From: Weihang Li <liweihang@huawei.com>
To: <dledford@redhat.com>, <jgg@ziepe.ca>
Cc: <leon@kernel.org>, <linux-rdma@vger.kernel.org>, <linuxarm@huawei.com>
Subject: [PATCH RFC v2 for-next 0/7] ib core support to send ib port link event
Date: Tue, 4 Feb 2020 16:24:01 +0800 [thread overview]
Message-ID: <20200204082408.18728-1-liweihang@huawei.com> (raw)
From: Lang Cheng <chenglang@huawei.com>
Some provider driver has realized this function, but these code are
implemented separately by each manufacturer. This series provides an
solution in ib_core, and remove the relevant codes of some manufacturers.
The key point is how to shield the port event of the backup port in the ib
bonding scenario. Since the active-backup control is judged by the vendor
driver, so the ops.query_port of vendor would determine the port role. And
there is no relevant data structure in ib_core, so modify struct
ib_port_cache to store this information.
Supports reporting port active time during device registration and sending
port error events when device is deregistered.
The previous discussion can be found at:
https://patchwork.kernel.org/cover/11335999/
Changes since v1:
- Fix comments from Leon and Jason.
- Move event processing flow into global notifier instead of one notifier
per device.
Lang Cheng (7):
RDMA/core: add inactive attribute of ib_port_cache
RDMA/mlx5: remove deliver net device event
qede: remove invalid notify operation
RDMA/qedr: remove deliver net device event
RDMA/vmw_pvrdma: remove deliver net device event
RDMA/core: support send port event
RDMA/core: report link status when register and deregister ib device
drivers/infiniband/core/cache.c | 16 ++++-
drivers/infiniband/core/device.c | 45 ++++++++++++
drivers/infiniband/core/roce_gid_mgmt.c | 45 ++++++++++++
drivers/infiniband/hw/mlx5/main.c | 95 +++-----------------------
drivers/infiniband/hw/qedr/main.c | 19 ------
drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c | 5 --
drivers/net/ethernet/qlogic/qede/qede_rdma.c | 4 --
include/rdma/ib_cache.h | 10 +++
include/rdma/ib_verbs.h | 2 +
9 files changed, 126 insertions(+), 115 deletions(-)
--
2.8.1
next reply other threads:[~2020-02-04 8:24 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-04 8:24 Weihang Li [this message]
2020-02-04 8:24 ` [PATCH RFC v2 for-next 1/7] RDMA/core: add inactive attribute of ib_port_cache Weihang Li
2020-02-19 21:01 ` Jason Gunthorpe
2020-02-20 3:19 ` Lang Cheng
2020-02-20 6:40 ` Leon Romanovsky
2020-02-04 8:24 ` [PATCH RFC v2 for-next 2/7] RDMA/mlx5: remove deliver net device event Weihang Li
2020-02-19 21:03 ` Jason Gunthorpe
2020-02-04 8:24 ` [PATCH RFC v2 for-next 3/7] qede: remove invalid notify operation Weihang Li
2020-02-19 21:04 ` Jason Gunthorpe
2020-02-20 4:18 ` Lang Cheng
2020-02-04 8:24 ` [PATCH RFC v2 for-next 4/7] RDMA/qedr: remove deliver net device event Weihang Li
2020-02-04 8:24 ` [PATCH RFC v2 for-next 5/7] RDMA/vmw_pvrdma: " Weihang Li
2020-02-04 8:24 ` [PATCH RFC v2 for-next 6/7] RDMA/core: support send port event Weihang Li
2020-02-19 21:07 ` Jason Gunthorpe
2020-02-20 8:48 ` Lang Cheng
2020-02-04 8:24 ` [PATCH RFC v2 for-next 7/7] RDMA/core: report link status when register and deregister ib device Weihang Li
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200204082408.18728-1-liweihang@huawei.com \
--to=liweihang@huawei.com \
--cc=dledford@redhat.com \
--cc=jgg@ziepe.ca \
--cc=leon@kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=linuxarm@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).