cygserver debug output patch

Robert Collins robert.collins@syncretize.net
Tue Jun 11 07:19:00 GMT 2002



> -----Original Message-----
> From: cygwin-patches-owner@cygwin.com 
> [mailto:cygwin-patches-owner@cygwin.com] On Behalf Of Conrad Scott
> Sent: Tuesday, 11 June 2002 5:58 AM


> This defines the same set of XXX_printf macros as does 
> sys/strace.h: of
> these, system_printf maps to a printf on stderr and the 
> others ditto if the
> DEBUGGING flag is given (i.e. --enable-debugging) and no-ops 
> otherwise. They
> also use __PRETTY_FUNCTION__ to give more information about 
> the problem
> location.
> 
> I've also added more debugging calls into the code, whenever 
> I got really
> lost, basically :-)

All cool. I'll actually review this tomorrow night, I've used my hacking
time tonight already :[.
 
> Apart from that, just a couple of minor changes to a pthread_once_t
> initialisation and some pure virtual functions on the
> cygserver_transport_base class.
> 
> I hope this is all fine (Robert et al). I'll continue with 
> some more hacking
> around fun. Umm . . . sorry, make that "careful development" :-)

It's great. Welcome to the club. There's a grand total of three
contributors to the cygserver code - Egor, myself... and you.

One nit: your changelog doesn't say *what* from
cygserver_transport_pipes.cc you are moving to woutsup.h. 

> p.s. I've attached the new "woutsup.h" file separately as I couldn't
> convince cvs diff to put it into the patch file. I thought the -N flag
> should do it, but I had no luck. Could someone give me a hint? Thanks.

You need CVS access to make that work right... then you can CVS add, and
use the -N flag, and then cvs remove without ever committing. 

Rob



More information about the Cygwin-patches mailing list