Re: PATCH: Cygwin: file mode support and ufs writecleanswap bugfix

----- Original Message -----
From: "Henrik Nordstrom" <hno@hem.passagen.se>
To: "Adrian Chadd" <adrian@creative.net.au>
Cc: <squid-dev@squid-cache.org>
Sent: Thursday, January 04, 2001 6:08 AM
Subject: Re: PATCH: Cygwin: file mode support and ufs writecleanswap bugfix

> Adrian Chadd wrote:
>
> > I don't meant to be evil, but I'm kind of curious now. Why is there
> > a cygwin port when Romeo has done a port of squid to native NT?
>
> Good question, especially considering that the cygwin port has some
> issues mainly due to lack of proper non-blocking support in cygwin...
> (think this only affects accept() now.. Robert should know more)

AFAIK there are no accept() blocking issues.. (although there is an issues in client_side that has a cygwin workaround - it's not on
accept but on read() however). Cygwin has had patches back to fcntl to interpret the O_NONBLOCK flags correctly. It needs a
polygraph session to be sure that there are none however.

>
> But the short answer is: A cygwin port is easy (almost a UNIX
> environment), and the native port requires a Microsoft compiler and
> quite extensive patching of Squid which is not yet in the
> distribution...
>
> Either way, any general problems the cygwin port sees will most likely
> also be seen by the native port as well, except for the few that is
> cygwin bugs/shortcomings rather than effects of the underlying OS.
>
> /Henrik
>
>
Received on Wed Jan 03 2001 - 16:30:03 MST