> - Memory management fixups. Handle out-of-memory conditions correctly,
> and handle high memory load much more gracefully.
I still see two mm/scheduling bugs with 2.1.127:
- kswapd is far too aggressive at killing the page cache. Assuming I run
a normal SMP X11 setup with 128MB and most of the memory used in the page
cache (about 60%). Now I start a kernel compile with -j4. The result is that
nearly all of the page cache is freed after a few seconds (it runs with >50%
free memory then), then when it continues compiling the page cache slowly
builds up again, then after about 30s is completely blow away again etc.
At the end of the kernel compile I have >50% of free memory and only a small
page cache.
I want the kernel to use my memory, not keep it free @)
This is easily observable when watching xosview.
- The scheduler is not fair enough. When I have a non-niced cpu hog running
on both CPUs all my xterms become very unresponsible. This didn't happen in
2.0, it is a 2.1 regression. When I nice the hogs it gets better, but I think
the kernel should be fairer to interactive processes even without user help.
The current situation makes Linux "feel slow".
Rik's scheduling big patch seems to help a bit here.
-Andi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/