On Wed, Jan 30, 2019 at 12:48:57PM +0200, Leon Romanovsky wrote: > From: Leon Romanovsky <leonro@xxxxxxxxxxxx> > > The additions of .doit callbacks posses new access pattern to the > resource entries by some user visible index. Back then, the legacy > DB was implemented as hash because per-index access wasn't needed > and XArray wasn't accepted yet. > > Acceptance of XArray together with per-index access requires the refresh > of DB implementation. > > Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxxxx> > drivers/infiniband/core/nldev.c | 3 +- > drivers/infiniband/core/restrack.c | 101 +++++++++++++++++++++-------- > include/rdma/restrack.h | 22 +++++-- > 3 files changed, 90 insertions(+), 36 deletions(-) > > diff --git a/drivers/infiniband/core/nldev.c b/drivers/infiniband/core/nldev.c > index 8ad1116c72de..ff222c12d06d 100644 > +++ b/drivers/infiniband/core/nldev.c > @@ -973,6 +973,7 @@ static int res_get_common_dumpit(struct sk_buff *skb, > int start = cb->args[0]; > bool has_cap_net_admin; > struct nlmsghdr *nlh; > + unsigned long id = 0; The new xa_for_each no longer needs the = 0's, they should be removed for clarity > @@ -234,12 +276,15 @@ void rdma_restrack_del(struct rdma_restrack_entry *res) > if (!dev) > return; > > + if (!xa_load(&dev->res.xa[res->type], res->id)) > + goto out; > + This xa_load seems out of place, at least it doesn't belong to this patch. Jason
