Re: [PATCH v2 2/4] mm: introduce fincore()

[Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>]

On Fri, Jul 04, 2014 at 05:15:59PM +0200, Cédric Villemain wrote:
> Le vendredi 4 juillet 2014 03:12:30 Christoph Hellwig a écrit :
> > On Thu, Jul 03, 2014 at 05:52:13PM -0400, Naoya Horiguchi wrote:
> > > This patch provides a new system call fincore(2), which provides
> > > mincore()- like information, i.e. page residency of a given file.
> > > But unlike mincore(), fincore() has a mode flag which allows us to
> > > extract detailed information about page cache like pfn and page
> > > flag. This kind of information is very helpful, for example when
> > > applications want to know the file cache status to control the IO
> > > on their own way.
> > 
> > It's still a nasty multiplexer for multiple different reporting
> > formats in a single system call.  How about your really just do a
> > fincore that mirrors mincore instead of piggybacking exports of
> > various internal flags (tags and page flags onto it.

We can do it in mincore-compatible way with FINCORE_BMAP mode.
If you choose it, you don't care about any details about other modes.
I don't make no default mode, but if we have a good reason, I'm OK
to set FINCORE_BMAP as default mode.

> The fincore à la mincore got some arguments against it too. It seems this 
> implementations try (I've not tested nor have a close look yet) to 
> answer both concerns : have details and also possible to have 
> aggregation function not too expansive.

Correct, that's the motivation of this non-trivial interface.
This could finally obsoletes messy /proc/kpage{flags,count} and/or
/proc/pid/pagemap kind of things, and we will not have to collect
information over all these interfaces (so that's less expensive.)

Thanks,
Naoya Horiguchi

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>