<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
--></style>
</head>
<body class='hmmessage'>
I thought about this a little..but..don't Modula-3 user threads preempt at arbitrary points?<br><br> - Jay<br><br><br>> From: dragisha@m3w.org<br>> To: jay.krell@cornell.edu<br>> Date: Fri, 18 Sep 2009 11:37:48 +0200<br>> CC: m3devel@elegosoft.com<br>> Subject: Re: [M3devel] formsedit crash<br>> <br>> I've had some funny race/thread-order conditions when I was switching to<br>> pthreads.... Someone lazy got too grooved in with fixed order of<br>> execution implied by how user threads work. It was in pp package and<br>> also somewhere I can't remember right now... Will look a bit through hm3<br>> svn (dead in house spinoff of pm3 from times when all mainstream m3 had<br>> was user threads).<br>> <br>> On Fri, 2009-09-18 at 07:49 +0000, Jay K wrote:<br>> > I have it further narrowed down to the last two weeks of 1/2007.<br>> > Which is just a few changes.<br>> > I fear it is the switch from user threads to pthreads on 1/23/2007.<br>> > I'll narrow it down further though, and then try user threads on<br>> > Solaris<br>> > (which will probably require repairing initialization order to make<br>> > them work<br>> > again anyway).<br>> > <br>> > - Jay<br>> > <br>> > <br>> > <br>> > ______________________________________________________________________<br>> > From: jay.krell@cornell.edu<br>> > To: m3devel@elegosoft.com<br>> > Subject: formsedit crash<br>> > Date: Wed, 16 Sep 2009 15:18:54 +0000<br>> > <br>> > The formsedit crash appears to have started between 12/1/2006 and<br>> > 3/1/2007.<br>> > I will confirm and further narrow this down over the next few days.<br>> > I've been building various dates/versions and seeing how they act.<br>> > <br>> > - Jay<br>> > <br>> > <br>> > <br>> > <br>> -- <br>> Dragiša Durić <dragisha@m3w.org><br>> <br> </body>
</html>