<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.7653.38">
<TITLE>RE: [Cook] Cook distributed build with load-balancing system</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<P><FONT SIZE=2>Glad to hear it works for you.<BR>
You probably also noticed that you could code up:<BR>
<BR>
#ifdef parallel<BR>
parallel_jobs = [parallel];<BR>
parallel_rsh = ./parsub;<BR>
parallel_hosts = host1 host2 host3 host4;<BR>
#endif<BR>
<BR>
And thus be able to say cook parallel=3 or whatever value you want.<BR>
<BR>
Jerry Pendergraft voice:651-523-6935<BR>
St.Jude Medical mobil:651-491-0163<BR>
Atrial Fibrillation Division email:jpendergraft@sjm.com<BR>
1350 Energy Lane<BR>
St.Paul, MN 55108<BR>
<BR>
<BR>
-----Original Message-----<BR>
From: flamtapgt@gmail.com on behalf of Ryan Thompson<BR>
Sent: Thu 04-Sep-08 08:57<BR>
To: Pendergraft, Jerry<BR>
Cc: cook-users@auug.org.au<BR>
Subject: Re: [Cook] Cook distributed build with load-balancing system<BR>
<BR>
Thanks, that works nicely.<BR>
<BR>
Now I can run 'cook <target> parallel=1' to enable the parallel features.<BR>
<BR>
-RT<BR>
<BR>
On Wed, Sep 3, 2008 at 5:36 PM, Pendergraft, Jerry <JPendergraft@sjm.com>wrote:<BR>
<BR>
> Try using the #ifdef mechanism ala:<BR>
><BR>
> #ifdef parallel<BR>
> parallel_jobs = 4;<BR>
> parallel_rsh = ./parsub;<BR>
> parallel_hosts = host1 host2 host3 host4;<BR>
> #endif<BR>
><BR>
> Jerry Pendergraft voice:651-523-6935<BR>
> St.Jude Medical mobil:651-491-0163<BR>
> Atrial Fibrillation Division email:jpendergraft@sjm.com<email%3Ajpendergraft@sjm.com><BR>
> 1350 Energy Lane<BR>
> St.Paul, MN 55108<BR>
><BR>
> -----Original Message-----<BR>
> From: flamtapgt@gmail.com on behalf of Ryan Thompson<BR>
> Sent: Wed 03-Sep-08 15:34<BR>
> To: Pendergraft, Jerry<BR>
> Cc: cook-users@auug.org.au<BR>
> Subject: Re: [Cook] Cook distributed build with load-balancing system<BR>
><BR>
> This works well with my environment. Now I would like to make using the<BR>
> parallel features optional.<BR>
><BR>
> So if the command is 'cook' the project will build using the normal<BR>
> options.<BR>
> If the command is 'cook parallel' the project will build using parallel<BR>
> builds.<BR>
><BR>
> So somehow I need to conditionally set the 'parallel_jobs', 'parallel_rsh',<BR>
> and 'parallel_hosts' variables only when the target "parallel" is specified<BR>
> on the command line.<BR>
><BR>
> I have tried a number of options but I can't get this to work. Please let<BR>
> me<BR>
> know how to do this, or at least where to look in the user's guide.<BR>
><BR>
> Here's the gist of what I want to do. This doesn't work though (it doesn't<BR>
> build in parallel).<BR>
><BR>
> ../bin/myprogram: [OBJ] [INCDIR]<BR>
> {<BR>
> [CC] -o ../bin/[target] [LIBS] [OBJ] ;<BR>
> }<BR>
><BR>
> parallel::<BR>
> {<BR>
> parallel_jobs = 4;<BR>
> parallel_rsh = ./parsub;<BR>
> parallel_hosts = host1 host2 host3 host4;<BR>
> cook ../bin/myprogram<BR>
> }<BR>
><BR>
> Thanks for any help,<BR>
> -Ryan Thompson<BR>
><BR>
><BR>
> On Wed, Sep 3, 2008 at 3:26 PM, Pendergraft, Jerry <JPendergraft@sjm.com<BR>
> >wrote:<BR>
><BR>
> > -----Original Message-----<BR>
> > From: cook-users-bounces+jpendergraft=sjm.com@auug.org.au on behalf of<BR>
> > Ryan Thompson<BR>
> > Sent: Wed 03-Sep-08 13:26<BR>
> > To: cook-users@auug.org.au<BR>
> > Subject: [Cook] Cook distributed build with load-balancing system<BR>
> ><BR>
> > > I am interested in using the "virtual machine" ability to run a<BR>
> parallel<BR>
> > > build in our batch compute environment (Platform LSF, similar to Sun<BR>
> > Grid,<BR>
> > > etc.).<BR>
> ><BR>
> > >Since this is not rsh or ssh I am not sure how it will work. I think I<BR>
> can<BR>
> > >point the parallel_rsh variable to a script which does the job<BR>
> submission<BR>
> > >(similar to using cook_rsh), but I am concerned by this paragraph in<BR>
> > >"limitations":<BR>
> ><BR>
> > >"The exit status of the remote command is not reported in the exit<BR>
> status<BR>
> > of<BR>
> > >the rsh(1) command. There are internal contortions used by Cook to<BR>
> obtain<BR>
> > >the exit status;"<BR>
> ><BR>
> > > Can you provide more information on what Cook will look for to<BR>
> determine<BR>
> > the<BR>
> > > exit status of the command named in "parallel_rsh"?<BR>
> ><BR>
> > The paragraph refers to the inability of rsh to correctly report exit<BR>
> > status, thus<BR>
> > the "internal contortions" used by cook were designed to ignore that and<BR>
> > work around it.<BR>
> > Thus you should see the correct exit status, no matter what actually<BR>
> farms<BR>
> > out the command.<BR>
> ><BR>
> > Jerry Pendergraft voice:651-523-6935<BR>
> > St.Jude Medical mobil:651-491-0163<BR>
> > Atrial Fibrillation Division email:jpendergraft@sjm.com<email%3Ajpendergraft@sjm.com><BR>
> <email%3Ajpendergraft@sjm.com <email%253Ajpendergraft@sjm.com>><BR>
> > 1350 Energy Lane<BR>
> > St.Paul, MN 55108<BR>
> ><BR>
> > >Thanks,<BR>
> > >-Ryan Thompson<BR>
> ><BR>
> ><BR>
><BR>
><BR>
<BR>
</FONT>
</P>
</BODY>
</HTML>