From: Jason Gunthorpe <jgg@ziepe.ca>
To: Leon Romanovsky <leon@kernel.org>
Cc: Jiri Pirko <jiri@resnulli.us>,
linux-rdma@vger.kernel.org, mrgolin@amazon.com,
gal.pressman@linux.dev, sleybo@amazon.com, parav@nvidia.com,
mbloch@nvidia.com, yanjun.zhu@linux.dev, wangliang74@huawei.com,
marco.crivellari@suse.com, roman.gushchin@linux.dev,
phaddad@nvidia.com, lirongqing@baidu.com, ynachum@amazon.com,
huangjunxian6@hisilicon.com, kalesh-anakkur.purayil@broadcom.com,
ohartoov@nvidia.com, michaelgur@nvidia.com, shayd@nvidia.com,
edwards@nvidia.com, sriharsha.basavapatna@broadcom.com,
andrew.gospodarek@broadcom.com, selvin.xavier@broadcom.com
Subject: Re: [PATCH rdma-next 01/10] RDMA/umem: Add reference counting to ib_umem
Date: Tue, 3 Feb 2026 13:01:37 -0400 [thread overview]
Message-ID: <20260203170137.GT2328995@ziepe.ca> (raw)
In-Reply-To: <20260203165600.GW34749@unreal>
On Tue, Feb 03, 2026 at 06:56:00PM +0200, Leon Romanovsky wrote:
> I opted for an even simpler approach: embed the umem directly in ib_cq, set
> cq->umem in ib_core’s create_cq, and clean it up in ib_core’s destroy_cq.
The issue is most drivers create the umem from parameters in their
driver data so the core code cannot do it at all.
The structure I gave is a way for the drivers to parse their driver
data while still keeping the umem out of the bowels of the driver.
You can lift the umems into the core structures too, but things like
the DBR with their driver-specific umems would be hard to deal with.
Jason
next prev parent reply other threads:[~2026-02-03 17:01 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-03 8:49 [PATCH rdma-next 00/10] RDMA: Extend uverbs umem support for QP buffers and doorbell records Jiri Pirko
2026-02-03 8:49 ` [PATCH rdma-next 01/10] RDMA/umem: Add reference counting to ib_umem Jiri Pirko
2026-02-03 10:03 ` Leon Romanovsky
2026-02-03 10:11 ` Jiri Pirko
2026-02-03 12:26 ` Leon Romanovsky
2026-02-03 12:46 ` Jiri Pirko
2026-02-03 13:03 ` Leon Romanovsky
2026-02-03 13:20 ` Jiri Pirko
2026-02-03 13:32 ` Leon Romanovsky
2026-02-03 14:31 ` Jiri Pirko
2026-02-03 13:49 ` Sriharsha Basavapatna
2026-02-03 14:29 ` Jiri Pirko
2026-02-03 14:49 ` Sriharsha Basavapatna
2026-02-03 14:51 ` Jason Gunthorpe
2026-02-03 15:39 ` Jiri Pirko
2026-02-03 16:59 ` Jason Gunthorpe
2026-02-04 7:01 ` Jiri Pirko
2026-02-04 15:38 ` Jiri Pirko
2026-02-04 17:46 ` Jason Gunthorpe
2026-02-04 17:54 ` Leon Romanovsky
2026-02-04 17:56 ` Jiri Pirko
2026-02-03 16:56 ` Leon Romanovsky
2026-02-03 17:01 ` Jason Gunthorpe [this message]
2026-02-03 8:49 ` [PATCH rdma-next 02/10] RDMA/uverbs: Use umem refcounting for CQ creation with external buffer Jiri Pirko
2026-02-03 8:49 ` [PATCH rdma-next 03/10] RDMA/mlx5: Add support for CQ creation with external umem buffer Jiri Pirko
2026-02-03 8:49 ` [PATCH rdma-next 04/10] RDMA/uverbs: Factor out common buffer umem parsing into helper Jiri Pirko
2026-02-03 8:49 ` [PATCH rdma-next 05/10] RDMA/core: Add support for QP buffer umem in QP creation Jiri Pirko
2026-02-03 8:49 ` [PATCH rdma-next 06/10] RDMA/mlx5: Add support for QP creation with external umem buffers Jiri Pirko
2026-02-03 8:49 ` [PATCH rdma-next 07/10] RDMA/uverbs: Add doorbell record umem support to CQ creation Jiri Pirko
2026-02-03 8:50 ` [PATCH rdma-next 08/10] RDMA/mlx5: Add external doorbell record umem support for CQ Jiri Pirko
2026-02-03 8:50 ` [PATCH rdma-next 09/10] RDMA/uverbs: Add doorbell record umem support to QP creation Jiri Pirko
2026-02-03 8:50 ` [PATCH rdma-next 10/10] RDMA/mlx5: Add external doorbell record umem support for QP Jiri Pirko
2026-02-03 9:59 ` [PATCH rdma-next 00/10] RDMA: Extend uverbs umem support for QP buffers and doorbell records Leon Romanovsky
2026-02-03 10:13 ` Jiri Pirko
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=20260203170137.GT2328995@ziepe.ca \
--to=jgg@ziepe.ca \
--cc=andrew.gospodarek@broadcom.com \
--cc=edwards@nvidia.com \
--cc=gal.pressman@linux.dev \
--cc=huangjunxian6@hisilicon.com \
--cc=jiri@resnulli.us \
--cc=kalesh-anakkur.purayil@broadcom.com \
--cc=leon@kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=lirongqing@baidu.com \
--cc=marco.crivellari@suse.com \
--cc=mbloch@nvidia.com \
--cc=michaelgur@nvidia.com \
--cc=mrgolin@amazon.com \
--cc=ohartoov@nvidia.com \
--cc=parav@nvidia.com \
--cc=phaddad@nvidia.com \
--cc=roman.gushchin@linux.dev \
--cc=selvin.xavier@broadcom.com \
--cc=shayd@nvidia.com \
--cc=sleybo@amazon.com \
--cc=sriharsha.basavapatna@broadcom.com \
--cc=wangliang74@huawei.com \
--cc=yanjun.zhu@linux.dev \
--cc=ynachum@amazon.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