> -----Original Message-----
> From: Jason Gunthorpe
> Sent: Tuesday, December 18, 2018 11:51 AM
> To: Parav Pandit <parav@xxxxxxxxxxxx>
> Cc: Leon Romanovsky <leon@xxxxxxxxxx>; Doug Ledford
> <dledford@xxxxxxxxxx>; Leon Romanovsky <leonro@xxxxxxxxxxxx>; RDMA
> mailing list <linux-rdma@xxxxxxxxxxxxxxx>; Ira Weiny <ira.weiny@xxxxxxxxx>;
> Jack Morgenstein <jackm@xxxxxxxxxxxxxxxxxx>
> Subject: Re: [PATCH rdma-next 1/6] IB/umad: Define and use macro for
> infiniband_mad class name
>
> On Tue, Dec 18, 2018 at 10:49:08AM -0700, Parav Pandit wrote:
> >
> >
> > > From: Jason Gunthorpe
> > > Sent: Tuesday, December 18, 2018 11:44 AM
> > > To: Leon Romanovsky <leon@xxxxxxxxxx>
> > > Cc: Doug Ledford <dledford@xxxxxxxxxx>; Leon Romanovsky
> > > <leonro@xxxxxxxxxxxx>; RDMA mailing list
> > > <linux-rdma@xxxxxxxxxxxxxxx>; Ira Weiny <ira.weiny@xxxxxxxxx>; Jack
> > > Morgenstein <jackm@xxxxxxxxxxxxxxxxxx>; Parav Pandit
> > > <parav@xxxxxxxxxxxx>
> > > Subject: Re: [PATCH rdma-next 1/6] IB/umad: Define and use macro for
> > > infiniband_mad class name
> > >
> > > On Tue, Dec 18, 2018 at 02:15:43PM +0200, Leon Romanovsky wrote:
> > > > From: Parav Pandit <parav@xxxxxxxxxxxx>
> > > >
> > > > Define a class name macro for infiniband_mad class and use it in
> > > > procedure calls (replacing the explicit string argument).
> > > >
> > > > Signed-off-by: Parav Pandit <parav@xxxxxxxxxxxx>
> > > > Reviewed-by: Jack Morgenstein <jackm@xxxxxxxxxxxxxxxxxx>
> > > > Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxxxx>
> > > > drivers/infiniband/core/user_mad.c | 8 +++++---
> > > > 1 file changed, 5 insertions(+), 3 deletions(-)
> > > >
> > > > diff --git a/drivers/infiniband/core/user_mad.c
> > > > b/drivers/infiniband/core/user_mad.c
> > > > index f55f48f6b272..6ca5b517a944 100644
> > > > +++ b/drivers/infiniband/core/user_mad.c
> > > > @@ -61,6 +61,8 @@ MODULE_AUTHOR("Roland Dreier");
> > > > MODULE_DESCRIPTION("InfiniBand userspace MAD packet access");
> > > > MODULE_LICENSE("Dual BSD/GPL");
> > > >
> > > > +#define IB_UMAD_CLASS_NAME "infiniband_mad"
> > > > +
> > > > enum {
> > > > IB_UMAD_MAX_PORTS = RDMA_MAX_PORTS,
> > > > IB_UMAD_MAX_AGENTS = 32,
> > > > @@ -1338,7 +1340,7 @@ static int __init ib_umad_init(void)
> > > >
> > > > ret = register_chrdev_region(base_umad_dev,
> > > > IB_UMAD_NUM_FIXED_MINOR * 2,
> > > > - "infiniband_mad");
> > > > + IB_UMAD_CLASS_NAME);
> > > > if (ret) {
> > > > pr_err("couldn't register device number\n");
> > > > goto out;
> > > > @@ -1346,14 +1348,14 @@ static int __init ib_umad_init(void)
> > > >
> > > > ret = alloc_chrdev_region(&dynamic_umad_dev, 0,
> > > > IB_UMAD_NUM_DYNAMIC_MINOR * 2,
> > > > - "infiniband_mad");
> > > > + IB_UMAD_CLASS_NAME);
> > > > if (ret) {
> > > > pr_err("couldn't register dynamic device number\n");
> > > > goto out_alloc;
> > > > }
> > > > dynamic_issm_dev = dynamic_umad_dev +
> > > IB_UMAD_NUM_DYNAMIC_MINOR;
> > > >
> > > > - umad_class = class_create(THIS_MODULE, "infiniband_mad");
> > > > + umad_class = class_create(THIS_MODULE, IB_UMAD_CLASS_NAME);
> > >
> > > Why not create the class first and just use umad_class->name for all
> > > the other stuff?
> > >
> > Yes. can be done, I didn't think it that way because it attracts more lines of
> code change, by changing ordering of class and other objects.
> > Let me know if that is ok.
>
> Well, once you do patch 3 no re-ordering is needed, umad_class.name is the
> right value.
>
Ok. will send v1 through Leon's queue.
