High memory pressure makes system nearly unresponsive. 325

On Sat, 4 Feb 2006 22:33:21 +0100, Michael Heiming

Well, at least I have confirmation that such a thing *can* be done without clobbering the UI "experience."

Yabut, I've tested on 2.6.15 with preemtible kernel and all that, and don't see any real difference... except that once, on 2.6.15, I thought the box was never going to recover. It kept "stuttering" for several minutes after the high activity was over with.

That would make a difference, but I still can't see why I have no issue under 'Doze, then. You say your box is nearly the same spec. Does it also have an nforce2 chipset? I've read some posts about the Linux driver for nforce2 being prone to lockups, etc. I really don't know how valid the copmplaints were, or whether the issues are resolved, though.

Yeah? That's interesting. I compile all my graphics and 3d software locally, with athlon optimization, but I wouldn't have expected it to make a lot of difference for awk. I'll have to try that.

Same here. I've compiled my own kernels for years, and have always felt that I had a finely tuned match for my systems. I've tried recompiling, starting from the stock Slackware .config as well as another I got from have left something vital out, somehow.

I'm really running out of things to try.

I dunno. Maybe it's time to upgrade to SATA anyhow.

Thanks for the input.

Jonathan

-- Don't just hit reply. Email address is broken. Thank your friendly neighborhood spammer. Email replies to: