[PATCH rdma-next v2 0/2] Align write() and ioctl() paths

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



From: Leon Romanovsky <leonro@xxxxxxxxxxxx>

Changelog:
v2:
 * Fixed warning reported by kbuild where I wrote same assignment twice.
v1:
https://lore.kernel.org/lkml/20200716103956.1422139-1-leon@xxxxxxxxxx
 * v0 revealed old bug
 * https://lore.kernel.org/linux-rdma/20200716102059.1420681-1-leon@xxxxxxxxxx
  that took a while to find.
 * create_cq() was rewritten to make sure that uobj is properly
 * initialized.
v0:
https://lore.kernel.org/lkml/20200708110554.1270613-1-leon@xxxxxxxxxx

----------------------------------------------------
Hi,

The discussion about RWQ table patch revealed incosistency with use of
usecnt, complex unwind flows without any reason and difference between
write() and ioctl() paths.

This series extends infrastructure to be consistent, reliable and
predicable in regards of commit/desotry uobject.

Thanks

Leon Romanovsky (2):
  RDMA/core: Align abort/commit object scheme for write() and ioctl()
    paths
  RDMA/core: Update write interface to use automatic object lifetime

 drivers/infiniband/core/uverbs_cmd.c          | 312 ++++++------------
 drivers/infiniband/core/uverbs_main.c         |   4 +
 .../infiniband/core/uverbs_std_types_device.c |   7 +-
 include/rdma/uverbs_ioctl.h                   |   1 +
 include/rdma/uverbs_std_types.h               |  14 +
 5 files changed, 118 insertions(+), 220 deletions(-)

--
2.26.2




[Index of Archives]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Photo]     [Yosemite News]     [Yosemite Photos]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux