On Wed, Sep 3, 2025 at 2:17 PM Darrick J. Wong <djwong@xxxxxxxxxx> wrote:
>
> On Fri, Aug 29, 2025 at 04:56:26PM -0700, Joanne Koong wrote:
> > Do readahead in fuse using iomap. This gives us granular uptodate
> > tracking for large folios, which optimizes how much data needs to be
> > read in. If some portions of the folio are already uptodate (eg through
> > a prior write), we only need to read in the non-uptodate portions.
> >
> > Signed-off-by: Joanne Koong <joannelkoong@xxxxxxxxx>
> > ---
> > fs/fuse/file.c | 214 +++++++++++++++++++++++++++----------------------
> > 1 file changed, 118 insertions(+), 96 deletions(-)
> >
> > diff --git a/fs/fuse/file.c b/fs/fuse/file.c
> > index bdfb13cdee4b..1659603f4cb6 100644
> > --- a/fs/fuse/file.c
> > +++ b/fs/fuse/file.c
> > @@ -2176,10 +2187,21 @@ static ssize_t fuse_iomap_writeback_range(struct iomap_writepage_ctx *wpc,
> > return -EIO;
> > }
> >
> > - if (wpa && fuse_writepage_need_send(fc, pos, len, ap, data)) {
> > - fuse_writepages_send(inode, data);
> > - data->wpa = NULL;
> > - data->nr_bytes = 0;
> > + if (wpa) {
> > + bool send = fuse_folios_need_send(fc, pos, len, ap, data->nr_bytes,
> > + true);
> > +
> > + if (!send) {
> > + /* Need to grow the pages array? If so, did the expansion fail? */
> > + send = (ap->num_folios == data->max_folios) &&
> > + !fuse_pages_realloc(data, fc->max_pages);
> > + }
>
> What purpose this code relocation serve? I gather the idea here is that
> writes need to reallocate the pages array, whereas readahead can simply
> constrain to whatever's already allocated?
>
I think it's more that for readahead there's more guiding info about
what size array is needed (eg looking at readahead_count(rac)) whereas
that's more uncertain for writeback, especially since the folios can
only be part of the same request if they're contiguous. The writeback
array gets reallocated exponentially up on a per-need basis starting
at 1. imo it seems better to start the writeback array at 2 or 4
folios but I also don't really have empirical data to back this theory
up.
Thanks,
Joanne
> --D
>
