Re: [PATCH v4 03/10] fs: Introduce ->corrupted_range() for superblock

Re: [PATCH v4 03/10] fs: Introduce ->corrupted_range() for superblock

[Dan Williams]

[ drop old linux-nvdimm@lists.01.org, add nvdimm@lists.linux.dev ]

On Thu, Jun 3, 2021 at 6:19 PM Shiyang Ruan <ruansy.fnst@fujitsu.com> wrote:
>
> Memory failure occurs in fsdax mode will finally be handled in
> filesystem.  We introduce this interface to find out files or metadata
> affected by the corrupted range, and try to recover the corrupted data
> if possiable.
>
> Signed-off-by: Shiyang Ruan <ruansy.fnst@fujitsu.com>
> ---
>  include/linux/fs.h | 2 ++
>  1 file changed, 2 insertions(+)
>
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index c3c88fdb9b2a..92af36c4225f 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -2176,6 +2176,8 @@ struct super_operations {
>                                   struct shrink_control *);
>         long (*free_cached_objects)(struct super_block *,
>                                     struct shrink_control *);
> +       int (*corrupted_range)(struct super_block *sb, struct block_device *bdev,
> +                              loff_t offset, size_t len, void *data);

Why does the superblock need a new operation? Wouldn't whatever
function is specified here just be specified to the dax_dev as the
->notify_failure() holder callback?

RE: [PATCH v4 03/10] fs: Introduce ->corrupted_range() for superblock

[ruansy.fnst]

> -----Original Message-----
> From: Dan Williams <dan.j.williams@intel.com>
> Subject: Re: [PATCH v4 03/10] fs: Introduce ->corrupted_range() for superblock
> 
> [ drop old linux-nvdimm@lists.01.org, add nvdimm@lists.linux.dev ]
> 
> On Thu, Jun 3, 2021 at 6:19 PM Shiyang Ruan <ruansy.fnst@fujitsu.com> wrote:
> >
> > Memory failure occurs in fsdax mode will finally be handled in
> > filesystem.  We introduce this interface to find out files or metadata
> > affected by the corrupted range, and try to recover the corrupted data
> > if possiable.
> >
> > Signed-off-by: Shiyang Ruan <ruansy.fnst@fujitsu.com>
> > ---
> >  include/linux/fs.h | 2 ++
> >  1 file changed, 2 insertions(+)
> >
> > diff --git a/include/linux/fs.h b/include/linux/fs.h index
> > c3c88fdb9b2a..92af36c4225f 100644
> > --- a/include/linux/fs.h
> > +++ b/include/linux/fs.h
> > @@ -2176,6 +2176,8 @@ struct super_operations {
> >                                   struct shrink_control *);
> >         long (*free_cached_objects)(struct super_block *,
> >                                     struct shrink_control *);
> > +       int (*corrupted_range)(struct super_block *sb, struct block_device
> *bdev,
> > +                              loff_t offset, size_t len, void *data);
> 
> Why does the superblock need a new operation? Wouldn't whatever function is
> specified here just be specified to the dax_dev as the
> ->notify_failure() holder callback?

Because we need to find out which file is effected by the given poison page so that memory-failure code can do collect_procs() and kill_procs() jobs.  And it needs filesystem to use its rmap feature to search the file from a given offset.  So, we need this implemented by the specified filesystem and called by dax_device's holder.

This is the call trace I described in cover letter:
memory_failure()
 * fsdax case
 pgmap->ops->memory_failure()      => pmem_pgmap_memory_failure()
  dax_device->holder_ops->corrupted_range() =>
                                      - fs_dax_corrupted_range()
                                      - md_dax_corrupted_range()
   sb->s_ops->currupted_range()    => xfs_fs_corrupted_range()  <== **HERE**
    xfs_rmap_query_range()
     xfs_currupt_helper()
      * corrupted on metadata
          try to recover data, call xfs_force_shutdown()
      * corrupted on file data
          try to recover data, call mf_dax_kill_procs()
 * normal case
 mf_generic_kill_procs()

As you can see, this new added operation is an important for the whole progress.


--
Thanks,
Ruan.

Re: [PATCH v4 03/10] fs: Introduce ->corrupted_range() for superblock

[Dan Williams]

On Wed, Jun 16, 2021 at 11:51 PM ruansy.fnst@fujitsu.com
<ruansy.fnst@fujitsu.com> wrote:
>
> > -----Original Message-----
> > From: Dan Williams <dan.j.williams@intel.com>
> > Subject: Re: [PATCH v4 03/10] fs: Introduce ->corrupted_range() for superblock
> >
> > [ drop old linux-nvdimm@lists.01.org, add nvdimm@lists.linux.dev ]
> >
> > On Thu, Jun 3, 2021 at 6:19 PM Shiyang Ruan <ruansy.fnst@fujitsu.com> wrote:
> > >
> > > Memory failure occurs in fsdax mode will finally be handled in
> > > filesystem.  We introduce this interface to find out files or metadata
> > > affected by the corrupted range, and try to recover the corrupted data
> > > if possiable.
> > >
> > > Signed-off-by: Shiyang Ruan <ruansy.fnst@fujitsu.com>
> > > ---
> > >  include/linux/fs.h | 2 ++
> > >  1 file changed, 2 insertions(+)
> > >
> > > diff --git a/include/linux/fs.h b/include/linux/fs.h index
> > > c3c88fdb9b2a..92af36c4225f 100644
> > > --- a/include/linux/fs.h
> > > +++ b/include/linux/fs.h
> > > @@ -2176,6 +2176,8 @@ struct super_operations {
> > >                                   struct shrink_control *);
> > >         long (*free_cached_objects)(struct super_block *,
> > >                                     struct shrink_control *);
> > > +       int (*corrupted_range)(struct super_block *sb, struct block_device
> > *bdev,
> > > +                              loff_t offset, size_t len, void *data);
> >
> > Why does the superblock need a new operation? Wouldn't whatever function is
> > specified here just be specified to the dax_dev as the
> > ->notify_failure() holder callback?
>
> Because we need to find out which file is effected by the given poison page so that memory-failure code can do collect_procs() and kill_procs() jobs.  And it needs filesystem to use its rmap feature to search the file from a given offset.  So, we need this implemented by the specified filesystem and called by dax_device's holder.
>
> This is the call trace I described in cover letter:
> memory_failure()
>  * fsdax case
>  pgmap->ops->memory_failure()      => pmem_pgmap_memory_failure()
>   dax_device->holder_ops->corrupted_range() =>
>                                       - fs_dax_corrupted_range()
>                                       - md_dax_corrupted_range()
>    sb->s_ops->currupted_range()    => xfs_fs_corrupted_range()  <== **HERE**
>     xfs_rmap_query_range()
>      xfs_currupt_helper()
>       * corrupted on metadata
>           try to recover data, call xfs_force_shutdown()
>       * corrupted on file data
>           try to recover data, call mf_dax_kill_procs()
>  * normal case
>  mf_generic_kill_procs()
>
> As you can see, this new added operation is an important for the whole progress.

I don't think you need either fs_dax_corrupted_range() nor
sb->s_ops->corrupted_range(). In fact that fs_dax_corrupted_range()
looks broken because the filesystem may not even be mounted on the
device associated with the error. The holder_data and holder_op should
be sufficient from communicating the stack of notifications:

pgmap->notify_memory_failure() => pmem_pgmap_notify_failure()
pmem_dax_dev->holder_ops->notify_failure(pmem_dax_dev) =>
md_dax_notify_failure()
md_dax_dev->holder_ops->notify_failure() => xfs_notify_failure()

I.e. the entire chain just walks dax_dev holder ops.

RE: [PATCH v4 03/10] fs: Introduce ->corrupted_range() for superblock

[ruansy.fnst]

> -----Original Message-----
> From: Dan Williams <dan.j.williams@intel.com>
> Subject: Re: [PATCH v4 03/10] fs: Introduce ->corrupted_range() for superblock
> 
> On Wed, Jun 16, 2021 at 11:51 PM ruansy.fnst@fujitsu.com
> <ruansy.fnst@fujitsu.com> wrote:
> >
> > > -----Original Message-----
> > > From: Dan Williams <dan.j.williams@intel.com>
> > > Subject: Re: [PATCH v4 03/10] fs: Introduce ->corrupted_range() for
> > > superblock
> > >
> > > [ drop old linux-nvdimm@lists.01.org, add nvdimm@lists.linux.dev ]
> > >
> > > On Thu, Jun 3, 2021 at 6:19 PM Shiyang Ruan <ruansy.fnst@fujitsu.com>
> wrote:
> > > >
> > > > Memory failure occurs in fsdax mode will finally be handled in
> > > > filesystem.  We introduce this interface to find out files or
> > > > metadata affected by the corrupted range, and try to recover the
> > > > corrupted data if possiable.
> > > >
> > > > Signed-off-by: Shiyang Ruan <ruansy.fnst@fujitsu.com>
> > > > ---
> > > >  include/linux/fs.h | 2 ++
> > > >  1 file changed, 2 insertions(+)
> > > >
> > > > diff --git a/include/linux/fs.h b/include/linux/fs.h index
> > > > c3c88fdb9b2a..92af36c4225f 100644
> > > > --- a/include/linux/fs.h
> > > > +++ b/include/linux/fs.h
> > > > @@ -2176,6 +2176,8 @@ struct super_operations {
> > > >                                   struct shrink_control *);
> > > >         long (*free_cached_objects)(struct super_block *,
> > > >                                     struct shrink_control *);
> > > > +       int (*corrupted_range)(struct super_block *sb, struct
> > > > + block_device
> > > *bdev,
> > > > +                              loff_t offset, size_t len, void
> > > > + *data);
> > >
> > > Why does the superblock need a new operation? Wouldn't whatever
> > > function is specified here just be specified to the dax_dev as the
> > > ->notify_failure() holder callback?
> >
> > Because we need to find out which file is effected by the given poison page so
> that memory-failure code can do collect_procs() and kill_procs() jobs.  And it
> needs filesystem to use its rmap feature to search the file from a given offset.
> So, we need this implemented by the specified filesystem and called by
> dax_device's holder.
> >
> > This is the call trace I described in cover letter:
> > memory_failure()
> >  * fsdax case
> >  pgmap->ops->memory_failure()      => pmem_pgmap_memory_failure()
> >   dax_device->holder_ops->corrupted_range() =>
> >                                       - fs_dax_corrupted_range()
> >                                       - md_dax_corrupted_range()
> >    sb->s_ops->currupted_range()    => xfs_fs_corrupted_range()  <==
> **HERE**
> >     xfs_rmap_query_range()
> >      xfs_currupt_helper()
> >       * corrupted on metadata
> >           try to recover data, call xfs_force_shutdown()
> >       * corrupted on file data
> >           try to recover data, call mf_dax_kill_procs()
> >  * normal case
> >  mf_generic_kill_procs()
> >
> > As you can see, this new added operation is an important for the whole
> progress.
> 
> I don't think you need either fs_dax_corrupted_range() nor
> sb->s_ops->corrupted_range(). In fact that fs_dax_corrupted_range()
> looks broken because the filesystem may not even be mounted on the device
> associated with the error. 

If filesystem is not mounted, then there won't be any process using the broken page and no one need to be killed in memory-failure.  So, I think we can just return and handle the error on driver level if needed.

> The holder_data and holder_op should be sufficient
> from communicating the stack of notifications:
> 
> pgmap->notify_memory_failure() => pmem_pgmap_notify_failure()
> pmem_dax_dev->holder_ops->notify_failure(pmem_dax_dev) =>
> md_dax_notify_failure()
> md_dax_dev->holder_ops->notify_failure() => xfs_notify_failure()
> 
> I.e. the entire chain just walks dax_dev holder ops.

Oh, I see.  Just need to implement holder_ops in filesystem or mapped_device directly.  I made the routine complicated.


--
Thanks,
Ruan.

Re: [PATCH v4 04/10] mm, fsdax: Refactor memory-failure handler for dax mapping

[Dan Williams]

[ drop old nvdimm list, add the new one ]

On Thu, Jun 3, 2021 at 6:19 PM Shiyang Ruan <ruansy.fnst@fujitsu.com> wrote:
>
> The current memory_failure_dev_pagemap() can only handle single-mapped
> dax page for fsdax mode.  The dax page could be mapped by multiple files
> and offsets if we let reflink feature & fsdax mode work together.  So,
> we refactor current implementation to support handle memory failure on
> each file and offset.

I don't understand this organization, perhaps because this patch
introduces mf_dax_kill_procs() without a user. However, my expectation
is that memory_failure() is mostly untouched save for an early check
via pgmap->notify_memory_failure(). If pgmap->notify_memory_failure()
indicates that the memory failure was handled by the pgmap->owner /
dax_dev holder stack, then the typical memory failure path is
short-circuited. Otherwise, for non-reflink filesystems where
page->mapping() is valid the legacy / existing memory_failure()
operates as it does currently. If reflink capable filesystems want to
share a common implementation to map pfns to files they can, but I
don't think that common code belongs in mm/memory-failure.c.

>
> Signed-off-by: Shiyang Ruan <ruansy.fnst@fujitsu.com>
> ---
>  fs/dax.c            |  21 ++++++++
>  include/linux/dax.h |   1 +
>  include/linux/mm.h  |   9 ++++
>  mm/memory-failure.c | 114 ++++++++++++++++++++++++++++----------------
>  4 files changed, 105 insertions(+), 40 deletions(-)
>
> diff --git a/fs/dax.c b/fs/dax.c
> index 62352cbcf0f4..58faca85455a 100644
> --- a/fs/dax.c
> +++ b/fs/dax.c
> @@ -389,6 +389,27 @@ static struct page *dax_busy_page(void *entry)
>         return NULL;
>  }
>
> +/*
> + * dax_load_pfn - Load pfn of the DAX entry corresponding to a page
> + * @mapping: The file whose entry we want to load
> + * @index:   The offset where the DAX entry located in
> + *
> + * Return:   pfn of the DAX entry
> + */
> +unsigned long dax_load_pfn(struct address_space *mapping, unsigned long index)
> +{
> +       XA_STATE(xas, &mapping->i_pages, index);
> +       void *entry;
> +       unsigned long pfn;
> +
> +       xas_lock_irq(&xas);
> +       entry = xas_load(&xas);
> +       pfn = dax_to_pfn(entry);
> +       xas_unlock_irq(&xas);

This looks racy, what happened to the locking afforded by dax_lock_page()?

> +
> +       return pfn;
> +}
> +
>  /*
>   * dax_lock_mapping_entry - Lock the DAX entry corresponding to a page
>   * @page: The page whose entry we want to lock
> diff --git a/include/linux/dax.h b/include/linux/dax.h
> index 1ce343a960ab..6e758daa5004 100644
> --- a/include/linux/dax.h
> +++ b/include/linux/dax.h
> @@ -158,6 +158,7 @@ int dax_writeback_mapping_range(struct address_space *mapping,
>
>  struct page *dax_layout_busy_page(struct address_space *mapping);
>  struct page *dax_layout_busy_page_range(struct address_space *mapping, loff_t start, loff_t end);
> +unsigned long dax_load_pfn(struct address_space *mapping, unsigned long index);
>  dax_entry_t dax_lock_page(struct page *page);
>  void dax_unlock_page(struct page *page, dax_entry_t cookie);
>  #else
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index c274f75efcf9..2b7527e93c77 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -1187,6 +1187,14 @@ static inline bool is_device_private_page(const struct page *page)
>                 page->pgmap->type == MEMORY_DEVICE_PRIVATE;
>  }
>
> +static inline bool is_device_fsdax_page(const struct page *page)
> +{
> +       return IS_ENABLED(CONFIG_DEV_PAGEMAP_OPS) &&
> +               IS_ENABLED(CONFIG_FS_DAX) &&
> +               is_zone_device_page(page) &&
> +               page->pgmap->type == MEMORY_DEVICE_FS_DAX;

Why is this necessary? The dax_dev holder is the one that knows the
nature of the pfn. The common memory_failure() code should not care
about fsdax vs devdax.

> +}
> +
>  static inline bool is_pci_p2pdma_page(const struct page *page)
>  {
>         return IS_ENABLED(CONFIG_DEV_PAGEMAP_OPS) &&
> @@ -3078,6 +3086,7 @@ enum mf_flags {
>         MF_MUST_KILL = 1 << 2,
>         MF_SOFT_OFFLINE = 1 << 3,
>  };
> +int mf_dax_kill_procs(struct address_space *mapping, pgoff_t index, int flags);
>  extern int memory_failure(unsigned long pfn, int flags);
>  extern void memory_failure_queue(unsigned long pfn, int flags);
>  extern void memory_failure_queue_kick(int cpu);
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> index 85ad98c00fd9..4377e727d478 100644
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -56,6 +56,7 @@
>  #include <linux/kfifo.h>
>  #include <linux/ratelimit.h>
>  #include <linux/page-isolation.h>
> +#include <linux/dax.h>
>  #include "internal.h"
>  #include "ras/ras_event.h"
>
> @@ -120,6 +121,13 @@ static int hwpoison_filter_dev(struct page *p)
>         if (PageSlab(p))
>                 return -EINVAL;
>
> +       if (pfn_valid(page_to_pfn(p))) {
> +               if (is_device_fsdax_page(p))

This is racy unless the page is pinned. Also, not clear why this is needed?

> +                       return 0;
> +               else
> +                       return -EINVAL;
> +       }
> +
>         mapping = page_mapping(p);
>         if (mapping == NULL || mapping->host == NULL)
>                 return -EINVAL;
> @@ -290,10 +298,9 @@ void shake_page(struct page *p, int access)
>  }
>  EXPORT_SYMBOL_GPL(shake_page);
>
> -static unsigned long dev_pagemap_mapping_shift(struct page *page,
> -               struct vm_area_struct *vma)
> +static unsigned long dev_pagemap_mapping_shift(struct vm_area_struct *vma,
> +                                              unsigned long address)
>  {
> -       unsigned long address = vma_address(page, vma);
>         pgd_t *pgd;
>         p4d_t *p4d;
>         pud_t *pud;
> @@ -333,9 +340,8 @@ static unsigned long dev_pagemap_mapping_shift(struct page *page,
>   * Schedule a process for later kill.
>   * Uses GFP_ATOMIC allocations to avoid potential recursions in the VM.
>   */
> -static void add_to_kill(struct task_struct *tsk, struct page *p,
> -                      struct vm_area_struct *vma,
> -                      struct list_head *to_kill)
> +static void add_to_kill(struct task_struct *tsk, struct page *p, pgoff_t pgoff,
> +                       struct vm_area_struct *vma, struct list_head *to_kill)
>  {
>         struct to_kill *tk;
>
> @@ -346,9 +352,12 @@ static void add_to_kill(struct task_struct *tsk, struct page *p,
>         }
>
>         tk->addr = page_address_in_vma(p, vma);
> -       if (is_zone_device_page(p))
> -               tk->size_shift = dev_pagemap_mapping_shift(p, vma);
> -       else
> +       if (is_zone_device_page(p)) {
> +               if (is_device_fsdax_page(p))
> +                       tk->addr = vma->vm_start +
> +                                       ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
> +               tk->size_shift = dev_pagemap_mapping_shift(vma, tk->addr);

What was wrong with the original code?

> +       } else
>                 tk->size_shift = page_shift(compound_head(p));
>
>         /*
> @@ -496,7 +505,7 @@ static void collect_procs_anon(struct page *page, struct list_head *to_kill,
>                         if (!page_mapped_in_vma(page, vma))
>                                 continue;
>                         if (vma->vm_mm == t->mm)
> -                               add_to_kill(t, page, vma, to_kill);
> +                               add_to_kill(t, page, 0, vma, to_kill);
>                 }
>         }
>         read_unlock(&tasklist_lock);
> @@ -506,24 +515,19 @@ static void collect_procs_anon(struct page *page, struct list_head *to_kill,
>  /*
>   * Collect processes when the error hit a file mapped page.
>   */
> -static void collect_procs_file(struct page *page, struct list_head *to_kill,
> -                               int force_early)
> +static void collect_procs_file(struct page *page, struct address_space *mapping,
> +               pgoff_t pgoff, struct list_head *to_kill, int force_early)
>  {

collect_procs() and kill_procs() are the only core memory_failure()
helpers I expect would be exported for a fileystem dax_dev holder to
call when it is trying to cleanup a memory_failure() on a reflink'd
mapping.

>         struct vm_area_struct *vma;
>         struct task_struct *tsk;
> -       struct address_space *mapping = page->mapping;
> -       pgoff_t pgoff;
>
>         i_mmap_lock_read(mapping);
>         read_lock(&tasklist_lock);
> -       pgoff = page_to_pgoff(page);
>         for_each_process(tsk) {
>                 struct task_struct *t = task_early_kill(tsk, force_early);
> -
>                 if (!t)
>                         continue;
> -               vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff,
> -                                     pgoff) {
> +               vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff, pgoff) {
>                         /*
>                          * Send early kill signal to tasks where a vma covers
>                          * the page but the corrupted page is not necessarily
> @@ -532,7 +536,7 @@ static void collect_procs_file(struct page *page, struct list_head *to_kill,
>                          * to be informed of all such data corruptions.
>                          */
>                         if (vma->vm_mm == t->mm)
> -                               add_to_kill(t, page, vma, to_kill);
> +                               add_to_kill(t, page, pgoff, vma, to_kill);
>                 }
>         }
>         read_unlock(&tasklist_lock);
> @@ -551,7 +555,8 @@ static void collect_procs(struct page *page, struct list_head *tokill,
>         if (PageAnon(page))
>                 collect_procs_anon(page, tokill, force_early);
>         else
> -               collect_procs_file(page, tokill, force_early);
> +               collect_procs_file(page, page_mapping(page), page_to_pgoff(page),
> +                                  tokill, force_early);
>  }
>
>  static const char *action_name[] = {
> @@ -1218,6 +1223,51 @@ static int try_to_split_thp_page(struct page *page, const char *msg)
>         return 0;
>  }
>
> +static void unmap_and_kill(struct list_head *to_kill, unsigned long pfn,
> +               struct address_space *mapping, pgoff_t index, int flags)
> +{
> +       struct to_kill *tk;
> +       unsigned long size = 0;
> +       loff_t start;
> +
> +       list_for_each_entry(tk, to_kill, nd)
> +               if (tk->size_shift)
> +                       size = max(size, 1UL << tk->size_shift);
> +       if (size) {
> +               /*
> +                * Unmap the largest mapping to avoid breaking up
> +                * device-dax mappings which are constant size. The
> +                * actual size of the mapping being torn down is
> +                * communicated in siginfo, see kill_proc()
> +                */
> +               start = (index << PAGE_SHIFT) & ~(size - 1);
> +               unmap_mapping_range(mapping, start, size, 0);
> +       }
> +
> +       kill_procs(to_kill, flags & MF_MUST_KILL, false, pfn, flags);
> +}
> +
> +int mf_dax_kill_procs(struct address_space *mapping, pgoff_t index, int flags)
> +{
> +       LIST_HEAD(to_kill);
> +       /* load the pfn of the dax mapping file */
> +       unsigned long pfn = dax_load_pfn(mapping, index);
> +
> +       /*
> +        * Unlike System-RAM there is no possibility to swap in a
> +        * different physical page at a given virtual address, so all
> +        * userspace consumption of ZONE_DEVICE memory necessitates
> +        * SIGBUS (i.e. MF_MUST_KILL)
> +        */
> +       flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
> +       collect_procs_file(pfn_to_page(pfn), mapping, index, &to_kill,
> +                          flags & MF_ACTION_REQUIRED);
> +
> +       unmap_and_kill(&to_kill, pfn, mapping, index, flags);
> +       return 0;
> +}
> +EXPORT_SYMBOL_GPL(mf_dax_kill_procs);
> +
>  static int memory_failure_hugetlb(unsigned long pfn, int flags)
>  {
>         struct page *p = pfn_to_page(pfn);
> @@ -1298,12 +1348,8 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
>                 struct dev_pagemap *pgmap)
>  {
>         struct page *page = pfn_to_page(pfn);
> -       const bool unmap_success = true;
> -       unsigned long size = 0;
> -       struct to_kill *tk;
> -       LIST_HEAD(tokill);
> +       LIST_HEAD(to_kill);
>         int rc = -EBUSY;
> -       loff_t start;
>         dax_entry_t cookie;
>
>         if (flags & MF_COUNT_INCREASED)
> @@ -1355,22 +1401,10 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
>          * SIGBUS (i.e. MF_MUST_KILL)
>          */
>         flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
> -       collect_procs(page, &tokill, flags & MF_ACTION_REQUIRED);
> +       collect_procs_file(page, page->mapping, page->index, &to_kill,
> +                          flags & MF_ACTION_REQUIRED);
>
> -       list_for_each_entry(tk, &tokill, nd)
> -               if (tk->size_shift)
> -                       size = max(size, 1UL << tk->size_shift);
> -       if (size) {
> -               /*
> -                * Unmap the largest mapping to avoid breaking up
> -                * device-dax mappings which are constant size. The
> -                * actual size of the mapping being torn down is
> -                * communicated in siginfo, see kill_proc()
> -                */
> -               start = (page->index << PAGE_SHIFT) & ~(size - 1);
> -               unmap_mapping_range(page->mapping, start, size, 0);
> -       }
> -       kill_procs(&tokill, flags & MF_MUST_KILL, !unmap_success, pfn, flags);
> +       unmap_and_kill(&to_kill, pfn, page->mapping, page->index, flags);

There's just too much change in this patch and not enough
justification of what is being refactored and why.

RE: [PATCH v4 04/10] mm, fsdax: Refactor memory-failure handler for dax mapping

[ruansy.fnst]

> -----Original Message-----
> From: Dan Williams <dan.j.williams@intel.com>
> Subject: Re: [PATCH v4 04/10] mm, fsdax: Refactor memory-failure handler for
> dax mapping
> 
> [ drop old nvdimm list, add the new one ]
> 
> On Thu, Jun 3, 2021 at 6:19 PM Shiyang Ruan <ruansy.fnst@fujitsu.com> wrote:
> >
> > The current memory_failure_dev_pagemap() can only handle single-mapped
> > dax page for fsdax mode.  The dax page could be mapped by multiple
> > files and offsets if we let reflink feature & fsdax mode work
> > together.  So, we refactor current implementation to support handle
> > memory failure on each file and offset.
> 
> I don't understand this organization, perhaps because this patch introduces
> mf_dax_kill_procs() without a user. 

Yes, I think I made it a mess... I should reorganize this whole patchset.

The mf_dax_kill_procs() is used by xfs in patch 9.  I was mean to refactor these code for the next patches usage.

> However, my expectation is that
> memory_failure() is mostly untouched save for an early check via
> pgmap->notify_memory_failure(). If pgmap->notify_memory_failure() indicates
> that the memory failure was handled by the pgmap->owner / dax_dev holder
> stack, then the typical memory failure path is short-circuited. Otherwise, for
> non-reflink filesystems where
> page->mapping() is valid the legacy / existing memory_failure()
> operates as it does currently.

You can find this logic in patch 5.

When it comes to memory-failure() and memory_failure_dev_pagemap(), after some check, it will try to call pgmap->ops->memory_failure().  If this interface is implemented, for example pgmap is pmem device, it will call the dax_dev holder stack. And finally, it comes to mf_dax_kill_procs().
However, if something wrong happens in this stack, such as feature not support or interface not implemented, it will roll back to normal memory-failure hanlder which is refactored as mf_generic_kill_porcs().

So, I think we are in agreement on this.

Let me reorganize these code.

> If reflink capable filesystems want to share a
> common implementation to map pfns to files they can, but I don't think that
> common code belongs in mm/memory-failure.c.
> 
> >
> > Signed-off-by: Shiyang Ruan <ruansy.fnst@fujitsu.com>
> > ---
> >  fs/dax.c            |  21 ++++++++
> >  include/linux/dax.h |   1 +
> >  include/linux/mm.h  |   9 ++++
> >  mm/memory-failure.c | 114
> > ++++++++++++++++++++++++++++----------------
> >  4 files changed, 105 insertions(+), 40 deletions(-)
> >
> > diff --git a/fs/dax.c b/fs/dax.c
> > index 62352cbcf0f4..58faca85455a 100644
> > --- a/fs/dax.c
> > +++ b/fs/dax.c
> > @@ -389,6 +389,27 @@ static struct page *dax_busy_page(void *entry)
> >         return NULL;
> >  }
> >
> > +/*
> > + * dax_load_pfn - Load pfn of the DAX entry corresponding to a page
> > + * @mapping: The file whose entry we want to load
> > + * @index:   The offset where the DAX entry located in
> > + *
> > + * Return:   pfn of the DAX entry
> > + */
> > +unsigned long dax_load_pfn(struct address_space *mapping, unsigned
> > +long index) {
> > +       XA_STATE(xas, &mapping->i_pages, index);
> > +       void *entry;
> > +       unsigned long pfn;
> > +
> > +       xas_lock_irq(&xas);
> > +       entry = xas_load(&xas);
> > +       pfn = dax_to_pfn(entry);
> > +       xas_unlock_irq(&xas);
> 
> This looks racy, what happened to the locking afforded by dax_lock_page()?
> 
> > +
> > +       return pfn;
> > +}
> > +
> >  /*
> >   * dax_lock_mapping_entry - Lock the DAX entry corresponding to a page
> >   * @page: The page whose entry we want to lock diff --git
> > a/include/linux/dax.h b/include/linux/dax.h index
> > 1ce343a960ab..6e758daa5004 100644
> > --- a/include/linux/dax.h
> > +++ b/include/linux/dax.h
> > @@ -158,6 +158,7 @@ int dax_writeback_mapping_range(struct
> > address_space *mapping,
> >
> >  struct page *dax_layout_busy_page(struct address_space *mapping);
> > struct page *dax_layout_busy_page_range(struct address_space *mapping,
> > loff_t start, loff_t end);
> > +unsigned long dax_load_pfn(struct address_space *mapping, unsigned
> > +long index);
> >  dax_entry_t dax_lock_page(struct page *page);  void
> > dax_unlock_page(struct page *page, dax_entry_t cookie);  #else diff
> > --git a/include/linux/mm.h b/include/linux/mm.h index
> > c274f75efcf9..2b7527e93c77 100644
> > --- a/include/linux/mm.h
> > +++ b/include/linux/mm.h
> > @@ -1187,6 +1187,14 @@ static inline bool is_device_private_page(const
> struct page *page)
> >                 page->pgmap->type == MEMORY_DEVICE_PRIVATE;  }
> >
> > +static inline bool is_device_fsdax_page(const struct page *page) {
> > +       return IS_ENABLED(CONFIG_DEV_PAGEMAP_OPS) &&
> > +               IS_ENABLED(CONFIG_FS_DAX) &&
> > +               is_zone_device_page(page) &&
> > +               page->pgmap->type == MEMORY_DEVICE_FS_DAX;
> 
> Why is this necessary? The dax_dev holder is the one that knows the nature of
> the pfn. The common memory_failure() code should not care about fsdax vs
> devdax.

add_to_kill() in collect_procs() needs this. Please see explanation at below.

> 
> > +}
> > +
> >  static inline bool is_pci_p2pdma_page(const struct page *page)  {
> >         return IS_ENABLED(CONFIG_DEV_PAGEMAP_OPS) && @@
> -3078,6
> > +3086,7 @@ enum mf_flags {
> >         MF_MUST_KILL = 1 << 2,
> >         MF_SOFT_OFFLINE = 1 << 3,
> >  };
> > +int mf_dax_kill_procs(struct address_space *mapping, pgoff_t index,
> > +int flags);
> >  extern int memory_failure(unsigned long pfn, int flags);  extern void
> > memory_failure_queue(unsigned long pfn, int flags);  extern void
> > memory_failure_queue_kick(int cpu); diff --git a/mm/memory-failure.c
> > b/mm/memory-failure.c index 85ad98c00fd9..4377e727d478 100644
> > --- a/mm/memory-failure.c
> > +++ b/mm/memory-failure.c
> > @@ -56,6 +56,7 @@
> >  #include <linux/kfifo.h>
> >  #include <linux/ratelimit.h>
> >  #include <linux/page-isolation.h>
> > +#include <linux/dax.h>
> >  #include "internal.h"
> >  #include "ras/ras_event.h"
> >
> > @@ -120,6 +121,13 @@ static int hwpoison_filter_dev(struct page *p)
> >         if (PageSlab(p))
> >                 return -EINVAL;
> >
> > +       if (pfn_valid(page_to_pfn(p))) {
> > +               if (is_device_fsdax_page(p))
> 
> This is racy unless the page is pinned. Also, not clear why this is needed?
> 
> > +                       return 0;
> > +               else
> > +                       return -EINVAL;
> > +       }
> > +
> >         mapping = page_mapping(p);
> >         if (mapping == NULL || mapping->host == NULL)
> >                 return -EINVAL;
> > @@ -290,10 +298,9 @@ void shake_page(struct page *p, int access)  }
> > EXPORT_SYMBOL_GPL(shake_page);
> >
> > -static unsigned long dev_pagemap_mapping_shift(struct page *page,
> > -               struct vm_area_struct *vma)
> > +static unsigned long dev_pagemap_mapping_shift(struct vm_area_struct
> *vma,
> > +                                              unsigned long
> address)
> >  {
> > -       unsigned long address = vma_address(page, vma);
> >         pgd_t *pgd;
> >         p4d_t *p4d;
> >         pud_t *pud;
> > @@ -333,9 +340,8 @@ static unsigned long
> dev_pagemap_mapping_shift(struct page *page,
> >   * Schedule a process for later kill.
> >   * Uses GFP_ATOMIC allocations to avoid potential recursions in the VM.
> >   */
> > -static void add_to_kill(struct task_struct *tsk, struct page *p,
> > -                      struct vm_area_struct *vma,
> > -                      struct list_head *to_kill)
> > +static void add_to_kill(struct task_struct *tsk, struct page *p, pgoff_t pgoff,
> > +                       struct vm_area_struct *vma, struct list_head
> > +*to_kill)
> >  {
> >         struct to_kill *tk;
> >
> > @@ -346,9 +352,12 @@ static void add_to_kill(struct task_struct *tsk, struct
> page *p,
> >         }
> >
> >         tk->addr = page_address_in_vma(p, vma);
> > -       if (is_zone_device_page(p))
> > -               tk->size_shift = dev_pagemap_mapping_shift(p, vma);
> > -       else
> > +       if (is_zone_device_page(p)) {
> > +               if (is_device_fsdax_page(p))
> > +                       tk->addr = vma->vm_start +
> > +                                       ((pgoff - vma->vm_pgoff) <<
> PAGE_SHIFT);
> > +               tk->size_shift = dev_pagemap_mapping_shift(vma,
> > + tk->addr);
> 
> What was wrong with the original code?

Here is the explanation: For normal page, it associate file's mapping and offset to page->mapping, page->index for rmap tracking.  But for fsdax, in order to support reflink, we no longer use this mechanism, using dax_device holder stack instead.  Thus, this dax page->mapping is NULL.  As a result, we need is_device_fsdax_page(p) to distinguish if a page is a fsdax page and calculate this tk->addr manually.

> 
> > +       } else
> >                 tk->size_shift = page_shift(compound_head(p));
> >
> >         /*
> > @@ -496,7 +505,7 @@ static void collect_procs_anon(struct page *page,
> struct list_head *to_kill,
> >                         if (!page_mapped_in_vma(page, vma))
> >                                 continue;
> >                         if (vma->vm_mm == t->mm)
> > -                               add_to_kill(t, page, vma, to_kill);
> > +                               add_to_kill(t, page, 0, vma, to_kill);
> >                 }
> >         }
> >         read_unlock(&tasklist_lock);
> > @@ -506,24 +515,19 @@ static void collect_procs_anon(struct page
> > *page, struct list_head *to_kill,
> >  /*
> >   * Collect processes when the error hit a file mapped page.
> >   */
> > -static void collect_procs_file(struct page *page, struct list_head *to_kill,
> > -                               int force_early)
> > +static void collect_procs_file(struct page *page, struct address_space
> *mapping,
> > +               pgoff_t pgoff, struct list_head *to_kill, int
> > +force_early)
> >  {
> 
> collect_procs() and kill_procs() are the only core memory_failure() helpers I
> expect would be exported for a fileystem dax_dev holder to call when it is trying
> to cleanup a memory_failure() on a reflink'd mapping.

Yes, they are the core we need.  But there are some small different when dealing with normal page and dax page.  So, I factor these two core functions into two helpers.  One is mf_generic_kill_procs() for normal page. Another is mf_dax_kill_procs() for dax page.

> 
> >         struct vm_area_struct *vma;
> >         struct task_struct *tsk;
> > -       struct address_space *mapping = page->mapping;
> > -       pgoff_t pgoff;
> >
> >         i_mmap_lock_read(mapping);
> >         read_lock(&tasklist_lock);
> > -       pgoff = page_to_pgoff(page);
> >         for_each_process(tsk) {
> >                 struct task_struct *t = task_early_kill(tsk,
> > force_early);
> > -
> >                 if (!t)
> >                         continue;
> > -               vma_interval_tree_foreach(vma, &mapping->i_mmap,
> pgoff,
> > -                                     pgoff) {
> > +               vma_interval_tree_foreach(vma, &mapping->i_mmap,
> > + pgoff, pgoff) {
> >                         /*
> >                          * Send early kill signal to tasks where a vma
> covers
> >                          * the page but the corrupted page is not
> > necessarily @@ -532,7 +536,7 @@ static void collect_procs_file(struct page
> *page, struct list_head *to_kill,
> >                          * to be informed of all such data corruptions.
> >                          */
> >                         if (vma->vm_mm == t->mm)
> > -                               add_to_kill(t, page, vma, to_kill);
> > +                               add_to_kill(t, page, pgoff, vma,
> > + to_kill);
> >                 }
> >         }
> >         read_unlock(&tasklist_lock);
> > @@ -551,7 +555,8 @@ static void collect_procs(struct page *page, struct
> list_head *tokill,
> >         if (PageAnon(page))
> >                 collect_procs_anon(page, tokill, force_early);
> >         else
> > -               collect_procs_file(page, tokill, force_early);
> > +               collect_procs_file(page, page_mapping(page),
> page_to_pgoff(page),
> > +                                  tokill, force_early);
> >  }
> >
> >  static const char *action_name[] = {
> > @@ -1218,6 +1223,51 @@ static int try_to_split_thp_page(struct page *page,
> const char *msg)
> >         return 0;
> >  }
> >
> > +static void unmap_and_kill(struct list_head *to_kill, unsigned long pfn,
> > +               struct address_space *mapping, pgoff_t index, int
> > +flags) {
> > +       struct to_kill *tk;
> > +       unsigned long size = 0;
> > +       loff_t start;
> > +
> > +       list_for_each_entry(tk, to_kill, nd)
> > +               if (tk->size_shift)
> > +                       size = max(size, 1UL << tk->size_shift);
> > +       if (size) {
> > +               /*
> > +                * Unmap the largest mapping to avoid breaking up
> > +                * device-dax mappings which are constant size. The
> > +                * actual size of the mapping being torn down is
> > +                * communicated in siginfo, see kill_proc()
> > +                */
> > +               start = (index << PAGE_SHIFT) & ~(size - 1);
> > +               unmap_mapping_range(mapping, start, size, 0);
> > +       }
> > +
> > +       kill_procs(to_kill, flags & MF_MUST_KILL, false, pfn, flags);
> > +}
> > +
> > +int mf_dax_kill_procs(struct address_space *mapping, pgoff_t index,
> > +int flags) {
> > +       LIST_HEAD(to_kill);
> > +       /* load the pfn of the dax mapping file */
> > +       unsigned long pfn = dax_load_pfn(mapping, index);
> > +
> > +       /*
> > +        * Unlike System-RAM there is no possibility to swap in a
> > +        * different physical page at a given virtual address, so all
> > +        * userspace consumption of ZONE_DEVICE memory necessitates
> > +        * SIGBUS (i.e. MF_MUST_KILL)
> > +        */
> > +       flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
> > +       collect_procs_file(pfn_to_page(pfn), mapping, index, &to_kill,
> > +                          flags & MF_ACTION_REQUIRED);
> > +
> > +       unmap_and_kill(&to_kill, pfn, mapping, index, flags);
> > +       return 0;
> > +}
> > +EXPORT_SYMBOL_GPL(mf_dax_kill_procs);
> > +
> >  static int memory_failure_hugetlb(unsigned long pfn, int flags)  {
> >         struct page *p = pfn_to_page(pfn); @@ -1298,12 +1348,8 @@
> > static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
> >                 struct dev_pagemap *pgmap)  {
> >         struct page *page = pfn_to_page(pfn);
> > -       const bool unmap_success = true;
> > -       unsigned long size = 0;
> > -       struct to_kill *tk;
> > -       LIST_HEAD(tokill);
> > +       LIST_HEAD(to_kill);
> >         int rc = -EBUSY;
> > -       loff_t start;
> >         dax_entry_t cookie;
> >
> >         if (flags & MF_COUNT_INCREASED) @@ -1355,22 +1401,10 @@
> static
> > int memory_failure_dev_pagemap(unsigned long pfn, int flags,
> >          * SIGBUS (i.e. MF_MUST_KILL)
> >          */
> >         flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
> > -       collect_procs(page, &tokill, flags & MF_ACTION_REQUIRED);
> > +       collect_procs_file(page, page->mapping, page->index, &to_kill,
> > +                          flags & MF_ACTION_REQUIRED);
> >
> > -       list_for_each_entry(tk, &tokill, nd)
> > -               if (tk->size_shift)
> > -                       size = max(size, 1UL << tk->size_shift);
> > -       if (size) {
> > -               /*
> > -                * Unmap the largest mapping to avoid breaking up
> > -                * device-dax mappings which are constant size. The
> > -                * actual size of the mapping being torn down is
> > -                * communicated in siginfo, see kill_proc()
> > -                */
> > -               start = (page->index << PAGE_SHIFT) & ~(size - 1);
> > -               unmap_mapping_range(page->mapping, start, size, 0);
> > -       }
> > -       kill_procs(&tokill, flags & MF_MUST_KILL, !unmap_success, pfn,
> flags);
> > +       unmap_and_kill(&to_kill, pfn, page->mapping, page->index,
> > + flags);
> 
> There's just too much change in this patch and not enough justification of what
> is being refactored and why.

Sorry again for the mess...


--
Thanks,
Ruan.

Re: [PATCH v4 05/10] mm, pmem: Implement ->memory_failure() in pmem driver

[Dan Williams]

On Thu, Jun 3, 2021 at 6:19 PM Shiyang Ruan <ruansy.fnst@fujitsu.com> wrote:
>
> Call the ->memory_failure() which is implemented by pmem driver, in
> order to finally notify filesystem to handle the corrupted data.  The
> handler which collects and kills processes are moved into
> mf_dax_kill_procs(), which will be called by filesystem.
>
> Keep the old handler in order to roll back if driver or filesystem
> does not support ->memory_failure()/->corrupted_range().
>
> Signed-off-by: Shiyang Ruan <ruansy.fnst@fujitsu.com>
> ---
>  block/genhd.c         | 30 ++++++++++++++++++
>  drivers/nvdimm/pmem.c | 14 +++++++++
>  include/linux/genhd.h |  1 +
>  mm/memory-failure.c   | 71 +++++++++++++++++++++++++++----------------

I would not expect a patch that converts the pmem driver to touch
mm/memory-failure.c. mf_generic_kill_procs() has nothing to do with
the pmem driver.

>  4 files changed, 90 insertions(+), 26 deletions(-)
>
> diff --git a/block/genhd.c b/block/genhd.c
> index 9f8cb7beaad1..75834bd057df 100644
> --- a/block/genhd.c
> +++ b/block/genhd.c
> @@ -718,6 +718,36 @@ struct block_device *bdget_disk(struct gendisk *disk, int partno)
>         return bdev;
>  }
>
> +/**
> + * bdget_disk_sector - get block device by given sector number
> + * @disk: gendisk of interest
> + * @sector: sector number
> + *
> + * RETURNS: the found block device where sector locates in
> + */
> +struct block_device *bdget_disk_sector(struct gendisk *disk, sector_t sector)
> +{
> +       struct block_device *part = NULL, *p;
> +       unsigned long idx;
> +
> +       rcu_read_lock();
> +       xa_for_each(&disk->part_tbl, idx, p) {
> +               if (p->bd_partno == 0)
> +                       continue;
> +               if (p->bd_start_sect <= sector &&
> +                       sector < p->bd_start_sect + bdev_nr_sectors(p)) {
> +                       part = p;
> +                       break;
> +               }
> +       }
> +       rcu_read_unlock();
> +       if (!part)
> +               part = disk->part0;
> +
> +       return bdget_disk(disk, part->bd_partno);
> +}
> +EXPORT_SYMBOL(bdget_disk_sector);

I can't see any justification for this function. The pmem driver does
not need it and the upper layer holders should have their own
mechanism to go from dax_dev offset to bdev if they need to, but
hopefully they don't. A dax_device failure should be independent of a
block_device failure.

> +
>  /*
>   * print a full list of all partitions - intended for places where the root
>   * filesystem can't be mounted and thus to give the victim some idea of what
> diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c
> index ed10a8b66068..98349e7d0a28 100644
> --- a/drivers/nvdimm/pmem.c
> +++ b/drivers/nvdimm/pmem.c
> @@ -364,9 +364,23 @@ static void pmem_release_disk(void *__pmem)
>         put_disk(pmem->disk);
>  }
>
> +static int pmem_pagemap_memory_failure(struct dev_pagemap *pgmap,
> +               unsigned long pfn, int flags)
> +{
> +       struct pmem_device *pdev =
> +                       container_of(pgmap, struct pmem_device, pgmap);

This local variable is called @pmem not @pdev in other parts of the driver.

> +       loff_t offset = PFN_PHYS(pfn) - pdev->phys_addr - pdev->data_offset;
> +       struct block_device *bdev =
> +                       bdget_disk_sector(pdev->disk, offset >> SECTOR_SHIFT);
> +
> +       return dax_corrupted_range(pdev->dax_dev, bdev, offset,
> +                                  page_size(pfn_to_page(pfn)), &flags);

Per previous comments this interface should be range based. Why is the
last argument &flags? That was passed in by value so any changes to it
will not be reflected back to memory_failure()?

> +}
> +
>  static const struct dev_pagemap_ops fsdax_pagemap_ops = {
>         .kill                   = pmem_pagemap_kill,
>         .cleanup                = pmem_pagemap_cleanup,
> +       .memory_failure         = pmem_pagemap_memory_failure,
>  };
>
>  static int pmem_attach_disk(struct device *dev,
> diff --git a/include/linux/genhd.h b/include/linux/genhd.h
> index 6fc26f7bdf71..2ad70c02c343 100644
> --- a/include/linux/genhd.h
> +++ b/include/linux/genhd.h
> @@ -219,6 +219,7 @@ static inline void add_disk_no_queue_reg(struct gendisk *disk)
>
>  extern void del_gendisk(struct gendisk *gp);
>  extern struct block_device *bdget_disk(struct gendisk *disk, int partno);
> +extern struct block_device *bdget_disk_sector(struct gendisk *disk, sector_t sector);
>
>  void set_disk_ro(struct gendisk *disk, bool read_only);
>
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> index 4377e727d478..43017d7f3918 100644
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -1247,6 +1247,36 @@ static void unmap_and_kill(struct list_head *to_kill, unsigned long pfn,
>         kill_procs(to_kill, flags & MF_MUST_KILL, false, pfn, flags);
>  }
>
> +static int mf_generic_kill_procs(unsigned long long pfn, int flags)
> +{
> +       struct page *page = pfn_to_page(pfn);
> +       LIST_HEAD(to_kill);
> +       dax_entry_t cookie;
> +
> +       /*
> +        * Prevent the inode from being freed while we are interrogating
> +        * the address_space, typically this would be handled by
> +        * lock_page(), but dax pages do not use the page lock. This
> +        * also prevents changes to the mapping of this pfn until
> +        * poison signaling is complete.
> +        */
> +       cookie = dax_lock_page(page);
> +       if (!cookie)
> +               return -EBUSY;
> +       /*
> +        * Unlike System-RAM there is no possibility to swap in a
> +        * different physical page at a given virtual address, so all
> +        * userspace consumption of ZONE_DEVICE memory necessitates
> +        * SIGBUS (i.e. MF_MUST_KILL)
> +        */
> +       flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
> +       collect_procs(page, &to_kill, flags & MF_ACTION_REQUIRED);
> +
> +       unmap_and_kill(&to_kill, pfn, page->mapping, page->index, flags);
> +       dax_unlock_page(page, cookie);
> +       return 0;
> +}
> +

Per above, I am surprised to find this in this patch, it belong in its
own patch if at all.

>  int mf_dax_kill_procs(struct address_space *mapping, pgoff_t index, int flags)
>  {
>         LIST_HEAD(to_kill);
> @@ -1348,9 +1378,7 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
>                 struct dev_pagemap *pgmap)
>  {
>         struct page *page = pfn_to_page(pfn);
> -       LIST_HEAD(to_kill);
>         int rc = -EBUSY;
> -       dax_entry_t cookie;
>
>         if (flags & MF_COUNT_INCREASED)
>                 /*
> @@ -1364,20 +1392,9 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
>                 goto out;
>         }
>
> -       /*
> -        * Prevent the inode from being freed while we are interrogating
> -        * the address_space, typically this would be handled by
> -        * lock_page(), but dax pages do not use the page lock. This
> -        * also prevents changes to the mapping of this pfn until
> -        * poison signaling is complete.
> -        */
> -       cookie = dax_lock_page(page);
> -       if (!cookie)
> -               goto out;
> -
>         if (hwpoison_filter(page)) {
>                 rc = 0;
> -               goto unlock;
> +               goto out;
>         }
>
>         if (pgmap->type == MEMORY_DEVICE_PRIVATE) {
> @@ -1385,7 +1402,7 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
>                  * TODO: Handle HMM pages which may need coordination
>                  * with device-side memory.
>                  */
> -               goto unlock;
> +               goto out;
>         }
>
>         /*
> @@ -1395,19 +1412,21 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
>         SetPageHWPoison(page);
>
>         /*
> -        * Unlike System-RAM there is no possibility to swap in a
> -        * different physical page at a given virtual address, so all
> -        * userspace consumption of ZONE_DEVICE memory necessitates
> -        * SIGBUS (i.e. MF_MUST_KILL)
> +        * Call driver's implementation to handle the memory failure,
> +        * otherwise roll back to generic handler.
>          */
> -       flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
> -       collect_procs_file(page, page->mapping, page->index, &to_kill,
> -                          flags & MF_ACTION_REQUIRED);
> +       if (pgmap->ops->memory_failure) {
> +               rc = pgmap->ops->memory_failure(pgmap, pfn, flags);
> +               /*
> +                * Roll back to generic handler too if operation is not
> +                * supported inside the driver/device/filesystem.
> +                */
> +               if (rc != EOPNOTSUPP)
> +                       goto out;
> +       }
> +
> +       rc = mf_generic_kill_procs(pfn, flags);
>
> -       unmap_and_kill(&to_kill, pfn, page->mapping, page->index, flags);
> -       rc = 0;
> -unlock:
> -       dax_unlock_page(page, cookie);
>  out:
>         /* drop pgmap ref acquired in caller */
>         put_dev_pagemap(pgmap);
> --
> 2.31.1
>
>
>