Does not compile because of ed25519

some of ed25519 logic does not have #ifdef ED25519_SUPPORT so this code does not compile as is

is anywhere alive here? I can offer a PR if there is a maintainer around

[encedo]

I have no clue what do you mean by "does not compile as is ". Download external ED25529 libs and link. I have not compiled this code recently (>2years) but should be just fine.

Hope I helped.

Here are more details: rusefi@28047fa

Minor issue:

// uncomment to add support for digital code signature using ED25519
#define  ED25519_SUPPORT

comment says UNCOMMENT while ED25519_SUPPORT is enabled by default and I have no idea what is ED25519_SUPPORT and I do not need it :)

Second issue is that https://github.com/encedo/hex2dfu/blob/master/hex2dfu.c#L126 shows conditional compilation while https://github.com/encedo/hex2dfu/blob/master/hex2dfu.c#L261 does not have similar conditional compilation

and minor issue with

uint32_t crc32(uint32_t crc, const void *buf, size_t size);

being kind of close but not exactly the same as

unsigned int crc32(unsigned int crc, const void *buf, size_t size) {

my weird compiler was happy with unsigned int but not happy with uint32_t

Great code since it works 4 years later :)

[encedo]

Now it makes sense to me.

The issue with conditional compilation is true - lack the second #ifdef clause (L261)- my mistake. I have missed this issue as I.... always compile with ED25519 support (I use this code in production to digitally sign firmware against bootloader).

The issue with uint32_t can be solved by adding #include <stdint.h> (my mingw32 was ok without it, new version require it explicitly). I have never compiled it under Linux/mac as I use Keil IDE.

I'm happy my code is useful :)

Chris

[encedo]

Hi

I will review all the proposals and merge with them.

Stay tuned.
Chris

[rusefillc]

new PR #9