[PATCH] dm: Deal with merge_bvec_fn in component devices better

[PATCH] dm: Deal with merge_bvec_fn in component devices better

[Ben Hutchings]

[-- Attachment #1.1: Type: text/plain, Size: 2101 bytes --]

This is analogous to commit 627a2d3c29427637f4c5d31ccc7fcbd8d312cd71
('md: deal with merge_bvec_fn in component devices better.'),
which does the same for md-devices at the top of the stack.  The
following explanation is taken from that commit.  Thanks to Neil Brown
<neilb@suse.de> for the advice.

If a component device has a merge_bvec_fn then as we never call it
we must ensure we never need to.  Currently this is done by setting
max_sector to 1 PAGE, however this does not stop a bio being created
with several sub-page iovecs that would violate the merge_bvec_fn.

So instead set max_segments to 1 and set the segment boundary to the
same as a page boundary to ensure there is only ever one single-page
segment of IO requested at a time.

This can particularly be an issue when 'xen' is used as it is
known to submit multiple small buffers in a single bio.

References: http://bugs.debian.org/604457
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
We've been carrying this in Debian for about 3 years and it's about time
it got reviewed...  It's basically Neil's work so should probably have
his sign-off as well.

Ben.

--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -536,13 +536,15 @@ int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
 		       (unsigned long long) start << SECTOR_SHIFT);
 
 	/*
-	 * Check if merge fn is supported.
-	 * If not we'll force DM to use PAGE_SIZE or
-	 * smaller I/O, just to be safe.
+	 * If we don't call merge_bvec_fn, we must never risk
+	 * violating it, so limit max_phys_segments to 1 lying within
+	 * a single page.
 	 */
-	if (dm_queue_merge_is_compulsory(q) && !ti->type->merge)
-		blk_limits_max_hw_sectors(limits,
-					  (unsigned int) (PAGE_SIZE >> 9));
+	if (dm_queue_merge_is_compulsory(q) && !ti->type->merge) {
+		limits->max_segments = 1;
+		limits->seg_boundary_mask = PAGE_CACHE_SIZE - 1;
+	}
+
 	return 0;
 }
 EXPORT_SYMBOL_GPL(dm_set_device_limits);

-- 
Ben Hutchings
The most exhausting thing in life is being insincere. - Anne Morrow Lindberg

[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

Bug#604457: [dm-devel] [PATCH] dm: Deal with merge_bvec_fn in component devices better

[Mikulas Patocka]

Hi

This was already fixed in 2.6.31 with commit 
8cbeb67ad50f7d68e5e83be2cb2284de8f9c03b5.

See this piece of code in dm.c:

/*
 * If the target doesn't support merge method and some of the devices
 * provided their merge_bvec method (we know this by looking at
 * queue_max_hw_sectors), then we can't allow bios with multiple vector
 * entries.  So always set max_size to 0, and the code below allows
 * just one page.
 */
else if (queue_max_hw_sectors(q) <= PAGE_SIZE >> 9)
        max_size = 0;

Mikulas


On Sun, 1 Sep 2013, Ben Hutchings wrote:

> This is analogous to commit 627a2d3c29427637f4c5d31ccc7fcbd8d312cd71
> ('md: deal with merge_bvec_fn in component devices better.'),
> which does the same for md-devices at the top of the stack.  The
> following explanation is taken from that commit.  Thanks to Neil Brown
> <neilb@suse.de> for the advice.
> 
> If a component device has a merge_bvec_fn then as we never call it
> we must ensure we never need to.  Currently this is done by setting
> max_sector to 1 PAGE, however this does not stop a bio being created
> with several sub-page iovecs that would violate the merge_bvec_fn.
> 
> So instead set max_segments to 1 and set the segment boundary to the
> same as a page boundary to ensure there is only ever one single-page
> segment of IO requested at a time.
> 
> This can particularly be an issue when 'xen' is used as it is
> known to submit multiple small buffers in a single bio.
> 
> References: http://bugs.debian.org/604457
> Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
> ---
> We've been carrying this in Debian for about 3 years and it's about time
> it got reviewed...  It's basically Neil's work so should probably have
> his sign-off as well.
> 
> Ben.
> 
> --- a/drivers/md/dm-table.c
> +++ b/drivers/md/dm-table.c
> @@ -536,13 +536,15 @@ int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
>  		       (unsigned long long) start << SECTOR_SHIFT);
>  
>  	/*
> -	 * Check if merge fn is supported.
> -	 * If not we'll force DM to use PAGE_SIZE or
> -	 * smaller I/O, just to be safe.
> +	 * If we don't call merge_bvec_fn, we must never risk
> +	 * violating it, so limit max_phys_segments to 1 lying within
> +	 * a single page.
>  	 */
> -	if (dm_queue_merge_is_compulsory(q) && !ti->type->merge)
> -		blk_limits_max_hw_sectors(limits,
> -					  (unsigned int) (PAGE_SIZE >> 9));
> +	if (dm_queue_merge_is_compulsory(q) && !ti->type->merge) {
> +		limits->max_segments = 1;
> +		limits->seg_boundary_mask = PAGE_CACHE_SIZE - 1;
> +	}
> +
>  	return 0;
>  }
>  EXPORT_SYMBOL_GPL(dm_set_device_limits);
> 
> -- 
> Ben Hutchings
> The most exhausting thing in life is being insincere. - Anne Morrow Lindberg
> 

Re: [PATCH] dm: Deal with merge_bvec_fn in component devices better

[Ben Hutchings]

[-- Attachment #1.1: Type: text/plain, Size: 3158 bytes --]

On Mon, 2013-09-02 at 10:32 -0400, Mikulas Patocka wrote:
> Hi
> 
> This was already fixed in 2.6.31 with commit 
> 8cbeb67ad50f7d68e5e83be2cb2284de8f9c03b5.
> 
> See this piece of code in dm.c:
> 
> /*
>  * If the target doesn't support merge method and some of the devices
>  * provided their merge_bvec method (we know this by looking at
>  * queue_max_hw_sectors), then we can't allow bios with multiple vector
>  * entries.  So always set max_size to 0, and the code below allows
>  * just one page.
>  */
> else if (queue_max_hw_sectors(q) <= PAGE_SIZE >> 9)
>         max_size = 0;

Thanks, I suspected it might have been fixed somehow.

Ben.

> Mikulas
> 
> 
> On Sun, 1 Sep 2013, Ben Hutchings wrote:
> 
> > This is analogous to commit 627a2d3c29427637f4c5d31ccc7fcbd8d312cd71
> > ('md: deal with merge_bvec_fn in component devices better.'),
> > which does the same for md-devices at the top of the stack.  The
> > following explanation is taken from that commit.  Thanks to Neil Brown
> > <neilb@suse.de> for the advice.
> > 
> > If a component device has a merge_bvec_fn then as we never call it
> > we must ensure we never need to.  Currently this is done by setting
> > max_sector to 1 PAGE, however this does not stop a bio being created
> > with several sub-page iovecs that would violate the merge_bvec_fn.
> > 
> > So instead set max_segments to 1 and set the segment boundary to the
> > same as a page boundary to ensure there is only ever one single-page
> > segment of IO requested at a time.
> > 
> > This can particularly be an issue when 'xen' is used as it is
> > known to submit multiple small buffers in a single bio.
> > 
> > References: http://bugs.debian.org/604457
> > Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
> > ---
> > We've been carrying this in Debian for about 3 years and it's about time
> > it got reviewed...  It's basically Neil's work so should probably have
> > his sign-off as well.
> > 
> > Ben.
> > 
> > --- a/drivers/md/dm-table.c
> > +++ b/drivers/md/dm-table.c
> > @@ -536,13 +536,15 @@ int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
> >  		       (unsigned long long) start << SECTOR_SHIFT);
> >  
> >  	/*
> > -	 * Check if merge fn is supported.
> > -	 * If not we'll force DM to use PAGE_SIZE or
> > -	 * smaller I/O, just to be safe.
> > +	 * If we don't call merge_bvec_fn, we must never risk
> > +	 * violating it, so limit max_phys_segments to 1 lying within
> > +	 * a single page.
> >  	 */
> > -	if (dm_queue_merge_is_compulsory(q) && !ti->type->merge)
> > -		blk_limits_max_hw_sectors(limits,
> > -					  (unsigned int) (PAGE_SIZE >> 9));
> > +	if (dm_queue_merge_is_compulsory(q) && !ti->type->merge) {
> > +		limits->max_segments = 1;
> > +		limits->seg_boundary_mask = PAGE_CACHE_SIZE - 1;
> > +	}
> > +
> >  	return 0;
> >  }
> >  EXPORT_SYMBOL_GPL(dm_set_device_limits);
> > 
> > -- 
> > Ben Hutchings
> > The most exhausting thing in life is being insincere. - Anne Morrow Lindberg
> > 

-- 
Ben Hutchings
If you seem to know what you are doing, you'll be given more to do.

[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]