Hi,
On Fri, Mar 18, 2016 at 05:44:03PM +0500, Roman Mamedov wrote:
> On Thu, 17 Mar 2016 15:16:15 -0700
> Liu Bo wrote:
>
> > For nocow/prealloc files, we try our best to not allocate space, however,
> > this ends up a huge performance regression since it's expensive to check
> > if data is shar
(sorry for any duplicates, vger.org hates gmail)
On Thu, Mar 17, 2016 at 11:16 PM, Liu Bo wrote:
> For nocow/prealloc files, we try our best to not allocate space, however,
> this ends up a huge performance regression since it's expensive to check
> if data is shared.
>
> Let's go back to only go
On Thu, 17 Mar 2016 15:16:15 -0700
Liu Bo wrote:
> For nocow/prealloc files, we try our best to not allocate space, however,
> this ends up a huge performance regression since it's expensive to check
> if data is shared.
>
> Let's go back to only go check shared data once we're not able to alloc
