Re: [RFC][PATCH V4] btrfs: preferred_metadata: preferred device for metadata

[Zygo Blaxell <ce3g8jdj@umail.furryterror.org>]

On Fri, May 29, 2020 at 06:37:27PM +0200, Goffredo Baroncelli wrote:
> On 5/28/20 11:59 PM, Hans van Kranenburg wrote:
> > Hi!
> > 
> > On 5/28/20 8:34 PM, Goffredo Baroncelli wrote:
[...]
> > > The patches set is composed by four patches:
> > > 
> > > - The first patch adds the ioctl to update the btrfs_dev_item.type field.
> > > The ioctl is generic to handle more fields, however now only the "type"
> > > field is supported.
> > 
> > What are your thoughts about the chicken/egg situation of changing these
> > properties only when the filesystem is mounted?
> 
> The logic is related only to a chunk allocation. I.e. if you have a not
> empty filesystem, after enabling the preferred_metadata "mode", in order
> to get the benefit a full balance is required.

Ideally this feature comes with a balance filter that selects block
groups that don't match the current preferred allocation policy, so
you can do a balance on those block groups and leave the rest alone.
There are existing balance filters by devid and stripe count to work from,
so it shouldn't be a lot of new kernel code.