Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-27 Thread Jeff Squyres
On Sep 25, 2010, at 10:29 AM, Ralph Castain wrote: > What I was trying (poorly) to propose is that this capability be done the > traditional way of an option. To mimic 'make', I propose that we add a -j > option to autogen.pl. If a value is given, then that is the number of > parallel automake

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-25 Thread Ralph Castain
First off, let me publicly apologize to Jeff - this email thread came across wrong. I wasn't mad or upset, but was speaking more tongue-in-cheek. Email is a bad medium for such nuances, and I should have realized that before attempting it. My concern was solely that we had introduced a new behavio

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-25 Thread Jeff Squyres
Everyone always complains to me about how long the build takes, so I took a step to reduce that time. I have been manually setting AUTOMAKE_JOBS for a long time and run lots of other things at the same time as automake such as mail, ppt, etc. and haven't seen any noticeable difference. Admit

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-24 Thread George Bosilca
I would accept this behavior, at the condition that the threads are running at the lowest priority. This will give us the best of the two worlds, parallel build if the node is empty, and not a significant disturbance if I'm still busy around the computer. George. "All the books in the world

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-24 Thread Paul H. Hargrove
I don't feel as strongly about this as Ralph, but do think the new behavior violates the "principle of least surprise". -Paul Ralph Castain wrote: Been thinking about this more today, and I actually find this new "feature" disturbing. It bothers me that OMPI is now dictating that it will do a

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-24 Thread Ralph Castain
Been thinking about this more today, and I actually find this new "feature" disturbing. It bothers me that OMPI is now dictating that it will do a parallel build without my knowledge unless I specifically tell it not to. If it were technically possible, would we next force "make -j4"?? How would th

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-24 Thread Ralph Castain
I hope you'll understand if I don't run that test while on the road...one battery yank per week is my limit :-) On Fri, Sep 24, 2010 at 4:40 AM, Jeff Squyres (jsquyres) wrote: > Also to clarify: > > - did autogen set am-jobs to 2 in your case? (it should do that if lstopo > is not found - it a

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-24 Thread Jeff Squyres (jsquyres)
Also to clarify: - did autogen set am-jobs to 2 in your case? (it should do that if lstopo is not found - it also limits itself to 4 at max) - in the same scenario, what happens if you manually set am-jobs to 1 and run autogen? Ie do you get the same heat/sluggishness? I have experienced vms

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-24 Thread Ralf Wildenhues
* Ralph Castain wrote on Fri, Sep 24, 2010 at 06:41:38AM CEST: > My point is just that it is unwise to assume that the OMPI build can utilize > all available processors. I'm sure it's fine for the MTT runs, especially on > Jeff's machines as they are dedicated to that purpose - just not a good > ge

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-24 Thread Ralph Castain
Sent to both for reference (see below) Just to clarify. It wasn't a deadlock situation, but rather that the machine was overloaded and running so hard that the response to keystrokes was multiple seconds. Thus, there was no way to shut it down from the keyboard or screen. Even a ctrl-c was just ge

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-24 Thread Ralf Wildenhues
Hello Ralph, wow, that's not good to hear. I knew the perl ithreads implementation wasn't all that efficient, but causing a deadlock sounds like you have more trouble than just perl; at least I hope so. For reference, can you send 'perl -V' output (if you like, to the bug-automake at gnu.org lis

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-23 Thread Ralph Castain
I found one major negative to this change - it assumes that my build is being done in exclusion of anything else on my computer. Unfortunately, this is never true. So my laptop hemorrhaged itself into frozen silence, overheated to the point of being burning hot, and had to have its battery yanked

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-22 Thread Jeff Squyres
On Sep 22, 2010, at 4:51 PM, Ralf Wildenhues wrote: > Thanks for the measurements! I'm a bit surprised that the speedup is > not higher. Do you have timings as to how much of the autogen.pl time > is spent inside automake? No, they didn't. I re-ran them to just time autoreconf (is there a way

Re: [OMPI devel] Setting AUTOMAKE_JOBS

2010-09-22 Thread Ralf Wildenhues
Hi Jeff, adding bug-automake in Cc: (non-subscribers can't post to the Open MPI list, so please remove that Cc: in case) * Jeff Squyres wrote on Wed, Sep 22, 2010 at 03:50:19PM CEST: > $AUTOMAKE_JOBS Total wall time > valueof autogen.pl > 83:0

[OMPI devel] Setting AUTOMAKE_JOBS

2010-09-22 Thread Jeff Squyres
Some of you may be unaware that recent versions of automake can run in parallel. That is, automake will run in parallel with a degree of (at most) $AUTOMAKE_JOBS. This can speed up the execution time of autogen.pl quite a bit on some platforms. On my cluster at cisco, here's a few quick timin