<div><span class="gmail_quote">On 10/9/08, <b class="gmail_sendername">Arnar Birgisson</b> <<a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:arnarbi@gmail.com" target="_blank">arnarbi@gmail.com</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">Hi Larry,<br><br>On Thu, Oct 9, 2008 at 22:06, Larry Dickson <<a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:ldickson@cuttedge.com" target="_blank">ldickson@cuttedge.com</a>> wrote:<br>
> Did my message (20081006) get through with the attached notes? I haven't<br>> heard any responses, and I'm wondering if the attachments (just short text<br>> things) somehow gummed things up.<br><br>Yes, they went through :)<br>
<br>I read your notes quickly, I didn't have any questions or comments.<br>I'm afraid I don't totally understand where this fits in and what<br>functionality it intends to replace.<br><br>cheers,<br>Arnar<br></blockquote>
</div>
<div> </div>
<div>It was in response to the problem you and David identified last month, namely ---</div>
<div> </div>
<div>On Fri, Sep 26 17:09:54 CEST 2008, Arnar Birgisson <arnarbi at <a onclick="return top.js.OpenExtLink(window,event,this)" href="http://gmail.com/" target="_blank">gmail.com</a>> wrote:<br>> Hey there,<br>><br>
> On Fri, Sep 26, 2008 at 15:50, David Naylor <dragonsa at <a onclick="return top.js.OpenExtLink(window,event,this)" href="http://highveldmail.co.za/" target="_blank">highveldmail.co.za</a>> wrote:<br>> >> > 2) Will stackless switch to another tasklet when entering blocking<br>
> >> > routines<br>> >> > (like IO), I know threads do this...<br>> >><br>> >> No it won't. In fact, it fundamentally *can't* (since it "runs" in<br>> >> only one thread).<br>
> ><br>> > Surely there is a way around this? Some kind of pooling select? If there is<br>> > no work around then I cannot see too much practical use for my thread library<br>> > [except having to avoid learning tasklets for someone who is familiar with<br>
> > threads]. As I understand it, due to the GIL the only real practical use for<br>> > threads is if one has blocking function calls (IO-type, etc)<br>><br>> The solution would be asynchronous I/O. There have been discussions<br>
> here occasionally about something generic, like wrapping libevent or<br>> similar in an interface that "looks" synchronous but in the background<br>> does async I/O and uses channels to make it look synchronous. I figure<br>
> such a thing would be an excellent component of your thread library.<br>><br>> > [Has the GIL restriction been fixed in 3k? As far as I know Jython does not<br>> > have this limitation...]<br>><br>
> The GIL has not been removed in Py 3.0, nor will it be removed any<br>> time soon. Jython does not have such a thing.<br> </div>
<div>This design solves all these problems, using only C/Unix select (which you pointed out is already used to do time.sleep) in the virtual machine; and it runs in only one thread. There is no need to remove the GIL.</div>
<div> </div>
<div>Larry<br> </div>