mhu/reflow-oven-control-sw
2
0
Fork 0
Code Issues 4 Pull Requests Releases 4 Wiki Activity

Add sha 256 library and adapt license sections in documentation

Browse Source
This commit is contained in:
Mario Hüttel
2021-08-24 22:19:12 +02:00
parent d468e52dea
commit d63761d016
8 changed files with 906 additions and 362 deletions
Download Patch File Download Diff File Expand all files Collapse all files

46
stm-firmware/updater/ram-code/3rd-party/sha256/LICENSE.md vendored Normal file
View File

@@ -0,0 +1,46 @@
# Licensing Information
Except as otherwise noted (below and/or in individual files), this project is
licensed under the [Unlicense](#the-unlicense)
(https://opensource.org/licenses/unlicense) or the [Zero Clause BSD
license](#zero-clause-bsd-license) (https://opensource.org/licenses/0bsd), at
your option.
## The Unlicense
This is free and unencumbered software released into the public domain.
Anyone is free to copy, modify, publish, use, compile, sell, or distribute this
software, either in source code form or as a compiled binary, for any purpose,
commercial or non-commercial, and by any means.
In jurisdictions that recognize copyright laws, the author or authors of this
software dedicate any and all copyright interest in the software to the public
domain. We make this dedication for the benefit of the public at large and to
the detriment of our heirs and successors. We intend this dedication to be an
overt act of relinquishment in perpetuity of all present and future rights to
this software under copyright law.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
For more information, please refer to <http://unlicense.org>
## Zero Clause BSD License
&copy; 2021 Alain Mosnier
Permission to use, copy, modify, and/or distribute this software for any
purpose with or without fee is hereby granted.
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH
REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT,
INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR
OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
PERFORMANCE OF THIS SOFTWARE.

134
stm-firmware/updater/ram-code/3rd-party/sha256/README.md vendored Normal file
View File

@@ -0,0 +1,134 @@
# sha-2 [![Build Status](https://travis-ci.com/amosnier/sha-2.svg?branch=master)](https://travis-ci.com/amosnier/sha-2)
## Disclaimer
The SHA256 implementation used in thsi project is taken from https://github.com/amosnier/sha-2. Check the appropriate license file.
## Contents
SHA-2 algorithm implementations.
At the moment, only SHA-256 is implemented.
## NEW: streaming API
In response to [an enhancement
request](https://github.com/amosnier/sha-2/issues/9), a new streaming
API has been created.
The following code is a silly but complete example:
```C
struct Sha_256 sha_256;
uint8_t hash[32];
sha_256_init(&sha_256, hash);
sha_256_write(&sha_256, "ab", strlen("ab"));
sha_256_write(&sha_256, "c", strlen("c"));
sha_256_close(&sha_256);
```
That is the equivalent of the (of course still supported) legacy way:
```C
uint8_t hash[32];
calc_sha_256(hash, "abc", strlen("abc"));
```
See [header file](https://github.com/amosnier/sha-2/blob/master/sha-256.h)
for more information.
Since the streaming API is a generalization of the non-streaming one,
the latter has been ported to the former, without measurable performance
impact. However, [a legacy
branch](https://github.com/amosnier/sha-2/tree/legacy) has been created
in order to make the legacy implementation easily available. If you can
measure a significant difference between the two, please post an issue.
## Design criteria
- Easy to test, include in any project, compile and link.
- ANSI C with as little specific C99 as possible (e.g. extended
integer types are used, but not bool).
- Portable. Makes no assumptions on the target system's endianess or
word size.
- The SHA-256 implementation is a straightforward implementation of
the algorithm specified on
[Wikipedia](https://en.wikipedia.org/wiki/SHA-2).
## Notes
The Makefile is as minimal as possible. No effort was put into making
it general. Its purpose is mainly to ease testing for the developer's
host machine. The actual implementation is however extremely easy to
include in any project, may it use GNU make or any other build tool.
## Code review
This code has been reviewed at [Stack Exchange CODE
REVIEW](https://codereview.stackexchange.com/questions/182812/self-contained-sha-256-implementation-in-c),
and the implementation has been improved accordingly.
## Testing
Testing is continuously performed on Travis CI (see above).
Apart from that, the implementation has been successfully tested on an x86-64 machine
under Linux as well as on a 16-bit DSP. On the x86-64 machine, all the
available NIST test vectors where successfully tested ([SHA-256
examples](https://csrc.nist.gov/CSRC/media/Projects/Cryptographic-Standards-and-Guidelines/documents/examples/SHA256.pdf)
and [SHA-2 Additional
examples](https://csrc.nist.gov/CSRC/media/Projects/Cryptographic-Standards-and-Guidelines/documents/examples/SHA2_Additional.pdf),
plus a few others).
In particular:
```
Input Message: "abc"
Message Digest is BA7816BF 8F01CFEA 414140DE 5DAE2223 B00361A3 96177A9C B410FF61 F20015AD
```
```
Input Message: "abcdbcdecdefdefgefghfghighijhijkijkljklmklmnlmnomnopnopq"
Message Digest is 248D6A61 D20638B8 E5C02693 0C3E6039 A33CE459 64FF2167 F6ECEDD4 19DB06C1
```
```
SHA-256 Test Data
#1) 1 byte 0xbd
68325720 aabd7c82 f30f554b 313d0570 c95accbb 7dc4b5aa e11204c0 8ffe732b
#2) 4 bytes 0xc98c8e55
7abc22c0 ae5af26c e93dbb94 433a0e0b 2e119d01 4f8e7f65 bd56c61c cccd9504
#3) 55 bytes of zeros
02779466 cdec1638 11d07881 5c633f21 90141308 1449002f 24aa3e80 f0b88ef7
#4) 56 bytes of zeros
d4817aa5 497628e7 c77e6b60 6107042b bba31308 88c5f47a 375e6179 be789fbb
#5) 57 bytes of zeros
65a16cb7 861335d5 ace3c607 18b5052e 44660726 da4cd13b b745381b 235a1785
#6) 64 bytes of zeros
f5a5fd42 d16a2030 2798ef6e d309979b 43003d23 20d9f0e8 ea9831a9 2759fb4b
#7) 1000 bytes of zeros
541b3e9d aa09b20b f85fa273 e5cbd3e8 0185aa4e c298e765 db87742b 70138a53
#8) 1000 bytes of 0x41 ‘A’
c2e68682 3489ced2 017f6059 b8b23931 8b6364f6 dcd835d0 a519105a 1eadd6e4
#9) 1005 bytes of 0x55 ‘U’
f4d62dde c0f3dd90 ea1380fa 16a5ff8d c4c54b21 740650f2 4afc4120 903552b0
#10) 1000000 bytes of zeros
d29751f2 649b32ff 572b5e0a 9f541ea6 60a50f94 ff0beedf b0b692b9 24cc8025
#11) 0x20000000 (536870912) bytes of 0x5a ‘Z’
15a1868c 12cc5395 1e182344 277447cd 0979536b adcc512a d24c67e9 b2d4f3dd
#12) 0x41000000 (1090519040) bytes of zeros
461c19a9 3bd4344f 9215f5ec 64357090 342bc66b 15a14831 7d276e31 cbc20b53
#13) 0x6000003e (1610612798) bytes of 0x42 ‘B’
c23ce8a7 895f4b21 ec0daf37 920ac0a2 62a22004 5a03eb2d fed48ef9 b05aabea
```
## License
This repository is made available under a permissive license. See
[LICENSE FILE](LICENSE.md).
## Reference implementation
I had missed that when I made this implementation but [RFC 6234, chapter 8](https://tools.ietf.org/html/rfc6234#section-8) actually includes a reference implementation in C that is (at least in ambition) broader in scope than this one. I have however neither compiled nor tested it.

226
stm-firmware/updater/ram-code/3rd-party/sha256/sha-256.c vendored Normal file
View File

@@ -0,0 +1,226 @@
#include "sha-256.h"
#define TOTAL_LEN_LEN 8
/*
* ABOUT bool: this file does not use bool in order to be as pre-C99 compatible as possible.
*/
/*
* Comments from pseudo-code at https://en.wikipedia.org/wiki/SHA-2 are reproduced here.
* When useful for clarification, portions of the pseudo-code are reproduced here too.
*/
/*
* @brief Rotate a 32-bit value by a number of bits to the right.
* @param value The value to be rotated.
* @param count The number of bits to rotate by.
* @return The rotated value.
*/
static inline uint32_t right_rot(uint32_t value, unsigned int count)
{
/*
* Defined behaviour in standard C for all count where 0 < count < 32, which is what we need here.
*/
return value >> count | value << (32 - count);
}
/*
* @brief Update a hash value under calculation with a new chunk of data.
* @param h Pointer to the first hash item, of a total of eight.
* @param p Pointer to the chunk data, which has a standard length.
*
* @note This is the SHA-256 work horse.
*/
static inline void consume_chunk(uint32_t *h, const uint8_t *p)
{
unsigned i, j;
uint32_t ah[8];
/* Initialize working variables to current hash value: */
for (i = 0; i < 8; i++)
ah[i] = h[i];
/*
* The w-array is really w[64], but since we only need 16 of them at a time, we save stack by
* calculating 16 at a time.
*
* This optimization was not there initially and the rest of the comments about w[64] are kept in their
* initial state.
*/
/*
* create a 64-entry message schedule array w[0..63] of 32-bit words (The initial values in w[0..63]
* don't matter, so many implementations zero them here) copy chunk into first 16 words w[0..15] of the
* message schedule array
*/
uint32_t w[16];
/* Compression function main loop: */
for (i = 0; i < 4; i++) {
for (j = 0; j < 16; j++) {
if (i == 0) {
w[j] =
(uint32_t)p[0] << 24 | (uint32_t)p[1] << 16 | (uint32_t)p[2] << 8 | (uint32_t)p[3];
p += 4;
} else {
/* Extend the first 16 words into the remaining 48 words w[16..63] of the
* message schedule array: */
const uint32_t s0 = right_rot(w[(j + 1) & 0xf], 7) ^ right_rot(w[(j + 1) & 0xf], 18) ^
(w[(j + 1) & 0xf] >> 3);
const uint32_t s1 = right_rot(w[(j + 14) & 0xf], 17) ^
right_rot(w[(j + 14) & 0xf], 19) ^ (w[(j + 14) & 0xf] >> 10);
w[j] = w[j] + s0 + w[(j + 9) & 0xf] + s1;
}
const uint32_t s1 = right_rot(ah[4], 6) ^ right_rot(ah[4], 11) ^ right_rot(ah[4], 25);
const uint32_t ch = (ah[4] & ah[5]) ^ (~ah[4] & ah[6]);
/*
* Initialize array of round constants:
* (first 32 bits of the fractional parts of the cube roots of the first 64 primes 2..311):
*/
static const uint32_t k[] = {
0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5, 0x3956c25b, 0x59f111f1, 0x923f82a4,
0xab1c5ed5, 0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3, 0x72be5d74, 0x80deb1fe,
0x9bdc06a7, 0xc19bf174, 0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc, 0x2de92c6f,
0x4a7484aa, 0x5cb0a9dc, 0x76f988da, 0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7,
0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967, 0x27b70a85, 0x2e1b2138, 0x4d2c6dfc,
0x53380d13, 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85, 0xa2bfe8a1, 0xa81a664b,
0xc24b8b70, 0xc76c51a3, 0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070, 0x19a4c116,
0x1e376c08, 0x2748774c, 0x34b0bcb5, 0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3,
0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208, 0x90befffa, 0xa4506ceb, 0xbef9a3f7,
0xc67178f2};
const uint32_t temp1 = ah[7] + s1 + ch + k[i << 4 | j] + w[j];
const uint32_t s0 = right_rot(ah[0], 2) ^ right_rot(ah[0], 13) ^ right_rot(ah[0], 22);
const uint32_t maj = (ah[0] & ah[1]) ^ (ah[0] & ah[2]) ^ (ah[1] & ah[2]);
const uint32_t temp2 = s0 + maj;
ah[7] = ah[6];
ah[6] = ah[5];
ah[5] = ah[4];
ah[4] = ah[3] + temp1;
ah[3] = ah[2];
ah[2] = ah[1];
ah[1] = ah[0];
ah[0] = temp1 + temp2;
}
}
/* Add the compressed chunk to the current hash value: */
for (i = 0; i < 8; i++)
h[i] += ah[i];
}
/*
* Public functions. See header file for documentation.
*/
void sha_256_init(struct Sha_256 *sha_256, uint8_t hash[SIZE_OF_SHA_256_HASH])
{
sha_256->hash = hash;
sha_256->chunk_pos = sha_256->chunk;
sha_256->space_left = SIZE_OF_SHA_256_CHUNK;
sha_256->total_len = 0;
/*
* Initialize hash values (first 32 bits of the fractional parts of the square roots of the first 8 primes
* 2..19):
*/
sha_256->h[0] = 0x6a09e667;
sha_256->h[1] = 0xbb67ae85;
sha_256->h[2] = 0x3c6ef372;
sha_256->h[3] = 0xa54ff53a;
sha_256->h[4] = 0x510e527f;
sha_256->h[5] = 0x9b05688c;
sha_256->h[6] = 0x1f83d9ab;
sha_256->h[7] = 0x5be0cd19;
}
void sha_256_write(struct Sha_256 *sha_256, const void *data, size_t len)
{
sha_256->total_len += len;
const uint8_t *p = data;
while (len > 0) {
/*
* If the input chunks have sizes that are multiples of the calculation chunk size, no copies are
* necessary. We operate directly on the input data instead.
*/
if (sha_256->space_left == SIZE_OF_SHA_256_CHUNK && len >= SIZE_OF_SHA_256_CHUNK) {
consume_chunk(sha_256->h, p);
len -= SIZE_OF_SHA_256_CHUNK;
p += SIZE_OF_SHA_256_CHUNK;
continue;
}
/* General case, no particular optimization. */
const size_t consumed_len = len < sha_256->space_left ? len : sha_256->space_left;
memcpy(sha_256->chunk_pos, p, consumed_len);
sha_256->space_left -= consumed_len;
len -= consumed_len;
p += consumed_len;
if (sha_256->space_left == 0) {
consume_chunk(sha_256->h, sha_256->chunk);
sha_256->chunk_pos = sha_256->chunk;
sha_256->space_left = SIZE_OF_SHA_256_CHUNK;
} else {
sha_256->chunk_pos += consumed_len;
}
}
}
uint8_t *sha_256_close(struct Sha_256 *sha_256)
{
uint8_t *pos = sha_256->chunk_pos;
size_t space_left = sha_256->space_left;
uint32_t *const h = sha_256->h;
/*
* The current chunk cannot be full. Otherwise, it would already have be consumed. I.e. there is space left for
* at least one byte. The next step in the calculation is to add a single one-bit to the data.
*/
*pos++ = 0x80;
--space_left;
/*
* Now, the last step is to add the total data length at the end of the last chunk, and zero padding before
* that. But we do not necessarily have enough space left. If not, we pad the current chunk with zeroes, and add
* an extra chunk at the end.
*/
if (space_left < TOTAL_LEN_LEN) {
memset(pos, 0x00, space_left);
consume_chunk(h, sha_256->chunk);
pos = sha_256->chunk;
space_left = SIZE_OF_SHA_256_CHUNK;
}
const size_t left = space_left - TOTAL_LEN_LEN;
memset(pos, 0x00, left);
pos += left;
size_t len = sha_256->total_len;
pos[7] = (uint8_t)(len << 3);
len >>= 5;
int i;
for (i = 6; i >= 0; --i) {
pos[i] = (uint8_t)len;
len >>= 8;
}
consume_chunk(h, sha_256->chunk);
/* Produce the final hash value (big-endian): */
int j;
uint8_t *const hash = sha_256->hash;
for (i = 0, j = 0; i < 8; i++) {
hash[j++] = (uint8_t)(h[i] >> 24);
hash[j++] = (uint8_t)(h[i] >> 16);
hash[j++] = (uint8_t)(h[i] >> 8);
hash[j++] = (uint8_t)h[i];
}
return sha_256->hash;
}
void calc_sha_256(uint8_t hash[SIZE_OF_SHA_256_HASH], const void *input, size_t len)
{
struct Sha_256 sha_256;
sha_256_init(&sha_256, hash);
sha_256_write(&sha_256, input, len);
(void)sha_256_close(&sha_256);
}

103
stm-firmware/updater/ram-code/3rd-party/sha256/sha-256.h vendored Normal file
View File

@@ -0,0 +1,103 @@
#ifndef SHA_256_H
#define SHA_256_H
#include <stdint.h>
#include <string.h>
#ifdef __cplusplus
extern "C" {
#endif
/*
* @brief Size of the SHA-256 sum. This times eight is 256 bits.
*/
#define SIZE_OF_SHA_256_HASH 32
/*
* @brief Size of the chunks used for the calculations.
*
* @note This should mostly be ignored by the user, although when using the streaming API, it has an impact for
* performance. Add chunks whose size is a multiple of this, and you will avoid a lot of superfluous copying in RAM!
*/
#define SIZE_OF_SHA_256_CHUNK 64
/*
* @brief The opaque SHA-256 type, that should be instantiated when using the streaming API.
*
* @note Although the details are exposed here, in order to make instantiation easy, you should refrain from directly
* accessing the fields, as they may change in the future.
*/
struct Sha_256 {
uint8_t *hash;
uint8_t chunk[SIZE_OF_SHA_256_CHUNK];
uint8_t *chunk_pos;
size_t space_left;
size_t total_len;
uint32_t h[8];
};
/*
* @brief The simple SHA-256 calculation function.
* @param hash Hash array, where the result is delivered.
* @param input Pointer to the data the hash shall be calculated on.
* @param len Length of the input data, in byte.
*
* @note If all of the data you are calculating the hash value on is available in a contiguous buffer in memory, this is
* the function you should use.
*
* @note If either of the passed pointers is NULL, the results are unpredictable.
*/
void calc_sha_256(uint8_t hash[SIZE_OF_SHA_256_HASH], const void *input, size_t len);
/*
* @brief Initialize a SHA-256 streaming calculation.
* @param sha_256 A pointer to a SHA-256 structure.
* @param hash Hash array, where the result will be delivered.
*
* @note If all of the data you are calculating the hash value on is not available in a contiguous buffer in memory, this is
* where you should start. Instantiate a SHA-256 structure, for instance by simply declaring it locally, make your hash
* buffer available, and invoke this function. Once a SHA-256 hash has been calculated (see further below) a SHA-256
* structure can be initialized again for the next calculation.
*
* @note If either of the passed pointers is NULL, the results are unpredictable.
*/
void sha_256_init(struct Sha_256 *sha_256, uint8_t hash[SIZE_OF_SHA_256_HASH]);
/*
* @brief Stream more input data for an on-going SHA-256 calculation.
* @param sha_256 A pointer to a previously initialized SHA-256 structure.
* @param data Pointer to the data to be added to the calculation.
* @param len Length of the data to add, in byte.
*
* @note This function may be invoked an arbitrary number of times between initialization and closing, but the maximum
* data length is limited by the SHA-256 algorithm: the total number of bits (i.e. the total number of bytes times
* eight) must be representable by a 64-bit unsigned integer. While that is not a practical limitation, the results are
* unpredictable if that limit is exceeded.
*
* @note This function may be invoked on empty data (zero length), although that obviously will not add any data.
*
* @note If either of the passed pointers is NULL, the results are unpredictable.
*/
void sha_256_write(struct Sha_256 *sha_256, const void *data, size_t len);
/*
* @brief Conclude a SHA-256 streaming calculation, making the hash value available.
* @param sha_256 A pointer to a previously initialized SHA-256 structure.
* @return Pointer to the hash array, where the result is delivered.
*
* @note After this function has been invoked, the result is available in the hash buffer that initially was provided. A
* pointer to the hash value is returned for convenience, but you should feel free to ignore it: it is simply a pointer
* to the first byte of your initially provided hash array.
*
* @note If the passed pointer is NULL, the results are unpredictable.
*
* @note Invoking this function for a calculation with no data (the writing function has never been invoked, or it only
* has been invoked with empty data) is legal. It will calculate the SHA-256 value of the empty string.
*/
uint8_t *sha_256_close(struct Sha_256 *sha_256);
#ifdef __cplusplus
}
#endif
#endif