FreeTDS Development Group

["ZIGLIO, Frediano, VF-IT" <Frediano.Ziglio AT vodafone.com>]

> 
> On Jan 11,  3:07pm, Frediano.Ziglio AT vodafone.com ("ZIGLIO, 
> Frediano, VF-IT") wrote:
> -- Subject: Re: [freetds] Heavy load patch for net.c
> 
> | This is an updated patch that contain the if as you 
> suggested to avoid
> | overflow.
> | It contains also fix for the other select in net.c. I didn't add
> | tds_poll cause
> | - it avoids to reinitialize (FD_ZERO) fd_set structs at 
> every call if
> | poll is not available
> | - it's quite small so less problems and easy to backport
> | It does not change pool but I don't think it's a big problem.
> | If nobody objects I would backport it in a couple of day... 
> you to get
> | RC2 :)
> | 
> 
> - Why do you undef USE_POLL?

I use this macro only in net.c to distinguish about poll or select. It
sum up all tests (HAVE_POLL and HAVE_POLL_H). Do not confuse USE_POLL
with HAVE_POLL. You can note that USE_POLL is always defined some rows
ago. It's just to make sure that USE_POLL is not defined.

> - From the poll(2) opengroup definition:
> 
>     POLLERR
>     An error has occurred on the device or stream. This flag is
>     only valid in the revents bitmask; it is ignored in the events
>     member.
> 
>     POLLHUP
>     The device has been disconnected. This event and POLLOUT are
>     mutually exclusive; a stream can never be writable if a hangup
>     has occurred. However, this event and POLLIN, POLLRDNORM,
>     POLLRDBAND or POLLPRI are not mutually exclusive. This flag is
>     only valid in the revents bitmask; it is ignored in the events
>     member.
> 
>     So you should not use POLLIN|POLHUP in the read mask, 
> just POLLIN is good

yes, so I should remove POLLHUP from macro.

>     enough, and in the error mask, 0 should be ok.
> 
> christos

What happen in case of error? Error is always returned?? If I put 0 in
events??

freddy77