On Thu, Apr 14, 2016 at 09:33:37AM -0400, Holger Freyther wrote: So, I want to change this: static inline void OCTET_STRING_noalloc(OCTET_STRING_t *s, const uint8_t *str, int size) { s->buf = str; s->size = size; } because OCTET_STRING_t.s is not const, and must not be const, will never be const. OCTET_STRING_noalloc() has no way to tell whether callers will subsequently modify the buffer or not. So that function must not secretly unconst: static inline void OCTET_STRING_noalloc(OCTET_STRING_t *s, const uint8_t *str, int size) { s->buf = (uint8_t*)str; s->size = size; } ^ this is bad. To get rid of compiler warnings, I want to change to more sensible: static inline void OCTET_STRING_noalloc(OCTET_STRING_t *s, uint8_t *str, int size) { s->buf = str; s->size = size; } All callers of OCTET_STRING_noalloc() pass a non-const pointer, except that one above. I just want to unconst that single caller. OCTET_STRING_t.buf is simply not const. So if we're not going to do an ugly unconst like this somewhere, we have to change ranap_new_msg_dt() to take a non-const uint8_t* nas. I didn't want to change that though, because the function does rightfully signal outwards that *nas is never changed, because its use of the unconsted value is contained entirely in that function (to encode a message) and nas is in fact not modified (given that the encoders work correctly). Admitted, I should also add a comment. ~Neels -- - Neels Hofmeyr <nhofmeyr(a)sysmocom.de> http://www.sysmocom.de/ ======================================================================= * sysmocom - systems for mobile communications GmbH * Alt-Moabit 93 * 10559 Berlin, Germany * Sitz / Registered office: Berlin, HRB 134158 B * Geschäftsführer / Managing Directors: Holger Freyther, Harald Welte