Patch to have clock_gettime(CLOCK_CPUTIME_ID) return ENOENT if HP timing is unreliable

Christoph Lameter clameter@sgi.com
Thu Jun 17 20:05:00 GMT 2004


Well, it would have been nice to get responses on the subject matter
rather than a threat. I was certainly only trying to help.

I still have no answer as to why it is acceptable for
clock_gettime(CLOCK_CPUTIME_ID) to return unreliable data.

Is it that applications should not rely on
clock_gettime(CLOCK_CPUTIME_ID) if clock_getclockcpuid() returns
ENOENT? If so is that documented somewhere?

On Thu, 17 Jun 2004, Ulrich Drepper wrote:

> Christoph Lameter wrote:
>
> > Would that be acceptable?
>
> No.  I've explained this before and just keeping on sending such mails
> will have only one result, namely, that you'll be ignored regardless of
> what you say.



More information about the Libc-alpha mailing list