5.4-stable request: 52e29e331070cd ('btrfs: don't set path->leave_spinning for truncate')

5.4-stable request: 52e29e331070cd ('btrfs: don't set path->leave_spinning for truncate')

[Holger Hoffstätte]

Hi,

I was just looking throught the current 5.4-stable queue and saw that
28553fa992cb28 ('Btrfs: fix race between shrinking truncate and fiemap')
is queued. Upstream has a follow-up fix for this: 52e29e331070cd aka
'btrfs: don't set path->leave_spinning for truncate'.

Would be nice to get those in together. I only looked at 5.4, don't
know about other queues.

thanks,
Holger

Re: 5.4-stable request: 52e29e331070cd ('btrfs: don't set path->leave_spinning for truncate')

[Sasha Levin]

On Tue, Feb 18, 2020 at 01:01:40PM +0100, Holger Hoffstätte wrote:
>Hi,
>
>I was just looking throught the current 5.4-stable queue and saw that
>28553fa992cb28 ('Btrfs: fix race between shrinking truncate and fiemap')
>is queued. Upstream has a follow-up fix for this: 52e29e331070cd aka
>'btrfs: don't set path->leave_spinning for truncate'.
>
>Would be nice to get those in together. I only looked at 5.4, don't
>know about other queues.

Since that fix just hit upstream, I'm going to remove 28553fa992cb28
from the queue now and work on getting both patches in the next release.

-- 
Thanks,
Sasha

Re: 5.4-stable request: 52e29e331070cd ('btrfs: don't set path->leave_spinning for truncate')

[Holger Hoffstätte]

Hi Sasha,

On 2/18/20 5:14 PM, Sasha Levin wrote:
> On Tue, Feb 18, 2020 at 01:01:40PM +0100, Holger Hoffstätte wrote:
>> Hi,
>>
>> I was just looking throught the current 5.4-stable queue and saw that
>> 28553fa992cb28 ('Btrfs: fix race between shrinking truncate and fiemap')
>> is queued. Upstream has a follow-up fix for this: 52e29e331070cd aka
>> 'btrfs: don't set path->leave_spinning for truncate'.
>>
>> Would be nice to get those in together. I only looked at 5.4, don't
>> know about other queues.
> 
> Since that fix just hit upstream, I'm going to remove 28553fa992cb28
> from the queue now and work on getting both patches in the next release.

Thanks, but maybe wait until [1] has also hit mainline? It's another fix
for 28553fa992cb28.

thanks,
Holger

[1] https://patchwork.kernel.org/patch/11394171/

Re: 5.4-stable request: 52e29e331070cd ('btrfs: don't set path->leave_spinning for truncate')

[Greg KH]

On Thu, Feb 20, 2020 at 04:12:24PM +0100, Holger Hoffstätte wrote:
> 
> Hi Sasha,
> 
> On 2/18/20 5:14 PM, Sasha Levin wrote:
> > On Tue, Feb 18, 2020 at 01:01:40PM +0100, Holger Hoffstätte wrote:
> > > Hi,
> > > 
> > > I was just looking throught the current 5.4-stable queue and saw that
> > > 28553fa992cb28 ('Btrfs: fix race between shrinking truncate and fiemap')
> > > is queued. Upstream has a follow-up fix for this: 52e29e331070cd aka
> > > 'btrfs: don't set path->leave_spinning for truncate'.
> > > 
> > > Would be nice to get those in together. I only looked at 5.4, don't
> > > know about other queues.
> > 
> > Since that fix just hit upstream, I'm going to remove 28553fa992cb28
> > from the queue now and work on getting both patches in the next release.
> 
> Thanks, but maybe wait until [1] has also hit mainline? It's another fix
> for 28553fa992cb28.
> 
> thanks,
> Holger
> 
> [1] https://patchwork.kernel.org/patch/11394171/

Ok, I think I have this all now queued up properly for 5.4.y and 5.5.y,
but older kernels will need backports.  Can someone send those series of
patches to stable@vger.kernel.org if it is needed in older kernels like
the commit says they are needed?

thanks,

greg k-h