[GIT PULL] Please pull RDMA subsystem changes

From: Jason Gunthorpe
Date: Fri Sep 09 2022 - 12:17:29 EST

Hi Linus,

This seems unusually big for the second pull request from RDMA. irdma
is the usual new driver fixes, but the rest seems to just be people
testing into obscure corners.

The out-of-subsystem change to fork.c is to export mmput_async() as we
can't do the synchronous put due to held locks and interaction with
mmu notifiers.


The following changes since commit b16de8b9e7d1aae169d059c3a0dd9a881a3c0d1d:

RDMA: Handle the return code from dma_resv_wait_timeout() properly (2022-08-16 17:13:23 +0300)

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 a261786fdc0a5bed2e5f994dcc0ffeeeb0d662c7:

RDMA/irdma: Report RNR NAK generation in device caps (2022-09-07 11:22:18 +0300)

v6.0 second rc pull request

Many bug fixes in several drivers:

- Fix misuse of the DMA API in rtrs

- Several irdma issues: hung task due to SQ flushing, incorrect capability
reporting to userspace, improper error handling for MW corners, touching
an uninitialized SGL for during invalidation.

- hns was using the wrong page size limits for the HW, an incorrect
calculation of wqe_shift causing WQE corruption, and mis computed
a timer id.

- Fix a crash in SRP triggered by blktests

- Fix compiler errors by calling virt_to_page() with the proper type in

- Userspace triggerable deadlock in ODP

- mlx5 could use the wrong profile due to some driver loading races,
counters were not working in some device configurations, and a crash on
error unwind.

Chengchang Tang (1):
RDMA/hns: Fix supported page size

Chris Mi (1):
RDMA/mlx5: Set local port to one when accessing counters

Jack Wang (2):
RDMA/rtrs-clt: Use the right sg_cnt after ib_dma_map_sg
RDMA/rtrs-srv: Pass the correct number of entries for dma mapped SGL

Linus Walleij (1):
RDMA/siw: Pass a pointer to virt_to_page()

Maher Sanalla (1):
RDMA/mlx5: Rely on RoCE fw cap instead of devlink when setting profile

Maor Gottlieb (1):
RDMA/mlx5: Fix UMR cleanup on error flow of driver init

Michael Guralnik (1):
RDMA/cma: Fix arguments order in net device validation

Shiraz Saleem (1):
RDMA/irdma: Fix drain SQ hang with no completion

Sindhu-Devale (5):
RDMA/irdma: Report the correct max cqes from query device
RDMA/irdma: Return error on MR deregister CQP failure
RDMA/irdma: Return correct WC error for bind operation failure
RDMA/irdma: Use s/g array in post send only when its valid
RDMA/irdma: Report RNR NAK generation in device caps

Wenpeng Liang (2):
RDMA/hns: Fix wrong fixed value of qp->rq.wqe_shift
MAINTAINERS: Update maintainers of HiSilicon RoCE

Yishai Hadas (1):
IB/core: Fix a nested dead lock as part of ODP flow

Yixing Liu (1):
RDMA/hns: Remove the num_qpc_timer variable

yangx.jy@xxxxxxxxxxx (1):
RDMA/srp: Set scmnd->result only when scmnd is not NULL

drivers/infiniband/core/cma.c | 4 ++--
drivers/infiniband/core/umem_odp.c | 2 +-
drivers/infiniband/hw/hns/hns_roce_device.h | 1 -
drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 3 +--
drivers/infiniband/hw/hns/hns_roce_hw_v2.h | 4 ++--
drivers/infiniband/hw/hns/hns_roce_main.c | 2 +-
drivers/infiniband/hw/hns/hns_roce_qp.c | 7 ++-----
drivers/infiniband/hw/irdma/uk.c | 7 +++++--
drivers/infiniband/hw/irdma/utils.c | 15 +++++++++------
drivers/infiniband/hw/irdma/verbs.c | 13 ++++++++++---
drivers/infiniband/hw/mlx5/mad.c | 6 ++++++
drivers/infiniband/hw/mlx5/main.c | 2 +-
drivers/infiniband/hw/mlx5/mlx5_ib.h | 1 +
drivers/infiniband/hw/mlx5/umr.c | 3 +++
drivers/infiniband/sw/siw/siw_qp_tx.c | 18 ++++++++++++++----
drivers/infiniband/ulp/rtrs/rtrs-clt.c | 9 +++++----
drivers/infiniband/ulp/rtrs/rtrs-srv.c | 14 +++++++-------
drivers/infiniband/ulp/srp/ib_srp.c | 3 ++-
drivers/net/ethernet/mellanox/mlx5/core/main.c | 23 +++++++++++++++++++++--
include/linux/mlx5/driver.h | 19 ++++++++++---------
kernel/fork.c | 1 +
22 files changed, 105 insertions(+), 54 deletions(-)

Attachment: signature.asc
Description: PGP signature