[runtime] why f90_mset need guard the NULL pointer

[vfdff]

we can see the define of f90_mset in file runtime/flang/miscsup_com.c, which has a condition if (d) , the detail snip code as following, I think the user should make sure don't pass a NULL pointer to the runtime API f90_mset ? so this condition can be deleted according performance.

5359 #define _MSET(n, t)                                                            \
5360   void ENTF90(MSET##n, mset##n)(void *d, void *v, SZ_T size)                   \
5361   {                                                                            \
5362     if (d && size > 0)                                                         \
5363       __c_mset##n(d, *((t *)v), size);                                         \
5364   }

[d-parks]

I don't believe that the routines you're referencing are user callable - you might want to first double check whether these routines are internal prior to changing the semantics to have the user validate the arguments prior to the call.

Instead of deleting the test, you might want to give consideration to expanding it to also include checking that v != NULL.

[vfdff]

thanks @d-parks very much , so do you mean that the __c_cset##n is not a general API, which don't like the memset in libc ?