lkml.org 
[lkml]   [2018]   [Mar]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
Subject[GIT PULL] Please pull RDMA subsystem changes
Hi Linus,

Fifth pull request for -rc.

It has been fairly silent lately on our -rc front. Big queue of patches on the
mailing list going to for-next though.

The following changes since commit e8980d67d6017c8eee8f9c35f782c4bd68e004c9:

RDMA/ucma: Ensure that CM_ID exists prior to access it (2018-03-20 11:07:21 -0600)

are available in the Git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus

for you to fetch changes up to 84652aefb347297aa08e91e283adf7b18f77c2d5:

RDMA/ucma: Introduce safer rdma_addr_size() variants (2018-03-28 16:13:36 -0600)

----------------------------------------------------------------
Fifth pull request for 4.16-rc

Bug fixes:
- qedr driver bugfixes causing application hangs, wrong uapi errnos, and a
race condition
- 3 syzkaller found bugfixes in the ucma uapi

Regression fixes for things introduced in 4.16:
- Crash on error introduced in mlx5 UMR flow
- Crash on module unload/etc introduced by bad interaction of restrack
and mlx5 patches this cycle
- Typo in a two line syzkaller bugfix causing a bad regression
- Coverity report of nonsense code in hns driver

----------------------------------------------------------------
Chien Tin Tung (1):
RDMA/ucma: Correct option size check using optlen

Colin Ian King (1):
RDMA/hns: ensure for-loop actually iterates and free's buffers

Jason Gunthorpe (1):
RDMA/rdma_cm: Fix use after free race with process_one_req

Kalderon, Michal (3):
RDMA/qedr: fix QP's ack timeout configuration
RDMA/qedr: Fix rc initialization on CNQ allocation failure
RDMA/qedr: Fix QP state initialization race

Leon Romanovsky (3):
RDMA/restrack: Move restrack_clean to be symmetrical to restrack_init
RDMA/ucma: Check that device is connected prior to access it
RDMA/ucma: Check that device exists prior to accessing it

Mark Bloch (1):
IB/mlx5: Don't clean uninitialized UMR resources

Roland Dreier (1):
RDMA/ucma: Introduce safer rdma_addr_size() variants

drivers/infiniband/core/addr.c | 25 ++++++++++++++++
drivers/infiniband/core/device.c | 3 +-
drivers/infiniband/core/ucma.c | 47 +++++++++++++++++-------------
drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 2 +-
drivers/infiniband/hw/mlx5/main.c | 12 ++++++--
drivers/infiniband/hw/mlx5/mr.c | 3 ++
drivers/infiniband/hw/qedr/main.c | 3 +-
drivers/infiniband/hw/qedr/verbs.c | 45 +++++++++++++++++++---------
include/rdma/ib_addr.h | 2 ++
9 files changed, 102 insertions(+), 40 deletions(-)

\
 
 \ /
  Last update: 2018-03-29 21:18    [W:0.030 / U:0.260 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site