Hi Linus,
This is my final pull request for the -rc cycle. It includes two bugs
against the newly added opa vnic that were found by turning on the
debug kernel options (one sleeping while holding a lock, so a one line
fix where they switched it from GFP_KERNEL allocation to a GFP_ATOMIC
allocation, the other a case where they had an isolated caller of their
code that could call them in an atomic context so they had to switch
their use of a mutex to a spinlock to be safe, so this was considerably
more lines of diff because all uses of that lock had to be switched),
one bug that was discussed with you already about an out of bounds
array access in ib_uverbs_modify_qp and ib_uverbs_create_ah and is only
7 lines of diff, and one fix to an earlier fix in the -rc cycle that
broke hfi1 and qib in regards to IPoIB (this one is, unfortunately,
larger than I would like for a -rc7 submission, but fixing the problem
required that we not treat all devices as though they had allocated a
netdev universally because it isn't true, and it took 70 lines of diff
to resolve the issue, but the final patch has been vetted by Intel and
Mellanox and they've both given their approval to the fix).
Here's the boilerplate:
The following changes since commit
d4702645838c8e04893383b50406249382b4e6bf:
rdma/cxgb4: Fix memory leaks during module exit (2017-06-14 15:24:50
-0400)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma.git
tags/for-linus
for you to fetch changes up to
8e959601996dc645f4ed7004482a1667c27deb39:
IB/core, opa_vnic, hfi1, mlx5: Properly free rdma_netdev (2017-07-05
17:11:00 -0400)
----------------------------------------------------------------
Fixes #3 for 4.12-rc
- 2 Fixes for OPA found by debug kernel
- 1 Fix for user supplied input causing kernel problems
- 1 Fix for the IPoIB fixes submitted around -rc4
----------------------------------------------------------------
Boris Pismenny (1):
RDMA/uverbs: Check port number supplied by user verbs cmds
Niranjana Vishwanathapura (1):
IB/core, opa_vnic, hfi1, mlx5: Properly free rdma_netdev
Vishwanathapura, Niranjana (2):
IB/opa_vnic: Use GFP_ATOMIC while sending trap
IB/opa_vnic: Use spinlock instead of mutex for stats_lock
drivers/infiniband/core/uverbs_cmd.c | 8 +++++++
drivers/infiniband/hw/hfi1/verbs.c | 1 -
drivers/infiniband/hw/hfi1/vnic.h | 1 -
drivers/infiniband/hw/hfi1/vnic_main.c | 19 +++++++-------
-
drivers/infiniband/hw/mlx5/main.c | 27
++++++++++++++--------
drivers/infiniband/ulp/ipoib/ipoib_main.c | 8 +++----
drivers/infiniband/ulp/opa_vnic/opa_vnic_ethtool.c | 4 ++--
.../infiniband/ulp/opa_vnic/opa_vnic_internal.h | 2 +-
drivers/infiniband/ulp/opa_vnic/opa_vnic_netdev.c | 16 ++++++-------
drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c | 2 +-
.../infiniband/ulp/opa_vnic/opa_vnic_vema_iface.c | 8 +++----
include/rdma/ib_verbs.h | 6 +++--
12 files changed, 58 insertions(+), 44 deletions(-)
--
Doug Ledford <dledford@xxxxxxxxxx>
GPG KeyID: B826A3330E572FDD
Key fingerprint = AE6B 1BDA 122B 23B4 265B 1274 B826 A333 0E57 2FDD
Attachment:
signature.asc
Description: This is a digitally signed message part
