On Thu, Feb 07, 2019 at 05:43:51PM +0200, Leon Romanovsky wrote: > On Tue, Feb 05, 2019 at 02:52:08PM -0700, Jason Gunthorpe wrote: > > On Sat, Feb 02, 2019 at 01:42:47PM +0200, Leon Romanovsky wrote: > > > From: Leon Romanovsky <leonro@xxxxxxxxxxxx> > > > > > > Implement doit callbacks and ensure that users won't provide port values > > > on resource entry allocated in per-device mode needed for .doit callback. > > > > > > Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxxxx> > > > drivers/infiniband/core/nldev.c | 186 +++++++++++++++++++++----------- > > > include/rdma/restrack.h | 2 +- > > > 2 files changed, 127 insertions(+), 61 deletions(-) > > > > > > diff --git a/drivers/infiniband/core/nldev.c b/drivers/infiniband/core/nldev.c > > > index 7bbc98080488..e5910bbbcec7 100644 > > > +++ b/drivers/infiniband/core/nldev.c > > > @@ -375,7 +375,6 @@ static int fill_res_qp_entry(struct sk_buff *msg, bool has_cap_net_admin, > > > struct ib_qp *qp = container_of(res, struct ib_qp, res); > > > struct ib_device *dev = qp->device; > > > struct ib_qp_init_attr qp_init_attr; > > > - struct nlattr *entry_attr; > > > struct ib_qp_attr qp_attr; > > > int ret; > > > > > > @@ -386,10 +385,6 @@ static int fill_res_qp_entry(struct sk_buff *msg, bool has_cap_net_admin, > > > if (port && port != qp_attr.port_num) > > > return 0; > > > > Now that the caller is checking if the port is needed or not these > > 'if (port)' and !port should be removed from the fill functions > > I'll check It is not so easy, because user has an option do not provide PORT for port information. In such case, we will return him him information about all ports in that specific device. It can be done, but maybe later. Thanks
Attachment:
signature.asc
Description: PGP signature
