Re: [tinydtls-dev] Current status of HAVE_ASSERT_H

[Gaëtan Harter <gaetan.harter@xxxxxxxx>]

Hi Raul,

I saw this problem before and I thought I addressed it but I must have 
forgotten.
Without HAVE_ASSERT_H the code does not compile.

Gaëtan
On 04/04/2017 02:40 PM, Raul Fuentes wrote:
> Hi people,
>
> I have a observation regarding *HAVE_ASSERT_H*, seems that Linux and 
> Contiki (and RIOT)  supports assert.h, but also this become almost 
> mandatory to run.
>
> Yet, I found discrepancies between ccm.c and sha2.c
>
> Those are the lines for sha2.c
>
>     /#ifdef HAVE_ASSERT_H
>     #include <assert.h>    /* assert() */
>     #else
>     #ifndef assert
>     #warning "assertions are disabled"
>     #  define assert(x)
>     #endif/
>
>
> Those are the lines for ccm.c
>
>     /#ifdef HAVE_ASSERT_H
>     # include <assert.h>
>     #endif/
>
>
> Mostly because both of them call assert() without any type of protection.
>
> Technically speaking, ccm.c should be the same than sha2.c. And I 
> think dtls.c could benefit of this too.
>
> What is the general opinion about this?
>
> P.S. I'm assuming this protection is for more restricted resources 
> such as applications running with FreeRTOS.
>
>
>
>
> _______________________________________________
> tinydtls-dev mailing list
> tinydtls-dev@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or unsubscribe from this list, visit
> https://dev.eclipse.org/mailman/listinfo/tinydtls-dev