Re: [PATCH rdma-next v1 14/15] RDMA/core: Allow port_groups to be used with namespaces

[Leon Romanovsky]

On Tue, Jun 08, 2021 at 05:32:56AM +0000, Parav Pandit wrote:
> 
> 
> > From: Leon Romanovsky <leon@xxxxxxxxxx>
> > Sent: Tuesday, June 8, 2021 10:25 AM
> > 
> > On Mon, Jun 07, 2021 at 01:29:58PM +0000, Parav Pandit wrote:
> > >
> > >
> > > > From: Leon Romanovsky <leon@xxxxxxxxxx>
> > > > Sent: Monday, June 7, 2021 1:48 PM
> > > >
> > > > From: Jason Gunthorpe <jgg@xxxxxxxxxx>
> > > >
> > > > Now that the port_groups data is being destroyed and managed by the
> > > > core code this restriction is no longer needed. All the
> > > > ib_port_attrs are compatible with the core's sysfs lifecycle.
> > > >
> > > > Signed-off-by: Jason Gunthorpe <jgg@xxxxxxxxxx>
> > > > Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxx>
> > > > ---
> > > >  drivers/infiniband/core/device.c | 10 ++++------
> > > > drivers/infiniband/core/sysfs.c  | 17 ++++++-----------
> > > >  2 files changed, 10 insertions(+), 17 deletions(-)
> > 
> > <...>
> > 
> > > > diff --git a/drivers/infiniband/core/sysfs.c
> > > > b/drivers/infiniband/core/sysfs.c index 09a2e1066df0..f42034fcf3d9
> > > > 100644
> > > > --- a/drivers/infiniband/core/sysfs.c
> > > > +++ b/drivers/infiniband/core/sysfs.c
> > > > @@ -1236,11 +1236,9 @@ static struct ib_port *setup_port(struct
> > > > ib_core_device *coredev, int port_num,
> > > >  	ret = sysfs_create_groups(&p->kobj, p->groups_list);
> > > >  	if (ret)
> > > >  		goto err_del;
> > > > -	if (is_full_dev) {
> > > > -		ret = sysfs_create_groups(&p->kobj, device-
> > > > >ops.port_groups);
> > > > -		if (ret)
> > > > -			goto err_groups;
> > > > -	}
> > > > +	ret = sysfs_create_groups(&p->kobj, device->ops.port_groups);
> > > > +	if (ret)
> > > > +		goto err_groups;
> > > >
> > > This will expose counters in all net namespaces in shared mode (default
> > case).
> > > Application running in one net namespace will be able to monitor counters
> > of other net namespace.
> > > This should be avoided.
> > 
> > In shared mode, we are sharing sysfs anyway and have two options to deal
> > with the port properties (counters):
> > 1. Show them in all namespaces as being global to port which is shared
> > anyway.
> > 2. Show them in init_net namespace only and applications that were left in
> > this namespace will see not their counters anyway.
> > 
> > Why should we avoid "item 1"?
> Because it is incorrect to show port counters updated by application running in net ns 1, to show to application running in net ns 2.
> Once/if there is per netns counters exist, than those counters can be shown using more modern rdma stats command.

ok, I see your point and have no strong position about it.

Thanks