Noah Roberts noted: I was going through the kernel changelog when I heard about this and it seems that may be the case. I'm getting used to using Fernando's patched kernels, which are working out pretty well; but it would be very nice to see this in the main kernel. I'm not excited about having another subsystem(pam) to tweak, but this seems to be something that can easily be set up during installation depending upon the type of system the user wants. Anyway, this might not be everything we wanted or need; but our needs did seem to get acknowledged.

On 5/8/05, Tracey Hytry <shakti(a)bayarea.net> wrote: I don't think it really makes a difference if the user wants it or not. I never saw an option in the kernel to turn it on so apparently it always is. So pam should know how to deal with it always and then limits.conf can just be set up for the user's needs.

Reuben Martin wrote: I have a question regarding using these patchsets on already patched kernels. A month or so ago I tried adding these patches above to the debian patched kernel, much like I assume you are doing with the gentoo patched kernel. I ran into problems with Ingo's patch and Andrews mm patch not applying cleanly. It seemed like it'd be a lot of work to clean them up by hand so I just gave up for the time being, as the regular 2.6.11 that I had (with debian patches) gave me acceptible latency for now. (Debian packages lsm module sources so it wasn't a problem getting that working) So, I'm wondering if anyone has any advice for doing this sort of thing. Do I have to just bite the bullet and just apply the patches by hand? I'm not completely averse do doing this when I get some free time, but if there is a better way I'd rather know about it. BTW, what exactly does "broken out" mean, in reference to andrew's mm patches? It seemed to me like its just all the individual patches from mm, but I didn't look very hard. Thanks, Dave Strock