofs/hol_light/x86_64/mlkem/mlkem_intt.S](proofs/hol_light/x86_64/mlkem/mlkem_intt.S)
325325
- [proofs/hol_light/x86_64/mlkem/mlkem_ntt.S](proofs/hol_light/x86_64/mlkem/mlkem_ntt.S)
@@ -362,14 +362,14 @@ source code and documentation.
362362
- [README.md](README.md)
363363
- [dev/README.md](dev/README.md)
364364
- [dev/aarch64_clean/README.md](dev/aarch64_clean/README.md)
365-
- [dev/aarch64_clean/src/intt.S](dev/aarch64_clean/src/intt.S)
366-
- [dev/aarch64_clean/src/ntt.S](dev/aarch64_clean/src/ntt.S)
365+
- [dev/aarch64_clean/src/intt_aarch64_asm.S](dev/aarch64_clean/src/intt_aarch64_asm.S)
366+
- [dev/aarch64_clean/src/ntt_aarch64_asm.S](dev/aarch64_clean/src/ntt_aarch64_asm.S)
367367
- [dev/aarch64_opt/README.md](dev/aarch64_opt/README.md)
368-
- [dev/aarch64_opt/src/intt.S](dev/aarch64_opt/src/intt.S)
369-
- [dev/aarch64_opt/src/ntt.S](dev/aarch64_opt/src/ntt.S)
368+
- [dev/aarch64_opt/src/intt_aarch64_asm.S](dev/aarch64_opt/src/intt_aarch64_asm.S)
369+
- [dev/aarch64_opt/src/ntt_aarch64_asm.S](dev/aarch64_opt/src/ntt_aarch64_asm.S)
370370
- [mlkem/src/native/aarch64/README.md](mlkem/src/native/aarch64/README.md)
371-
- [mlkem/src/native/aarch64/src/intt.S](mlkem/src/native/aarch64/src/intt.S)
372-
- [mlkem/src/native/aarch64/src/ntt.S](mlkem/src/native/aarch64/src/ntt.S)
371+
- [mlkem/src/native/aarch64/src/intt_aarch64_asm.S](mlkem/src/native/aarch64/src/intt_aarch64_asm.S)
372+
- [mlkem/src/native/aarch64/src/ntt_aarch64_asm.S](mlkem/src/native/aarch64/src/ntt_aarch64_asm.S)
373373
- [proofs/hol_light/aarch64/mlkem/mlkem_intt.S](proofs/hol_light/aarch64/mlkem/mlkem_intt.S)
374374
- [proofs/hol_light/aarch64/mlkem/mlkem_ntt.S](proofs/hol_light/aarch64/mlkem/mlkem_ntt.S)
375375

README.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -99,7 +99,7 @@ mlkem-native currently offers the following backends:
9999
If you'd like contribute new backends, please reach out or just open a PR.
100100

101101
Our AArch64 assembly is developed using the [SLOTHY](https://github.com/slothy-optimizer/slothy) superoptimizer, following the approach described in the SLOTHY paper[^SLOTHY_Paper]:
102-
We write 'clean' assembly by hand and automate micro-optimizations (e.g. see the [clean](dev/aarch64_clean/src/ntt.S) vs [optimized](dev/aarch64_opt/src/ntt.S) AArch64 NTT).
102+
We write 'clean' assembly by hand and automate micro-optimizations (e.g. see the [clean](dev/aarch64_clean/src/ntt_aarch64_asm.S) vs [optimized](dev/aarch64_opt/src/ntt_aarch64_asm.S) AArch64 NTT).
103103
See [dev/README.md](dev/README.md) for more details.
104104

105105
## Test Vectors

SOUNDNESS.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -310,7 +310,7 @@ arithmetic bounds, and constant tables.
310310
satisfies `abs(ival zi) <= &23594` and `(ival zi == forward_ntt (ival o x) i) (mod &3329)`.
311311
In other words, we provide a description of the underlying modular arithmetic function (here, the NTT),
312312
plus a bound on the concrete being computed.
313-
- The CBMC contract on `mlk_ntt_asm` simplifies this to the mere bounds assertions
313+
- The CBMC contract on `mlk_ntt_aarch64_asm` simplifies this to the mere bounds assertions
314314
`requires(array_abs_bound(p, 0, MLKEM_N, 8192))`
315315
and `ensures(array_abs_bound(p, 0, MLKEM_N, 23595))`, omitting the description of the
316316
functional behavior.

dev/aarch64_clean/meta.h

Lines changed: 15 additions & 13 deletions
Original file line numberDiff line numberDiff line change
@@ -28,39 +28,40 @@
2828
MLK_MUST_CHECK_RETURN_VALUE
2929
static MLK_INLINE int mlk_ntt_native(int16_t data[MLKEM_N])
3030
{
31-
mlk_ntt_asm(data, mlk_aarch64_ntt_zetas_layer12345,
32-
mlk_aarch64_ntt_zetas_layer67);
31+
mlk_ntt_aarch64_asm(data, mlk_aarch64_ntt_zetas_layer12345,
32+
mlk_aarch64_ntt_zetas_layer67);
3333
return MLK_NATIVE_FUNC_SUCCESS;
3434
}
3535

3636
MLK_MUST_CHECK_RETURN_VALUE
3737
static MLK_INLINE int mlk_intt_native(int16_t data[MLKEM_N])
3838
{
39-
mlk_intt_asm(data, mlk_aarch64_invntt_zetas_layer12345,
40-
mlk_aarch64_invntt_zetas_layer67);
39+
mlk_intt_aarch64_asm(data, mlk_aarch64_invntt_zetas_layer12345,
40+
mlk_aarch64_invntt_zetas_layer67);
4141
return MLK_NATIVE_FUNC_SUCCESS;
4242
}
4343

4444
MLK_MUST_CHECK_RETURN_VALUE
4545
static MLK_INLINE int mlk_poly_reduce_native(int16_t data[MLKEM_N])
4646
{
47-
mlk_poly_reduce_asm(data);
47+
mlk_poly_reduce_aarch64_asm(data);
4848
return MLK_NATIVE_FUNC_SUCCESS;
4949
}
5050

5151
MLK_MUST_CHECK_RETURN_VALUE
5252
static MLK_INLINE int mlk_poly_tomont_native(int16_t data[MLKEM_N])
5353
{
54-
mlk_poly_tomont_asm(data);
54+
mlk_poly_tomont_aarch64_asm(data);
5555
return MLK_NATIVE_FUNC_SUCCESS;
5656
}
5757

5858
MLK_MUST_CHECK_RETURN_VALUE
5959
static MLK_INLINE int mlk_poly_mulcache_compute_native(int16_t x[MLKEM_N / 2],
6060
const int16_t y[MLKEM_N])
6161
{
62-
mlk_poly_mulcache_compute_asm(x, y, mlk_aarch64_zetas_mulcache_native,
63-
mlk_aarch64_zetas_mulcache_twisted_native);
62+
mlk_poly_mulcache_compute_aarch64_asm(
63+
x, y, mlk_aarch64_zetas_mulcache_native,
64+
mlk_aarch64_zetas_mulcache_twisted_native);
6465
return MLK_NATIVE_FUNC_SUCCESS;
6566
}
6667

@@ -70,7 +71,7 @@ static MLK_INLINE int mlk_polyvec_basemul_acc_montgomery_cached_k2_native(
7071
int16_t r[MLKEM_N], const int16_t a[2 * MLKEM_N],
7172
const int16_t b[2 * MLKEM_N], const int16_t b_cache[2 * (MLKEM_N / 2)])
7273
{
73-
mlk_polyvec_basemul_acc_montgomery_cached_asm_k2(r, a, b, b_cache);
74+
mlk_polyvec_basemul_acc_montgomery_cached_k2_aarch64_asm(r, a, b, b_cache);
7475
return MLK_NATIVE_FUNC_SUCCESS;
7576
}
7677
#endif /* MLK_CONFIG_MULTILEVEL_WITH_SHARED || MLKEM_K == 2 */
@@ -81,7 +82,7 @@ static MLK_INLINE int mlk_polyvec_basemul_acc_montgomery_cached_k3_native(
8182
int16_t r[MLKEM_N], const int16_t a[3 * MLKEM_N],
8283
const int16_t b[3 * MLKEM_N], const int16_t b_cache[3 * (MLKEM_N / 2)])
8384
{
84-
mlk_polyvec_basemul_acc_montgomery_cached_asm_k3(r, a, b, b_cache);
85+
mlk_polyvec_basemul_acc_montgomery_cached_k3_aarch64_asm(r, a, b, b_cache);
8586
return MLK_NATIVE_FUNC_SUCCESS;
8687
}
8788
#endif /* MLK_CONFIG_MULTILEVEL_WITH_SHARED || MLKEM_K == 3 */
@@ -92,7 +93,7 @@ static MLK_INLINE int mlk_polyvec_basemul_acc_montgomery_cached_k4_native(
9293
int16_t r[MLKEM_N], const int16_t a[4 * MLKEM_N],
9394
const int16_t b[4 * MLKEM_N], const int16_t b_cache[4 * (MLKEM_N / 2)])
9495
{
95-
mlk_polyvec_basemul_acc_montgomery_cached_asm_k4(r, a, b, b_cache);
96+
mlk_polyvec_basemul_acc_montgomery_cached_k4_aarch64_asm(r, a, b, b_cache);
9697
return MLK_NATIVE_FUNC_SUCCESS;
9798
}
9899
#endif /* MLK_CONFIG_MULTILEVEL_WITH_SHARED || MLKEM_K == 4 */
@@ -101,7 +102,7 @@ MLK_MUST_CHECK_RETURN_VALUE
101102
static MLK_INLINE int mlk_poly_tobytes_native(uint8_t r[MLKEM_POLYBYTES],
102103
const int16_t a[MLKEM_N])
103104
{
104-
mlk_poly_tobytes_asm(r, a);
105+
mlk_poly_tobytes_aarch64_asm(r, a);
105106
return MLK_NATIVE_FUNC_SUCCESS;
106107
}
107108

@@ -115,7 +116,8 @@ static MLK_INLINE int mlk_rej_uniform_native(int16_t *r, unsigned len,
115116
{
116117
return MLK_NATIVE_FUNC_FALLBACK;
117118
}
118-
return (int)mlk_rej_uniform_asm(r, buf, buflen, mlk_rej_uniform_table);
119+
return (int)mlk_rej_uniform_aarch64_asm(r, buf, buflen,
120+
mlk_rej_uniform_table);
119121
}
120122
#endif /* !__ASSEMBLER__ */
121123

0 commit comments

Comments
 (0)