> > I'm also pretty sure last few development kernels have some big
> > problems with memory. Looks like kernel looses big amount of memory
> > (7-8 MB or even more) and never finds it back.
> >
> > It's enough to execute ls -alR / to loose massive amount of memory In
> > fact, on 32MB machine, memory problem is so obvious that you don't
> > need any tool to see it (top, free or something). Machine gets slower
> > and you can notice it.
>
> The memory "leak" is probably just excessive growth of the dcache and
> inodes. [...]
nope, when we are short of memory, we zap all dentries pretty fast, down
to zero. And the SLAB should free their pages subsequently. You can see
that happen in the HashInfo output. Eg. doing an "ls -aR /" zaps the whole
dentry tree 4-5 times on my system, before completion.
the memory leak thing i personally see is a pretty slow but steady growth
in kernel-allocated nonswappable memory, which renders a normal desktop
system unusable after about 1 day uptime.
is that memory leak detection patch still around somewhere ... does it
detect SLAB leaks too?
-- mingo