On Tue, Nov 12, 2024 at 9:32 AM Joanne Koong <joannelkoong@xxxxxxxxx> wrote:
>
> On Fri, Nov 8, 2024 at 4:13 PM Joanne Koong <joannelkoong@xxxxxxxxx> wrote:
> >
> > Add support for folios larger than one page size for non-writeback
>
> I'll change this naming from "non-writeback" to "writethrough"
>
> > writes.
> >
> > Signed-off-by: Joanne Koong <joannelkoong@xxxxxxxxx>
> > ---
> > fs/fuse/file.c | 29 ++++++++++++++++++-----------
> > 1 file changed, 18 insertions(+), 11 deletions(-)
> >
> > diff --git a/fs/fuse/file.c b/fs/fuse/file.c
> > index a89fdc55a40b..6ee23ab9b7f2 100644
> > --- a/fs/fuse/file.c
> > +++ b/fs/fuse/file.c
> > @@ -1146,19 +1146,15 @@ static ssize_t fuse_fill_write_pages(struct fuse_io_args *ia,
> > num = min(num, max_pages << PAGE_SHIFT);
> >
> > ap->args.in_pages = true;
> > - ap->descs[0].offset = offset;
> >
> > while (num) {
> > size_t tmp;
> > struct folio *folio;
> > pgoff_t index = pos >> PAGE_SHIFT;
> > - unsigned int bytes = min(PAGE_SIZE - offset, num);
> > -
> > - again:
> > - err = -EFAULT;
> > - if (fault_in_iov_iter_readable(ii, bytes))
> > - break;
> > + unsigned int bytes;
> > + unsigned int folio_offset;
> >
> > + again:
> > folio = __filemap_get_folio(mapping, index, FGP_WRITEBEGIN,
> > mapping_gfp_mask(mapping));
>
> This __filemap_get_folio() call (and the one in fuse_write_begin() as
> well) needs to also set the order on the fgf flag to enable large
> folios, else all folios returned will be order 0.
>
> I'll fix this in v2.
Ran some benchmarks and trying to get the largest folios possible from
__filemap_get_folio() is an over-optimization and slows down writes
significantly. I'll leave this as is for v2, and we could look into
optimizing this in the future.
>
> > if (IS_ERR(folio)) {
> > @@ -1166,10 +1162,20 @@ static ssize_t fuse_fill_write_pages(struct fuse_io_args *ia,
> > break;
> > }
> >
> > + folio_offset = ((index - folio->index) << PAGE_SHIFT) + offset;
> > + bytes = min(folio_size(folio) - folio_offset, num);
> > +
> > + err = -EFAULT;
> > + if (fault_in_iov_iter_readable(ii, bytes)) {
> > + folio_unlock(folio);
> > + folio_put(folio);
> > + break;
> > + }
> > +
> > if (mapping_writably_mapped(mapping))
> > flush_dcache_folio(folio);
> >
> > - tmp = copy_folio_from_iter_atomic(folio, offset, bytes, ii);
> > + tmp = copy_folio_from_iter_atomic(folio, folio_offset, bytes, ii);
> > flush_dcache_folio(folio);
> >
> > if (!tmp) {
> > @@ -1180,6 +1186,7 @@ static ssize_t fuse_fill_write_pages(struct fuse_io_args *ia,
> >
> > err = 0;
> > ap->folios[ap->num_folios] = folio;
> > + ap->descs[ap->num_folios].offset = folio_offset;
> > ap->descs[ap->num_folios].length = tmp;
> > ap->num_folios++;
> >
> > @@ -1187,11 +1194,11 @@ static ssize_t fuse_fill_write_pages(struct fuse_io_args *ia,
> > pos += tmp;
> > num -= tmp;
> > offset += tmp;
> > - if (offset == PAGE_SIZE)
> > + if (offset == folio_size(folio))
> > offset = 0;
> >
> > - /* If we copied full page, mark it uptodate */
> > - if (tmp == PAGE_SIZE)
> > + /* If we copied full folio, mark it uptodate */
> > + if (tmp == folio_size(folio))
> > folio_mark_uptodate(folio);
> >
> > if (folio_test_uptodate(folio)) {
> > --
> > 2.43.5
> >
