infernap12/jdk
1
0
Fork 0
mirror of https://github.com/openjdk/jdk.git synced 2026-01-28 03:58:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

8374351: RISC-V: Small refactoring for crypto macro-assembler routines

Browse Source 
Reviewed-by: fyang, fjiang
This commit is contained in:
Anjian Wen 2025-12-28 09:13:09 +00:00
parent 9512a43e82
commit 5e685f6f2c
1 changed files with 55 additions and 45 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

100
src/hotspot/cpu/riscv/stubGenerator_riscv.cpp
View File

@ -2436,7 +2436,7 @@ class StubGenerator: public StubCodeGenerator {
StubRoutines::_unsafe_setmemory = generate_unsafe_setmemory();
}
void generate_aes_loadkeys(const Register &key, VectorRegister *working_vregs, int rounds) {
void aes_load_keys(const Register &key, VectorRegister *working_vregs, int rounds) {
const int step = 16;
for (int i = 0; i < rounds; i++) {
__ vle32_v(working_vregs[i], key);
@ -2448,7 +2448,7 @@ class StubGenerator: public StubCodeGenerator {
}
}
void generate_aes_encrypt(const VectorRegister &res, VectorRegister *working_vregs, int rounds) {
void aes_encrypt(const VectorRegister &res, VectorRegister *working_vregs, int rounds) {
assert(rounds <= 15, "rounds should be less than or equal to working_vregs size");
__ vxor_vv(res, res, working_vregs[0]);
@ -2499,8 +2499,8 @@ class StubGenerator: public StubCodeGenerator {
// Else we fallthrough to the biggest case (256-bit key size)
// Note: the following function performs key += 15*16
generate_aes_loadkeys(key, working_vregs, 15);
generate_aes_encrypt(res, working_vregs, 15);
aes_load_keys(key, working_vregs, 15);
aes_encrypt(res, working_vregs, 15);
__ vse32_v(res, to);
__ mv(c_rarg0, 0);
__ leave();
@ -2508,8 +2508,8 @@ class StubGenerator: public StubCodeGenerator {
__ bind(L_aes192);
// Note: the following function performs key += 13*16
generate_aes_loadkeys(key, working_vregs, 13);
generate_aes_encrypt(res, working_vregs, 13);
aes_load_keys(key, working_vregs, 13);
aes_encrypt(res, working_vregs, 13);
__ vse32_v(res, to);
__ mv(c_rarg0, 0);
__ leave();
@ -2517,8 +2517,8 @@ class StubGenerator: public StubCodeGenerator {
__ bind(L_aes128);
// Note: the following function performs key += 11*16
generate_aes_loadkeys(key, working_vregs, 11);
generate_aes_encrypt(res, working_vregs, 11);
aes_load_keys(key, working_vregs, 11);
aes_encrypt(res, working_vregs, 11);
__ vse32_v(res, to);
__ mv(c_rarg0, 0);
__ leave();
@ -2527,7 +2527,7 @@ class StubGenerator: public StubCodeGenerator {
return start;
}
void generate_aes_decrypt(const VectorRegister &res, VectorRegister *working_vregs, int rounds) {
void aes_decrypt(const VectorRegister &res, VectorRegister *working_vregs, int rounds) {
assert(rounds <= 15, "rounds should be less than or equal to working_vregs size");
__ vxor_vv(res, res, working_vregs[rounds - 1]);
@ -2578,8 +2578,8 @@ class StubGenerator: public StubCodeGenerator {
// Else we fallthrough to the biggest case (256-bit key size)
// Note: the following function performs key += 15*16
generate_aes_loadkeys(key, working_vregs, 15);
generate_aes_decrypt(res, working_vregs, 15);
aes_load_keys(key, working_vregs, 15);
aes_decrypt(res, working_vregs, 15);
__ vse32_v(res, to);
__ mv(c_rarg0, 0);
__ leave();
@ -2587,8 +2587,8 @@ class StubGenerator: public StubCodeGenerator {
__ bind(L_aes192);
// Note: the following function performs key += 13*16
generate_aes_loadkeys(key, working_vregs, 13);
generate_aes_decrypt(res, working_vregs, 13);
aes_load_keys(key, working_vregs, 13);
aes_decrypt(res, working_vregs, 13);
__ vse32_v(res, to);
__ mv(c_rarg0, 0);
__ leave();
@ -2596,8 +2596,8 @@ class StubGenerator: public StubCodeGenerator {
__ bind(L_aes128);
// Note: the following function performs key += 11*16
generate_aes_loadkeys(key, working_vregs, 11);
generate_aes_decrypt(res, working_vregs, 11);
aes_load_keys(key, working_vregs, 11);
aes_decrypt(res, working_vregs, 11);
__ vse32_v(res, to);
__ mv(c_rarg0, 0);
__ leave();
@ -2622,14 +2622,14 @@ class StubGenerator: public StubCodeGenerator {
__ vsetivli(x0, 4, Assembler::e32, Assembler::m1);
__ vle32_v(v16, rvec);
generate_aes_loadkeys(key, working_vregs, round);
aes_load_keys(key, working_vregs, round);
Label L_enc_loop;
__ bind(L_enc_loop);
// Encrypt from source by block size
__ vle32_v(v17, from);
__ addi(from, from, BLOCK_SIZE);
__ vxor_vv(v16, v16, v17);
generate_aes_encrypt(v16, working_vregs, round);
aes_encrypt(v16, working_vregs, round);
__ vse32_v(v16, to);
__ addi(to, to, BLOCK_SIZE);
__ subi(len, len, BLOCK_SIZE);
@ -2709,14 +2709,14 @@ class StubGenerator: public StubCodeGenerator {
__ vsetivli(x0, 4, Assembler::e32, Assembler::m1);
__ vle32_v(v16, rvec);
generate_aes_loadkeys(key, working_vregs, round);
aes_load_keys(key, working_vregs, round);
Label L_dec_loop;
// Decrypt from source by block size
__ bind(L_dec_loop);
__ vle32_v(v17, from);
__ addi(from, from, BLOCK_SIZE);
__ vmv_v_v(v18, v17);
generate_aes_decrypt(v17, working_vregs, round);
aes_decrypt(v17, working_vregs, round);
__ vxor_vv(v17, v17, v16);
__ vse32_v(v17, to);
__ vmv_v_v(v16, v18);
@ -2811,7 +2811,7 @@ class StubGenerator: public StubCodeGenerator {
Register input_len, Register saved_encrypted_ctr, Register used_ptr) {
// Algorithm:
//
// generate_aes_loadkeys();
// aes_load_keys();
// load_counter_128(counter_hi, counter_lo, counter);
//
// L_next:
@ -2825,7 +2825,7 @@ class StubGenerator: public StubCodeGenerator {
//
// L_main_loop:
// if (len == 0) goto L_exit;
// saved_encrypted_ctr = generate_aes_encrypt(counter);
// saved_encrypted_ctr = aes_encrypt(counter);
//
// add_counter_128(counter_hi, counter_lo);
// be_store_counter_128(counter_hi, counter_lo, counter);
@ -2869,7 +2869,7 @@ class StubGenerator: public StubCodeGenerator {
__ mv(block_size, BLOCK_SIZE);
// load keys to working_vregs according to round
generate_aes_loadkeys(key, working_vregs, round);
aes_load_keys(key, working_vregs, round);
// 128-bit big-endian load
be_load_counter_128(counter_hi, counter_lo, counter);
@ -2902,7 +2902,7 @@ class StubGenerator: public StubCodeGenerator {
__ vle32_v(v16, counter);
// encrypt counter according to round
generate_aes_encrypt(v16, working_vregs, round);
aes_encrypt(v16, working_vregs, round);
__ vse32_v(v16, saved_encrypted_ctr);
@ -2997,6 +2997,34 @@ class StubGenerator: public StubCodeGenerator {
return start;
}
void ghash_loop(Register state, Register subkeyH, Register data, Register blocks,
VectorRegister vtmp1, VectorRegister vtmp2, VectorRegister vtmp3) {
VectorRegister partial_hash = vtmp1;
VectorRegister hash_subkey = vtmp2;
VectorRegister cipher_text = vtmp3;
const unsigned int BLOCK_SIZE = 16;
__ vsetivli(x0, 2, Assembler::e64, Assembler::m1);
__ vle64_v(hash_subkey, subkeyH);
__ vrev8_v(hash_subkey, hash_subkey);
__ vle64_v(partial_hash, state);
__ vrev8_v(partial_hash, partial_hash);
__ vsetivli(x0, 4, Assembler::e32, Assembler::m1);
Label L_ghash_loop;
__ bind(L_ghash_loop);
__ vle32_v(cipher_text, data);
__ addi(data, data, BLOCK_SIZE);
__ vghsh_vv(partial_hash, hash_subkey, cipher_text);
__ subi(blocks, blocks, 1);
__ bnez(blocks, L_ghash_loop);
__ vsetivli(x0, 2, Assembler::e64, Assembler::m1);
__ vrev8_v(partial_hash, partial_hash);
__ vse64_v(partial_hash, state);
}
/**
* Arguments:
*
@ -3024,30 +3052,12 @@ class StubGenerator: public StubCodeGenerator {
Register data = c_rarg2;
Register blocks = c_rarg3;
VectorRegister partial_hash = v1;
VectorRegister hash_subkey = v2;
VectorRegister cipher_text = v3;
VectorRegister vtmp1 = v1;
VectorRegister vtmp2 = v2;
VectorRegister vtmp3 = v3;
const unsigned int BLOCK_SIZE = 16;
ghash_loop(state, subkeyH, data, blocks, vtmp1, vtmp2, vtmp3);
__ vsetivli(x0, 2, Assembler::e64, Assembler::m1);
__ vle64_v(hash_subkey, subkeyH);
__ vrev8_v(hash_subkey, hash_subkey);
__ vle64_v(partial_hash, state);
__ vrev8_v(partial_hash, partial_hash);
__ vsetivli(x0, 4, Assembler::e32, Assembler::m1);
Label L_ghash_loop;
__ bind(L_ghash_loop);
__ vle32_v(cipher_text, data);
__ addi(data, data, BLOCK_SIZE);
__ vghsh_vv(partial_hash, hash_subkey, cipher_text);
__ subi(blocks, blocks, 1);
__ bnez(blocks, L_ghash_loop);
__ vsetivli(x0, 2, Assembler::e64, Assembler::m1);
__ vrev8_v(partial_hash, partial_hash);
__ vse64_v(partial_hash, state);
__ leave();
__ ret();