ps performance sucks (was Re: dcache_rcu [performance results])

["Martin J. Bligh" <mbligh@aracnet.com>]

> Clearly ps could do with a cleanup.  There is no reason to
> read environ if it wasn't asked for.   Deciding which files
> are needed based on the command line options would be a
> start.
> 
> I'm thinking that ps, top and company are good reasons to
> make an exception of one value per file in proc.  Clearly
> open+read+close of 3-5 "files" each extracting data from
> task_struct isn't more efficient than one "file" that
> generates the needed data one field per line.

I think it's pretty trivial to make /proc/<pid>/psinfo, which
dumps the garbage from all five files in one place. Which makes
it 5 times better, but it still sucks.
 
> Don't get me wrong.  I believe in the one field per file
> rule but ps &co are the exception that proves (tests) the
> rule.  Especially on the heavily laden systems with
> tens of thousands of tasks.  We could do with a something
> between /dev/kmem and five files per pid.

I had a very brief think about this at the weekend, seeing
if I could make a big melting pot /proc/psinfo file that did
seqfile and read everything out in one go, using seq_file
internally to interate over the tasklist. The most obvious
problem that sprung to mind seems to be the tasklist locking -
you obviously can't just hold a lock over the whole thing.
As I know very little about that, I'll let someone else suggest
how to do this, but I'm prepared to do the grunt work of implementing
it if need be.

M.