Re: [PATCH v2 24/46] mm/writeback: Add __folio_end_writeback()

[Matthew Wilcox]

On Wed, Jun 23, 2021 at 11:15:55AM +0200, Christoph Hellwig wrote:
> On Tue, Jun 22, 2021 at 01:15:29PM +0100, Matthew Wilcox (Oracle) wrote:
> > test_clear_page_writeback() is actually an mm-internal function, although
> > it's named as if it's a pagecache function.  Move it to mm/internal.h,
> > rename it to __folio_end_writeback() and change the return type to bool.
> > 
> > The conversion from page to folio is mostly about accounting the number
> > of pages being written back, although it does eliminate a couple of
> > calls to compound_head().
> 
> While this looks good, I think the whole abstraction is wrong.  I think
> test_clear_page_writeback should just be merged into it's only caller.

I'm not opposed to doing that, but something else has to get
un-static'ed in order to make that happen.

folio_end_writeback (exported, filemap.c)
 -> folio_wake (static, filemap.c)
     -> folio_wake_bit (static, filemap.c)
 -> __folio_end_writeback (non-static, page-writeback.c)
     -> __wb_writeout_add (static, page-writeback.c)

I'm not sure there's an obviously better split than where it is right
now.

> But if that is somehow not on the table this change looks ok:
> 
> Reviewed-by: Christoph Hellwig <hch@xxxxxx>