From mjbaysek at cs.cmu.edu Tue Apr 6 13:12:41 2010 From: mjbaysek at cs.cmu.edu (Michael J. Baysek) Date: Tue, 06 Apr 2010 13:12:41 -0400 Subject: [auton-users] Update your gmake-magic Message-ID: <4BBB6B89.4050309@cs.cmu.edu> Hi Lab, Please update your gmake-magic right away. 1) Log into cpu.autonlab.org 2) cd h/gmake-magic (or wherever your gmake-magic(s) is/are) 3) cvs up This changes the way rmath_new, and other static .a files are handled within CVS. It allows you to use rmath_new without needing to check out 970 MB of extra bits into your ~/h. If you had rmath_new checked out in your ~/h , I have taken the liberty of performing the gmake-magic update for you, and removing the old rmath_new. The next time you build something which requires rmath_new, it will be checked out afresh. If you have h/ directories on your local disks, you should also update the gmake-magic and delete rmath_new. The change has been tested for the most common cases, and should also work universally. However, if you encounter a problem we did not account for, please email me or Steve mjbaysek at cs.cmu.edu sbrudene at cs.cmu.edu Best, Mike P.S. In general, it is good practice to run cvs up on your whole h/ from time to time. Doing so makes sure that your code will compile and run correctly against the latest changes that have been committed to CVS. You can take this email as a recommendation to do the full h/ update if you haven't done so recently. To do the full h/ update: cd ~/h cvs up Happy coding -- Michael J. Baysek, Systems Analyst Carnegie Mellon University, Auton Lab 412-268-8939 Office mjbaysek at cs.cmu.edu http://www.autonlab.org From mjbaysek at cs.cmu.edu Mon Apr 19 17:31:51 2010 From: mjbaysek at cs.cmu.edu (Michael J. Baysek) Date: Mon, 19 Apr 2010 17:31:51 -0400 Subject: [auton-users] Memory issues on Compute Machines Message-ID: <4BCCCBC7.4080308@cs.cmu.edu> Hi Lab, Recently, some of you have experienced errors when mallocking larger amounts of RAM on various machines. I've made further adjustments to the vm subsystems on the compute nodes to minimize this effect. Please continue to let me know if you experience any oddities. Best, Mike -- Michael J. Baysek, Systems Analyst Carnegie Mellon University, Auton Lab 412-268-8939 Office mjbaysek at cs.cmu.edu http://www.autonlab.org