>
> On 8/8/19 4:41 PM, Ira Weiny wrote:
> > On Thu, Aug 08, 2019 at 03:59:15PM -0700, John Hubbard wrote:
> >> On 8/8/19 12:20 PM, John Hubbard wrote:
> >>> On 8/8/19 4:09 AM, Vlastimil Babka wrote:
> >>>> On 8/8/19 8:21 AM, Michal Hocko wrote:
> >>>>> On Wed 07-08-19 16:32:08, John Hubbard wrote:
> >>>>>> On 8/7/19 4:01 AM, Michal Hocko wrote:
> >>>>>>> On Mon 05-08-19 15:20:17, john.hubbard@xxxxxxxxx wrote:
> ...
> >> Oh, and meanwhile, I'm leaning toward a cheap fix: just use
> >> gup_fast() instead of get_page(), and also fix the releasing code. So
> >> this incremental patch, on top of the existing one, should do it:
> >>
> >> diff --git a/mm/mlock.c b/mm/mlock.c
> >> index b980e6270e8a..2ea272c6fee3 100644
> >> --- a/mm/mlock.c
> >> +++ b/mm/mlock.c
> >> @@ -318,18 +318,14 @@ static void __munlock_pagevec(struct pagevec
> *pvec, struct zone *zone)
> >> /*
> >> * We won't be munlocking this page in the next phase
> >> * but we still need to release the follow_page_mask()
> >> - * pin. We cannot do it under lru_lock however. If it's
> >> - * the last pin, __page_cache_release() would deadlock.
> >> + * pin.
> >> */
> >> - pagevec_add(&pvec_putback, pvec->pages[i]);
> >> + put_user_page(pages[i]);
>
> correction, make that:
> put_user_page(pvec->pages[i]);
>
> (This is not fully tested yet.)
>
> >> pvec->pages[i] = NULL;
> >> }
> >> __mod_zone_page_state(zone, NR_MLOCK, delta_munlocked);
> >> spin_unlock_irq(&zone->zone_pgdat->lru_lock);
> >>
> >> - /* Now we can release pins of pages that we are not munlocking */
> >> - pagevec_release(&pvec_putback);
> >> -
> >
> > I'm not an expert but this skips a call to lru_add_drain(). Is that ok?
>
> Yes: unless I'm missing something, there is no reason to go through
> lru_add_drain in this case. These are gup'd pages that are not going to get
> any further processing.
>
> >
> >> /* Phase 2: page munlock */
> >> for (i = 0; i < nr; i++) {
> >> struct page *page = pvec->pages[i]; @@ -394,6 +390,8
> >> @@ static unsigned long __munlock_pagevec_fill(struct pagevec *pvec,
> >> start += PAGE_SIZE;
> >> while (start < end) {
> >> struct page *page = NULL;
> >> + int ret;
> >> +
> >> pte++;
> >> if (pte_present(*pte))
> >> page = vm_normal_page(vma, start, *pte); @@
> >> -411,7 +409,13 @@ static unsigned long __munlock_pagevec_fill(struct
> pagevec *pvec,
> >> if (PageTransCompound(page))
> >> break;
> >>
> >> - get_page(page);
> >> + /*
> >> + * Use get_user_pages_fast(), instead of get_page() so that the
> >> + * releasing code can unconditionally call put_user_page().
> >> + */
> >> + ret = get_user_pages_fast(start, 1, 0, &page);
> >> + if (ret != 1)
> >> + break;
> >
> > I like the idea of making this a get/put pair but I'm feeling uneasy
> > about how this is really supposed to work.
> >
> > For sure the GUP/PUP was supposed to be separate from [get|put]_page.
> >
>
> Actually, they both take references on the page. And it is absolutely OK to call
> them both on the same page.
>
> But anyway, we're not mixing them up here. If you follow the code paths,
> either gup or follow_page_mask() is used, and then put_user_page()
> releases.
>
> So...you haven't actually pointed to a bug here, right? :)
No... no bug.
sorry this was just a general comment on semantics. But in keeping with the semantics discussion it is further confusing that follow_page_mask() is also mixed in here...
Which is where my comment was driving toward. If you call GUP there should be a PUP. Get_page/put_page... follow_page/unfollow_page... ??? ;-) Ok now I'm off the rails... but that was the point...
I think Jan and Michal are onto something here WRT internal vs external interfaces.
Ira
