Re: [PATCH]: Adding a counter in vma to indicate the number of physical_pages_backing it

[Albert Cahalan]

Quoting two different people:

> BTW, what is smaps used for (who uses it), anyway?
...
> smaps is only a debugging kludge anyways and it's
> not a good idea to we bloat core data structures for it.

I'd be using it in procps for the pmap command if it
were not so horribly nasty. I may eventually get around
to using it, but maybe it's just too gross to tolerate.
That mess should never have slipped into the kernel.
Just take a look at /proc/self/smaps some time. Wow.

A month or two ago I supplied a patch to replace smaps
with something sanely parsable. I was essentially told
that we already have this lovely smaps dungheap that I
should just use, but a couple people were eager to see
the patch go in.

Anyway, I need smaps stuff plus info about locked memory
and page sizes. Solaris provides this. People seem to
like it. I guess it's for performance tuning of app code or
maybe for scalibility predictions.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/