From: Jason Gunthorpe <jgg@nvidia.com>
To: Yishai Hadas <yishaih@nvidia.com>
Cc: <linux-rdma@vger.kernel.org>, Bob Pearson <rpearsonhpe@gmail.com>,
<maorg@nvidia.com>
Subject: Re: [PATCH 3/9] mlx5: Move context initialization out of mlx5_query_device_ex()
Date: Tue, 17 Nov 2020 15:08:37 -0400 [thread overview]
Message-ID: <20201117190837.GW917484@nvidia.com> (raw)
In-Reply-To: <7f9d950b-329d-17d0-a25f-4e6f5851c26f@nvidia.com>
On Tue, Nov 17, 2020 at 07:24:41PM +0200, Yishai Hadas wrote:
> > +void mlx5_query_device_ctx(struct mlx5_context *mctx)
> > +{
> > + struct ibv_device_attr_ex device_attr;
> > + struct mlx5_query_device_ex_resp resp;
> > + size_t resp_size = sizeof(resp);
> > +
> > + get_lag_caps(mctx);
> > +
> > + if (!(mctx->cmds_supp_uhw & MLX5_USER_CMDS_SUPP_UHW_QUERY_DEVICE))
> > + return;
> > +
> Any reason to not read some applicable context fields (e.g. max_dm_size)
> over uverbs by reducing the resp_size to the ib part ?
No, I missed the device_attr detail
> > + if (ibv_cmd_query_device_any(&mctx->ibv_ctx.context, NULL, &device_attr,
> > + sizeof(device_attr), &resp.ibv_resp,
> > + &resp_size))
> > + return;
> > +
> > + mctx->cached_device_cap_flags = device_attr.orig_attr.device_cap_flags;
> > + mctx->atomic_cap = device_attr.orig_attr.atomic_cap;
> > + mctx->cached_tso_caps = device_attr.tso_caps;
>
> The device_attr.tso_caps is not set over uverbs / cmd_device.c, it
> comes as UHW. The below should be done instead.
Got it
I updated the github
Thanks,
Jason
next prev parent reply other threads:[~2020-11-17 19:08 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-16 20:23 [PATCH 0/9] Simplify query_device() in libibverbs Jason Gunthorpe
2020-11-16 20:23 ` [PATCH 1/9] verbs: Simplify query_device_ex Jason Gunthorpe
2020-11-16 20:23 ` [PATCH 2/9] verbs: Add ibv_cmd_query_device_any() Jason Gunthorpe
2020-11-18 12:43 ` Gal Pressman
2020-11-18 12:45 ` Jason Gunthorpe
2020-11-16 20:23 ` [PATCH 3/9] mlx5: Move context initialization out of mlx5_query_device_ex() Jason Gunthorpe
2020-11-17 17:24 ` Yishai Hadas
2020-11-17 19:08 ` Jason Gunthorpe [this message]
2020-11-16 20:23 ` [PATCH 4/9] efa: Move the context intialization out of efa_query_device_ex() Jason Gunthorpe
2020-11-18 7:39 ` Gal Pressman
2020-11-18 8:07 ` Gal Pressman
2020-11-18 12:45 ` Gal Pressman
2020-11-18 20:59 ` Jason Gunthorpe
2020-11-16 20:23 ` [PATCH 5/9] mlx4: Move the context intialization out of mlx4_query_device_ex() Jason Gunthorpe
2020-11-16 20:23 ` [PATCH 6/9] providers: Remove normal query_device() from providers that have _ex Jason Gunthorpe
2020-11-18 12:47 ` Gal Pressman
2020-11-16 20:23 ` [PATCH 7/9] providers: Convert all providers to implement query_device_ex Jason Gunthorpe
2020-11-16 20:23 ` [PATCH 8/9] verbs: Remove dead code Jason Gunthorpe
2020-11-18 12:46 ` Gal Pressman
2020-11-18 12:53 ` Jason Gunthorpe
2020-11-16 20:23 ` [PATCH 9/9] verbs: Delete query_device() internal support Jason Gunthorpe
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=20201117190837.GW917484@nvidia.com \
--to=jgg@nvidia.com \
--cc=linux-rdma@vger.kernel.org \
--cc=maorg@nvidia.com \
--cc=rpearsonhpe@gmail.com \
--cc=yishaih@nvidia.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).