On Tue, 2008-11-04 at 17:46 -0500, Peter Memishian wrote:
> > > Along the way, I also made some minor modifications to ipmpstat and
>  > > in.mpathd for consistency and elegance.  Could someone take a look?
>  > > 
>  > >   http://zhadum.east/ws/clearview/clearview-ipmpdev
>  > 
>  > This looks fine to me, only two minor comments:
> 
> Thanks for the review Seb.
> 
>  > usr/src/cmd/cmd-inet/usr.lib/in.mpathd/mpd_tables.c
>  > and
>  > usr/src/cmd/cmd-inet/usr.sbin/ipmpstat/ipmpstat.c
>  > 
>  > * I'm not particularly fond of using the global "errno" to store the
>  > return value of the sysevent_evc_*() functions (stemming from MT
>  > paranoia), but I don't have a strong feeling about it.
> 
> I'm not sure what the MT issue would be -- even in a multithreaded program,
> errno is per-thread.

There's no issue, I just wouldn't use errno as a convenience within a
local function myself.

> 
>  > usr/src/cmd/cmd-inet/usr.sbin/ipmpstat/ipmpstat.c
>  > 
>  > * 912,916: I can't immediately tell what the ramifications would be of
>  > actually returning an error in this callback here.
> 
> The GPEC interface documents either returning 0 or EAGAIN; the latter is
> supposed to be used when you want the framework to give you the event
> again.  Since we just want to ignore the event, we return 0.

I see, thanks.

-Seb



Reply via email to