[lkml]   [2012]   [Feb]   [17]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH 2/2] [RFC] fadvise: Add _VOLATILE,_ISVOLATILE, and _NONVOLATILE flags
On Fri, 2012-02-17 at 15:45 +1100, Dave Chinner wrote:
> On Wed, Feb 15, 2012 at 12:37:50PM +1100, NeilBrown wrote:
> > On Tue, 14 Feb 2012 16:29:10 -0800 John Stultz <> wrote:
> >
> > > But I'm open to other ideas and arguments.
> >
> > I didn't notice the original patch, but found it at
> >
> > and had a look.
> >
> > My first comment is -ENODOC. A bit background always helps, so let me try to
> > construct that:
> >
> > The goal is to allow applications to interact with the kernel's cache
> > management infrastructure. In particular an application can say "this
> > memory contains data that might be useful in the future, but can be
> > reconstructed if necessary, and it is cheaper to reconstruct it than to read
> > it back from disk, so don't bother writing it out".
> >
> > The proposed mechanism - at a high level - is for user-space to be able to
> > say "This memory is volatile" and then later "this memory is no longer
> > volatile". If the content of the memory is still available the second
> > request succeeds. If not, it fails.. Well, actually it succeeds but reports
> > that some content has been lost. (not sure what happens then - can the app do
> > a binary search to find which pages it still has or something).
> >
> > (technically we should probably include the cost to reconstruct the page,
> > which the kernel measures as 'seeks' but maybe that isn't necessary).
> >
> > This is implemented by using files in a 'tmpfs' filesystem. These file
> > support three new flags to fadvise:
> >
> > POSIX_FADV_VOLATILE - this marks a range of pages as 'volatile'. They may be
> > removed from the page cache as needed, even if they are not 'clean'.
> > POSIX_FADV_NONVOLATILE - this marks a range of pages as non-volatile.
> > If any pages in the range were previously volatile but have since been
> > removed, then a status is returned reporting this.
> > POSIX_FADV_ISVOLATILE - this does not actually give any advice to the kernel
> > but rather asks a question: Are any of these pages volatile?
> What about for files that aren't on tmpfs? the fadvise() interface
> is not tmpfs specific, and given that everyone is talking about
> volatility of page cache pages, I fail to see what is tmpfs specific
> about this proposal.
> So what are the semantics that are supposed to apply to a file that
> is on a filesystem with stable storage that is cached in the page
> cache?

Indeed, this is probably the most awkward case. So currently, we use
vmtruncate_range, which should punch a hole in the file. If I switch to
invalidate_inode_pages2_range(), then I think dirty data is dropped and
the backed page remains (I'm currently reading over that now).

> If this is tmpfs specific behaviour that is required, then IMO
> fadvise is not the correct interface to use here because fadvise is
> supposed to be a generic interface to controlling the page cache
> behaviour on any given file....
> > As a counter-point, this is my first thought of an implementation approach
> > (-ENOPATCH, sorry)
> >
> > - new mount option for tmpfs e.g. 'volatile'. Any file in a filesystem
> > mounted with that option and which is not currently open by any process can
> > have blocks removed at any time. The file name must remain, and the file
> > size must not change.
> > - lseek can be used to determine if anything has been purged with 'SEEK_DATA'
> > and 'SEEK_HOLE'.
> >
> > So you can only mark volatility on a whole-file granularity (hence the
> > question above).
> > 'open' says "NONVOLATILE".
> > 'close' says "VOLATILE".
> > 'lseek' is used to check if anything was discarded.
> >
> > This approach would allow multiple processes to share a cache (might this be
> > valuable?) as it doesn't become truly volatile until all processes close
> > their handles.
> If this functionality is only useful for tmpfs, then I'd much prefer
> a tmpfs specific approach like this....

Since, as I think more on this, this seems to map closer to file hole
punching, would fallocate be the right interface? FALLOC_FL_PUNCH_HOLE
isn't supported by all filesystems, after all.



 \ /
  Last update: 2012-02-17 06:41    [W:0.096 / U:1.052 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site