From: Leon Romanovsky <leon@kernel.org>
To: Gal Pressman <galpress@amazon.com>
Cc: Doug Ledford <dledford@redhat.com>,
Jason Gunthorpe <jgg@nvidia.com>,
linux-rdma@vger.kernel.org,
"majd@mellanox.com" <majd@mellanox.com>,
Matan Barak <matanb@mellanox.com>,
Sagi Grimberg <sagi@grimberg.me>,
Yishai Hadas <yishaih@mellanox.com>
Subject: Re: [PATCH rdma-next 3/6] RDMA/mlx5: Use mlx5_umem_find_best_quantized_pgoff() for WQ
Date: Tue, 27 Oct 2020 08:03:20 +0200 [thread overview]
Message-ID: <20201027060320.GE4821@unreal> (raw)
In-Reply-To: <94d30486-1909-f044-b59c-ba52e0b1e0e9@amazon.com>
On Mon, Oct 26, 2020 at 04:42:12PM +0200, Gal Pressman wrote:
> On 26/10/2020 15:26, Leon Romanovsky wrote:
> > From: Jason Gunthorpe <jgg@nvidia.com>
> >
> > This fixes a subtle bug, the WQ mailbox has only 5 bits to describe the
> > page_offset, while mlx5_ib_get_buf_offset() is hard wired to only work
> > with 6 bit page_offsets.
> >
> > Thus it did not properly reject badly aligned buffers.
> >
> > YISHAI: WTF? Why does this PRM command only have 5 bits? We must force 4k
> > alignment for WQ umems in the userspace?
>
> You forgot to remove those :).
Yeah, sorry, this is what is happening if you are keeping patches in
submission queue for months :(.
Thanks
next prev parent reply other threads:[~2020-10-27 6:03 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-26 13:26 [PATCH rdma-next 0/6] Use ib_umem_find_best_pgsz() for all umems Leon Romanovsky
2020-10-26 13:26 ` [PATCH rdma-next 1/6] RDMA/mlx5: Use ib_umem_find_best_pgsz() for devx Leon Romanovsky
2020-10-26 13:26 ` [PATCH rdma-next 2/6] RDMA/mlx5: Use ib_umem_find_best_pgoff() for SRQ Leon Romanovsky
2020-10-26 13:26 ` [PATCH rdma-next 3/6] RDMA/mlx5: Use mlx5_umem_find_best_quantized_pgoff() for WQ Leon Romanovsky
2020-10-26 14:42 ` Gal Pressman
2020-10-27 6:03 ` Leon Romanovsky [this message]
2020-10-26 13:26 ` [PATCH rdma-next 4/6] RDMA/mlx5: Use mlx5_umem_find_best_quantized_pgoff() for QP Leon Romanovsky
2020-10-26 13:26 ` [PATCH rdma-next 5/6] RDMA/mlx5: mlx5_umem_find_best_quantized_pgoff() for CQ Leon Romanovsky
2020-10-26 13:26 ` [PATCH rdma-next 6/6] RDMA/mlx5: Lower setting the umem's PAS for SRQ Leon Romanovsky
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=20201027060320.GE4821@unreal \
--to=leon@kernel.org \
--cc=dledford@redhat.com \
--cc=galpress@amazon.com \
--cc=jgg@nvidia.com \
--cc=linux-rdma@vger.kernel.org \
--cc=majd@mellanox.com \
--cc=matanb@mellanox.com \
--cc=sagi@grimberg.me \
--cc=yishaih@mellanox.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).