On Sun, 28 Apr 2019 23:05:38 +0200 Michal Hocko <mhocko@xxxxxxxx> wrote:
> On Fri 26-04-19 11:25:42, Andrew Morton wrote:
> > On Fri, 26 Apr 2019 18:26:42 +0800 Yafang Shao <laoar.shao@xxxxxxxxx> wrote:
> [...]
> > > +/*
> > > + * Wait for a page to complete writeback
> > > + */
> > > +void wait_on_page_writeback(struct page *page)
> > > +{
> > > + if (PageWriteback(page)) {
> > > + trace_wait_on_page_writeback(page, page_mapping(page));
> > > + wait_on_page_bit(page, PG_writeback);
> > > + }
> > > +}
> > > +EXPORT_SYMBOL_GPL(wait_on_page_writeback);
> >
> > But this is a stealth change to the wait_on_page_writeback() licensing.
>
> Why do we have to put that out of line in the first place?
Seems like a good thing to do from a size and icache-footprint POV.
wait_on_page_writeback() has a ton of callsites and the allmodconfig
out-of-line version is around 600 bytes of code (gack).
> Btw. wait_on_page_bit is EXPORT_SYMBOL...
OK, I'll leave wait_on_page_writeback() as EXPROT_SYMBOL().
