515 lines
16 KiB
C
515 lines
16 KiB
C
// Copyright 2015-2016 Espressif Systems (Shanghai) PTE LTD
|
|
//
|
|
// 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.
|
|
|
|
#include <stdlib.h>
|
|
#include <assert.h>
|
|
#include <string.h>
|
|
#include <stdio.h>
|
|
#include <sys/param.h> // For MIN/MAX(a, b)
|
|
|
|
#include <freertos/FreeRTOS.h>
|
|
#include <freertos/task.h>
|
|
#include <freertos/semphr.h>
|
|
#include <rom/spi_flash.h>
|
|
#include <rom/cache.h>
|
|
#include <soc/soc.h>
|
|
#include <soc/dport_reg.h>
|
|
#include "sdkconfig.h"
|
|
#include "esp_ipc.h"
|
|
#include "esp_attr.h"
|
|
#include "esp_spi_flash.h"
|
|
#include "esp_log.h"
|
|
#include "cache_utils.h"
|
|
|
|
/* bytes erased by SPIEraseBlock() ROM function */
|
|
#define BLOCK_ERASE_SIZE 65536
|
|
|
|
/* Limit number of bytes written/read in a single SPI operation,
|
|
as these operations disable all higher priority tasks from running.
|
|
*/
|
|
#define MAX_WRITE_CHUNK 8192
|
|
#define MAX_READ_CHUNK 16384
|
|
|
|
#if CONFIG_SPI_FLASH_ENABLE_COUNTERS
|
|
static const char* TAG = "spi_flash";
|
|
static spi_flash_counters_t s_flash_stats;
|
|
|
|
#define COUNTER_START() uint32_t ts_begin = xthal_get_ccount()
|
|
#define COUNTER_STOP(counter) \
|
|
do{ \
|
|
s_flash_stats.counter.count++; \
|
|
s_flash_stats.counter.time += (xthal_get_ccount() - ts_begin) / (XT_CLOCK_FREQ / 1000000); \
|
|
} while(0)
|
|
|
|
#define COUNTER_ADD_BYTES(counter, size) \
|
|
do { \
|
|
s_flash_stats.counter.bytes += size; \
|
|
} while (0)
|
|
|
|
#else
|
|
#define COUNTER_START()
|
|
#define COUNTER_STOP(counter)
|
|
#define COUNTER_ADD_BYTES(counter, size)
|
|
|
|
#endif //CONFIG_SPI_FLASH_ENABLE_COUNTERS
|
|
|
|
static esp_err_t spi_flash_translate_rc(SpiFlashOpResult rc);
|
|
|
|
const DRAM_ATTR spi_flash_guard_funcs_t g_flash_guard_default_ops = {
|
|
.start = spi_flash_disable_interrupts_caches_and_other_cpu,
|
|
.end = spi_flash_enable_interrupts_caches_and_other_cpu,
|
|
.op_lock = spi_flash_op_lock,
|
|
.op_unlock = spi_flash_op_unlock
|
|
};
|
|
|
|
const DRAM_ATTR spi_flash_guard_funcs_t g_flash_guard_no_os_ops = {
|
|
.start = spi_flash_disable_interrupts_caches_and_other_cpu_no_os,
|
|
.end = spi_flash_enable_interrupts_caches_no_os,
|
|
.op_lock = 0,
|
|
.op_unlock = 0
|
|
};
|
|
|
|
static const spi_flash_guard_funcs_t *s_flash_guard_ops;
|
|
|
|
void spi_flash_init()
|
|
{
|
|
spi_flash_init_lock();
|
|
#if CONFIG_SPI_FLASH_ENABLE_COUNTERS
|
|
spi_flash_reset_counters();
|
|
#endif
|
|
}
|
|
|
|
void IRAM_ATTR spi_flash_guard_set(const spi_flash_guard_funcs_t* funcs)
|
|
{
|
|
s_flash_guard_ops = funcs;
|
|
}
|
|
|
|
size_t IRAM_ATTR spi_flash_get_chip_size()
|
|
{
|
|
return g_rom_flashchip.chip_size;
|
|
}
|
|
|
|
static inline void IRAM_ATTR spi_flash_guard_start()
|
|
{
|
|
if (s_flash_guard_ops && s_flash_guard_ops->start) {
|
|
s_flash_guard_ops->start();
|
|
}
|
|
}
|
|
|
|
static inline void IRAM_ATTR spi_flash_guard_end()
|
|
{
|
|
if (s_flash_guard_ops && s_flash_guard_ops->end) {
|
|
s_flash_guard_ops->end();
|
|
}
|
|
}
|
|
|
|
static inline void IRAM_ATTR spi_flash_guard_op_lock()
|
|
{
|
|
if (s_flash_guard_ops && s_flash_guard_ops->op_lock) {
|
|
s_flash_guard_ops->op_lock();
|
|
}
|
|
}
|
|
|
|
static inline void IRAM_ATTR spi_flash_guard_op_unlock()
|
|
{
|
|
if (s_flash_guard_ops && s_flash_guard_ops->op_unlock) {
|
|
s_flash_guard_ops->op_unlock();
|
|
}
|
|
}
|
|
|
|
static SpiFlashOpResult IRAM_ATTR spi_flash_unlock()
|
|
{
|
|
static bool unlocked = false;
|
|
if (!unlocked) {
|
|
spi_flash_guard_start();
|
|
SpiFlashOpResult rc = SPIUnlock();
|
|
spi_flash_guard_end();
|
|
if (rc != SPI_FLASH_RESULT_OK) {
|
|
return rc;
|
|
}
|
|
unlocked = true;
|
|
}
|
|
return SPI_FLASH_RESULT_OK;
|
|
}
|
|
|
|
esp_err_t IRAM_ATTR spi_flash_erase_sector(size_t sec)
|
|
{
|
|
return spi_flash_erase_range(sec * SPI_FLASH_SEC_SIZE, SPI_FLASH_SEC_SIZE);
|
|
}
|
|
|
|
esp_err_t IRAM_ATTR spi_flash_erase_range(uint32_t start_addr, uint32_t size)
|
|
{
|
|
if (start_addr % SPI_FLASH_SEC_SIZE != 0) {
|
|
return ESP_ERR_INVALID_ARG;
|
|
}
|
|
if (size % SPI_FLASH_SEC_SIZE != 0) {
|
|
return ESP_ERR_INVALID_SIZE;
|
|
}
|
|
if (size + start_addr > spi_flash_get_chip_size()) {
|
|
return ESP_ERR_INVALID_SIZE;
|
|
}
|
|
size_t start = start_addr / SPI_FLASH_SEC_SIZE;
|
|
size_t end = start + size / SPI_FLASH_SEC_SIZE;
|
|
const size_t sectors_per_block = BLOCK_ERASE_SIZE / SPI_FLASH_SEC_SIZE;
|
|
COUNTER_START();
|
|
SpiFlashOpResult rc = spi_flash_unlock();
|
|
if (rc == SPI_FLASH_RESULT_OK) {
|
|
for (size_t sector = start; sector != end && rc == SPI_FLASH_RESULT_OK; ) {
|
|
spi_flash_guard_start();
|
|
if (sector % sectors_per_block == 0 && end - sector > sectors_per_block) {
|
|
rc = SPIEraseBlock(sector / sectors_per_block);
|
|
sector += sectors_per_block;
|
|
COUNTER_ADD_BYTES(erase, sectors_per_block * SPI_FLASH_SEC_SIZE);
|
|
} else {
|
|
rc = SPIEraseSector(sector);
|
|
++sector;
|
|
COUNTER_ADD_BYTES(erase, SPI_FLASH_SEC_SIZE);
|
|
}
|
|
spi_flash_guard_end();
|
|
}
|
|
}
|
|
COUNTER_STOP(erase);
|
|
return spi_flash_translate_rc(rc);
|
|
}
|
|
|
|
esp_err_t IRAM_ATTR spi_flash_write(size_t dst, const void *srcv, size_t size)
|
|
{
|
|
// Out of bound writes are checked in ROM code, but we can give better
|
|
// error code here
|
|
if (dst + size > g_rom_flashchip.chip_size) {
|
|
return ESP_ERR_INVALID_SIZE;
|
|
}
|
|
if (size == 0) {
|
|
return ESP_OK;
|
|
}
|
|
|
|
SpiFlashOpResult rc = SPI_FLASH_RESULT_OK;
|
|
COUNTER_START();
|
|
const char *srcc = (const char *) srcv;
|
|
/*
|
|
* Large operations are split into (up to) 3 parts:
|
|
* - Left padding: 4 bytes up to the first 4-byte aligned destination offset.
|
|
* - Middle part
|
|
* - Right padding: 4 bytes from the last 4-byte aligned offset covered.
|
|
*/
|
|
size_t left_off = dst & ~3U;
|
|
size_t left_size = MIN(((dst + 3) & ~3U) - dst, size);
|
|
size_t mid_off = left_size;
|
|
size_t mid_size = (size - left_size) & ~3U;
|
|
size_t right_off = left_size + mid_size;
|
|
size_t right_size = size - mid_size - left_size;
|
|
rc = spi_flash_unlock();
|
|
if (rc != SPI_FLASH_RESULT_OK) {
|
|
goto out;
|
|
}
|
|
if (left_size > 0) {
|
|
uint32_t t = 0xffffffff;
|
|
memcpy(((uint8_t *) &t) + (dst - left_off), srcc, left_size);
|
|
spi_flash_guard_start();
|
|
rc = SPIWrite(left_off, &t, 4);
|
|
spi_flash_guard_end();
|
|
if (rc != SPI_FLASH_RESULT_OK) {
|
|
goto out;
|
|
}
|
|
COUNTER_ADD_BYTES(write, 4);
|
|
}
|
|
if (mid_size > 0) {
|
|
/* If src buffer is 4-byte aligned as well and is not in a region that requires cache access to be enabled, we
|
|
* can write directly without buffering in RAM. */
|
|
#ifdef ESP_PLATFORM
|
|
bool direct_write = ( (uintptr_t) srcc >= 0x3FFAE000
|
|
&& (uintptr_t) srcc < 0x40000000
|
|
&& ((uintptr_t) srcc + mid_off) % 4 == 0 );
|
|
#else
|
|
bool direct_write = true;
|
|
#endif
|
|
while(mid_size > 0 && rc == SPI_FLASH_RESULT_OK) {
|
|
uint32_t write_buf[8];
|
|
uint32_t write_size;
|
|
const uint32_t *write_src = (const uint32_t *) (srcc + mid_off);
|
|
if (direct_write) {
|
|
write_size = MIN(mid_size, MAX_WRITE_CHUNK); /* Write in chunks, to avoid starving other CPU/tasks */
|
|
} else {
|
|
write_size = MIN(mid_size, sizeof(write_buf));
|
|
memcpy(write_buf, write_src, write_size);
|
|
write_src = write_buf;
|
|
}
|
|
spi_flash_guard_start();
|
|
rc = SPIWrite(dst + mid_off, write_src, write_size);
|
|
spi_flash_guard_end();
|
|
COUNTER_ADD_BYTES(write, write_size);
|
|
mid_size -= write_size;
|
|
mid_off += write_size;
|
|
}
|
|
if (rc != SPI_FLASH_RESULT_OK) {
|
|
goto out;
|
|
}
|
|
}
|
|
|
|
if (right_size > 0) {
|
|
uint32_t t = 0xffffffff;
|
|
memcpy(&t, srcc + right_off, right_size);
|
|
spi_flash_guard_start();
|
|
rc = SPIWrite(dst + right_off, &t, 4);
|
|
spi_flash_guard_end();
|
|
if (rc != SPI_FLASH_RESULT_OK) {
|
|
goto out;
|
|
}
|
|
COUNTER_ADD_BYTES(write, 4);
|
|
}
|
|
out:
|
|
COUNTER_STOP(write);
|
|
|
|
spi_flash_guard_op_lock();
|
|
spi_flash_mark_modified_region(dst, size);
|
|
spi_flash_guard_op_unlock();
|
|
|
|
return spi_flash_translate_rc(rc);
|
|
}
|
|
|
|
esp_err_t IRAM_ATTR spi_flash_write_encrypted(size_t dest_addr, const void *src, size_t size)
|
|
{
|
|
const uint8_t *ssrc = (const uint8_t *)src;
|
|
if ((dest_addr % 16) != 0) {
|
|
return ESP_ERR_INVALID_ARG;
|
|
}
|
|
if ((size % 16) != 0) {
|
|
return ESP_ERR_INVALID_SIZE;
|
|
}
|
|
|
|
COUNTER_START();
|
|
SpiFlashOpResult rc = spi_flash_unlock();
|
|
|
|
if (rc == SPI_FLASH_RESULT_OK) {
|
|
/* SPI_Encrypt_Write encrypts data in RAM as it writes,
|
|
so copy to a temporary buffer - 32 bytes at a time.
|
|
|
|
Each call to SPI_Encrypt_Write takes a 32 byte "row" of
|
|
data to encrypt, and each row is two 16 byte AES blocks
|
|
that share a key (as derived from flash address).
|
|
*/
|
|
uint8_t encrypt_buf[32] __attribute__((aligned(4)));
|
|
uint32_t row_size;
|
|
for (size_t i = 0; i < size; i += row_size) {
|
|
uint32_t row_addr = dest_addr + i;
|
|
if (i == 0 && (row_addr % 32) != 0) {
|
|
/* writing to second block of a 32 byte row */
|
|
row_size = 16;
|
|
row_addr -= 16;
|
|
/* copy to second block in buffer */
|
|
memcpy(encrypt_buf + 16, ssrc + i, 16);
|
|
/* decrypt the first block from flash, will reencrypt to same bytes */
|
|
spi_flash_read_encrypted(row_addr, encrypt_buf, 16);
|
|
}
|
|
else if (size - i == 16) {
|
|
/* 16 bytes left, is first block of a 32 byte row */
|
|
row_size = 16;
|
|
/* copy to first block in buffer */
|
|
memcpy(encrypt_buf, ssrc + i, 16);
|
|
/* decrypt the second block from flash, will reencrypt to same bytes */
|
|
spi_flash_read_encrypted(row_addr + 16, encrypt_buf + 16, 16);
|
|
}
|
|
else {
|
|
/* Writing a full 32 byte row (2 blocks) */
|
|
row_size = 32;
|
|
memcpy(encrypt_buf, ssrc + i, 32);
|
|
}
|
|
|
|
spi_flash_guard_start();
|
|
rc = SPI_Encrypt_Write(row_addr, (uint32_t *)encrypt_buf, 32);
|
|
spi_flash_guard_end();
|
|
if (rc != SPI_FLASH_RESULT_OK) {
|
|
break;
|
|
}
|
|
}
|
|
bzero(encrypt_buf, sizeof(encrypt_buf));
|
|
}
|
|
COUNTER_ADD_BYTES(write, size);
|
|
COUNTER_STOP(write);
|
|
|
|
spi_flash_guard_op_lock();
|
|
spi_flash_mark_modified_region(dest_addr, size);
|
|
spi_flash_guard_op_unlock();
|
|
|
|
return spi_flash_translate_rc(rc);
|
|
}
|
|
|
|
esp_err_t IRAM_ATTR spi_flash_read(size_t src, void *dstv, size_t size)
|
|
{
|
|
// Out of bound reads are checked in ROM code, but we can give better
|
|
// error code here
|
|
if (src + size > g_rom_flashchip.chip_size) {
|
|
return ESP_ERR_INVALID_SIZE;
|
|
}
|
|
if (size == 0) {
|
|
return ESP_OK;
|
|
}
|
|
|
|
SpiFlashOpResult rc = SPI_FLASH_RESULT_OK;
|
|
COUNTER_START();
|
|
spi_flash_guard_start();
|
|
/* To simplify boundary checks below, we handle small reads separately. */
|
|
if (size < 16) {
|
|
uint32_t t[6]; /* Enough for 16 bytes + 4 on either side for padding. */
|
|
uint32_t read_src = src & ~3U;
|
|
uint32_t left_off = src & 3U;
|
|
uint32_t read_size = (left_off + size + 3) & ~3U;
|
|
rc = SPIRead(read_src, t, read_size);
|
|
if (rc != SPI_FLASH_RESULT_OK) {
|
|
goto out;
|
|
}
|
|
COUNTER_ADD_BYTES(read, read_size);
|
|
memcpy(dstv, ((char *) t) + left_off, size);
|
|
goto out;
|
|
}
|
|
char *dstc = (char *) dstv;
|
|
intptr_t dsti = (intptr_t) dstc;
|
|
/*
|
|
* Large operations are split into (up to) 3 parts:
|
|
* - The middle part: from the first 4-aligned position in src to the first
|
|
* 4-aligned position in dst.
|
|
*/
|
|
size_t src_mid_off = (src % 4 == 0 ? 0 : 4 - (src % 4));
|
|
size_t dst_mid_off = (dsti % 4 == 0 ? 0 : 4 - (dsti % 4));
|
|
size_t mid_size = (size - MAX(src_mid_off, dst_mid_off)) & ~3U;
|
|
/*
|
|
* - Once the middle part is in place, src_mid_off bytes from the preceding
|
|
* 4-aligned source location are added on the left.
|
|
*/
|
|
size_t pad_left_src = src & ~3U;
|
|
size_t pad_left_size = src_mid_off;
|
|
/*
|
|
* - Finally, the right part is added: from the end of the middle part to
|
|
* the end. Depending on the alignment of source and destination, this may
|
|
* be a 4 or 8 byte read from pad_right_src.
|
|
*/
|
|
size_t pad_right_src = (src + pad_left_size + mid_size) & ~3U;
|
|
size_t pad_right_off = (pad_right_src - src);
|
|
size_t pad_right_size = (size - pad_right_off);
|
|
if (mid_size > 0) {
|
|
uint32_t mid_remaining = mid_size;
|
|
uint32_t mid_read = 0;
|
|
while (mid_remaining > 0) {
|
|
uint32_t read_size = MIN(mid_remaining, MAX_READ_CHUNK);
|
|
rc = SPIRead(src + src_mid_off + mid_read, (uint32_t *) (dstc + dst_mid_off + mid_read), read_size);
|
|
if (rc != SPI_FLASH_RESULT_OK) {
|
|
goto out;
|
|
}
|
|
mid_remaining -= read_size;
|
|
mid_read += read_size;
|
|
if (mid_remaining > 0) {
|
|
/* Drop guard momentarily, allows other tasks to preempt */
|
|
spi_flash_guard_end();
|
|
spi_flash_guard_start();
|
|
}
|
|
}
|
|
COUNTER_ADD_BYTES(read, mid_size);
|
|
/*
|
|
* If offsets in src and dst are different, perform an in-place shift
|
|
* to put destination data into its final position.
|
|
* Note that the shift can be left (src_mid_off < dst_mid_off) or right.
|
|
*/
|
|
if (src_mid_off != dst_mid_off) {
|
|
memmove(dstc + src_mid_off, dstc + dst_mid_off, mid_size);
|
|
}
|
|
}
|
|
if (pad_left_size > 0) {
|
|
uint32_t t;
|
|
rc = SPIRead(pad_left_src, &t, 4);
|
|
if (rc != SPI_FLASH_RESULT_OK) {
|
|
goto out;
|
|
}
|
|
COUNTER_ADD_BYTES(read, 4);
|
|
memcpy(dstc, ((uint8_t *) &t) + (4 - pad_left_size), pad_left_size);
|
|
}
|
|
if (pad_right_size > 0) {
|
|
uint32_t t[2];
|
|
int32_t read_size = (pad_right_size <= 4 ? 4 : 8);
|
|
rc = SPIRead(pad_right_src, t, read_size);
|
|
if (rc != SPI_FLASH_RESULT_OK) {
|
|
goto out;
|
|
}
|
|
COUNTER_ADD_BYTES(read, read_size);
|
|
memcpy(dstc + pad_right_off, t, pad_right_size);
|
|
}
|
|
out:
|
|
spi_flash_guard_end();
|
|
COUNTER_STOP(read);
|
|
return spi_flash_translate_rc(rc);
|
|
}
|
|
|
|
esp_err_t IRAM_ATTR spi_flash_read_encrypted(size_t src, void *dstv, size_t size)
|
|
{
|
|
if (src + size > g_rom_flashchip.chip_size) {
|
|
return ESP_ERR_INVALID_SIZE;
|
|
}
|
|
if (size == 0) {
|
|
return ESP_OK;
|
|
}
|
|
|
|
esp_err_t err;
|
|
const uint8_t *map;
|
|
spi_flash_mmap_handle_t map_handle;
|
|
size_t map_src = src & ~(SPI_FLASH_MMU_PAGE_SIZE-1);
|
|
size_t map_size = size + (src - map_src);
|
|
|
|
err = spi_flash_mmap(map_src, map_size, SPI_FLASH_MMAP_DATA, (const void **)&map, &map_handle);
|
|
if (err != ESP_OK) {
|
|
return err;
|
|
}
|
|
memcpy(dstv, map + (src - map_src), size);
|
|
spi_flash_munmap(map_handle);
|
|
return err;
|
|
}
|
|
|
|
|
|
static esp_err_t IRAM_ATTR spi_flash_translate_rc(SpiFlashOpResult rc)
|
|
{
|
|
switch (rc) {
|
|
case SPI_FLASH_RESULT_OK:
|
|
return ESP_OK;
|
|
case SPI_FLASH_RESULT_TIMEOUT:
|
|
return ESP_ERR_FLASH_OP_TIMEOUT;
|
|
case SPI_FLASH_RESULT_ERR:
|
|
default:
|
|
return ESP_ERR_FLASH_OP_FAIL;
|
|
}
|
|
}
|
|
|
|
#if CONFIG_SPI_FLASH_ENABLE_COUNTERS
|
|
|
|
static inline void dump_counter(spi_flash_counter_t* counter, const char* name)
|
|
{
|
|
ESP_LOGI(TAG, "%s count=%8d time=%8dus bytes=%8d\n", name,
|
|
counter->count, counter->time, counter->bytes);
|
|
}
|
|
|
|
const spi_flash_counters_t* spi_flash_get_counters()
|
|
{
|
|
return &s_flash_stats;
|
|
}
|
|
|
|
void spi_flash_reset_counters()
|
|
{
|
|
memset(&s_flash_stats, 0, sizeof(s_flash_stats));
|
|
}
|
|
|
|
void spi_flash_dump_counters()
|
|
{
|
|
dump_counter(&s_flash_stats.read, "read ");
|
|
dump_counter(&s_flash_stats.write, "write");
|
|
dump_counter(&s_flash_stats.erase, "erase");
|
|
}
|
|
|
|
#endif //CONFIG_SPI_FLASH_ENABLE_COUNTERS
|