Date: Sun, 27 Jun 2010 16:35:36 +0200 From: Gabor Kovesdan <gabor@FreeBSD.org> To: soc-status@freebsd.org Subject: Collective resource limits status report #4 Message-ID: <4C2761B8.9030800@FreeBSD.org> In-Reply-To: <4C230A0B.3080700@FreeBSD.org> References: <4C1BCB96.4040608@FreeBSD.org> <AANLkTikwUYWCIvlDQ60L4L8gMcEeDFIV6850csEwuH8E@mail.gmail.com> <4C21CAF0.2040607@FreeBSD.org> <AANLkTikLoskpABjaUlufnBGOBeU-Z62CiuQfH5sDyY1Z@mail.gmail.com> <4C230A0B.3080700@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Hello, since the last status report, I've added the getjlimit() and setjlimit() calls. They are fully implemented (unless I didn't consider some details) in a general way that doesn't depend tightly on particular kinds of resource limits. I've also started to implement the first limit type, JLIMIT_NUMPROC, which is quite straightforward. But there is a bug I'm facing in the internal structures I use for accounting. I have a linked list of jobs, and each job entry has a head of another kind of linked list, which stores entries of processes inside the given job. I have no problem with the outer list but the inner one makes my kernel crash. I double-checked those code parts and I still think they are correct, so I couldn't figure out the problem so far. This week I haven't merged to Perforce yet, first I'd like to make the code a bit more stable. I hope I can accomplish it in 1-2 days. While debugging my code, I've also written some additions to my test program, which is supposed to be general enough and accepts various paramteres. My intention is to use it as a regression test inside a shell script, which will call it with different parameter sets to test different cases. Regards, Gabor
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4C2761B8.9030800>