This is the mail archive of the
cygwin-patches@cygwin.com
mailing list for the Cygwin project.
Re: /dev/dsp
- To: cygwin-patches at cygwin dot com
- Subject: Re: /dev/dsp
- From: Christopher Faylor <cgf at redhat dot com>
- Date: Tue, 17 Apr 2001 12:24:30 -0400
- References: <F23AZndKaQf0Stamg6d00000c37@hotmail.com>
- Reply-To: cygwin-patches at cygwin dot com
On Tue, Apr 17, 2001 at 03:36:17PM -0000, Andy Younger wrote:
>I have just been looking through the applied patch after failing to get the
>CVS version of /dev/dsp working on my "at work" machine.
>
>It seems to be to do with the FH_OSS_DSP enumeration in fhandler.h, this has
>been set up as a fast device, where as in the original patch, it was a slow
>one. What exactly is the difference between slow & fast devices. Should it
>not be a slow device being that calls to write() can block for a good
>fraction of a second (depending on the size & number of fragments).
>
>Making it a slow device enables it to work. Is this maybe just exposing a
>problem somewhere else? I don't have time to look at this right now, maybe
>later on, any hints welcomed.
>
>As to how I got it too work last night, I'm not sure - maybe it was just
>tiredness.. Bahhh..
This doesn't qualify as a slow device unless it can be interrupted.
That means that you have to provide the appropriate framework for doing
so in select.cc like you see with other slow devices. If/when you do
that we can make it slow again.
The problem that you're seeing is due to the fact that I forgot to move
the \dev\dsp string in path.cc. I have rectified this.
cgf