On Thu, Feb 8, 2024 at 2:05 PM Kairui Song <ryncsn@xxxxxxxxx> wrote:
>
> On Thu, Feb 8, 2024 at 2:31 AM Minchan Kim <minchan@xxxxxxxxxx> wrote:
> >
> > On Wed, Feb 07, 2024 at 12:06:15PM +0800, Kairui Song wrote:
> > > On Wed, Feb 7, 2024 at 12:02 PM Chris Li <chrisl@xxxxxxxxxx> wrote:
> > > >
> > > > On Tue, Feb 6, 2024 at 6:21 PM Kairui Song <ryncsn@xxxxxxxxx> wrote:
> > > > >
> > > > > On Wed, Feb 7, 2024 at 10:03 AM Chris Li <chrisl@xxxxxxxxxx> wrote:
> > > > > >
> > > > > > On Tue, Feb 6, 2024 at 4:43 PM Barry Song <21cnbao@xxxxxxxxx> wrote:
> > > > > > >
> > > > > > > On Wed, Feb 7, 2024 at 7:18 AM Chris Li <chrisl@xxxxxxxxxx> wrote:
> > > > > > > >
> > > > > > > > Hi Kairui,
> > > > > > > >
> > > > > > > > Sorry replying to your patch V1 late, I will reply on the V2 thread.
> > > > > > > >
> > > > > > > > On Tue, Feb 6, 2024 at 10:28 AM Kairui Song <ryncsn@xxxxxxxxx> wrote:
> > > > > > > > >
> > > > > > > > > From: Kairui Song <kasong@xxxxxxxxxxx>
> > > > > > > > >
> > > > > > > > > When skipping swapcache for SWP_SYNCHRONOUS_IO, if two or more threads
> > > > > > > > > swapin the same entry at the same time, they get different pages (A, B).
> > > > > > > > > Before one thread (T0) finishes the swapin and installs page (A)
> > > > > > > > > to the PTE, another thread (T1) could finish swapin of page (B),
> > > > > > > > > swap_free the entry, then swap out the possibly modified page
> > > > > > > > > reusing the same entry. It breaks the pte_same check in (T0) because
> > > > > > > > > PTE value is unchanged, causing ABA problem. Thread (T0) will
> > > > > > > > > install a stalled page (A) into the PTE and cause data corruption.
> > > > > > > > >
> > > > > > > > > One possible callstack is like this:
> > > > > > > > >
> > > > > > > > > CPU0 CPU1
> > > > > > > > > ---- ----
> > > > > > > > > do_swap_page() do_swap_page() with same entry
> > > > > > > > > <direct swapin path> <direct swapin path>
> > > > > > > > > <alloc page A> <alloc page B>
> > > > > > > > > swap_read_folio() <- read to page A swap_read_folio() <- read to page B
> > > > > > > > > <slow on later locks or interrupt> <finished swapin first>
> > > > > > > > > ... set_pte_at()
> > > > > > > > > swap_free() <- entry is free
> > > > > > > > > <write to page B, now page A stalled>
> > > > > > > > > <swap out page B to same swap entry>
> > > > > > > > > pte_same() <- Check pass, PTE seems
> > > > > > > > > unchanged, but page A
> > > > > > > > > is stalled!
> > > > > > > > > swap_free() <- page B content lost!
> > > > > > > > > set_pte_at() <- staled page A installed!
> > > > > > > > >
> > > > > > > > > And besides, for ZRAM, swap_free() allows the swap device to discard
> > > > > > > > > the entry content, so even if page (B) is not modified, if
> > > > > > > > > swap_read_folio() on CPU0 happens later than swap_free() on CPU1,
> > > > > > > > > it may also cause data loss.
> > > > > > > > >
> > > > > > > > > To fix this, reuse swapcache_prepare which will pin the swap entry using
> > > > > > > > > the cache flag, and allow only one thread to pin it. Release the pin
> > > > > > > > > after PT unlocked. Racers will simply busy wait since it's a rare
> > > > > > > > > and very short event.
> > > > > > > > >
> > > > > > > > > Other methods like increasing the swap count don't seem to be a good
> > > > > > > > > idea after some tests, that will cause racers to fall back to use the
> > > > > > > > > swap cache again. Parallel swapin using different methods leads to
> > > > > > > > > a much more complex scenario.
> > > > > > > > >
> > > > > > > > > Reproducer:
> > > > > > > > >
> > > > > > > > > This race issue can be triggered easily using a well constructed
> > > > > > > > > reproducer and patched brd (with a delay in read path) [1]:
> > > > > > > > >
> > > > > > > > > With latest 6.8 mainline, race caused data loss can be observed easily:
> > > > > > > > > $ gcc -g -lpthread test-thread-swap-race.c && ./a.out
> > > > > > > > > Polulating 32MB of memory region...
> > > > > > > > > Keep swapping out...
> > > > > > > > > Starting round 0...
> > > > > > > > > Spawning 65536 workers...
> > > > > > > > > 32746 workers spawned, wait for done...
> > > > > > > > > Round 0: Error on 0x5aa00, expected 32746, got 32743, 3 data loss!
> > > > > > > > > Round 0: Error on 0x395200, expected 32746, got 32743, 3 data loss!
> > > > > > > > > Round 0: Error on 0x3fd000, expected 32746, got 32737, 9 data loss!
> > > > > > > > > Round 0 Failed, 15 data loss!
> > > > > > > > >
> > > > > > > > > This reproducer spawns multiple threads sharing the same memory region
> > > > > > > > > using a small swap device. Every two threads updates mapped pages one by
> > > > > > > > > one in opposite direction trying to create a race, with one dedicated
> > > > > > > > > thread keep swapping out the data out using madvise.
> > > > > > > > >
> > > > > > > > > The reproducer created a reproduce rate of about once every 5 minutes,
> > > > > > > > > so the race should be totally possible in production.
> > > > > > > > >
> > > > > > > > > After this patch, I ran the reproducer for over a few hundred rounds
> > > > > > > > > and no data loss observed.
> > > > > > > > >
> > > > > > > > > Performance overhead is minimal, microbenchmark swapin 10G from 32G
> > > > > > > > > zram:
> > > > > > > > >
> > > > > > > > > Before: 10934698 us
> > > > > > > > > After: 11157121 us
> > > > > > > > > Non-direct: 13155355 us (Dropping SWP_SYNCHRONOUS_IO flag)
> > > > > > > > >
> > > > > > > > > Fixes: 0bcac06f27d7 ("mm, swap: skip swapcache for swapin of synchronous device")
> > > > > > > > > Reported-by: "Huang, Ying" <ying.huang@xxxxxxxxx>
> > > > > > > > > Closes: https://lore.kernel.org/lkml/87bk92gqpx.fsf_-_@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/
> > > > > > > > > Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> > > > > > > > > Signed-off-by: Kairui Song <kasong@xxxxxxxxxxx>
> > > > > > > > > Reviewed-by: "Huang, Ying" <ying.huang@xxxxxxxxx>
> > > > > > > > > Acked-by: Yu Zhao <yuzhao@xxxxxxxxxx>
> > > > > > > > >
> > > > > > > > > ---
> > > > > > > > > Update from V1:
> > > > > > > > > - Add some words on ZRAM case, it will discard swap content on swap_free so the race window is a bit different but cure is the same. [Barry Song]
> > > > > > > > > - Update comments make it cleaner [Huang, Ying]
> > > > > > > > > - Add a function place holder to fix CONFIG_SWAP=n built [SeongJae Park]
> > > > > > > > > - Update the commit message and summary, refer to SWP_SYNCHRONOUS_IO instead of "direct swapin path" [Yu Zhao]
> > > > > > > > > - Update commit message.
> > > > > > > > > - Collect Review and Acks.
> > > > > > > > >
> > > > > > > > > include/linux/swap.h | 5 +++++
> > > > > > > > > mm/memory.c | 15 +++++++++++++++
> > > > > > > > > mm/swap.h | 5 +++++
> > > > > > > > > mm/swapfile.c | 13 +++++++++++++
> > > > > > > > > 4 files changed, 38 insertions(+)
> > > > > > > > >
> > > > > > > > > diff --git a/include/linux/swap.h b/include/linux/swap.h
> > > > > > > > > index 4db00ddad261..8d28f6091a32 100644
> > > > > > > > > --- a/include/linux/swap.h
> > > > > > > > > +++ b/include/linux/swap.h
> > > > > > > > > @@ -549,6 +549,11 @@ static inline int swap_duplicate(swp_entry_t swp)
> > > > > > > > > return 0;
> > > > > > > > > }
> > > > > > > > >
> > > > > > > > > +static inline int swapcache_prepare(swp_entry_t swp)
> > > > > > > > > +{
> > > > > > > > > + return 0;
> > > > > > > > > +}
> > > > > > > > > +
> > > > > > > > > static inline void swap_free(swp_entry_t swp)
> > > > > > > > > {
> > > > > > > > > }
> > > > > > > > > diff --git a/mm/memory.c b/mm/memory.c
> > > > > > > > > index 7e1f4849463a..1749c700823d 100644
> > > > > > > > > --- a/mm/memory.c
> > > > > > > > > +++ b/mm/memory.c
> > > > > > > > > @@ -3867,6 +3867,16 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> > > > > > > > > if (!folio) {
> > > > > > > > > if (data_race(si->flags & SWP_SYNCHRONOUS_IO) &&
> > > > > > > > > __swap_count(entry) == 1) {
> > > > > > > > > + /*
> > > > > > > > > + * Prevent parallel swapin from proceeding with
> > > > > > > > > + * the cache flag. Otherwise, another thread may
> > > > > > > > > + * finish swapin first, free the entry, and swapout
> > > > > > > > > + * reusing the same entry. It's undetectable as
> > > > > > > > > + * pte_same() returns true due to entry reuse.
> > > > > > > > > + */
> > > > > > > > > + if (swapcache_prepare(entry))
> > > > > > > > > + goto out;
> > > > > > > > > +
> > > > > > > >
> > > > > > > > I am puzzled by this "goto out". If I understand this correctly, you
> > > > > > > > have two threads CPU1 and CPU2 racing to set the flag SWAP_HAS_CACHE.
> > > > > > > > The CPU1 will succeed in adding the flag and the CPU2 will get
> > > > > > > > "-EEXIST" from "swapcache_prepare(entry)". Am I understanding it
> > > > > > > > correctly so far?
> > > > > > > >
> > > > > > > > Then the goto out seems wrong to me. For the CPU2, the page fault will
> > > > > > > > return *unhandled*. Even worse, the "-EEXIST" error is not preserved,
> > > > > > > > CPU2 does not even know the page fault is not handled, it will resume
> > > > > > > > from the page fault instruction, possibly generate another page fault
> > > > > > > > at the exact same location. That page fault loop will repeat until
> > > > > > > > CPU1 install the new pte on that faulting virtual address and pick up
> > > > > > > > by CPU2.
> > > > > > > >
> > > > > > > > Am I missing something obvious there?
> > > > > > >
> > > > > > > I feel you are right. any concurrent page faults at the same pte
> > > > > > > will increase the count of page faults for a couple of times now.
> > > > > > >
> > > > > > > >
> > > > > > > > I just re-read your comment: "Racers will simply busy wait since it's
> > > > > > > > a rare and very short event." That might be referring to the above
> > > > > > > > CPU2 page fault looping situation. I consider the page fault looping
> > > > > > > > on CPU2 not acceptable. For one it will mess up the page fault
> > > > > > > > statistics.
> > > > > > > > In my mind, having an explicit loop for CPU2 waiting for the PTE to
> > > > > > > > show up is still better than this page fault loop. You can have more
> > > > > > > > CPU power friendly loops.
> > > > > > >
> > > > > > > I assume you mean something like
> > > > > > >
> > > > > > > while(!pte_same())
> > > > > > > cpu_relax();
> > > > > > >
> > > > > > > then we still have a chance to miss the change of B.
> > > > > > >
> > > > > > > For example, another thread is changing pte to A->B->A, our loop can
> > > > > > > miss B. Thus we will trap into an infinite loop. this is even worse.
> > > > > >
> > > > > > Yes. You are right, it is worse. Thanks for catching that. That is why
> > > > > > I say this needs more discussion, I haven't fully thought it through
> > > > > > :-)
> > > > >
> > > > > Hi Chris and Barry,
> > > > >
> > > > > Thanks for the comments!
> > > > >
> > > > > The worst thing I know of returning in do_swap_page without handling
> > > > > the swap, is an increase of some statistic counters, note it will not
> > > > > cause major page fault counters to grow, only things like perf counter
> > > > > and vma lock statistic are affected.
> > > > >
> > > > > And actually there are multiple already existing return points in
> > > > > do_swap_page that will return without handling it, which may
> > > > > re-trigger the page fault.
> > > >
> > > > Thanks for pointing that out. I take a look at those, which seems
> > > > different than the case here. In those cases, it truely can not make
> > > > forward progress.
> > > > Here we actually have all the data it needs to complete the page
> > > > fault. Just a data synchronization issue preventing making forward
> > > > progress.
> > > > Ideally we can have some clever data structure to solve the
> > > > synchronization issue and make forward progress.
> > > >
> > > > > When do_swap_page is called, many pre-checks have been applied, and
> > > > > they could all be invalidated if something raced, simply looping
> > > > > inside here could miss a lot of corner cases, so we have to go through
> > > > > that again.
> > > >
> > > > Actually, I think about it. Looping it here seems worse in the sense
> > > > that it is already holding some locks. Return and retry the page fault
> > > > at least release those locks and let others have a chance to make
> > > > progress.
> > > >
> > > > >
> > > > > This patch did increase the chance of false positive increase of some
> > > > > counters, maybe something like returning a VM_FAULT_RETRY could make
> > > > > it better, but code is more complex and will cause other counters to
> > > > > grow.
> > > >
> > > > This is certainly not ideal. It might upset the feedback loop that
> > > > uses the swap fault statistic as input to adjust the swapping
> > > > behavior.
> > > >
> > > > Chris
> > >
> > > Hi Chris,
> > >
> > > Thanks for the reply.
> > >
> > > So I think the thing is, it's getting complex because this patch
> > > wanted to make it simple and just reuse the swap cache flags.
> >
> > I agree that a simple fix would be the important at this point.
> >
> > Considering your description, here's my understanding of the other idea:
> > Other method, such as increasing the swap count, haven't proven effective
> > in your tests. The approach risk forcing racers to rely on the swap cache
> > again and the potential performance loss in race scenario.
> >
> > While I understand that simplicity is important, and performance loss
> > in this case may be infrequent, I believe swap_count approach could be a
> > suitable solution. What do you think?
>
> Hi Minchan
>
> Yes, my main concern was about simplicity and performance.
>
> Increasing swap_count here will also race with another process from
> releasing swap_count to 0 (swapcache was able to sync callers in other
> call paths but we skipped swapcache here).
Just like a refcount, releasing to 0 should be ok. the last one who sets
refcount to 0 will finally release the swap entry.
Increasing swap_count from 1 to 2 might cause concurrent threads
go into non-sync-io path(full swapcache things), which was Minchan's
original patch wanted to avoid. i am not sure if this will negatively affect
performance. If not, it seems a feasible way.
> So the right step is: 1. Lock the cluster/swap lock; 2. Check if still
> have swap_count == 1, bail out if not; 3. Set it to 2;
> __swap_duplicate can be modified to support this, it's similar to
> existing logics for SWAP_HAS_CACHE.
>
> And swap freeing path will do more things, swapcache clean up needs to
> be handled even in the bypassing path since the racer may add it to
> swapcache.
>
> Reusing SWAP_HAS_CACHE seems to make it much simpler and avoided many
> overhead, so I used that way in this patch, the only issue is
> potentially repeated page faults now.
>
> I'm currently trying to add a SWAP_MAP_LOCK (or SWAP_MAP_SYNC, I'm bad
> at naming it) special value, so any racer can just spin on it to avoid
> all the problems, how do you think about this?
Thanks
Barry
