On Sat, Sep 22, 2018 at 1:11 AM Leon Romanovsky <leon@xxxxxxxxxx> wrote: > > On Fri, Sep 21, 2018 at 03:37:23AM +0000, Adit Ranadive wrote: > > > The current code has two copies of the device name, ibdev->dev and > > > dev_name(&ibdev->dev), and they are setup at different times, which is > > > very confusing. > > > > > > Set them both up at the same time and make dev_name() the lead name, which > > > is the proper use of the driver core APIs. To make it very clear that the > > > name is not valid until registration pass it in to the > > > ib_register_device() call rather than messing with ibdev->name directly. > > > > > > Also the reorganization now checks that dev_name is unique even if it does > > > not contain a %. > > > > > > Signed-off-by: Jason Gunthorpe <jgg@xxxxxxxxxxxx> > > > --- > > <...> > > > > > Is this supposed to be complementary to Leon's device rename patch - > > https://patchwork.kernel.org/patch/10607477/? > > > > It wasn't designed but it looks like yes. > > Both of us come to the same conclusion that alloc_names() and > ibdev->name needs to be changed, but from different reasons. > > I came to this conclusion while working on IB persistence naming > task (first step device rename, second rdma-core autodiscovery > over netlink, third possible step some changes to mlx4 and > fourth change will be change of names). > > And Jason came to this conclusion while tried to solve the race and > lockdep warning reported by Bart and Mark. > > > Otherwise, vmw_pvrdma looks fine. > > > > Acked-by: Adit Ranadive <aditr@xxxxxxxxxx> Looks good for bnxt_re: Acked-by: Devesh Sharma <devesh.sharma@xxxxxxxxxxxx>
