[v3] mm/swap: fix race when skipping swapcache

Commit Message

Kairui Song Feb. 16, 2024, 9:51 a.m. UTC

From: Kairui Song <[e-mail]>

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 wait since it's a rare and very
short event. A schedule() call is added to avoid wasting too much CPU
or adding too much noise to perf statistics

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")
Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
Reported-by: "Huang, Ying" <[e-mail]>
Closes: https://lore.kernel.org/lkml/[e-mail]/
Signed-off-by: Kairui Song <[e-mail]>
Cc: [e-mail]

---
Update from V2:
- Add a schedule() if raced to prevent repeated page faults wasting CPU
  and add noise to perf statistics.
- Use a bool to state the special case instead of reusing existing
  variables fixing error handling [Minchan Kim].

V2: https://lore.kernel.org/all/[e-mail]/

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.

V1: https://lore.kernel.org/all/[e-mail]/

 include/linux/swap.h |  5 +++++
 mm/memory.c          | 20 ++++++++++++++++++++
 mm/swap.h            |  5 +++++
 mm/swapfile.c        | 13 +++++++++++++
 4 files changed, 43 insertions(+)

Comments

Barry Song Feb. 16, 2024, 10:14 a.m. UTC | #1

On Fri, Feb 16, 2024 at 10:53 PM Kairui Song <[e-mail]> wrote:
>
> From: Kairui Song <[e-mail]>
>
> 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 wait since it's a rare and very
> short event. A schedule() call is added to avoid wasting too much CPU
> or adding too much noise to perf statistics
>
> 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")
> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> Reported-by: "Huang, Ying" <[e-mail]>
> Closes: https://lore.kernel.org/lkml/[e-mail]/
> Signed-off-by: Kairui Song <[e-mail]>
> Cc: [e-mail]
>
> ---
> Update from V2:
> - Add a schedule() if raced to prevent repeated page faults wasting CPU
>   and add noise to perf statistics.
> - Use a bool to state the special case instead of reusing existing
>   variables fixing error handling [Minchan Kim].
>
> V2: https://lore.kernel.org/all/[e-mail]/
>
> 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.
>
> V1: https://lore.kernel.org/all/[e-mail]/
>
>  include/linux/swap.h |  5 +++++
>  mm/memory.c          | 20 ++++++++++++++++++++
>  mm/swap.h            |  5 +++++
>  mm/swapfile.c        | 13 +++++++++++++
>  4 files changed, 43 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..7059230d0a54 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>         struct page *page;
>         struct swap_info_struct *si = NULL;
>         rmap_t rmap_flags = RMAP_NONE;
> +       bool need_clear_cache = false;
>         bool exclusive = false;
>         swp_entry_t entry;
>         pte_t pte;
> @@ -3867,6 +3868,20 @@ 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)) {
> +                               /* Relax a bit to prevent rapid repeated page faults */
> +                               schedule();
> +                               goto out;
> +                       }
> +                       need_clear_cache = true;

Hi Kairui,
i remember Ying had a suggestion to move swapcache_prepare after
swap_read_folio() to decrease race window. does that one even help
more to somehow "fix" the counting issue?


> +
>                         /* skip swapcache */
>                         folio = vma_alloc_folio(GFP_HIGHUSER_MOVABLE, 0,
>                                                 vma, vmf->address, false);
> @@ -4117,6 +4132,9 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>         if (vmf->pte)
>                 pte_unmap_unlock(vmf->pte, vmf->ptl);
>  out:
> +       /* Clear the swap cache pin for direct swapin after PTL unlock */
> +       if (need_clear_cache)
> +               swapcache_clear(si, entry);
>         if (si)
>                 put_swap_device(si);
>         return ret;
> @@ -4131,6 +4149,8 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>                 folio_unlock(swapcache);
>                 folio_put(swapcache);
>         }
> +       if (need_clear_cache)
> +               swapcache_clear(si, entry);
>         if (si)
>                 put_swap_device(si);
>         return ret;
> diff --git a/mm/swap.h b/mm/swap.h
> index 758c46ca671e..fc2f6ade7f80 100644
> --- a/mm/swap.h
> +++ b/mm/swap.h
> @@ -41,6 +41,7 @@ void __delete_from_swap_cache(struct folio *folio,
>  void delete_from_swap_cache(struct folio *folio);
>  void clear_shadow_from_swap_cache(int type, unsigned long begin,
>                                   unsigned long end);
> +void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry);
>  struct folio *swap_cache_get_folio(swp_entry_t entry,
>                 struct vm_area_struct *vma, unsigned long addr);
>  struct folio *filemap_get_incore_folio(struct address_space *mapping,
> @@ -97,6 +98,10 @@ static inline int swap_writepage(struct page *p, struct writeback_control *wbc)
>         return 0;
>  }
>
> +static inline void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry)
> +{
> +}
> +
>  static inline struct folio *swap_cache_get_folio(swp_entry_t entry,
>                 struct vm_area_struct *vma, unsigned long addr)
>  {
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 556ff7347d5f..746aa9da5302 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -3365,6 +3365,19 @@ int swapcache_prepare(swp_entry_t entry)
>         return __swap_duplicate(entry, SWAP_HAS_CACHE);
>  }
>
> +void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry)
> +{
> +       struct swap_cluster_info *ci;
> +       unsigned long offset = swp_offset(entry);
> +       unsigned char usage;
> +
> +       ci = lock_cluster_or_swap_info(si, offset);
> +       usage = __swap_entry_free_locked(si, offset, SWAP_HAS_CACHE);
> +       unlock_cluster_or_swap_info(si, ci);
> +       if (!usage)
> +               free_swap_slot(entry);
> +}
> +
>  struct swap_info_struct *swp_swap_info(swp_entry_t entry)
>  {
>         return swap_type_to_swap_info(swp_type(entry));
> --
> 2.43.0
>
>

Thanks
Barry

Kairui Song Feb. 16, 2024, 10:55 a.m. UTC | #2

On Fri, Feb 16, 2024 at 6:15 PM Barry Song <[e-mail]> wrote:
>
> On Fri, Feb 16, 2024 at 10:53 PM Kairui Song <[e-mail]> wrote:
> >
> > From: Kairui Song <[e-mail]>
> >
> > 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 wait since it's a rare and very
> > short event. A schedule() call is added to avoid wasting too much CPU
> > or adding too much noise to perf statistics
> >
> > 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")
> > Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> > Reported-by: "Huang, Ying" <[e-mail]>
> > Closes: https://lore.kernel.org/lkml/[e-mail]/
> > Signed-off-by: Kairui Song <[e-mail]>
> > Cc: [e-mail]
> >
> > ---
> > Update from V2:
> > - Add a schedule() if raced to prevent repeated page faults wasting CPU
> >   and add noise to perf statistics.
> > - Use a bool to state the special case instead of reusing existing
> >   variables fixing error handling [Minchan Kim].
> >
> > V2: https://lore.kernel.org/all/[e-mail]/
> >
> > 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.
> >
> > V1: https://lore.kernel.org/all/[e-mail]/
> >
> >  include/linux/swap.h |  5 +++++
> >  mm/memory.c          | 20 ++++++++++++++++++++
> >  mm/swap.h            |  5 +++++
> >  mm/swapfile.c        | 13 +++++++++++++
> >  4 files changed, 43 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..7059230d0a54 100644
> > --- a/mm/memory.c
> > +++ b/mm/memory.c
> > @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> >         struct page *page;
> >         struct swap_info_struct *si = NULL;
> >         rmap_t rmap_flags = RMAP_NONE;
> > +       bool need_clear_cache = false;
> >         bool exclusive = false;
> >         swp_entry_t entry;
> >         pte_t pte;
> > @@ -3867,6 +3868,20 @@ 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)) {
> > +                               /* Relax a bit to prevent rapid repeated page faults */
> > +                               schedule();
> > +                               goto out;
> > +                       }
> > +                       need_clear_cache = true;
>
> Hi Kairui,
> i remember Ying had a suggestion to move swapcache_prepare after
> swap_read_folio() to decrease race window. does that one even help
> more to somehow "fix" the counting issue?
>

Hi Barry,

Thanks for the comments!

Yes, that's one of the suggestions, I found the result already looks
good enough in test scenario I've posted after adding scheduler()
change. After moving swapcache_prepare actually waste more CPU as it
need to alloc/free extra folios now, and test result for the counting
issue barely changed, so I kept the original code.

David Hildenbrand Feb. 16, 2024, 4:53 p.m. UTC | #3

On 16.02.24 10:51, Kairui Song wrote:
> From: Kairui Song <[e-mail]>
> 
> 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 wait since it's a rare and very
> short event. A schedule() call is added to avoid wasting too much CPU
> or adding too much noise to perf statistics
> 
> 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")
> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> Reported-by: "Huang, Ying" <[e-mail]>
> Closes: https://lore.kernel.org/lkml/[e-mail]/
> Signed-off-by: Kairui Song <[e-mail]>
> Cc: [e-mail]
> 
> ---
> Update from V2:
> - Add a schedule() if raced to prevent repeated page faults wasting CPU
>    and add noise to perf statistics.
> - Use a bool to state the special case instead of reusing existing
>    variables fixing error handling [Minchan Kim].
> 
> V2: https://lore.kernel.org/all/[e-mail]/
> 
> 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.
> 
> V1: https://lore.kernel.org/all/[e-mail]/
> 
>   include/linux/swap.h |  5 +++++
>   mm/memory.c          | 20 ++++++++++++++++++++
>   mm/swap.h            |  5 +++++
>   mm/swapfile.c        | 13 +++++++++++++
>   4 files changed, 43 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..7059230d0a54 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>   	struct page *page;
>   	struct swap_info_struct *si = NULL;
>   	rmap_t rmap_flags = RMAP_NONE;
> +	bool need_clear_cache = false;
>   	bool exclusive = false;
>   	swp_entry_t entry;
>   	pte_t pte;
> @@ -3867,6 +3868,20 @@ 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)) {
> +				/* Relax a bit to prevent rapid repeated page faults */
> +				schedule();
> +				goto out;
> +			}
> +			need_clear_cache = true;
> +

I took a closer look at __read_swap_cache_async() and it essentially
does something similar.

Instead of returning, it keeps retrying until it finds that
swapcache_prepare() fails for another reason than -EEXISTS (e.g.,
freed concurrently) or it finds the entry in the swapcache.

So if you would succeed here on a freed+reused swap entry,
__read_swap_cache_async() would simply retry.

It spells that out:

		/*
		 * We might race against __delete_from_swap_cache(), and
		 * stumble across a swap_map entry whose SWAP_HAS_CACHE
		 * has not yet been cleared.  Or race against another
		 * __read_swap_cache_async(), which has set SWAP_HAS_CACHE
		 * in swap_map, but not yet added its folio to swap cache.
		 */

Whereby we could not race against this code here as well where we
speculatively set SWAP_HAS_CACHE and might never add something to the swap
cache.


I'd probably avoid the wrong returns and do something even closer to
__read_swap_cache_async().

while (true) {
	/*
	 * Fake that we are trying to insert a page into the swapcache, to
	 * serialize against concurrent threads wanting to do the same.
	 * [more from your description]
	 */
	ret = swapcache_prepare(entry);
	if (likely(!ret)
		/*
		 * Move forward with swapin, we'll recheck if the PTE hasn't
		 * changed later.
		 */
		break;
	else if (ret != -EEXIST)
		goto out;
	
	/*
          * See __read_swap_cache_async(). We might either have raced against
          * another thread, or the entry could have been freed and reused in the
	 * meantime. Make sure that the PTE did not change, to detect freeing.
	 */
	vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
				       vmf->address, &vmf->ptl);
	if (!vmf->pte || !pte_same(ptep_get(vmf->pte), vmf->orig_pte))
		goto unlock;


	schedule();
}



I was skeptical about the schedule(), but __read_swap_cache_async() does it
already because there is no better way to wait for the event to happen.

With something like above you would no longer depend on the speed of schedule() to
determine how often you would retry the fault, which would likely make sense.

I do wonder about the schedule() vs. schedule_timeout_uninterruptible(), though.
No expert on that area, do you have any idea?

David Hildenbrand Feb. 16, 2024, 6:01 p.m. UTC | #4

On 16.02.24 17:53, David Hildenbrand wrote:
> On 16.02.24 10:51, Kairui Song wrote:
>> From: Kairui Song <[e-mail]>
>>
>> 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 wait since it's a rare and very
>> short event. A schedule() call is added to avoid wasting too much CPU
>> or adding too much noise to perf statistics
>>
>> 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")
>> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
>> Reported-by: "Huang, Ying" <[e-mail]>
>> Closes: https://lore.kernel.org/lkml/[e-mail]/
>> Signed-off-by: Kairui Song <[e-mail]>
>> Cc: [e-mail]
>>
>> ---
>> Update from V2:
>> - Add a schedule() if raced to prevent repeated page faults wasting CPU
>>     and add noise to perf statistics.
>> - Use a bool to state the special case instead of reusing existing
>>     variables fixing error handling [Minchan Kim].
>>
>> V2: https://lore.kernel.org/all/[e-mail]/
>>
>> 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.
>>
>> V1: https://lore.kernel.org/all/[e-mail]/
>>
>>    include/linux/swap.h |  5 +++++
>>    mm/memory.c          | 20 ++++++++++++++++++++
>>    mm/swap.h            |  5 +++++
>>    mm/swapfile.c        | 13 +++++++++++++
>>    4 files changed, 43 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..7059230d0a54 100644
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>>    	struct page *page;
>>    	struct swap_info_struct *si = NULL;
>>    	rmap_t rmap_flags = RMAP_NONE;
>> +	bool need_clear_cache = false;
>>    	bool exclusive = false;
>>    	swp_entry_t entry;
>>    	pte_t pte;
>> @@ -3867,6 +3868,20 @@ 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)) {
>> +				/* Relax a bit to prevent rapid repeated page faults */
>> +				schedule();
>> +				goto out;
>> +			}
>> +			need_clear_cache = true;
>> +
> 
> I took a closer look at __read_swap_cache_async() and it essentially
> does something similar.
> 
> Instead of returning, it keeps retrying until it finds that
> swapcache_prepare() fails for another reason than -EEXISTS (e.g.,
> freed concurrently) or it finds the entry in the swapcache.
> 
> So if you would succeed here on a freed+reused swap entry,
> __read_swap_cache_async() would simply retry.
> 
> It spells that out:
> 
> 		/*
> 		 * We might race against __delete_from_swap_cache(), and
> 		 * stumble across a swap_map entry whose SWAP_HAS_CACHE
> 		 * has not yet been cleared.  Or race against another
> 		 * __read_swap_cache_async(), which has set SWAP_HAS_CACHE
> 		 * in swap_map, but not yet added its folio to swap cache.
> 		 */
> 
> Whereby we could not race against this code here as well where we
> speculatively set SWAP_HAS_CACHE and might never add something to the swap
> cache.
> 
> 
> I'd probably avoid the wrong returns and do something even closer to
> __read_swap_cache_async().
> 
> while (true) {
> 	/*
> 	 * Fake that we are trying to insert a page into the swapcache, to
> 	 * serialize against concurrent threads wanting to do the same.
> 	 * [more from your description]
> 	 */
> 	ret = swapcache_prepare(entry);
> 	if (likely(!ret)
> 		/*
> 		 * Move forward with swapin, we'll recheck if the PTE hasn't
> 		 * changed later.
> 		 */
> 		break;
> 	else if (ret != -EEXIST)
> 		goto out;
> 	
> 	/*
>            * See __read_swap_cache_async(). We might either have raced against
>            * another thread, or the entry could have been freed and reused in the
> 	 * meantime. Make sure that the PTE did not change, to detect freeing.
> 	 */
> 	vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
> 				       vmf->address, &vmf->ptl);
> 	if (!vmf->pte || !pte_same(ptep_get(vmf->pte), vmf->orig_pte))
> 		goto unlock;
> 
> 
> 	schedule();
> }
> 
> 
> 
> I was skeptical about the schedule(), but __read_swap_cache_async() does it
> already because there is no better way to wait for the event to happen.
> 
> With something like above you would no longer depend on the speed of schedule() to
> determine how often you would retry the fault, which would likely make sense.
> 
> I do wonder about the schedule() vs. schedule_timeout_uninterruptible(), though.
> No expert on that area, do you have any idea?
> 


Forgot to add

Acked-by: David Hildenbrand <[e-mail]>

But I suspect we do not want to not rely on schedule() to actually 
sleep, and instead keep retrying until the other thread finished, 
similar to above.

Huang, Ying Feb. 18, 2024, 7:59 a.m. UTC | #5

David Hildenbrand <[e-mail]> writes:

> On 16.02.24 10:51, Kairui Song wrote:
>> From: Kairui Song <[e-mail]>
>> 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 wait since it's a rare and very
>> short event. A schedule() call is added to avoid wasting too much CPU
>> or adding too much noise to perf statistics
>> 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")
>> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
>> Reported-by: "Huang, Ying" <[e-mail]>
>> Closes: https://lore.kernel.org/lkml/[e-mail]/
>> Signed-off-by: Kairui Song <[e-mail]>
>> Cc: [e-mail]
>> ---
>> Update from V2:
>> - Add a schedule() if raced to prevent repeated page faults wasting CPU
>>    and add noise to perf statistics.
>> - Use a bool to state the special case instead of reusing existing
>>    variables fixing error handling [Minchan Kim].
>> V2:
>> https://lore.kernel.org/all/[e-mail]/
>> 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.
>> V1:
>> https://lore.kernel.org/all/[e-mail]/
>>   include/linux/swap.h |  5 +++++
>>   mm/memory.c          | 20 ++++++++++++++++++++
>>   mm/swap.h            |  5 +++++
>>   mm/swapfile.c        | 13 +++++++++++++
>>   4 files changed, 43 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..7059230d0a54 100644
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>>   	struct page *page;
>>   	struct swap_info_struct *si = NULL;
>>   	rmap_t rmap_flags = RMAP_NONE;
>> +	bool need_clear_cache = false;
>>   	bool exclusive = false;
>>   	swp_entry_t entry;
>>   	pte_t pte;
>> @@ -3867,6 +3868,20 @@ 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)) {
>> +				/* Relax a bit to prevent rapid repeated page faults */
>> +				schedule();
>> +				goto out;
>> +			}
>> +			need_clear_cache = true;
>> +
>
> I took a closer look at __read_swap_cache_async() and it essentially
> does something similar.
>
> Instead of returning, it keeps retrying until it finds that
> swapcache_prepare() fails for another reason than -EEXISTS (e.g.,
> freed concurrently) or it finds the entry in the swapcache.
>
> So if you would succeed here on a freed+reused swap entry,
> __read_swap_cache_async() would simply retry.
>
> It spells that out:
>
> 		/*
> 		 * We might race against __delete_from_swap_cache(), and
> 		 * stumble across a swap_map entry whose SWAP_HAS_CACHE
> 		 * has not yet been cleared.  Or race against another
> 		 * __read_swap_cache_async(), which has set SWAP_HAS_CACHE
> 		 * in swap_map, but not yet added its folio to swap cache.
> 		 */
>
> Whereby we could not race against this code here as well where we
> speculatively set SWAP_HAS_CACHE and might never add something to the swap
> cache.
>
>
> I'd probably avoid the wrong returns and do something even closer to
> __read_swap_cache_async().
>
> while (true) {
> 	/*
> 	 * Fake that we are trying to insert a page into the swapcache, to
> 	 * serialize against concurrent threads wanting to do the same.
> 	 * [more from your description]
> 	 */
> 	ret = swapcache_prepare(entry);
> 	if (likely(!ret)
> 		/*
> 		 * Move forward with swapin, we'll recheck if the PTE hasn't
> 		 * changed later.
> 		 */
> 		break;
> 	else if (ret != -EEXIST)
> 		goto out;

The swap entry may be kept in swap cache for long time.  For example, it
may be read into swap cache via MADV_WILLNEED.

--
Best Regards,
Huang, Ying

> 	
> 	/*
>          * See __read_swap_cache_async(). We might either have raced against
>          * another thread, or the entry could have been freed and reused in the
> 	 * meantime. Make sure that the PTE did not change, to detect freeing.
> 	 */
> 	vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
> 				       vmf->address, &vmf->ptl);
> 	if (!vmf->pte || !pte_same(ptep_get(vmf->pte), vmf->orig_pte))
> 		goto unlock;
>
>
> 	schedule();
> }
>
>
>
> I was skeptical about the schedule(), but __read_swap_cache_async() does it
> already because there is no better way to wait for the event to happen.
>
> With something like above you would no longer depend on the speed of schedule() to
> determine how often you would retry the fault, which would likely make sense.
>
> I do wonder about the schedule() vs. schedule_timeout_uninterruptible(), though.
> No expert on that area, do you have any idea?

Barry Song Feb. 18, 2024, 8:19 a.m. UTC | #6

On Sun, Feb 18, 2024 at 9:02 PM Huang, Ying <[e-mail]> wrote:
>
> David Hildenbrand <[e-mail]> writes:
>
> > On 16.02.24 10:51, Kairui Song wrote:
> >> From: Kairui Song <[e-mail]>
> >> 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 wait since it's a rare and very
> >> short event. A schedule() call is added to avoid wasting too much CPU
> >> or adding too much noise to perf statistics
> >> 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")
> >> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> >> Reported-by: "Huang, Ying" <[e-mail]>
> >> Closes: https://lore.kernel.org/lkml/[e-mail]/
> >> Signed-off-by: Kairui Song <[e-mail]>
> >> Cc: [e-mail]
> >> ---
> >> Update from V2:
> >> - Add a schedule() if raced to prevent repeated page faults wasting CPU
> >>    and add noise to perf statistics.
> >> - Use a bool to state the special case instead of reusing existing
> >>    variables fixing error handling [Minchan Kim].
> >> V2:
> >> https://lore.kernel.org/all/[e-mail]/
> >> 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.
> >> V1:
> >> https://lore.kernel.org/all/[e-mail]/
> >>   include/linux/swap.h |  5 +++++
> >>   mm/memory.c          | 20 ++++++++++++++++++++
> >>   mm/swap.h            |  5 +++++
> >>   mm/swapfile.c        | 13 +++++++++++++
> >>   4 files changed, 43 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..7059230d0a54 100644
> >> --- a/mm/memory.c
> >> +++ b/mm/memory.c
> >> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> >>      struct page *page;
> >>      struct swap_info_struct *si = NULL;
> >>      rmap_t rmap_flags = RMAP_NONE;
> >> +    bool need_clear_cache = false;
> >>      bool exclusive = false;
> >>      swp_entry_t entry;
> >>      pte_t pte;
> >> @@ -3867,6 +3868,20 @@ 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)) {
> >> +                            /* Relax a bit to prevent rapid repeated page faults */
> >> +                            schedule();
> >> +                            goto out;
> >> +                    }
> >> +                    need_clear_cache = true;
> >> +
> >
> > I took a closer look at __read_swap_cache_async() and it essentially
> > does something similar.
> >
> > Instead of returning, it keeps retrying until it finds that
> > swapcache_prepare() fails for another reason than -EEXISTS (e.g.,
> > freed concurrently) or it finds the entry in the swapcache.
> >
> > So if you would succeed here on a freed+reused swap entry,
> > __read_swap_cache_async() would simply retry.
> >
> > It spells that out:
> >
> >               /*
> >                * We might race against __delete_from_swap_cache(), and
> >                * stumble across a swap_map entry whose SWAP_HAS_CACHE
> >                * has not yet been cleared.  Or race against another
> >                * __read_swap_cache_async(), which has set SWAP_HAS_CACHE
> >                * in swap_map, but not yet added its folio to swap cache.
> >                */
> >
> > Whereby we could not race against this code here as well where we
> > speculatively set SWAP_HAS_CACHE and might never add something to the swap
> > cache.
> >
> >
> > I'd probably avoid the wrong returns and do something even closer to
> > __read_swap_cache_async().
> >
> > while (true) {
> >       /*
> >        * Fake that we are trying to insert a page into the swapcache, to
> >        * serialize against concurrent threads wanting to do the same.
> >        * [more from your description]
> >        */
> >       ret = swapcache_prepare(entry);
> >       if (likely(!ret)
> >               /*
> >                * Move forward with swapin, we'll recheck if the PTE hasn't
> >                * changed later.
> >                */
> >               break;
> >       else if (ret != -EEXIST)
> >               goto out;
>
> The swap entry may be kept in swap cache for long time.  For example, it
> may be read into swap cache via MADV_WILLNEED.

This seems fine.

if swapcache has data from WILLNEED, the new page fault will hit it. Thus,
we won't go into the SYNC_IO path any more?

>
> --
> Best Regards,
> Huang, Ying
>
> >
> >       /*
> >          * See __read_swap_cache_async(). We might either have raced against
> >          * another thread, or the entry could have been freed and reused in the
> >        * meantime. Make sure that the PTE did not change, to detect freeing.
> >        */
> >       vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
> >                                      vmf->address, &vmf->ptl);
> >       if (!vmf->pte || !pte_same(ptep_get(vmf->pte), vmf->orig_pte))
> >               goto unlock;
> >
> >
> >       schedule();
> > }
> >
> >
> >
> > I was skeptical about the schedule(), but __read_swap_cache_async() does it
> > already because there is no better way to wait for the event to happen.
> >
> > With something like above you would no longer depend on the speed of schedule() to
> > determine how often you would retry the fault, which would likely make sense.
> >
> > I do wonder about the schedule() vs. schedule_timeout_uninterruptible(), though.
> > No expert on that area, do you have any idea?
>

Kairui Song Feb. 18, 2024, 8:29 a.m. UTC | #7

On Sat, Feb 17, 2024 at 2:02 AM David Hildenbrand <[e-mail]> wrote:
> On 16.02.24 17:53, David Hildenbrand wrote:
> > On 16.02.24 10:51, Kairui Song wrote:
> >> From: Kairui Song <[e-mail]>
> >>
> >> 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 wait since it's a rare and very
> >> short event. A schedule() call is added to avoid wasting too much CPU
> >> or adding too much noise to perf statistics
> >>
> >> 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")
> >> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> >> Reported-by: "Huang, Ying" <[e-mail]>
> >> Closes: https://lore.kernel.org/lkml/[e-mail]/
> >> Signed-off-by: Kairui Song <[e-mail]>
> >> Cc: [e-mail]
> >>
> >> ---
> >> Update from V2:
> >> - Add a schedule() if raced to prevent repeated page faults wasting CPU
> >>     and add noise to perf statistics.
> >> - Use a bool to state the special case instead of reusing existing
> >>     variables fixing error handling [Minchan Kim].
> >>
> >> V2: https://lore.kernel.org/all/[e-mail]/
> >>
> >> 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.
> >>
> >> V1: https://lore.kernel.org/all/[e-mail]/
> >>
> >>    include/linux/swap.h |  5 +++++
> >>    mm/memory.c          | 20 ++++++++++++++++++++
> >>    mm/swap.h            |  5 +++++
> >>    mm/swapfile.c        | 13 +++++++++++++
> >>    4 files changed, 43 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..7059230d0a54 100644
> >> --- a/mm/memory.c
> >> +++ b/mm/memory.c
> >> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> >>      struct page *page;
> >>      struct swap_info_struct *si = NULL;
> >>      rmap_t rmap_flags = RMAP_NONE;
> >> +    bool need_clear_cache = false;
> >>      bool exclusive = false;
> >>      swp_entry_t entry;
> >>      pte_t pte;
> >> @@ -3867,6 +3868,20 @@ 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)) {
> >> +                            /* Relax a bit to prevent rapid repeated page faults */
> >> +                            schedule();
> >> +                            goto out;
> >> +                    }
> >> +                    need_clear_cache = true;
> >> +

Hi David

Thanks for the review! I saw you added more replies so I'll just post
reply on your last mail.

> >
> > I took a closer look at __read_swap_cache_async() and it essentially
> > does something similar.
> >
> > Instead of returning, it keeps retrying until it finds that
> > swapcache_prepare() fails for another reason than -EEXISTS (e.g.,
> > freed concurrently) or it finds the entry in the swapcache.
> >
> > So if you would succeed here on a freed+reused swap entry,
> > __read_swap_cache_async() would simply retry.
> >
> > It spells that out:
> >
> >               /*
> >                * We might race against __delete_from_swap_cache(), and
> >                * stumble across a swap_map entry whose SWAP_HAS_CACHE
> >                * has not yet been cleared.  Or race against another
> >                * __read_swap_cache_async(), which has set SWAP_HAS_CACHE
> >                * in swap_map, but not yet added its folio to swap cache.
> >                */
> >
> > Whereby we could not race against this code here as well where we
> > speculatively set SWAP_HAS_CACHE and might never add something to the swap
> > cache.
> >
> >
> > I'd probably avoid the wrong returns and do something even closer to
> > __read_swap_cache_async().
> >
> > while (true) {
> >       /*
> >        * Fake that we are trying to insert a page into the swapcache, to
> >        * serialize against concurrent threads wanting to do the same.
> >        * [more from your description]
> >        */
> >       ret = swapcache_prepare(entry);
> >       if (likely(!ret)
> >               /*
> >                * Move forward with swapin, we'll recheck if the PTE hasn't
> >                * changed later.
> >                */
> >               break;
> >       else if (ret != -EEXIST)
> >               goto out;
> >
> >       /*
> >            * See __read_swap_cache_async(). We might either have raced against
> >            * another thread, or the entry could have been freed and reused in the
> >        * meantime. Make sure that the PTE did not change, to detect freeing.
> >        */
> >       vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
> >                                      vmf->address, &vmf->ptl);
> >       if (!vmf->pte || !pte_same(ptep_get(vmf->pte), vmf->orig_pte))
> >               goto unlock;
> >
> >
> >       schedule();
> > }

It was discussed earlier about looping in the page fault. One issue is
about swap entry may stuck in swapcache for a long time, so at lease
an extra cache loop up is need. To be safe we need to implement a
similar loop like the one in mm/swap_state.c, which I doubt is
necessary...

> >
> > I was skeptical about the schedule(), but __read_swap_cache_async() does it
> > already because there is no better way to wait for the event to happen.
> >
> > With something like above you would no longer depend on the speed of schedule() to
> > determine how often you would retry the fault, which would likely make sense.
> >
> > I do wonder about the schedule() vs. schedule_timeout_uninterruptible(), though.
> > No expert on that area, do you have any idea?
> >

schedule_timeout_uninterruptible seems more reasonable here from its
name (delay a bit to wait). My idea here is that SWP_SYNCHRONOUS_IO
devices are supposed to be super fast, so usually a second try will
just work (when tested with a less stressed test case and that seems
to be always true), and the race itself is rare enough to be ignore
for 7 years.

But when system is really stressed (eg. the reproducer I provided), it
may take longer to finish (SWP_SYNCHRONOUS_IO devices are CPU bound).
So a schedule() can help to avoid one task from looping page fault,
for better statistic and CPU usage.

Previous test results:
https://lore.kernel.org/lkml/CAMgjq7BvTJmxrWQOJvkLt4g_jnvmx07NdU63sGeRMGde4Ov=[e-mail]/
It showed schedule() works fine here.

Huang, Ying Feb. 18, 2024, 8:32 a.m. UTC | #8

Kairui Song <[e-mail]> writes:

> From: Kairui Song <[e-mail]>
>
> 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 wait since it's a rare and very
> short event. A schedule() call is added to avoid wasting too much CPU
> or adding too much noise to perf statistics
>
> 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.

The swap entry may be put in swap cache by some parallel code path
anyway.  So, we always need to consider that when reasoning the code.

> 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")
> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> Reported-by: "Huang, Ying" <[e-mail]>
> Closes: https://lore.kernel.org/lkml/[e-mail]/
> Signed-off-by: Kairui Song <[e-mail]>
> Cc: [e-mail]
>
> ---
> Update from V2:
> - Add a schedule() if raced to prevent repeated page faults wasting CPU
>   and add noise to perf statistics.
> - Use a bool to state the special case instead of reusing existing
>   variables fixing error handling [Minchan Kim].
>
> V2: https://lore.kernel.org/all/[e-mail]/
>
> 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.
>
> V1: https://lore.kernel.org/all/[e-mail]/
>
>  include/linux/swap.h |  5 +++++
>  mm/memory.c          | 20 ++++++++++++++++++++
>  mm/swap.h            |  5 +++++
>  mm/swapfile.c        | 13 +++++++++++++
>  4 files changed, 43 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..7059230d0a54 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>  	struct page *page;
>  	struct swap_info_struct *si = NULL;
>  	rmap_t rmap_flags = RMAP_NONE;
> +	bool need_clear_cache = false;
>  	bool exclusive = false;
>  	swp_entry_t entry;
>  	pte_t pte;
> @@ -3867,6 +3868,20 @@ 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)) {
> +				/* Relax a bit to prevent rapid repeated page faults */
> +				schedule();

The current task may be chosen in schedule().  So, I think that we
should use cond_resched() here.

--
Best Regards,
Huang, Ying

> +				goto out;
> +			}
> +			need_clear_cache = true;
> +
>  			/* skip swapcache */
>  			folio = vma_alloc_folio(GFP_HIGHUSER_MOVABLE, 0,
>  						vma, vmf->address, false);
> @@ -4117,6 +4132,9 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>  	if (vmf->pte)
>  		pte_unmap_unlock(vmf->pte, vmf->ptl);
>  out:
> +	/* Clear the swap cache pin for direct swapin after PTL unlock */
> +	if (need_clear_cache)
> +		swapcache_clear(si, entry);
>  	if (si)
>  		put_swap_device(si);
>  	return ret;
> @@ -4131,6 +4149,8 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>  		folio_unlock(swapcache);
>  		folio_put(swapcache);
>  	}
> +	if (need_clear_cache)
> +		swapcache_clear(si, entry);
>  	if (si)
>  		put_swap_device(si);
>  	return ret;
> diff --git a/mm/swap.h b/mm/swap.h
> index 758c46ca671e..fc2f6ade7f80 100644
> --- a/mm/swap.h
> +++ b/mm/swap.h
> @@ -41,6 +41,7 @@ void __delete_from_swap_cache(struct folio *folio,
>  void delete_from_swap_cache(struct folio *folio);
>  void clear_shadow_from_swap_cache(int type, unsigned long begin,
>  				  unsigned long end);
> +void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry);
>  struct folio *swap_cache_get_folio(swp_entry_t entry,
>  		struct vm_area_struct *vma, unsigned long addr);
>  struct folio *filemap_get_incore_folio(struct address_space *mapping,
> @@ -97,6 +98,10 @@ static inline int swap_writepage(struct page *p, struct writeback_control *wbc)
>  	return 0;
>  }
>  
> +static inline void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry)
> +{
> +}
> +
>  static inline struct folio *swap_cache_get_folio(swp_entry_t entry,
>  		struct vm_area_struct *vma, unsigned long addr)
>  {
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 556ff7347d5f..746aa9da5302 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -3365,6 +3365,19 @@ int swapcache_prepare(swp_entry_t entry)
>  	return __swap_duplicate(entry, SWAP_HAS_CACHE);
>  }
>  
> +void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry)
> +{
> +	struct swap_cluster_info *ci;
> +	unsigned long offset = swp_offset(entry);
> +	unsigned char usage;
> +
> +	ci = lock_cluster_or_swap_info(si, offset);
> +	usage = __swap_entry_free_locked(si, offset, SWAP_HAS_CACHE);
> +	unlock_cluster_or_swap_info(si, ci);
> +	if (!usage)
> +		free_swap_slot(entry);
> +}
> +
>  struct swap_info_struct *swp_swap_info(swp_entry_t entry)
>  {
>  	return swap_type_to_swap_info(swp_type(entry));

Huang, Ying Feb. 18, 2024, 8:39 a.m. UTC | #9

Barry Song <[e-mail]> writes:

> On Sun, Feb 18, 2024 at 9:02 PM Huang, Ying <[e-mail]> wrote:
>>
>> David Hildenbrand <[e-mail]> writes:
>>
>> > On 16.02.24 10:51, Kairui Song wrote:
>> >> From: Kairui Song <[e-mail]>
>> >> 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 wait since it's a rare and very
>> >> short event. A schedule() call is added to avoid wasting too much CPU
>> >> or adding too much noise to perf statistics
>> >> 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")
>> >> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
>> >> Reported-by: "Huang, Ying" <[e-mail]>
>> >> Closes: https://lore.kernel.org/lkml/[e-mail]/
>> >> Signed-off-by: Kairui Song <[e-mail]>
>> >> Cc: [e-mail]
>> >> ---
>> >> Update from V2:
>> >> - Add a schedule() if raced to prevent repeated page faults wasting CPU
>> >>    and add noise to perf statistics.
>> >> - Use a bool to state the special case instead of reusing existing
>> >>    variables fixing error handling [Minchan Kim].
>> >> V2:
>> >> https://lore.kernel.org/all/[e-mail]/
>> >> 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.
>> >> V1:
>> >> https://lore.kernel.org/all/[e-mail]/
>> >>   include/linux/swap.h |  5 +++++
>> >>   mm/memory.c          | 20 ++++++++++++++++++++
>> >>   mm/swap.h            |  5 +++++
>> >>   mm/swapfile.c        | 13 +++++++++++++
>> >>   4 files changed, 43 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..7059230d0a54 100644
>> >> --- a/mm/memory.c
>> >> +++ b/mm/memory.c
>> >> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>> >>      struct page *page;
>> >>      struct swap_info_struct *si = NULL;
>> >>      rmap_t rmap_flags = RMAP_NONE;
>> >> +    bool need_clear_cache = false;
>> >>      bool exclusive = false;
>> >>      swp_entry_t entry;
>> >>      pte_t pte;
>> >> @@ -3867,6 +3868,20 @@ 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)) {
>> >> +                            /* Relax a bit to prevent rapid repeated page faults */
>> >> +                            schedule();
>> >> +                            goto out;
>> >> +                    }
>> >> +                    need_clear_cache = true;
>> >> +
>> >
>> > I took a closer look at __read_swap_cache_async() and it essentially
>> > does something similar.
>> >
>> > Instead of returning, it keeps retrying until it finds that
>> > swapcache_prepare() fails for another reason than -EEXISTS (e.g.,
>> > freed concurrently) or it finds the entry in the swapcache.
>> >
>> > So if you would succeed here on a freed+reused swap entry,
>> > __read_swap_cache_async() would simply retry.
>> >
>> > It spells that out:
>> >
>> >               /*
>> >                * We might race against __delete_from_swap_cache(), and
>> >                * stumble across a swap_map entry whose SWAP_HAS_CACHE
>> >                * has not yet been cleared.  Or race against another
>> >                * __read_swap_cache_async(), which has set SWAP_HAS_CACHE
>> >                * in swap_map, but not yet added its folio to swap cache.
>> >                */
>> >
>> > Whereby we could not race against this code here as well where we
>> > speculatively set SWAP_HAS_CACHE and might never add something to the swap
>> > cache.
>> >
>> >
>> > I'd probably avoid the wrong returns and do something even closer to
>> > __read_swap_cache_async().
>> >
>> > while (true) {
>> >       /*
>> >        * Fake that we are trying to insert a page into the swapcache, to
>> >        * serialize against concurrent threads wanting to do the same.
>> >        * [more from your description]
>> >        */
>> >       ret = swapcache_prepare(entry);
>> >       if (likely(!ret)
>> >               /*
>> >                * Move forward with swapin, we'll recheck if the PTE hasn't
>> >                * changed later.
>> >                */
>> >               break;
>> >       else if (ret != -EEXIST)
>> >               goto out;
>>
>> The swap entry may be kept in swap cache for long time.  For example, it
>> may be read into swap cache via MADV_WILLNEED.
>
> This seems fine.
>
> if swapcache has data from WILLNEED, the new page fault will hit it. Thus,
> we won't go into the SYNC_IO path any more?

They may happen in parallel.  That is, one task is busy looping, while
another task read the swap entry into swap cache.

--
Best Regards,
Huang, Ying

>>
>> --
>> Best Regards,
>> Huang, Ying
>>
>> >
>> >       /*
>> >          * See __read_swap_cache_async(). We might either have raced against
>> >          * another thread, or the entry could have been freed and reused in the
>> >        * meantime. Make sure that the PTE did not change, to detect freeing.
>> >        */
>> >       vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
>> >                                      vmf->address, &vmf->ptl);
>> >       if (!vmf->pte || !pte_same(ptep_get(vmf->pte), vmf->orig_pte))
>> >               goto unlock;
>> >
>> >
>> >       schedule();
>> > }
>> >
>> >
>> >
>> > I was skeptical about the schedule(), but __read_swap_cache_async() does it
>> > already because there is no better way to wait for the event to happen.
>> >
>> > With something like above you would no longer depend on the speed of schedule() to
>> > determine how often you would retry the fault, which would likely make sense.
>> >
>> > I do wonder about the schedule() vs. schedule_timeout_uninterruptible(), though.
>> > No expert on that area, do you have any idea?
>>

Huang, Ying Feb. 18, 2024, 8:42 a.m. UTC | #10

Kairui Song <[e-mail]> writes:

> On Fri, Feb 16, 2024 at 6:15 PM Barry Song <[e-mail]> wrote:
>>
>> On Fri, Feb 16, 2024 at 10:53 PM Kairui Song <[e-mail]> wrote:
>> >
>> > From: Kairui Song <[e-mail]>
>> >
>> > 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 wait since it's a rare and very
>> > short event. A schedule() call is added to avoid wasting too much CPU
>> > or adding too much noise to perf statistics
>> >
>> > 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")
>> > Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
>> > Reported-by: "Huang, Ying" <[e-mail]>
>> > Closes: https://lore.kernel.org/lkml/[e-mail]/
>> > Signed-off-by: Kairui Song <[e-mail]>
>> > Cc: [e-mail]
>> >
>> > ---
>> > Update from V2:
>> > - Add a schedule() if raced to prevent repeated page faults wasting CPU
>> >   and add noise to perf statistics.
>> > - Use a bool to state the special case instead of reusing existing
>> >   variables fixing error handling [Minchan Kim].
>> >
>> > V2: https://lore.kernel.org/all/[e-mail]/
>> >
>> > 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.
>> >
>> > V1: https://lore.kernel.org/all/[e-mail]/
>> >
>> >  include/linux/swap.h |  5 +++++
>> >  mm/memory.c          | 20 ++++++++++++++++++++
>> >  mm/swap.h            |  5 +++++
>> >  mm/swapfile.c        | 13 +++++++++++++
>> >  4 files changed, 43 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..7059230d0a54 100644
>> > --- a/mm/memory.c
>> > +++ b/mm/memory.c
>> > @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>> >         struct page *page;
>> >         struct swap_info_struct *si = NULL;
>> >         rmap_t rmap_flags = RMAP_NONE;
>> > +       bool need_clear_cache = false;
>> >         bool exclusive = false;
>> >         swp_entry_t entry;
>> >         pte_t pte;
>> > @@ -3867,6 +3868,20 @@ 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)) {
>> > +                               /* Relax a bit to prevent rapid repeated page faults */
>> > +                               schedule();
>> > +                               goto out;
>> > +                       }
>> > +                       need_clear_cache = true;
>>
>> Hi Kairui,
>> i remember Ying had a suggestion to move swapcache_prepare after
>> swap_read_folio() to decrease race window. does that one even help
>> more to somehow "fix" the counting issue?
>>
>
> Hi Barry,
>
> Thanks for the comments!
>
> Yes, that's one of the suggestions, I found the result already looks
> good enough in test scenario I've posted after adding scheduler()
> change. After moving swapcache_prepare actually waste more CPU as it
> need to alloc/free extra folios now, and test result for the counting
> issue barely changed, so I kept the original code.

The time to allocate folios is unbounded, especially when direct
reclaiming is triggered.  And we can clear swap cache flag earlier to
reduce the race window too.

--
Best Regards,
Huang, Ying

Barry Song Feb. 18, 2024, 8:47 a.m. UTC | #11

On Sun, Feb 18, 2024 at 9:41 PM Huang, Ying <[e-mail]> wrote:
>
> Barry Song <[e-mail]> writes:
>
> > On Sun, Feb 18, 2024 at 9:02 PM Huang, Ying <[e-mail]> wrote:
> >>
> >> David Hildenbrand <[e-mail]> writes:
> >>
> >> > On 16.02.24 10:51, Kairui Song wrote:
> >> >> From: Kairui Song <[e-mail]>
> >> >> 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 wait since it's a rare and very
> >> >> short event. A schedule() call is added to avoid wasting too much CPU
> >> >> or adding too much noise to perf statistics
> >> >> 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")
> >> >> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> >> >> Reported-by: "Huang, Ying" <[e-mail]>
> >> >> Closes: https://lore.kernel.org/lkml/[e-mail]/
> >> >> Signed-off-by: Kairui Song <[e-mail]>
> >> >> Cc: [e-mail]
> >> >> ---
> >> >> Update from V2:
> >> >> - Add a schedule() if raced to prevent repeated page faults wasting CPU
> >> >>    and add noise to perf statistics.
> >> >> - Use a bool to state the special case instead of reusing existing
> >> >>    variables fixing error handling [Minchan Kim].
> >> >> V2:
> >> >> https://lore.kernel.org/all/[e-mail]/
> >> >> 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.
> >> >> V1:
> >> >> https://lore.kernel.org/all/[e-mail]/
> >> >>   include/linux/swap.h |  5 +++++
> >> >>   mm/memory.c          | 20 ++++++++++++++++++++
> >> >>   mm/swap.h            |  5 +++++
> >> >>   mm/swapfile.c        | 13 +++++++++++++
> >> >>   4 files changed, 43 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..7059230d0a54 100644
> >> >> --- a/mm/memory.c
> >> >> +++ b/mm/memory.c
> >> >> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> >> >>      struct page *page;
> >> >>      struct swap_info_struct *si = NULL;
> >> >>      rmap_t rmap_flags = RMAP_NONE;
> >> >> +    bool need_clear_cache = false;
> >> >>      bool exclusive = false;
> >> >>      swp_entry_t entry;
> >> >>      pte_t pte;
> >> >> @@ -3867,6 +3868,20 @@ 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)) {
> >> >> +                            /* Relax a bit to prevent rapid repeated page faults */
> >> >> +                            schedule();
> >> >> +                            goto out;
> >> >> +                    }
> >> >> +                    need_clear_cache = true;
> >> >> +
> >> >
> >> > I took a closer look at __read_swap_cache_async() and it essentially
> >> > does something similar.
> >> >
> >> > Instead of returning, it keeps retrying until it finds that
> >> > swapcache_prepare() fails for another reason than -EEXISTS (e.g.,
> >> > freed concurrently) or it finds the entry in the swapcache.
> >> >
> >> > So if you would succeed here on a freed+reused swap entry,
> >> > __read_swap_cache_async() would simply retry.
> >> >
> >> > It spells that out:
> >> >
> >> >               /*
> >> >                * We might race against __delete_from_swap_cache(), and
> >> >                * stumble across a swap_map entry whose SWAP_HAS_CACHE
> >> >                * has not yet been cleared.  Or race against another
> >> >                * __read_swap_cache_async(), which has set SWAP_HAS_CACHE
> >> >                * in swap_map, but not yet added its folio to swap cache.
> >> >                */
> >> >
> >> > Whereby we could not race against this code here as well where we
> >> > speculatively set SWAP_HAS_CACHE and might never add something to the swap
> >> > cache.
> >> >
> >> >
> >> > I'd probably avoid the wrong returns and do something even closer to
> >> > __read_swap_cache_async().
> >> >
> >> > while (true) {
> >> >       /*
> >> >        * Fake that we are trying to insert a page into the swapcache, to
> >> >        * serialize against concurrent threads wanting to do the same.
> >> >        * [more from your description]
> >> >        */
> >> >       ret = swapcache_prepare(entry);
> >> >       if (likely(!ret)
> >> >               /*
> >> >                * Move forward with swapin, we'll recheck if the PTE hasn't
> >> >                * changed later.
> >> >                */
> >> >               break;
> >> >       else if (ret != -EEXIST)
> >> >               goto out;
> >>
> >> The swap entry may be kept in swap cache for long time.  For example, it
> >> may be read into swap cache via MADV_WILLNEED.
> >
> > This seems fine.
> >
> > if swapcache has data from WILLNEED, the new page fault will hit it. Thus,
> > we won't go into the SYNC_IO path any more?
>
> They may happen in parallel.  That is, one task is busy looping, while
> another task read the swap entry into swap cache.

do_swap_page isn't busy looping swapcache_prepare, if it fails, it exits,
then we have a completely new page fault. this new page fault will
lookup swapcache and find it, going into the path to set swapcache
to ptes. so the new page fault won't do swapcache_prepare any more.

>
> --
> Best Regards,
> Huang, Ying
>
> >>
> >> --
> >> Best Regards,
> >> Huang, Ying
> >>
> >> >
> >> >       /*
> >> >          * See __read_swap_cache_async(). We might either have raced against
> >> >          * another thread, or the entry could have been freed and reused in the
> >> >        * meantime. Make sure that the PTE did not change, to detect freeing.
> >> >        */
> >> >       vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
> >> >                                      vmf->address, &vmf->ptl);
> >> >       if (!vmf->pte || !pte_same(ptep_get(vmf->pte), vmf->orig_pte))
> >> >               goto unlock;
> >> >
> >> >
> >> >       schedule();
> >> > }
> >> >
> >> >
> >> >
> >> > I was skeptical about the schedule(), but __read_swap_cache_async() does it
> >> > already because there is no better way to wait for the event to happen.
> >> >
> >> > With something like above you would no longer depend on the speed of schedule() to
> >> > determine how often you would retry the fault, which would likely make sense.
> >> >
> >> > I do wonder about the schedule() vs. schedule_timeout_uninterruptible(), though.
> >> > No expert on that area, do you have any idea?
> >>

Kairui Song Feb. 18, 2024, 9:20 a.m. UTC | #12

On Sun, Feb 18, 2024 at 4:34 PM Huang, Ying <[e-mail]> wrote:
>
> Kairui Song <[e-mail]> writes:
>
> > From: Kairui Song <[e-mail]>
> >
> > 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 wait since it's a rare and very
> > short event. A schedule() call is added to avoid wasting too much CPU
> > or adding too much noise to perf statistics
> >
> > 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.
>
> The swap entry may be put in swap cache by some parallel code path
> anyway.  So, we always need to consider that when reasoning the code.
>
> > 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")
> > Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> > Reported-by: "Huang, Ying" <[e-mail]>
> > Closes: https://lore.kernel.org/lkml/[e-mail]/
> > Signed-off-by: Kairui Song <[e-mail]>
> > Cc: [e-mail]
> >
> > ---
> > Update from V2:
> > - Add a schedule() if raced to prevent repeated page faults wasting CPU
> >   and add noise to perf statistics.
> > - Use a bool to state the special case instead of reusing existing
> >   variables fixing error handling [Minchan Kim].
> >
> > V2: https://lore.kernel.org/all/[e-mail]/
> >
> > 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.
> >
> > V1: https://lore.kernel.org/all/[e-mail]/
> >
> >  include/linux/swap.h |  5 +++++
> >  mm/memory.c          | 20 ++++++++++++++++++++
> >  mm/swap.h            |  5 +++++
> >  mm/swapfile.c        | 13 +++++++++++++
> >  4 files changed, 43 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..7059230d0a54 100644
> > --- a/mm/memory.c
> > +++ b/mm/memory.c
> > @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> >       struct page *page;
> >       struct swap_info_struct *si = NULL;
> >       rmap_t rmap_flags = RMAP_NONE;
> > +     bool need_clear_cache = false;
> >       bool exclusive = false;
> >       swp_entry_t entry;
> >       pte_t pte;
> > @@ -3867,6 +3868,20 @@ 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)) {
> > +                             /* Relax a bit to prevent rapid repeated page faults */
> > +                             schedule();
>
> The current task may be chosen in schedule().  So, I think that we
> should use cond_resched() here.
>

I think if we are worried about current task got chosen again we can
use schedule_timeout_uninterruptible(1) here. Isn't cond_resched still
__schedule() and and it can even get omitted, so it should be "weaker"
IIUC.

Kairui Song Feb. 18, 2024, 9:30 a.m. UTC | #13

On Sun, Feb 18, 2024 at 4:47 PM Barry Song <[e-mail]> wrote:
>
> On Sun, Feb 18, 2024 at 9:41 PM Huang, Ying <[e-mail]> wrote:
> >
> > Barry Song <[e-mail]> writes:
> >
> > > On Sun, Feb 18, 2024 at 9:02 PM Huang, Ying <[e-mail]> wrote:
> > >>
> > >> David Hildenbrand <[e-mail]> writes:
> > >>
> > >> > On 16.02.24 10:51, Kairui Song wrote:
> > >> >> From: Kairui Song <[e-mail]>
> > >> >> 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 wait since it's a rare and very
> > >> >> short event. A schedule() call is added to avoid wasting too much CPU
> > >> >> or adding too much noise to perf statistics
> > >> >> 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")
> > >> >> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> > >> >> Reported-by: "Huang, Ying" <[e-mail]>
> > >> >> Closes: https://lore.kernel.org/lkml/[e-mail]/
> > >> >> Signed-off-by: Kairui Song <[e-mail]>
> > >> >> Cc: [e-mail]
> > >> >> ---
> > >> >> Update from V2:
> > >> >> - Add a schedule() if raced to prevent repeated page faults wasting CPU
> > >> >>    and add noise to perf statistics.
> > >> >> - Use a bool to state the special case instead of reusing existing
> > >> >>    variables fixing error handling [Minchan Kim].
> > >> >> V2:
> > >> >> https://lore.kernel.org/all/[e-mail]/
> > >> >> 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.
> > >> >> V1:
> > >> >> https://lore.kernel.org/all/[e-mail]/
> > >> >>   include/linux/swap.h |  5 +++++
> > >> >>   mm/memory.c          | 20 ++++++++++++++++++++
> > >> >>   mm/swap.h            |  5 +++++
> > >> >>   mm/swapfile.c        | 13 +++++++++++++
> > >> >>   4 files changed, 43 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..7059230d0a54 100644
> > >> >> --- a/mm/memory.c
> > >> >> +++ b/mm/memory.c
> > >> >> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> > >> >>      struct page *page;
> > >> >>      struct swap_info_struct *si = NULL;
> > >> >>      rmap_t rmap_flags = RMAP_NONE;
> > >> >> +    bool need_clear_cache = false;
> > >> >>      bool exclusive = false;
> > >> >>      swp_entry_t entry;
> > >> >>      pte_t pte;
> > >> >> @@ -3867,6 +3868,20 @@ 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)) {
> > >> >> +                            /* Relax a bit to prevent rapid repeated page faults */
> > >> >> +                            schedule();
> > >> >> +                            goto out;
> > >> >> +                    }
> > >> >> +                    need_clear_cache = true;
> > >> >> +
> > >> >
> > >> > I took a closer look at __read_swap_cache_async() and it essentially
> > >> > does something similar.
> > >> >
> > >> > Instead of returning, it keeps retrying until it finds that
> > >> > swapcache_prepare() fails for another reason than -EEXISTS (e.g.,
> > >> > freed concurrently) or it finds the entry in the swapcache.
> > >> >
> > >> > So if you would succeed here on a freed+reused swap entry,
> > >> > __read_swap_cache_async() would simply retry.
> > >> >
> > >> > It spells that out:
> > >> >
> > >> >               /*
> > >> >                * We might race against __delete_from_swap_cache(), and
> > >> >                * stumble across a swap_map entry whose SWAP_HAS_CACHE
> > >> >                * has not yet been cleared.  Or race against another
> > >> >                * __read_swap_cache_async(), which has set SWAP_HAS_CACHE
> > >> >                * in swap_map, but not yet added its folio to swap cache.
> > >> >                */
> > >> >
> > >> > Whereby we could not race against this code here as well where we
> > >> > speculatively set SWAP_HAS_CACHE and might never add something to the swap
> > >> > cache.
> > >> >
> > >> >
> > >> > I'd probably avoid the wrong returns and do something even closer to
> > >> > __read_swap_cache_async().
> > >> >
> > >> > while (true) {
> > >> >       /*
> > >> >        * Fake that we are trying to insert a page into the swapcache, to
> > >> >        * serialize against concurrent threads wanting to do the same.
> > >> >        * [more from your description]
> > >> >        */
> > >> >       ret = swapcache_prepare(entry);
> > >> >       if (likely(!ret)
> > >> >               /*
> > >> >                * Move forward with swapin, we'll recheck if the PTE hasn't
> > >> >                * changed later.
> > >> >                */
> > >> >               break;
> > >> >       else if (ret != -EEXIST)
> > >> >               goto out;
> > >>
> > >> The swap entry may be kept in swap cache for long time.  For example, it
> > >> may be read into swap cache via MADV_WILLNEED.
> > >
> > > This seems fine.
> > >
> > > if swapcache has data from WILLNEED, the new page fault will hit it. Thus,
> > > we won't go into the SYNC_IO path any more?
> >
> > They may happen in parallel.  That is, one task is busy looping, while
> > another task read the swap entry into swap cache.
>
> do_swap_page isn't busy looping swapcache_prepare, if it fails, it exits,
> then we have a completely new page fault. this new page fault will
> lookup swapcache and find it, going into the path to set swapcache
> to ptes. so the new page fault won't do swapcache_prepare any more.
>

Hi Barry

The issue here with this code snip is that we could have
swapcache_prepare(entry) == -EEXIST and pte_same == true, not
necessarily WILLNEED, a concurrent fault and swapout could cause that.
Then we are stuck with while(true) here.

It can be fixed, but still there are other potential issues, we end up
with a similar loop in swap_state.c, may still need to bail out the
fault for some cases, things seem not improved here.

Chris Li Feb. 18, 2024, 6:37 p.m. UTC | #14

On Sun, Feb 18, 2024 at 12:41 AM Huang, Ying <[e-mail]> wrote:
>
> Barry Song <[e-mail]> writes:
>
> > On Sun, Feb 18, 2024 at 9:02 PM Huang, Ying <[e-mail]> wrote:
> >>
> >> David Hildenbrand <[e-mail]> writes:
> >>
> >> > On 16.02.24 10:51, Kairui Song wrote:
> >> >> From: Kairui Song <[e-mail]>
> >> >> 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 wait since it's a rare and very
> >> >> short event. A schedule() call is added to avoid wasting too much CPU
> >> >> or adding too much noise to perf statistics
> >> >> 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")
> >> >> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> >> >> Reported-by: "Huang, Ying" <[e-mail]>
> >> >> Closes: https://lore.kernel.org/lkml/[e-mail]/
> >> >> Signed-off-by: Kairui Song <[e-mail]>
> >> >> Cc: [e-mail]
> >> >> ---
> >> >> Update from V2:
> >> >> - Add a schedule() if raced to prevent repeated page faults wasting CPU
> >> >>    and add noise to perf statistics.
> >> >> - Use a bool to state the special case instead of reusing existing
> >> >>    variables fixing error handling [Minchan Kim].
> >> >> V2:
> >> >> https://lore.kernel.org/all/[e-mail]/
> >> >> 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.
> >> >> V1:
> >> >> https://lore.kernel.org/all/[e-mail]/
> >> >>   include/linux/swap.h |  5 +++++
> >> >>   mm/memory.c          | 20 ++++++++++++++++++++
> >> >>   mm/swap.h            |  5 +++++
> >> >>   mm/swapfile.c        | 13 +++++++++++++
> >> >>   4 files changed, 43 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..7059230d0a54 100644
> >> >> --- a/mm/memory.c
> >> >> +++ b/mm/memory.c
> >> >> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> >> >>      struct page *page;
> >> >>      struct swap_info_struct *si = NULL;
> >> >>      rmap_t rmap_flags = RMAP_NONE;
> >> >> +    bool need_clear_cache = false;
> >> >>      bool exclusive = false;
> >> >>      swp_entry_t entry;
> >> >>      pte_t pte;
> >> >> @@ -3867,6 +3868,20 @@ 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)) {
> >> >> +                            /* Relax a bit to prevent rapid repeated page faults */
> >> >> +                            schedule();
> >> >> +                            goto out;
> >> >> +                    }
> >> >> +                    need_clear_cache = true;
> >> >> +
> >> >
> >> > I took a closer look at __read_swap_cache_async() and it essentially
> >> > does something similar.
> >> >
> >> > Instead of returning, it keeps retrying until it finds that
> >> > swapcache_prepare() fails for another reason than -EEXISTS (e.g.,
> >> > freed concurrently) or it finds the entry in the swapcache.
> >> >
> >> > So if you would succeed here on a freed+reused swap entry,
> >> > __read_swap_cache_async() would simply retry.
> >> >
> >> > It spells that out:
> >> >
> >> >               /*
> >> >                * We might race against __delete_from_swap_cache(), and
> >> >                * stumble across a swap_map entry whose SWAP_HAS_CACHE
> >> >                * has not yet been cleared.  Or race against another
> >> >                * __read_swap_cache_async(), which has set SWAP_HAS_CACHE
> >> >                * in swap_map, but not yet added its folio to swap cache.
> >> >                */
> >> >
> >> > Whereby we could not race against this code here as well where we
> >> > speculatively set SWAP_HAS_CACHE and might never add something to the swap
> >> > cache.
> >> >
> >> >
> >> > I'd probably avoid the wrong returns and do something even closer to
> >> > __read_swap_cache_async().
> >> >
> >> > while (true) {
> >> >       /*
> >> >        * Fake that we are trying to insert a page into the swapcache, to
> >> >        * serialize against concurrent threads wanting to do the same.
> >> >        * [more from your description]
> >> >        */
> >> >       ret = swapcache_prepare(entry);
> >> >       if (likely(!ret)
> >> >               /*
> >> >                * Move forward with swapin, we'll recheck if the PTE hasn't
> >> >                * changed later.
> >> >                */
> >> >               break;
> >> >       else if (ret != -EEXIST)
> >> >               goto out;
> >>
> >> The swap entry may be kept in swap cache for long time.  For example, it
> >> may be read into swap cache via MADV_WILLNEED.

I am trying to find the alternative path which can cause the swap
fault to bring in the swap cache in page fault while the SYNC IO is
looping for HAS_SWAP_CACHE. Kairui was able to identify the in the
current code in do_page_fault() path, the rmap and fork case wouldn't
be able to modify the swap cache causing a problem. The MADV_WILLNEED
is an excellent example. Thank you for finding this example.

> >
> > This seems fine.
> >
> > if swapcache has data from WILLNEED, the new page fault will hit it. Thus,
> > we won't go into the SYNC_IO path any more?
>
> They may happen in parallel.  That is, one task is busy looping, while
> another task read the swap entry into swap cache.

Agree.

Chris

Huang, Ying Feb. 19, 2024, 12:39 a.m. UTC | #15

Kairui Song <[e-mail]> writes:

> On Sun, Feb 18, 2024 at 4:34 PM Huang, Ying <[e-mail]> wrote:
>>
>> Kairui Song <[e-mail]> writes:
>>
>> > From: Kairui Song <[e-mail]>
>> >
>> > 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 wait since it's a rare and very
>> > short event. A schedule() call is added to avoid wasting too much CPU
>> > or adding too much noise to perf statistics
>> >
>> > 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.
>>
>> The swap entry may be put in swap cache by some parallel code path
>> anyway.  So, we always need to consider that when reasoning the code.
>>
>> > 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")
>> > Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
>> > Reported-by: "Huang, Ying" <[e-mail]>
>> > Closes: https://lore.kernel.org/lkml/[e-mail]/
>> > Signed-off-by: Kairui Song <[e-mail]>
>> > Cc: [e-mail]
>> >
>> > ---
>> > Update from V2:
>> > - Add a schedule() if raced to prevent repeated page faults wasting CPU
>> >   and add noise to perf statistics.
>> > - Use a bool to state the special case instead of reusing existing
>> >   variables fixing error handling [Minchan Kim].
>> >
>> > V2: https://lore.kernel.org/all/[e-mail]/
>> >
>> > 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.
>> >
>> > V1: https://lore.kernel.org/all/[e-mail]/
>> >
>> >  include/linux/swap.h |  5 +++++
>> >  mm/memory.c          | 20 ++++++++++++++++++++
>> >  mm/swap.h            |  5 +++++
>> >  mm/swapfile.c        | 13 +++++++++++++
>> >  4 files changed, 43 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..7059230d0a54 100644
>> > --- a/mm/memory.c
>> > +++ b/mm/memory.c
>> > @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>> >       struct page *page;
>> >       struct swap_info_struct *si = NULL;
>> >       rmap_t rmap_flags = RMAP_NONE;
>> > +     bool need_clear_cache = false;
>> >       bool exclusive = false;
>> >       swp_entry_t entry;
>> >       pte_t pte;
>> > @@ -3867,6 +3868,20 @@ 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)) {
>> > +                             /* Relax a bit to prevent rapid repeated page faults */
>> > +                             schedule();
>>
>> The current task may be chosen in schedule().  So, I think that we
>> should use cond_resched() here.
>>
>
> I think if we are worried about current task got chosen again we can
> use schedule_timeout_uninterruptible(1) here. Isn't cond_resched still
> __schedule() and and it can even get omitted, so it should be "weaker"
> IIUC.

schedule_timeout_uninterruptible(1) will introduce 1ms latency for the
second task.  That may kill performance of some workloads.

--
Best Regards,
Huang, Ying

Huang, Ying Feb. 19, 2024, 2:33 a.m. UTC | #16

"Huang, Ying" <[e-mail]> writes:

> Kairui Song <[e-mail]> writes:
>
>> On Sun, Feb 18, 2024 at 4:34 PM Huang, Ying <[e-mail]> wrote:
>>>
>>> Kairui Song <[e-mail]> writes:
>>>
>>> > From: Kairui Song <[e-mail]>
>>> >
>>> > 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 wait since it's a rare and very
>>> > short event. A schedule() call is added to avoid wasting too much CPU
>>> > or adding too much noise to perf statistics
>>> >
>>> > 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.
>>>
>>> The swap entry may be put in swap cache by some parallel code path
>>> anyway.  So, we always need to consider that when reasoning the code.
>>>
>>> > 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")
>>> > Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
>>> > Reported-by: "Huang, Ying" <[e-mail]>
>>> > Closes: https://lore.kernel.org/lkml/[e-mail]/
>>> > Signed-off-by: Kairui Song <[e-mail]>
>>> > Cc: [e-mail]
>>> >
>>> > ---
>>> > Update from V2:
>>> > - Add a schedule() if raced to prevent repeated page faults wasting CPU
>>> >   and add noise to perf statistics.
>>> > - Use a bool to state the special case instead of reusing existing
>>> >   variables fixing error handling [Minchan Kim].
>>> >
>>> > V2: https://lore.kernel.org/all/[e-mail]/
>>> >
>>> > 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.
>>> >
>>> > V1: https://lore.kernel.org/all/[e-mail]/
>>> >
>>> >  include/linux/swap.h |  5 +++++
>>> >  mm/memory.c          | 20 ++++++++++++++++++++
>>> >  mm/swap.h            |  5 +++++
>>> >  mm/swapfile.c        | 13 +++++++++++++
>>> >  4 files changed, 43 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..7059230d0a54 100644
>>> > --- a/mm/memory.c
>>> > +++ b/mm/memory.c
>>> > @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>>> >       struct page *page;
>>> >       struct swap_info_struct *si = NULL;
>>> >       rmap_t rmap_flags = RMAP_NONE;
>>> > +     bool need_clear_cache = false;
>>> >       bool exclusive = false;
>>> >       swp_entry_t entry;
>>> >       pte_t pte;
>>> > @@ -3867,6 +3868,20 @@ 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)) {
>>> > +                             /* Relax a bit to prevent rapid repeated page faults */
>>> > +                             schedule();
>>>
>>> The current task may be chosen in schedule().  So, I think that we
>>> should use cond_resched() here.
>>>
>>
>> I think if we are worried about current task got chosen again we can
>> use schedule_timeout_uninterruptible(1) here. Isn't cond_resched still
>> __schedule() and and it can even get omitted, so it should be "weaker"
>> IIUC.
>
> schedule_timeout_uninterruptible(1) will introduce 1ms latency for the
> second task.  That may kill performance of some workloads.

Just found that the cond_sched() in __read_swap_cache_async() has been
changed to schedule_timeout_uninterruptible(1) to fix some live lock.
Details are in the description of commit 029c4628b2eb ("mm: swap: get
rid of livelock in swapin readahead").  I think the similar issue may
happen here too.  So, we must use schedule_timeout_uninterruptible(1)
here until some other better idea becomes available.

--
Best Regards,
Huang, Ying

Kairui Song Feb. 19, 2024, 3:09 a.m. UTC | #17

On Mon, Feb 19, 2024 at 10:35 AM Huang, Ying <[e-mail]> wrote:
> "Huang, Ying" <[e-mail]> writes:
> > Kairui Song <[e-mail]> writes:
> >
> >> On Sun, Feb 18, 2024 at 4:34 PM Huang, Ying <[e-mail]> wrote:
> >>>
> >>> Kairui Song <[e-mail]> writes:
> >>>
> >>> > From: Kairui Song <[e-mail]>
> >>> >
> >>> > 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 wait since it's a rare and very
> >>> > short event. A schedule() call is added to avoid wasting too much CPU
> >>> > or adding too much noise to perf statistics
> >>> >
> >>> > 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.
> >>>
> >>> The swap entry may be put in swap cache by some parallel code path
> >>> anyway.  So, we always need to consider that when reasoning the code.
> >>>
> >>> > 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")
> >>> > Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> >>> > Reported-by: "Huang, Ying" <[e-mail]>
> >>> > Closes: https://lore.kernel.org/lkml/[e-mail]/
> >>> > Signed-off-by: Kairui Song <[e-mail]>
> >>> > Cc: [e-mail]
> >>> >
> >>> > ---
> >>> > Update from V2:
> >>> > - Add a schedule() if raced to prevent repeated page faults wasting CPU
> >>> >   and add noise to perf statistics.
> >>> > - Use a bool to state the special case instead of reusing existing
> >>> >   variables fixing error handling [Minchan Kim].
> >>> >
> >>> > V2: https://lore.kernel.org/all/[e-mail]/
> >>> >
> >>> > 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.
> >>> >
> >>> > V1: https://lore.kernel.org/all/[e-mail]/
> >>> >
> >>> >  include/linux/swap.h |  5 +++++
> >>> >  mm/memory.c          | 20 ++++++++++++++++++++
> >>> >  mm/swap.h            |  5 +++++
> >>> >  mm/swapfile.c        | 13 +++++++++++++
> >>> >  4 files changed, 43 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..7059230d0a54 100644
> >>> > --- a/mm/memory.c
> >>> > +++ b/mm/memory.c
> >>> > @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> >>> >       struct page *page;
> >>> >       struct swap_info_struct *si = NULL;
> >>> >       rmap_t rmap_flags = RMAP_NONE;
> >>> > +     bool need_clear_cache = false;
> >>> >       bool exclusive = false;
> >>> >       swp_entry_t entry;
> >>> >       pte_t pte;
> >>> > @@ -3867,6 +3868,20 @@ 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)) {
> >>> > +                             /* Relax a bit to prevent rapid repeated page faults */
> >>> > +                             schedule();
> >>>
> >>> The current task may be chosen in schedule().  So, I think that we
> >>> should use cond_resched() here.
> >>>
> >>
> >> I think if we are worried about current task got chosen again we can
> >> use schedule_timeout_uninterruptible(1) here. Isn't cond_resched still
> >> __schedule() and and it can even get omitted, so it should be "weaker"
> >> IIUC.
> >
> > schedule_timeout_uninterruptible(1) will introduce 1ms latency for the
> > second task.  That may kill performance of some workloads.

It actually calls schedule_timeout so it should be a 1 jiffy latency,
not 1ms, right?

/**
 * schedule_timeout - sleep until timeout
 * @timeout: timeout value in jiffies
...

But I think what we really want here is actually the set_current_state
to force yield CPU for a short period. The latency should be mild.

>
> Just found that the cond_sched() in __read_swap_cache_async() has been
> changed to schedule_timeout_uninterruptible(1) to fix some live lock.
> Details are in the description of commit 029c4628b2eb ("mm: swap: get
> rid of livelock in swapin readahead").  I think the similar issue may
> happen here too.  So, we must use schedule_timeout_uninterruptible(1)
> here until some other better idea becomes available.

Indeed, I'll switch to schedule_timeout_uninterruptible(1). I've
tested and posted the result with schedule_timeout_uninterruptible(1)
before, it looked fine, or even better.

Kairui Song Feb. 19, 2024, 7:29 a.m. UTC | #18

On Mon, Feb 19, 2024 at 11:09 AM Kairui Song <[e-mail]> wrote:
>
> On Mon, Feb 19, 2024 at 10:35 AM Huang, Ying <[e-mail]> wrote:
> > "Huang, Ying" <[e-mail]> writes:
> > > Kairui Song <[e-mail]> writes:
> > >
> > >> On Sun, Feb 18, 2024 at 4:34 PM Huang, Ying <[e-mail]> wrote:
> > >>>
> > >>> Kairui Song <[e-mail]> writes:
> > >>>
> > >>> > From: Kairui Song <[e-mail]>
> > >>> >
> > >>> > 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 wait since it's a rare and very
> > >>> > short event. A schedule() call is added to avoid wasting too much CPU
> > >>> > or adding too much noise to perf statistics
> > >>> >
> > >>> > 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.
> > >>>
> > >>> The swap entry may be put in swap cache by some parallel code path
> > >>> anyway.  So, we always need to consider that when reasoning the code.
> > >>>
> > >>> > 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")
> > >>> > Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
> > >>> > Reported-by: "Huang, Ying" <[e-mail]>
> > >>> > Closes: https://lore.kernel.org/lkml/[e-mail]/
> > >>> > Signed-off-by: Kairui Song <[e-mail]>
> > >>> > Cc: [e-mail]
> > >>> >
> > >>> > ---
> > >>> > Update from V2:
> > >>> > - Add a schedule() if raced to prevent repeated page faults wasting CPU
> > >>> >   and add noise to perf statistics.
> > >>> > - Use a bool to state the special case instead of reusing existing
> > >>> >   variables fixing error handling [Minchan Kim].
> > >>> >
> > >>> > V2: https://lore.kernel.org/all/[e-mail]/
> > >>> >
> > >>> > 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.
> > >>> >
> > >>> > V1: https://lore.kernel.org/all/[e-mail]/
> > >>> >
> > >>> >  include/linux/swap.h |  5 +++++
> > >>> >  mm/memory.c          | 20 ++++++++++++++++++++
> > >>> >  mm/swap.h            |  5 +++++
> > >>> >  mm/swapfile.c        | 13 +++++++++++++
> > >>> >  4 files changed, 43 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..7059230d0a54 100644
> > >>> > --- a/mm/memory.c
> > >>> > +++ b/mm/memory.c
> > >>> > @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> > >>> >       struct page *page;
> > >>> >       struct swap_info_struct *si = NULL;
> > >>> >       rmap_t rmap_flags = RMAP_NONE;
> > >>> > +     bool need_clear_cache = false;
> > >>> >       bool exclusive = false;
> > >>> >       swp_entry_t entry;
> > >>> >       pte_t pte;
> > >>> > @@ -3867,6 +3868,20 @@ 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)) {
> > >>> > +                             /* Relax a bit to prevent rapid repeated page faults */
> > >>> > +                             schedule();
> > >>>
> > >>> The current task may be chosen in schedule().  So, I think that we
> > >>> should use cond_resched() here.
> > >>>
> > >>
> > >> I think if we are worried about current task got chosen again we can
> > >> use schedule_timeout_uninterruptible(1) here. Isn't cond_resched still
> > >> __schedule() and and it can even get omitted, so it should be "weaker"
> > >> IIUC.
> > >
> > > schedule_timeout_uninterruptible(1) will introduce 1ms latency for the
> > > second task.  That may kill performance of some workloads.
>
> It actually calls schedule_timeout so it should be a 1 jiffy latency,
> not 1ms, right?
>
> /**
>  * schedule_timeout - sleep until timeout
>  * @timeout: timeout value in jiffies
> ...
>
> But I think what we really want here is actually the set_current_state
> to force yield CPU for a short period. The latency should be mild.

I just forgot 1 jiffy >= 1 ms here, and uninterruptible should make it
unable to wakeup until timeout...

> > Just found that the cond_sched() in __read_swap_cache_async() has been
> > changed to schedule_timeout_uninterruptible(1) to fix some live lock.
> > Details are in the description of commit 029c4628b2eb ("mm: swap: get
> > rid of livelock in swapin readahead").  I think the similar issue may
> > happen here too.  So, we must use schedule_timeout_uninterruptible(1)
> > here until some other better idea becomes available.
>
> Indeed, I'll switch to schedule_timeout_uninterruptible(1). I've
> tested and posted the result with schedule_timeout_uninterruptible(1)
> before, it looked fine, or even better.

But this should be still the same though, the minor/major fault ratio
in previous test result [1] shows the race on ZRAM even with threads
set to race on purpose, the chance is low, and thanks for the info on
mentioning another commit!

[1] https://lore.kernel.org/all/CAMgjq7BvTJmxrWQOJvkLt4g_jnvmx07NdU63sGeRMGde4Ov=[e-mail]/

David Hildenbrand Feb. 19, 2024, 9:01 a.m. UTC | #19

On 18.02.24 08:59, Huang, Ying wrote:
> David Hildenbrand <[e-mail]> writes:
> 
>> On 16.02.24 10:51, Kairui Song wrote:
>>> From: Kairui Song <[e-mail]>
>>> 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 wait since it's a rare and very
>>> short event. A schedule() call is added to avoid wasting too much CPU
>>> or adding too much noise to perf statistics
>>> 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")
>>> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress-race [1]
>>> Reported-by: "Huang, Ying" <[e-mail]>
>>> Closes: https://lore.kernel.org/lkml/[e-mail]/
>>> Signed-off-by: Kairui Song <[e-mail]>
>>> Cc: [e-mail]
>>> ---
>>> Update from V2:
>>> - Add a schedule() if raced to prevent repeated page faults wasting CPU
>>>     and add noise to perf statistics.
>>> - Use a bool to state the special case instead of reusing existing
>>>     variables fixing error handling [Minchan Kim].
>>> V2:
>>> https://lore.kernel.org/all/[e-mail]/
>>> 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.
>>> V1:
>>> https://lore.kernel.org/all/[e-mail]/
>>>    include/linux/swap.h |  5 +++++
>>>    mm/memory.c          | 20 ++++++++++++++++++++
>>>    mm/swap.h            |  5 +++++
>>>    mm/swapfile.c        | 13 +++++++++++++
>>>    4 files changed, 43 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..7059230d0a54 100644
>>> --- a/mm/memory.c
>>> +++ b/mm/memory.c
>>> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>>>    	struct page *page;
>>>    	struct swap_info_struct *si = NULL;
>>>    	rmap_t rmap_flags = RMAP_NONE;
>>> +	bool need_clear_cache = false;
>>>    	bool exclusive = false;
>>>    	swp_entry_t entry;
>>>    	pte_t pte;
>>> @@ -3867,6 +3868,20 @@ 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)) {
>>> +				/* Relax a bit to prevent rapid repeated page faults */
>>> +				schedule();
>>> +				goto out;
>>> +			}
>>> +			need_clear_cache = true;
>>> +
>>
>> I took a closer look at __read_swap_cache_async() and it essentially
>> does something similar.
>>
>> Instead of returning, it keeps retrying until it finds that
>> swapcache_prepare() fails for another reason than -EEXISTS (e.g.,
>> freed concurrently) or it finds the entry in the swapcache.
>>
>> So if you would succeed here on a freed+reused swap entry,
>> __read_swap_cache_async() would simply retry.
>>
>> It spells that out:
>>
>> 		/*
>> 		 * We might race against __delete_from_swap_cache(), and
>> 		 * stumble across a swap_map entry whose SWAP_HAS_CACHE
>> 		 * has not yet been cleared.  Or race against another
>> 		 * __read_swap_cache_async(), which has set SWAP_HAS_CACHE
>> 		 * in swap_map, but not yet added its folio to swap cache.
>> 		 */
>>
>> Whereby we could not race against this code here as well where we
>> speculatively set SWAP_HAS_CACHE and might never add something to the swap
>> cache.
>>
>>
>> I'd probably avoid the wrong returns and do something even closer to
>> __read_swap_cache_async().
>>
>> while (true) {
>> 	/*
>> 	 * Fake that we are trying to insert a page into the swapcache, to
>> 	 * serialize against concurrent threads wanting to do the same.
>> 	 * [more from your description]
>> 	 */
>> 	ret = swapcache_prepare(entry);
>> 	if (likely(!ret)
>> 		/*
>> 		 * Move forward with swapin, we'll recheck if the PTE hasn't
>> 		 * changed later.
>> 		 */
>> 		break;
>> 	else if (ret != -EEXIST)
>> 		goto out;
> 
> The swap entry may be kept in swap cache for long time.  For example, it
> may be read into swap cache via MADV_WILLNEED.

Right, we'd have to check for the swapcache.

I briefly thought about just factoring out what we have in 
__read_swap_cache_async() and reusing here. Similar problem to solve, 
and quite a lot of duplicate code.

But not worth the churn in a simple fix. We could explore that option
as a cleanup on top.

Patch

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..7059230d0a54 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3799,6 +3799,7 @@  vm_fault_t do_swap_page(struct vm_fault *vmf)
 	struct page *page;
 	struct swap_info_struct *si = NULL;
 	rmap_t rmap_flags = RMAP_NONE;
+	bool need_clear_cache = false;
 	bool exclusive = false;
 	swp_entry_t entry;
 	pte_t pte;
@@ -3867,6 +3868,20 @@  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)) {
+				/* Relax a bit to prevent rapid repeated page faults */
+				schedule();
+				goto out;
+			}
+			need_clear_cache = true;
+
 			/* skip swapcache */
 			folio = vma_alloc_folio(GFP_HIGHUSER_MOVABLE, 0,
 						vma, vmf->address, false);
@@ -4117,6 +4132,9 @@  vm_fault_t do_swap_page(struct vm_fault *vmf)
 	if (vmf->pte)
 		pte_unmap_unlock(vmf->pte, vmf->ptl);
 out:
+	/* Clear the swap cache pin for direct swapin after PTL unlock */
+	if (need_clear_cache)
+		swapcache_clear(si, entry);
 	if (si)
 		put_swap_device(si);
 	return ret;
@@ -4131,6 +4149,8 @@  vm_fault_t do_swap_page(struct vm_fault *vmf)
 		folio_unlock(swapcache);
 		folio_put(swapcache);
 	}
+	if (need_clear_cache)
+		swapcache_clear(si, entry);
 	if (si)
 		put_swap_device(si);
 	return ret;
diff --git a/mm/swap.h b/mm/swap.h
index 758c46ca671e..fc2f6ade7f80 100644
--- a/mm/swap.h
+++ b/mm/swap.h
@@ -41,6 +41,7 @@  void __delete_from_swap_cache(struct folio *folio,
 void delete_from_swap_cache(struct folio *folio);
 void clear_shadow_from_swap_cache(int type, unsigned long begin,
 				  unsigned long end);
+void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry);
 struct folio *swap_cache_get_folio(swp_entry_t entry,
 		struct vm_area_struct *vma, unsigned long addr);
 struct folio *filemap_get_incore_folio(struct address_space *mapping,
@@ -97,6 +98,10 @@  static inline int swap_writepage(struct page *p, struct writeback_control *wbc)
 	return 0;
 }
 
+static inline void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry)
+{
+}
+
 static inline struct folio *swap_cache_get_folio(swp_entry_t entry,
 		struct vm_area_struct *vma, unsigned long addr)
 {
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 556ff7347d5f..746aa9da5302 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -3365,6 +3365,19 @@  int swapcache_prepare(swp_entry_t entry)
 	return __swap_duplicate(entry, SWAP_HAS_CACHE);
 }
 
+void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry)
+{
+	struct swap_cluster_info *ci;
+	unsigned long offset = swp_offset(entry);
+	unsigned char usage;
+
+	ci = lock_cluster_or_swap_info(si, offset);
+	usage = __swap_entry_free_locked(si, offset, SWAP_HAS_CACHE);
+	unlock_cluster_or_swap_info(si, ci);
+	if (!usage)
+		free_swap_slot(entry);
+}
+
 struct swap_info_struct *swp_swap_info(swp_entry_t entry)
 {
 	return swap_type_to_swap_info(swp_type(entry));