nanogui: Re: [linuxce-devel] Microwindows architecture explained


Previous by date: 31 Oct 1999 21:23:28 -0000 Nano-X now documented, Greg Haerr
Next by date: 31 Oct 1999 21:23:28 -0000 Re: Microwindows architecture explained, Thomas Stewart
Previous in thread: 31 Oct 1999 21:23:28 -0000 Re: [linuxce-devel] Microwindows architecture explained, Greg Haerr
Next in thread:

Subject: Re: [linuxce-devel] Microwindows architecture explained
From: Alan Cox ####@####.####
Date: 31 Oct 1999 21:23:28 -0000
Message-Id: <E11i2EX-0001lV-00@the-village.bc.nu>

> it's UNIX sockets...) so there's not a big need to error check the status.
> I'll take a look at the X protocol and see whether there's a wire-level
> synchronization
> capability in the rare case that things really do get hosed up.

X leaves reliability problems to the transport layer (which IMHO is the right
thing to do). If your tcp connection dies the app goes pop too.

> Although more complicated, an async error event is a good idea.  If we're using
> UNIX sockets, it's unlikely this will ever happen.

The X idea is that most operations dont fail ever - eg drawing a line, so these
calls are void and queue the operation. Instead X can return an error to
those calls that is delayed from the call

> 


Previous by date: 31 Oct 1999 21:23:28 -0000 Nano-X now documented, Greg Haerr
Next by date: 31 Oct 1999 21:23:28 -0000 Re: Microwindows architecture explained, Thomas Stewart
Previous in thread: 31 Oct 1999 21:23:28 -0000 Re: [linuxce-devel] Microwindows architecture explained, Greg Haerr
Next in thread:


Powered by ezmlm-browse 0.20.