Re: native win32 aufs

----- Original Message -----
From: "Adrian Chadd" <adrian@creative.net.au>
To: "Robert Collins" <robert.collins@itdomain.com.au>
Cc: <squid-dev@squid-cache.org>
Sent: Thursday, March 15, 2001 9:05 PM
Subject: Re: native win32 aufs

> On Thu, Mar 15, 2001, Robert Collins wrote:
> > I've got a native win32 aufs working.
> >
> > Adrian - you're mainly to blame for aufs?
>
> Nope, its Stewart Forester and Henrik.
>
> > Here's your just desserts: Thank you!
> >
> > The native win32 one is trivially different - different calls for
all
> > the pthread_cond_* and so on. It would be very easy for me to make a
> > single aufs with some #defined code to build on unix as pthreads and
> > cygwin as win32 threads.
> >
> > I'm planning if/when time arises to extend the cygwin pthreads
interface
> > to make this obsolete... so here's my question:
> >
> > Is it better to make a new fs module e.g. awin32 and then when
cygwin is
> > extended obsolete that module, or to extend aufs with #defined code
and
> > then remove that code later?
>
> Hrm. See, I'm willing to bet that you're using the same basic
> stuff in store_dir_foo and store_io_foo that exists in
> the aufs versions. If so, I think its time to tidy up this
> code a little and have a "unified" ufs directory type
> with different IO types.

I didn't even touch those files, other than to change "aufs" to "awin32"
so the module registered and linked properly.

> What do the rest of y'all think?
>
Yes yes yes.

Rob
Received on Thu Mar 15 2001 - 03:10:21 MST