From: Leon Romanovsky <leon@kernel.org>
To: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Doug Ledford <dledford@redhat.com>,
Maor Gottlieb <maorg@mellanox.com>,
linux-rdma@vger.kernel.org
Subject: Re: [PATCH rdma-next 06/14] RDMA/core: Add restrack dummy ops
Date: Mon, 25 May 2020 19:23:26 +0300 [thread overview]
Message-ID: <20200525162326.GC10591@unreal> (raw)
In-Reply-To: <20200525143650.GA21729@ziepe.ca>
On Mon, May 25, 2020 at 11:36:50AM -0300, Jason Gunthorpe wrote:
> On Wed, May 13, 2020 at 12:50:26PM +0300, Leon Romanovsky wrote:
> > From: Maor Gottlieb <maorg@mellanox.com>
> >
> > When fill_res_entry is not implemented by the vendor, then we just
> > need to return 0. Reduce some code and make it more clear by
> > set dummy ops.
> >
> > Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
> > Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
> > ---
> > drivers/infiniband/core/nldev.c | 46 ++++--------------------------
> > drivers/infiniband/core/restrack.c | 13 +++++++++
> > 2 files changed, 19 insertions(+), 40 deletions(-)
> >
> > diff --git a/drivers/infiniband/core/nldev.c b/drivers/infiniband/core/nldev.c
> > index 8548f09746ab..8b4115bc26b2 100644
> > --- a/drivers/infiniband/core/nldev.c
> > +++ b/drivers/infiniband/core/nldev.c
> > @@ -446,22 +446,6 @@ static int fill_res_name_pid(struct sk_buff *msg,
> > return err ? -EMSGSIZE : 0;
> > }
> >
> > -static bool fill_res_entry(struct ib_device *dev, struct sk_buff *msg,
> > - struct rdma_restrack_entry *res)
> > -{
> > - if (!dev->ops.fill_res_entry)
> > - return false;
> > - return dev->ops.fill_res_entry(msg, res);
> > -}
> > -
> > -static bool fill_stat_entry(struct ib_device *dev, struct sk_buff *msg,
> > - struct rdma_restrack_entry *res)
> > -{
> > - if (!dev->ops.fill_stat_entry)
> > - return false;
> > - return dev->ops.fill_stat_entry(msg, res);
> > -}
> > -
> > static int fill_res_qp_entry(struct sk_buff *msg, bool has_cap_net_admin,
> > struct rdma_restrack_entry *res, uint32_t port)
> > {
> > @@ -515,10 +499,7 @@ static int fill_res_qp_entry(struct sk_buff *msg, bool has_cap_net_admin,
> > if (fill_res_name_pid(msg, res))
> > goto err;
> >
> > - if (fill_res_entry(dev, msg, res))
> > - goto err;
> > -
> > - return 0;
> > + return dev->ops.fill_res_entry(msg, res);
> >
> > err: return -EMSGSIZE;
> > }
> > @@ -568,10 +549,7 @@ static int fill_res_cm_id_entry(struct sk_buff *msg, bool has_cap_net_admin,
> > if (fill_res_name_pid(msg, res))
> > goto err;
> >
> > - if (fill_res_entry(dev, msg, res))
> > - goto err;
> > -
> > - return 0;
> > + return dev->ops.fill_res_entry(msg, res);
> >
> > err: return -EMSGSIZE;
> > }
> > @@ -606,10 +584,7 @@ static int fill_res_cq_entry(struct sk_buff *msg, bool has_cap_net_admin,
> > if (fill_res_name_pid(msg, res))
> > goto err;
> >
> > - if (fill_res_entry(dev, msg, res))
> > - goto err;
> > -
> > - return 0;
> > + return dev->ops.fill_res_entry(msg, res);
> >
> > err: return -EMSGSIZE;
> > }
> > @@ -641,10 +616,7 @@ static int fill_res_mr_entry(struct sk_buff *msg, bool has_cap_net_admin,
> > if (fill_res_name_pid(msg, res))
> > goto err;
> >
> > - if (fill_res_entry(dev, msg, res))
> > - goto err;
> > -
> > - return 0;
> > + return dev->ops.fill_res_entry(msg, res);
> >
> > err: return -EMSGSIZE;
> > }
> > @@ -784,15 +756,9 @@ static int fill_stat_mr_entry(struct sk_buff *msg, bool has_cap_net_admin,
> > struct ib_device *dev = mr->pd->device;
> >
> > if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_MRN, res->id))
> > - goto err;
> > -
> > - if (fill_stat_entry(dev, msg, res))
> > - goto err;
> > -
> > - return 0;
> > + return -EMSGSIZE;
> >
> > -err:
> > - return -EMSGSIZE;
> > + return dev->ops.fill_stat_entry(msg, res);
> > }
> >
> > static int fill_stat_counter_hwcounters(struct sk_buff *msg,
> > diff --git a/drivers/infiniband/core/restrack.c b/drivers/infiniband/core/restrack.c
> > index 62fbb0ae9cb4..093b27c0bbe6 100644
> > --- a/drivers/infiniband/core/restrack.c
> > +++ b/drivers/infiniband/core/restrack.c
> > @@ -14,6 +14,17 @@
> > #include "cma_priv.h"
> > #include "restrack.h"
> >
> > +static int fill_res_dummy(struct sk_buff *msg,
> > + struct rdma_restrack_entry *entry)
> > +{
> > + return 0;
> > +}
> > +
> > +static const struct ib_device_ops restrack_dummy_ops = {
> > + .fill_res_entry = fill_res_dummy,
> > + .fill_stat_entry = fill_res_dummy,
> > +};
>
> If you are going to do this then you should do it for substantially
> everything, as we did in rdma-core. I don't want to see easy stuff
> like this half done..
>
> And this should be a broken out series or two as it really has nothing
> to do with 'raw format dumps'
Maor will be committed to do it. Can we merge this series because I see it is
an independent task?
Thanks
>
> Jason
next prev parent reply other threads:[~2020-05-25 16:23 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-13 9:50 [PATCH rdma-next 00/14] RAW format dumps through RDMAtool Leon Romanovsky
2020-05-13 9:50 ` [PATCH mlx5-next 01/14] net/mlx5: Export resource dump interface Leon Romanovsky
2020-05-25 14:24 ` Jason Gunthorpe
2020-05-25 15:25 ` Maor Gottlieb
2020-05-13 9:50 ` [PATCH mlx5-next 02/14] net/mlx5: Add support in query QP, CQ and MKEY segments Leon Romanovsky
2020-05-13 9:50 ` [PATCH rdma-next 03/14] RDMA/core: Fix double put of resource Leon Romanovsky
2020-05-13 9:50 ` [PATCH rdma-next 04/14] RDMA/core: Allow to override device op Leon Romanovsky
2020-05-25 14:26 ` Jason Gunthorpe
2020-05-25 16:28 ` Leon Romanovsky
2020-05-25 23:21 ` Kamal Heib
2020-05-26 5:53 ` Leon Romanovsky
2020-05-26 8:07 ` Kamal Heib
2020-05-26 9:44 ` Leon Romanovsky
2020-05-13 9:50 ` [PATCH rdma-next 05/14] RDMA/core: Don't call fill_res_entry for PD Leon Romanovsky
2020-05-13 9:50 ` [PATCH rdma-next 06/14] RDMA/core: Add restrack dummy ops Leon Romanovsky
2020-05-25 14:36 ` Jason Gunthorpe
2020-05-25 16:23 ` Leon Romanovsky [this message]
2020-05-13 9:50 ` [PATCH rdma-next 07/14] RDMA: Add dedicated MR resource tracker function Leon Romanovsky
2020-05-13 9:50 ` [PATCH rdma-next 08/14] RDMA: Add a dedicated CQ " Leon Romanovsky
2020-05-13 9:50 ` [PATCH rdma-next 09/14] RDMA: Add a dedicated QP " Leon Romanovsky
2020-05-25 14:34 ` Jason Gunthorpe
2020-05-25 16:26 ` Leon Romanovsky
2020-05-13 9:50 ` [PATCH rdma-next 10/14] RDMA: Add dedicated cm id " Leon Romanovsky
2020-05-13 9:50 ` [PATCH rdma-next 11/14] RDMA: Add support to dump resource tracker in RAW format Leon Romanovsky
2020-05-25 14:41 ` Jason Gunthorpe
2020-05-25 16:21 ` Leon Romanovsky
2020-05-27 6:14 ` Maor Gottlieb
2020-05-13 9:50 ` [PATCH rdma-next 12/14] RDMA/mlx5: Add support to get QP resource in raw format Leon Romanovsky
2020-05-13 9:50 ` [PATCH rdma-next 13/14] RDMA/mlx5: Add support to get CQ resource in RAW format Leon Romanovsky
2020-05-13 9:50 ` [PATCH rdma-next 14/14] RDMA/mlx5: Add support to get MR " Leon Romanovsky
2020-05-25 14:50 ` Jason Gunthorpe
2020-05-13 10:15 ` [PATCH rdma-next 00/14] RAW format dumps through RDMAtool 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=20200525162326.GC10591@unreal \
--to=leon@kernel.org \
--cc=dledford@redhat.com \
--cc=jgg@ziepe.ca \
--cc=linux-rdma@vger.kernel.org \
--cc=maorg@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).