[ntp:hackers] Segfaults in ntp-dev

David L. Mills mills at udel.edu
Sun Jun 3 22:24:05 PDT 2007


Paul,

This is bogus. The original architecture of Algol 60 and eventually 
Algol 68 was principle driven and well conceived. Nonsense creation of 
structures by implication is evil. What happens if I record the pointers 
and later attempt to free them? It might even work, but that is not an 
appropriate architectural model.

Dave

Paul Vixie wrote:

> there's no reason in principle (given ANSI C as a requirement) why
> the function can't just return a struct timeval. the days when
> functions couldn't return aggregates are long long gone.
>
> [fh:amd64] cc -c -ansi mills-test.c
> [fh:amd64] cat mills-test.c
> #include <sys/time.h>
> #include <time.h>
>
> struct timeval
> mills_test(struct timeval input) {
> return input;
> }
>
> how a compiler decides to implement this, stack, register, heap, paper
> airplane, etc, is not a concern except when doing this in a kernel.




More information about the hackers mailing list