Re: [PATCH v3 1/1] Create debugfs file with hyper-v balloon usage information

From: Wei Liu
Date: Mon Sep 05 2022 - 12:21:12 EST


On Sat, Sep 03, 2022 at 04:37:12AM +0000, Michael Kelley (LINUX) wrote:
> From: Wei Liu <wei.liu@xxxxxxxxxx> Sent: Wednesday, July 13, 2022 8:19 AM
> >
> > On Tue, Jul 12, 2022 at 04:24:12PM +0000, Michael Kelley (LINUX) wrote:
> > > From: Alexander Atanasov <alexander.atanasov@xxxxxxxxxxxxx> Sent: Monday, July
> > 11, 2022 11:18 AM
> > > >
> > > > Allow the guest to know how much it is ballooned by the host.
> > > > It is useful when debugging out of memory conditions.
> > > >
> > > > When host gets back memory from the guest it is accounted
> > > > as used memory in the guest but the guest have no way to know
> > > > how much it is actually ballooned.
> > > >
> > > > Expose current state, flags and max possible memory to the guest.
> > > > While at it - fix a 10+ years old typo.
> > > >
> > > > Signed-off-by: Alexander Atanasov <alexander.atanasov@xxxxxxxxxxxxx>
> > > > ---
> > [...]
> > >
> > > Reviewed-by: Michael Kelley <mikelley@xxxxxxxxxxxxx>
> >
> > I added "Drivers: hv:" prefix to the subject line and applied it to
> > hyperv-next. Thanks.
>
> Alexander -- I finally caught up on the long discussion of balloon
> driver reporting that occurred over much of August. I think your
> original plan had been for each of the balloon driver to report
> useful information in debugfs. But as a result of the discussion,
> it looks like virtio-balloon will be putting the information in
> /proc/meminfo. If that's the case, it seems like we should
> drop these changes to the Hyper-V balloon driver, and have
> the Hyper-V balloon driver take the same approach as
> virtio-balloon.

The debugfs interface contains far more information than what can be put
into meminfo.

That being said, I can send a PR to revert the changes if we need time
to wait for the other discussion to come to an conclusion.

>
> These Hyper-V balloon driver changes have already gone
> into 6.0-rc1. If we're going to drop them, we should do
> the revert before 6.0 is done.
>

Agreed.

Michael, let me know what you think.

Thanks,
Wei.

> Thoughts?
>
> Michael