Re: always fall back to buffered I/O after invalidation failures, was: Re: [PATCH 2/6] iomap: IOMAP_DIO_RWF_NO_STALE_PAGECACHE return if page invalidation fails

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Jul 07, 2020 at 08:49:52AM -0500, Goldwyn Rodrigues wrote:
> On 13:57 07/07, Matthew Wilcox wrote:
> > On Tue, Jul 07, 2020 at 07:43:46AM -0500, Goldwyn Rodrigues wrote:
> > > On  9:53 01/07, Christoph Hellwig wrote:
> > > > On Mon, Jun 29, 2020 at 02:23:49PM -0500, Goldwyn Rodrigues wrote:
> > > > > From: Goldwyn Rodrigues <rgoldwyn@xxxxxxxx>
> > > > > 
> > > > > For direct I/O, add the flag IOMAP_DIO_RWF_NO_STALE_PAGECACHE to indicate
> > > > > that if the page invalidation fails, return back control to the
> > > > > filesystem so it may fallback to buffered mode.
> > > > > 
> > > > > Reviewed-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
> > > > > Signed-off-by: Goldwyn Rodrigues <rgoldwyn@xxxxxxxx>
> > > > 
> > > > I'd like to start a discussion of this shouldn't really be the
> > > > default behavior.  If we have page cache that can't be invalidated it
> > > > actually makes a whole lot of sense to not do direct I/O, avoid the
> > > > warnings, etc.
> > > > 
> > > > Adding all the relevant lists.
> > > 
> > > Since no one responded so far, let me see if I can stir the cauldron :)
> > > 
> > > What error should be returned in case of such an error? I think the
> > 
> > Christoph's message is ambiguous.  I don't know if he means "fail the
> > I/O with an error" or "satisfy the I/O through the page cache".  I'm
> > strongly in favour of the latter.  Indeed, I'm in favour of not invalidating
> > the page cache at all for direct I/O.  For reads, I think the page cache
> > should be used to satisfy any portion of the read which is currently
> 
> That indeed would make reads faster. How about if the pages are dirty
> during DIO reads?
> Should a direct I/O read be responsible for making sure that the dirty
> pages are written back. Technically direct I/O reads is that we are
> reading from the device.

The filemap_write_and_wait_range should persist that data, right?

> > cached.  For writes, I think we should write into the page cache pages
> > which currently exist, and then force those pages to be written back,
> > but left in cache.
> 
> Yes, that makes sense.
> If this is implemented, what would be the difference between O_DIRECT
> and O_DSYNC, if any?

Presumably a direct write would proceed as it does today if there's no
pagecache at all?

--D

> -- 
> Goldwyn



[Index of Archives]     [XFS Filesystem Development (older mail)]     [Linux Filesystem Development]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux RAID]     [Linux SCSI]


  Powered by Linux