On Wed, Jun 23, 2021 at 07:51:18PM +0100, Matthew Wilcox wrote:
> But truncate_inode_pages_range() is also called by, for example,
> truncate(). In that case, nobody calls filemap_write_and_wait_range(),
> so we can't discard the page because it might still be dirty.
> Is that an acceptable way to choose behaviour -- if the split fails,
> discard the page if it's clean and keep it if it's dirty? I'll
> put a great big comment on it, because it's not entirely obvious.
It seems to be working ...
if (!folio_multi(folio))
return true;
// if (split_huge_page(&folio->page) == 0)
// return true;
if (folio_dirty(folio))
return false;
truncate_inode_folio(mapping, folio);
return true;
No additional xfstests failures from commenting out those two lines.
