2024-02-27 23:57:07

by NeilBrown

[permalink] [raw]
Subject: [PATCH stable 6.6 and 6.7] NFS: Fix data corruption caused by congestion.


when AOP_WRITEPAGE_ACTIVATE is returned (as NFS does when it detects
congestion) it is important that the folio is redirtied.
nfs_writepage_locked() doesn't do this, so files can become corrupted as
writes can be lost.

Note that this is not needed in v6.8 as AOP_WRITEPAGE_ACTIVATE cannot be
returned. It is needed for kernels v5.18..v6.7. Prior to 6.3 the patch
is different as it needs to mention "page", not "folio".

Reported-and-tested-by: Jacek Tomaka <[email protected]>
Fixes: 6df25e58532b ("nfs: remove reliance on bdi congestion")
Signed-off-by: NeilBrown <[email protected]>
---
fs/nfs/write.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index b664caea8b4e..9e345d3c305a 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -668,8 +668,10 @@ static int nfs_writepage_locked(struct folio *folio,
int err;

if (wbc->sync_mode == WB_SYNC_NONE &&
- NFS_SERVER(inode)->write_congested)
+ NFS_SERVER(inode)->write_congested) {
+ folio_redirty_for_writepage(wbc, folio);
return AOP_WRITEPAGE_ACTIVATE;
+ }

nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE);
nfs_pageio_init_write(&pgio, inode, 0, false,
--
2.43.0



2024-03-06 13:42:29

by Jeff Layton

[permalink] [raw]
Subject: Re: [PATCH stable 6.6 and 6.7] NFS: Fix data corruption caused by congestion.

On Wed, 2024-02-28 at 10:23 +1100, NeilBrown wrote:
> when AOP_WRITEPAGE_ACTIVATE is returned (as NFS does when it detects
> congestion) it is important that the folio is redirtied.
> nfs_writepage_locked() doesn't do this, so files can become corrupted as
> writes can be lost.
>
> Note that this is not needed in v6.8 as AOP_WRITEPAGE_ACTIVATE cannot be
> returned. It is needed for kernels v5.18..v6.7. Prior to 6.3 the patch
> is different as it needs to mention "page", not "folio".
>

Neil, I have a question about the above statement. In Linus's tree as of
this morning (v6.8-rc7-ish), it does this in nfs_writepages_locked:

if (wbc->sync_mode == WB_SYNC_NONE &&
NFS_SERVER(inode)->write_congested)
return AOP_WRITEPAGE_ACTIVATE;

The only caller of nfs_writepages_locked, and I don't see where it
redirties the page. Why don't we need this in v6.8?


> Reported-and-tested-by: Jacek Tomaka <[email protected]>
> Fixes: 6df25e58532b ("nfs: remove reliance on bdi congestion")
> Signed-off-by: NeilBrown <[email protected]>
> ---
> fs/nfs/write.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/fs/nfs/write.c b/fs/nfs/write.c
> index b664caea8b4e..9e345d3c305a 100644
> --- a/fs/nfs/write.c
> +++ b/fs/nfs/write.c
> @@ -668,8 +668,10 @@ static int nfs_writepage_locked(struct folio *folio,
> int err;
>
> if (wbc->sync_mode == WB_SYNC_NONE &&
> - NFS_SERVER(inode)->write_congested)
> + NFS_SERVER(inode)->write_congested) {
> + folio_redirty_for_writepage(wbc, folio);
> return AOP_WRITEPAGE_ACTIVATE;
> + }
>
> nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE);
> nfs_pageio_init_write(&pgio, inode, 0, false,

--
Jeff Layton <[email protected]>

2024-03-06 17:17:29

by Jeff Layton

[permalink] [raw]
Subject: Re: [PATCH stable 6.6 and 6.7] NFS: Fix data corruption caused by congestion.

On Wed, 2024-03-06 at 08:42 -0500, Jeff Layton wrote:
> On Wed, 2024-02-28 at 10:23 +1100, NeilBrown wrote:
> > when AOP_WRITEPAGE_ACTIVATE is returned (as NFS does when it detects
> > congestion) it is important that the folio is redirtied.
> > nfs_writepage_locked() doesn't do this, so files can become corrupted as
> > writes can be lost.
> >
> > Note that this is not needed in v6.8 as AOP_WRITEPAGE_ACTIVATE cannot be
> > returned. It is needed for kernels v5.18..v6.7. Prior to 6.3 the patch
> > is different as it needs to mention "page", not "folio".
> >
>
> Neil, I have a question about the above statement. In Linus's tree as of
> this morning (v6.8-rc7-ish), it does this in nfs_writepages_locked:
>
> if (wbc->sync_mode == WB_SYNC_NONE &&
> NFS_SERVER(inode)->write_congested)
> return AOP_WRITEPAGE_ACTIVATE;
>

Sorry, I meant to say:

The only caller of nfs_writepages_locked is nfs_wb_folio, and I don't
see where it redirties the folio. Why don't we need this in v6.8?


>
> > Reported-and-tested-by: Jacek Tomaka <[email protected]>
> > Fixes: 6df25e58532b ("nfs: remove reliance on bdi congestion")
> > Signed-off-by: NeilBrown <[email protected]>
> > ---
> > fs/nfs/write.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/nfs/write.c b/fs/nfs/write.c
> > index b664caea8b4e..9e345d3c305a 100644
> > --- a/fs/nfs/write.c
> > +++ b/fs/nfs/write.c
> > @@ -668,8 +668,10 @@ static int nfs_writepage_locked(struct folio *folio,
> > int err;
> >
> > if (wbc->sync_mode == WB_SYNC_NONE &&
> > - NFS_SERVER(inode)->write_congested)
> > + NFS_SERVER(inode)->write_congested) {
> > + folio_redirty_for_writepage(wbc, folio);
> > return AOP_WRITEPAGE_ACTIVATE;
> > + }
> >
> > nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE);
> > nfs_pageio_init_write(&pgio, inode, 0, false,
>

--
Jeff Layton <[email protected]>

2024-03-07 12:27:08

by NeilBrown

[permalink] [raw]
Subject: Re: [PATCH stable 6.6 and 6.7] NFS: Fix data corruption caused by congestion.

On Thu, 07 Mar 2024, Jeff Layton wrote:
> On Wed, 2024-02-28 at 10:23 +1100, NeilBrown wrote:
> > when AOP_WRITEPAGE_ACTIVATE is returned (as NFS does when it detects
> > congestion) it is important that the folio is redirtied.
> > nfs_writepage_locked() doesn't do this, so files can become corrupted as
> > writes can be lost.
> >
> > Note that this is not needed in v6.8 as AOP_WRITEPAGE_ACTIVATE cannot be
> > returned. It is needed for kernels v5.18..v6.7. Prior to 6.3 the patch
> > is different as it needs to mention "page", not "folio".
> >
>
> Neil, I have a question about the above statement. In Linus's tree as of
> this morning (v6.8-rc7-ish), it does this in nfs_writepages_locked:
>
> if (wbc->sync_mode == WB_SYNC_NONE &&
> NFS_SERVER(inode)->write_congested)
> return AOP_WRITEPAGE_ACTIVATE;
>
> The only caller of nfs_writepages_locked, and I don't see where it
> redirties the page. Why don't we need this in v6.8?

You are right - it doesn't redirty anything. But there is no bug
here....
I didn't see it at first either, but the only caller of
nfs_writepage_locked() is nfs_wb_folio() (as you say) and that always
passes a wbc with .sync_mode = WB_SYNC_ALL. So sync_mode is never
WB_SYNC_NODE and the code snippet you included above is dead code. I've
already posted a patch to Trond and Anna to remove that code.

Thanks for the review!

NeilBrown

>
>
> > Reported-and-tested-by: Jacek Tomaka <[email protected]>
> > Fixes: 6df25e58532b ("nfs: remove reliance on bdi congestion")
> > Signed-off-by: NeilBrown <[email protected]>
> > ---
> > fs/nfs/write.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/nfs/write.c b/fs/nfs/write.c
> > index b664caea8b4e..9e345d3c305a 100644
> > --- a/fs/nfs/write.c
> > +++ b/fs/nfs/write.c
> > @@ -668,8 +668,10 @@ static int nfs_writepage_locked(struct folio *folio,
> > int err;
> >
> > if (wbc->sync_mode == WB_SYNC_NONE &&
> > - NFS_SERVER(inode)->write_congested)
> > + NFS_SERVER(inode)->write_congested) {
> > + folio_redirty_for_writepage(wbc, folio);
> > return AOP_WRITEPAGE_ACTIVATE;
> > + }
> >
> > nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE);
> > nfs_pageio_init_write(&pgio, inode, 0, false,
>
> --
> Jeff Layton <[email protected]>
>


2024-03-07 12:30:36

by Jeff Layton

[permalink] [raw]
Subject: Re: [PATCH stable 6.6 and 6.7] NFS: Fix data corruption caused by congestion.

On Thu, 2024-03-07 at 22:41 +1100, NeilBrown wrote:
> On Thu, 07 Mar 2024, Jeff Layton wrote:
> > On Wed, 2024-02-28 at 10:23 +1100, NeilBrown wrote:
> > > when AOP_WRITEPAGE_ACTIVATE is returned (as NFS does when it detects
> > > congestion) it is important that the folio is redirtied.
> > > nfs_writepage_locked() doesn't do this, so files can become corrupted as
> > > writes can be lost.
> > >
> > > Note that this is not needed in v6.8 as AOP_WRITEPAGE_ACTIVATE cannot be
> > > returned. It is needed for kernels v5.18..v6.7. Prior to 6.3 the patch
> > > is different as it needs to mention "page", not "folio".
> > >
> >
> > Neil, I have a question about the above statement. In Linus's tree as of
> > this morning (v6.8-rc7-ish), it does this in nfs_writepages_locked:
> >
> > if (wbc->sync_mode == WB_SYNC_NONE &&
> > NFS_SERVER(inode)->write_congested)
> > return AOP_WRITEPAGE_ACTIVATE;
> >
> > The only caller of nfs_writepages_locked, and I don't see where it
> > redirties the page. Why don't we need this in v6.8?
>
> You are right - it doesn't redirty anything. But there is no bug
> here....
> I didn't see it at first either, but the only caller of
> nfs_writepage_locked() is nfs_wb_folio() (as you say) and that always
> passes a wbc with .sync_mode = WB_SYNC_ALL. So sync_mode is never
> WB_SYNC_NODE and the code snippet you included above is dead code. I've
> already posted a patch to Trond and Anna to remove that code.
>
> Thanks for the review!
>

Thanks Neil,

I missed that bit about the sync_mode. I sent a R-b for your other patch
too.

Cheers,
Jeff


> >
> >
> > > Reported-and-tested-by: Jacek Tomaka <[email protected]>
> > > Fixes: 6df25e58532b ("nfs: remove reliance on bdi congestion")
> > > Signed-off-by: NeilBrown <[email protected]>
> > > ---
> > > fs/nfs/write.c | 4 +++-
> > > 1 file changed, 3 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/fs/nfs/write.c b/fs/nfs/write.c
> > > index b664caea8b4e..9e345d3c305a 100644
> > > --- a/fs/nfs/write.c
> > > +++ b/fs/nfs/write.c
> > > @@ -668,8 +668,10 @@ static int nfs_writepage_locked(struct folio *folio,
> > > int err;
> > >
> > > if (wbc->sync_mode == WB_SYNC_NONE &&
> > > - NFS_SERVER(inode)->write_congested)
> > > + NFS_SERVER(inode)->write_congested) {
> > > + folio_redirty_for_writepage(wbc, folio);
> > > return AOP_WRITEPAGE_ACTIVATE;
> > > + }
> > >
> > > nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE);
> > > nfs_pageio_init_write(&pgio, inode, 0, false,
> >
> > --
> > Jeff Layton <[email protected]>
> >
>

--
Jeff Layton <[email protected]>