From: Jason Gunthorpe <jgg@nvidia.com>
To: liweihang <liweihang@huawei.com>
Cc: "leon@kernel.org" <leon@kernel.org>,
"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
Linuxarm <linuxarm@huawei.com>
Subject: Re: [PATCH rdma-core 4/4] libhns: Add support for direct wqe
Date: Wed, 16 Jun 2021 16:14:13 -0300 [thread overview]
Message-ID: <20210616191413.GO1002214@nvidia.com> (raw)
In-Reply-To: <3fa07c87b91047a79402a9871e4e932a@huawei.com>
On Wed, Jun 16, 2021 at 09:55:45AM +0000, liweihang wrote:
> If there is still any issues in this process, could you please tell us where to
> add the barrier? Thank you :)
I don't know ARM perfectly well, but generally look at
1) Do these special stores barrier with the spin unlock protecting
the post send? Allowing them to leak out will get things out of
order
2) ARM MMIO stores are not ordered, so that DB store the ST4 store
are not guaranteed to execute in program order without a barrier.
The spinlock is not a MMIO barrier
You could ignore some of this when the DB rings were basically
idempotent, but if you are xfering data it is more tricky. This is why
we always see a barrier after a WC store to put all future MMIO
strongly in order with the store.
Jason
next prev parent reply other threads:[~2021-06-16 19:14 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-28 9:32 [PATCH rdma-core 0/4] libhns: Add support for direct WQE Weihang Li
2021-05-28 9:32 ` [PATCH rdma-core 1/4] Update kernel headers Weihang Li
2021-05-28 9:32 ` [PATCH rdma-core 2/4] libhns: Refactor hns uar mmap flow Weihang Li
2021-05-28 9:32 ` [PATCH rdma-core 3/4] libhns: Fixes data type when writing doorbell Weihang Li
2021-06-04 14:43 ` Jason Gunthorpe
2021-06-09 3:35 ` liweihang
2021-05-28 9:32 ` [PATCH rdma-core 4/4] libhns: Add support for direct wqe Weihang Li
2021-06-04 14:50 ` Jason Gunthorpe
2021-06-11 9:20 ` liweihang
2021-06-11 11:31 ` Jason Gunthorpe
2021-06-16 9:55 ` liweihang
2021-06-16 19:14 ` Jason Gunthorpe [this message]
2021-06-18 7:23 ` liweihang
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=20210616191413.GO1002214@nvidia.com \
--to=jgg@nvidia.com \
--cc=leon@kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=linuxarm@huawei.com \
--cc=liweihang@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.