Re: Better pagecache statistics ?

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Re: Better pagecache statistics ?

Frank Ch. Eigler
Hi -

(Redirected from http://lkml.org/lkml/2005/12/1/182)

Badari Pulavarty wrote:

> [...]  Is there a way another user-level program/utility access some
> of the data maintained in those arrays ?

Not really.  One possibility is an on-demand /proc interface outlined
in systemtap bug #1154.

> [...]
> Does this mean that I can do something like
> page_cache[0xffff8100c4c6b298] = $mapping->nrpages ?
> And this won't generate bloated arrays ?

If by "bloat" you mean "trying to allocate 2**64 elements", then no,
it won't do that.  Systemtap associative arrays are more like hash
tables.

> [...]  Unfortunately, I can't capture whatever happend before
> inserting the problem. So it won't give me information about all
> whats there in the pagecache.

Until other mechanisms become available, one could perhaps start the
probe early on during boot.

> BTW, if you prefer - we can move the discussion to systemtap.

Done.

- FChE