"Frank Behrens" <[EMAIL PROTECTED]> writes:
> Juliusz Chroboczek <[EMAIL PROTECTED]> wrote on 2 Dec 2007 16:02:
>> Any ideas to make timeval_cmp more efficient are warmly welcomed.
>
> In FreeBSD sources I read in sys/time.h
>
> #define timevalcmp(tvp, uvp, cmp) \
> (((tvp)->tv_sec == (uvp)->tv_sec) ? \
> ((tvp)->tv_usec cmp (uvp)->tv_usec) : \
> ((tvp)->tv_sec cmp (uvp)->tv_sec))
>
> It seems to have an diffent API and I don't know if it uses less operations
> on default path.
It looks like it could be significantly faster in the general case, but
possibly not if most of the compared times differ in the tv_sec field.
A non-macro variant could be something like
static inline int
timeval_before(struct timeval *t1, struct timeval *t2)
{
return (t1->tv_sec != t2->tv_sec ?
t1->tv_sec < t2->tv_sec :
t1->tv_usec < t2->tv_usec);
}
where the sense of the first test is switched to optimise for the >1s apart
case, as it is in current polipo.
But to be honest I'm surprised this function shows up at all, it's just a
few instructions in any case. I guess it means that polipo is quite zippy
already.
-j.
-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
Polipo-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/polipo-users
