2020-03-09 05:58:05 +00:00
|
|
|
/**
|
|
|
|
* \brief Multi-precision integer library, ESP-IDF hardware accelerated parts
|
|
|
|
*
|
|
|
|
* based on mbedTLS implementation
|
|
|
|
*
|
|
|
|
* Copyright (C) 2006-2015, ARM Limited, All Rights Reserved
|
|
|
|
* Additions Copyright (C) 2016-2020, Espressif Systems (Shanghai) PTE Ltd
|
|
|
|
* SPDX-License-Identifier: Apache-2.0
|
|
|
|
*
|
|
|
|
* Licensed under the Apache License, Version 2.0 (the "License"); you may
|
|
|
|
* not use this file except in compliance with the License.
|
|
|
|
* You may obtain a copy of the License at
|
|
|
|
*
|
|
|
|
* http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
*
|
|
|
|
* Unless required by applicable law or agreed to in writing, software
|
|
|
|
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
|
|
|
|
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
* See the License for the specific language governing permissions and
|
|
|
|
* limitations under the License.
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
|
2020-03-03 07:01:19 +00:00
|
|
|
#include "soc/hwcrypto_periph.h"
|
2020-11-26 08:56:13 +00:00
|
|
|
#include "soc/dport_reg.h"
|
2021-10-25 09:13:46 +00:00
|
|
|
#include "esp_private/periph_ctrl.h"
|
2020-03-03 07:01:19 +00:00
|
|
|
#include <mbedtls/bignum.h>
|
2020-03-09 05:58:05 +00:00
|
|
|
#include "bignum_impl.h"
|
|
|
|
#include <sys/param.h>
|
2020-04-07 04:30:00 +00:00
|
|
|
#include <sys/lock.h>
|
|
|
|
|
|
|
|
static _lock_t mpi_lock;
|
2020-03-03 07:01:19 +00:00
|
|
|
|
|
|
|
/* Round up number of words to nearest
|
|
|
|
512 bit (16 word) block count.
|
|
|
|
*/
|
2020-03-09 05:58:05 +00:00
|
|
|
size_t esp_mpi_hardware_words(size_t words)
|
2020-03-03 07:01:19 +00:00
|
|
|
{
|
|
|
|
return (words + 0xF) & ~0xF;
|
|
|
|
}
|
|
|
|
|
2020-03-09 05:58:05 +00:00
|
|
|
void esp_mpi_enable_hardware_hw_op( void )
|
2020-03-03 07:01:19 +00:00
|
|
|
{
|
2020-04-07 04:30:00 +00:00
|
|
|
/* newlib locks lazy initialize on ESP-IDF */
|
|
|
|
_lock_acquire(&mpi_lock);
|
|
|
|
|
2020-03-03 07:01:19 +00:00
|
|
|
/* Enable RSA hardware */
|
|
|
|
periph_module_enable(PERIPH_RSA_MODULE);
|
|
|
|
DPORT_REG_CLR_BIT(DPORT_RSA_PD_CTRL_REG, DPORT_RSA_PD);
|
|
|
|
|
2020-03-09 05:58:05 +00:00
|
|
|
while (DPORT_REG_READ(RSA_CLEAN_REG) != 1)
|
|
|
|
{ }
|
2020-03-03 07:01:19 +00:00
|
|
|
// Note: from enabling RSA clock to here takes about 1.3us
|
|
|
|
}
|
|
|
|
|
2020-03-09 05:58:05 +00:00
|
|
|
void esp_mpi_disable_hardware_hw_op( void )
|
2020-03-03 07:01:19 +00:00
|
|
|
{
|
|
|
|
DPORT_REG_SET_BIT(DPORT_RSA_PD_CTRL_REG, DPORT_RSA_PD);
|
|
|
|
|
|
|
|
/* Disable RSA hardware */
|
|
|
|
periph_module_disable(PERIPH_RSA_MODULE);
|
2020-04-07 04:30:00 +00:00
|
|
|
|
|
|
|
_lock_release(&mpi_lock);
|
2020-03-03 07:01:19 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2021-12-28 04:04:05 +00:00
|
|
|
void esp_mpi_interrupt_enable( bool enable )
|
|
|
|
{
|
|
|
|
DPORT_REG_WRITE(RSA_INTERRUPT_REG, enable);
|
|
|
|
}
|
|
|
|
|
|
|
|
void esp_mpi_interrupt_clear( void )
|
|
|
|
{
|
|
|
|
DPORT_REG_WRITE(RSA_CLEAR_INTERRUPT_REG, 1);
|
|
|
|
}
|
|
|
|
|
2020-03-03 07:01:19 +00:00
|
|
|
/* Copy mbedTLS MPI bignum 'mpi' to hardware memory block at 'mem_base'.
|
|
|
|
|
|
|
|
If hw_words is higher than the number of words in the bignum then
|
|
|
|
these additional words will be zeroed in the memory buffer.
|
|
|
|
|
|
|
|
*/
|
|
|
|
static inline void mpi_to_mem_block(uint32_t mem_base, const mbedtls_mpi *mpi, size_t hw_words)
|
|
|
|
{
|
|
|
|
uint32_t *pbase = (uint32_t *)mem_base;
|
2020-03-09 05:58:05 +00:00
|
|
|
uint32_t copy_words = MIN(hw_words, mpi->n);
|
2020-03-03 07:01:19 +00:00
|
|
|
|
|
|
|
/* Copy MPI data to memory block registers */
|
2020-11-17 04:48:35 +00:00
|
|
|
for (uint32_t i = 0; i < copy_words; i++) {
|
2020-03-03 07:01:19 +00:00
|
|
|
pbase[i] = mpi->p[i];
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Zero any remaining memory block data */
|
2020-11-17 04:48:35 +00:00
|
|
|
for (uint32_t i = copy_words; i < hw_words; i++) {
|
2020-03-03 07:01:19 +00:00
|
|
|
pbase[i] = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Read mbedTLS MPI bignum back from hardware memory block.
|
|
|
|
|
|
|
|
Reads num_words words from block.
|
|
|
|
|
|
|
|
Bignum 'x' should already be grown to at least num_words by caller (can be done while
|
|
|
|
calculation is in progress, to save some cycles)
|
|
|
|
*/
|
2020-11-17 04:48:35 +00:00
|
|
|
static inline void mem_block_to_mpi(mbedtls_mpi *x, uint32_t mem_base, size_t num_words)
|
2020-03-03 07:01:19 +00:00
|
|
|
{
|
|
|
|
assert(x->n >= num_words);
|
|
|
|
|
|
|
|
/* Copy data from memory block registers */
|
|
|
|
esp_dport_access_read_buffer(x->p, mem_base, num_words);
|
|
|
|
|
|
|
|
/* Zero any remaining limbs in the bignum, if the buffer is bigger
|
|
|
|
than num_words */
|
2020-03-09 05:58:05 +00:00
|
|
|
for (size_t i = num_words; i < x->n; i++) {
|
2020-03-03 07:01:19 +00:00
|
|
|
x->p[i] = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/* Begin an RSA operation. op_reg specifies which 'START' register
|
|
|
|
to write to.
|
|
|
|
*/
|
|
|
|
static inline void start_op(uint32_t op_reg)
|
|
|
|
{
|
|
|
|
/* Clear interrupt status */
|
|
|
|
DPORT_REG_WRITE(RSA_INTERRUPT_REG, 1);
|
|
|
|
|
|
|
|
/* Note: above REG_WRITE includes a memw, so we know any writes
|
|
|
|
to the memory blocks are also complete. */
|
|
|
|
|
|
|
|
DPORT_REG_WRITE(op_reg, 1);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Wait for an RSA operation to complete.
|
|
|
|
*/
|
2020-03-09 05:58:05 +00:00
|
|
|
static inline void wait_op_complete(void)
|
2020-03-03 07:01:19 +00:00
|
|
|
{
|
2020-03-09 05:58:05 +00:00
|
|
|
while (DPORT_REG_READ(RSA_INTERRUPT_REG) != 1)
|
|
|
|
{ }
|
2020-03-03 07:01:19 +00:00
|
|
|
|
|
|
|
/* clear the interrupt */
|
|
|
|
DPORT_REG_WRITE(RSA_INTERRUPT_REG, 1);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Read result from last MPI operation */
|
2020-03-09 05:58:05 +00:00
|
|
|
void esp_mpi_read_result_hw_op(mbedtls_mpi *Z, size_t z_words)
|
2020-03-03 07:01:19 +00:00
|
|
|
{
|
|
|
|
wait_op_complete();
|
|
|
|
mem_block_to_mpi(Z, RSA_MEM_Z_BLOCK_BASE, z_words);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Z = (X * Y) mod M */
|
2020-03-09 05:58:05 +00:00
|
|
|
void esp_mpi_mul_mpi_mod_hw_op(const mbedtls_mpi *X, const mbedtls_mpi *Y, const mbedtls_mpi *M, const mbedtls_mpi *Rinv, mbedtls_mpi_uint Mprime, size_t hw_words)
|
2020-03-03 07:01:19 +00:00
|
|
|
{
|
|
|
|
/* Load M, X, Rinv, Mprime (Mprime is mod 2^32) */
|
|
|
|
mpi_to_mem_block(RSA_MEM_M_BLOCK_BASE, M, hw_words);
|
|
|
|
mpi_to_mem_block(RSA_MEM_X_BLOCK_BASE, X, hw_words);
|
|
|
|
mpi_to_mem_block(RSA_MEM_RB_BLOCK_BASE, Rinv, hw_words);
|
|
|
|
DPORT_REG_WRITE(RSA_M_DASH_REG, (uint32_t)Mprime);
|
|
|
|
|
|
|
|
/* "mode" register loaded with number of 512-bit blocks, minus 1 */
|
|
|
|
DPORT_REG_WRITE(RSA_MULT_MODE_REG, (hw_words / 16) - 1);
|
|
|
|
|
|
|
|
/* Execute first stage montgomery multiplication */
|
|
|
|
start_op(RSA_MULT_START_REG);
|
|
|
|
|
2020-03-09 05:58:05 +00:00
|
|
|
wait_op_complete();
|
2020-03-03 07:01:19 +00:00
|
|
|
|
|
|
|
/* execute second stage */
|
2020-03-09 05:58:05 +00:00
|
|
|
/* Load Y to X input memory block, rerun */
|
2020-03-03 07:01:19 +00:00
|
|
|
mpi_to_mem_block(RSA_MEM_X_BLOCK_BASE, Y, hw_words);
|
|
|
|
|
|
|
|
start_op(RSA_MULT_START_REG);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Z = X * Y */
|
2020-03-09 05:58:05 +00:00
|
|
|
void esp_mpi_mul_mpi_hw_op(const mbedtls_mpi *X, const mbedtls_mpi *Y, size_t hw_words)
|
2020-03-03 07:01:19 +00:00
|
|
|
{
|
|
|
|
/* Copy X (right-extended) & Y (left-extended) to memory block */
|
|
|
|
mpi_to_mem_block(RSA_MEM_X_BLOCK_BASE, X, hw_words);
|
|
|
|
mpi_to_mem_block(RSA_MEM_Z_BLOCK_BASE + hw_words * 4, Y, hw_words);
|
|
|
|
/* NB: as Y is left-extended, we don't zero the bottom words_mult words of Y block.
|
|
|
|
This is OK for now because zeroing is done by hardware when we do esp_mpi_acquire_hardware().
|
|
|
|
*/
|
|
|
|
|
|
|
|
DPORT_REG_WRITE(RSA_M_DASH_REG, 0);
|
|
|
|
|
|
|
|
/* "mode" register loaded with number of 512-bit blocks in result,
|
|
|
|
plus 7 (for range 9-12). (this is ((N~ / 32) - 1) + 8))
|
|
|
|
*/
|
|
|
|
DPORT_REG_WRITE(RSA_MULT_MODE_REG, ((hw_words * 2) / 16) + 7);
|
|
|
|
|
|
|
|
start_op(RSA_MULT_START_REG);
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2020-03-09 05:58:05 +00:00
|
|
|
int esp_mont_hw_op(mbedtls_mpi *Z, const mbedtls_mpi *X, const mbedtls_mpi *Y, const mbedtls_mpi *M,
|
|
|
|
mbedtls_mpi_uint Mprime,
|
|
|
|
size_t hw_words,
|
|
|
|
bool again)
|
2020-03-03 07:01:19 +00:00
|
|
|
{
|
|
|
|
// Note Z may be the same pointer as X or Y
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
// montgomery mult prepare
|
|
|
|
if (again == false) {
|
|
|
|
mpi_to_mem_block(RSA_MEM_M_BLOCK_BASE, M, hw_words);
|
|
|
|
DPORT_REG_WRITE(RSA_M_DASH_REG, Mprime);
|
|
|
|
DPORT_REG_WRITE(RSA_MULT_MODE_REG, hw_words / 16 - 1);
|
|
|
|
}
|
|
|
|
|
|
|
|
mpi_to_mem_block(RSA_MEM_X_BLOCK_BASE, X, hw_words);
|
|
|
|
mpi_to_mem_block(RSA_MEM_RB_BLOCK_BASE, Y, hw_words);
|
|
|
|
|
|
|
|
start_op(RSA_MULT_START_REG);
|
2020-07-06 18:01:44 +00:00
|
|
|
Z->s = 1; // The sign of Z will be = M->s (but M->s is always 1)
|
2020-03-03 07:01:19 +00:00
|
|
|
MBEDTLS_MPI_CHK( mbedtls_mpi_grow(Z, hw_words) );
|
|
|
|
|
|
|
|
wait_op_complete();
|
|
|
|
|
|
|
|
/* Read back the result */
|
|
|
|
mem_block_to_mpi(Z, RSA_MEM_Z_BLOCK_BASE, hw_words);
|
|
|
|
|
|
|
|
|
|
|
|
/* from HAC 14.36 - 3. If Z >= M then Z = Z - M */
|
|
|
|
if (mbedtls_mpi_cmp_mpi(Z, M) >= 0) {
|
|
|
|
MBEDTLS_MPI_CHK(mbedtls_mpi_sub_mpi(Z, Z, M));
|
|
|
|
}
|
2020-03-09 05:58:05 +00:00
|
|
|
cleanup:
|
2020-03-03 07:01:19 +00:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/* Special-case of mbedtls_mpi_mult_mpi(), where we use hardware montgomery mod
|
|
|
|
multiplication to calculate an mbedtls_mpi_mult_mpi result where either
|
|
|
|
A or B are >2048 bits so can't use the standard multiplication method.
|
|
|
|
|
|
|
|
Result (z_words, based on A bits + B bits) must still be less than 4096 bits.
|
|
|
|
|
|
|
|
This case is simpler than the general case modulo multiply of
|
|
|
|
esp_mpi_mul_mpi_mod() because we can control the other arguments:
|
|
|
|
|
|
|
|
* Modulus is chosen with M=(2^num_bits - 1) (ie M=R-1), so output
|
|
|
|
isn't actually modulo anything.
|
|
|
|
* Mprime and Rinv are therefore predictable as follows:
|
|
|
|
Mprime = 1
|
|
|
|
Rinv = 1
|
|
|
|
|
|
|
|
(See RSA Accelerator section in Technical Reference for more about Mprime, Rinv)
|
|
|
|
*/
|
2020-03-09 05:58:05 +00:00
|
|
|
void esp_mpi_mult_mpi_failover_mod_mult_hw_op(const mbedtls_mpi *X, const mbedtls_mpi *Y, size_t num_words)
|
2020-03-03 07:01:19 +00:00
|
|
|
{
|
|
|
|
size_t hw_words = num_words;
|
|
|
|
|
|
|
|
/* M = 2^num_words - 1, so block is entirely FF */
|
2020-11-17 04:48:35 +00:00
|
|
|
for (size_t i = 0; i < hw_words; i++) {
|
2020-03-03 07:01:19 +00:00
|
|
|
DPORT_REG_WRITE(RSA_MEM_M_BLOCK_BASE + i * 4, UINT32_MAX);
|
|
|
|
}
|
|
|
|
/* Mprime = 1 */
|
|
|
|
DPORT_REG_WRITE(RSA_M_DASH_REG, 1);
|
|
|
|
|
|
|
|
/* "mode" register loaded with number of 512-bit blocks, minus 1 */
|
|
|
|
DPORT_REG_WRITE(RSA_MULT_MODE_REG, (hw_words / 16) - 1);
|
|
|
|
|
|
|
|
/* Load X */
|
|
|
|
mpi_to_mem_block(RSA_MEM_X_BLOCK_BASE, X, hw_words);
|
|
|
|
|
2020-03-09 05:58:05 +00:00
|
|
|
/* Rinv = 1, write first word */
|
2020-03-03 07:01:19 +00:00
|
|
|
DPORT_REG_WRITE(RSA_MEM_RB_BLOCK_BASE, 1);
|
2020-03-09 05:58:05 +00:00
|
|
|
|
|
|
|
/* Zero out rest of the Rinv words */
|
2020-11-17 04:48:35 +00:00
|
|
|
for (size_t i = 1; i < hw_words; i++) {
|
2020-03-03 07:01:19 +00:00
|
|
|
DPORT_REG_WRITE(RSA_MEM_RB_BLOCK_BASE + i * 4, 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
start_op(RSA_MULT_START_REG);
|
|
|
|
|
2020-03-09 05:58:05 +00:00
|
|
|
wait_op_complete();
|
2020-03-03 07:01:19 +00:00
|
|
|
|
|
|
|
/* finish the modular multiplication */
|
|
|
|
/* Load Y to X input memory block, rerun */
|
|
|
|
mpi_to_mem_block(RSA_MEM_X_BLOCK_BASE, Y, hw_words);
|
|
|
|
|
|
|
|
start_op(RSA_MULT_START_REG);
|
|
|
|
|
2020-11-10 07:40:01 +00:00
|
|
|
}
|