Re: [Qemu-block] [Qemu-devel] [PATCH RFC for-2.6 0/3] block: Add meta dirty bitmap for migration/persistence

[Vladimir Sementsov-Ogievskiy]

On 08.12.2015 04:36, Fam Zheng wrote:
> On Mon, 12/07 18:47, John Snow wrote:
> > On 12/07/2015 12:59 AM, Fam Zheng wrote:
> > > Vladimir,
> > >
> > > This is what I propose to implement meta bitmap. It's implemented in the
> > > HBitmap level to be more efficient, and the interface slightly varies too.
> > I missed it: What was wrong with Vladimir's approach / what are the
> > benefits of this approach?
> The only real difference with this series is, only actual bit toggling will
> mark meta dirty. Vladimir's approach was in BdrvDirtyBitmap level which can't
> tell bit toggling from repetitive bit set/unset. This is from his patch:

Ok, you are right. It's truth.

> @@ -3390,6 +3428,9 @@ void bdrv_set_dirty_bitmap(BdrvDirtyBitmap *bitmap,
>   {
>       assert(bdrv_dirty_bitmap_enabled(bitmap));
>       hbitmap_set(bitmap->bitmap, cur_sector, nr_sectors);
> +    if (bitmap->meta_bitmap) {
> +        hbitmap_set(bitmap->meta_bitmap, cur_sector, nr_sectors);
> +    }
>   }
>
>   void bdrv_reset_dirty_bitmap(BdrvDirtyBitmap *bitmap,
> @@ -3397,6 +3438,9 @@ void bdrv_reset_dirty_bitmap(BdrvDirtyBitmap *bitmap,
>   {
>       assert(bdrv_dirty_bitmap_enabled(bitmap));
>       hbitmap_reset(bitmap->bitmap, cur_sector, nr_sectors);
> +    if (bitmap->meta_bitmap) {
> +        hbitmap_set(bitmap->meta_bitmap, cur_sector, nr_sectors);
> +    }
>   }
>
>   void bdrv_clear_dirty_bitmap(BdrvDirtyBitmap *bitmap)
>
> > > I'd like to use these operations to make dirty bitmap persistence more
> > > efficient too: unchanged dirty bits don't need to be flushed to disk. So I'm
> > > posting this as a separate series for a common base for both sides.
> > This is a reasonable use of the meta-bitmap strategy in general.
> >
> > Keep in mind Vladimir's approach to Meta bitmaps used a different
> > granularity such that 1 physical bit implied 1 sector needed to be
> > re-transmitted.
> Yes, I can fix the meta bitmap granularity.
>
> > A meta-bitmap that keeps track of disk flushes may require a different
> > granularity than one used for migration.
> >
> > > Posting as RFC as 2.6 dev phase is just starting, we can still tweak the
> > > interface and/or implementation to fit the need.
> > >
> > > Fam Zheng (3):
> > >    HBitmap: Introduce "meta" bitmap to track bit changes
> > >    tests: Add test code for meta bitmap
> > >    block: Support meta dirty bitmap
> > >
> > >   block.c                | 46 ++++++++++++++++++++++++++++++-
> > >   block/mirror.c         |  3 +-
> > >   blockdev.c             |  3 +-
> > >   include/block/block.h  | 11 ++++++++
> > >   include/qemu/hbitmap.h |  7 +++++
> > >   migration/block.c      |  2 +-
> > >   tests/test-hbitmap.c   | 74 ++++++++++++++++++++++++++++++++++++++++++++++++++
> > >   util/hbitmap.c         | 22 +++++++++++++++
> > >   8 files changed, 164 insertions(+), 4 deletions(-)


--
Best regards,
Vladimir
* now, @virtuozzo.com instead of @parallels.com. Sorry for this inconvenience.