Hollywood forums

Ok, it's fixed now. I hope I've found all OpenSSL functions that do callbacks. Here's a test binary: http://www.softwarefailure.de/tmp/hurl.lha

jPV wrote: ↑

Mon Dec 23, 2019 8:58 am

How about linking OpenSSL statically to the hURL plugin for MorphOS? hURL has different requirements for different platforms anyway, so maybe it could be ok?

This would bloat hurl.hwp to a 5MB+ binary that is loaded with the start of every Hollywood app. Not something very desirable IMHO...

FWIW, I've also opened an issue at Github now.

Ok, figured it out, works correctly now. Thanks a lot for your help!

It seems AmiSSL doesn't specify the calling convention. In fact, it depends on the compiler specifics (gcc in this case). What a horrible design. Well, another reason for MorphOS team to come up with an in-house solution that uses a clear specification ;) Stack grows upwards (pushing values to stac...

I don't see any of this documented in AmiSSL. Only the register convention of the LVOs is obviously documented. For callback functions I see no such specification. But since the parameters aren't in the registers, they surely must be on the stack. But how exactly are they stored in the stack? I've t...

Hmm, doesn't seem to work. I'm trying to wrap this callback: static int select_next_proto_cb(SSL *ssl, unsigned char **out, unsigned char *outlen, const unsigned char *in, unsigned int inlen, void *arg); This is the wrapper: static ULONG select_next_proto_cb_GATE(void) { printf("ADDRESS: %p %p %p %p...

Thanks for the explanation! Just to get this straight, cURL does the following for example: static int passwd_callback(char *buf, int num, int encrypting, void *global_passwd) { ... } SSL_CTX_set_default_passwd_cb(ctx, passwd_callback); So should this really be migrated to: static const struct EmulL...