DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Gaëtan Rivet" <gaetan.rivet@6wind.com>
To: Andrew Rybchenko <arybchenko@solarflare.com>
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] [PATCH v1 12/13] ethdev: process declarative eth devargs
Date: Fri, 31 Aug 2018 14:16:46 +0200	[thread overview]
Message-ID: <20180831121646.x46u43rhfyi45rng@bidouze.vm.6wind.com> (raw)
In-Reply-To: <4538a50d-d6fa-bca2-2dc5-5c9ee9a2c24d@solarflare.com>

On Fri, Aug 31, 2018 at 01:10:48PM +0300, Andrew Rybchenko wrote:
> On 08/30/2018 04:42 PM, Gaetan Rivet wrote:
> > Process the class-specific arguments in a devargs.
> > This processing takes the form of setting the proper eth_dev fields when
> > relevant.
> > 
> > Signed-off-by: Gaetan Rivet <gaetan.rivet@6wind.com>
> > ---
> >   lib/librte_ethdev/eth_private.h   |  5 +++
> >   lib/librte_ethdev/rte_class_eth.c | 62 +++++++++++++++++++++++++++++++
> >   lib/librte_ethdev/rte_ethdev.c    |  7 ++++
> >   3 files changed, 74 insertions(+)
> > 
> > diff --git a/lib/librte_ethdev/eth_private.h b/lib/librte_ethdev/eth_private.h
> > index 0f5c6d5c4..c0c065165 100644
> > --- a/lib/librte_ethdev/eth_private.h
> > +++ b/lib/librte_ethdev/eth_private.h
> > @@ -19,6 +19,11 @@ struct rte_eth_dev *
> >   eth_find_device(const struct rte_eth_dev *_start, rte_eth_cmp_t cmp,
> >   		const void *data);
> > +/* Generic rte_eth_dev parameters processor. */
> > +int
> > +rte_eth_dev_args_parse(struct rte_eth_dev *eth_dev,
> > +		       struct rte_devargs *da);
> > +
> >   #ifdef __cplusplus
> >   }
> >   #endif
> > diff --git a/lib/librte_ethdev/rte_class_eth.c b/lib/librte_ethdev/rte_class_eth.c
> > index d8d8e8845..18fdef605 100644
> > --- a/lib/librte_ethdev/rte_class_eth.c
> > +++ b/lib/librte_ethdev/rte_class_eth.c
> 
> <...>
> 
> > @@ -79,6 +93,54 @@ eth_dev_iterate(const void *start,
> >   	return edev;
> >   }
> > +static int
> > +eth_dev_set_name(struct rte_eth_dev *edev,
> > +		 const char *value)
> > +{
> > +	snprintf(edev->data->name,
> > +		 sizeof(edev->data->name),
> > +		 "%s", value);
> 
> strlcpy()?

Ok.

> Shouldn't we return error if name does fit in name buffer?

Good idea.

> 
> > +	return 0;
> > +}
> > +
> > +static int
> > +ethdev_args_process(const char *key,
> > +		    const char *value,
> > +		    void *_edev)
> > +{
> > +	static eth_dev_set_t eth_dev_set[] = {
> > +		[RTE_ETH_PARAMS_NAME] = eth_dev_set_name,
> > +		[RTE_ETH_PARAMS_MAX] = NULL,
> > +	};
> > +	struct rte_eth_dev *edev = _edev;
> > +	int param;
> > +
> > +	param = ethdev_param_id(key);
> > +	if (eth_dev_set[param])
> > +		return eth_dev_set[param](edev, value);
> > +	return 0;
> > +}
> > +
> > +int
> > +rte_eth_dev_args_parse(struct rte_eth_dev *edev,
> > +		       struct rte_devargs *devargs)
> > +{
> > +	struct rte_kvargs *kvargs = NULL;
> > +
> > +	if (devargs == NULL || devargs->cls_str == NULL)
> > +		return 0;
> > +
> > +	kvargs = rte_kvargs_parse_delim(devargs->cls_str, eth_params_keys, "/");
> > +	if (kvargs == NULL) {
> > +		RTE_LOG(ERR, EAL, "cannot parse argument list\n");
> > +		return -EINVAL;
> > +	}
> > +	if (rte_kvargs_process(kvargs, NULL, ethdev_args_process, edev))
> 
> Shouldn't we free kvargs here as well?
> 

Indeed, silly mistake.
Thanks

> > +		return -1;
> > +	rte_kvargs_free(kvargs);
> > +	return 0;
> > +}
> > +
> >   struct rte_class rte_class_eth = {
> >   	.dev_iterate = eth_dev_iterate,
> >   };
> 
> <...>

-- 
Gaëtan Rivet
6WIND

  reply	other threads:[~2018-08-31 12:17 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-30 13:41 [dpdk-dev] [PATCH v1 00/13] Implement new devargs framework Gaetan Rivet
2018-08-30 13:41 ` [dpdk-dev] [PATCH v1 01/13] bus/pci: implement device iteration and comparison Gaetan Rivet
2018-08-30 13:41 ` [dpdk-dev] [PATCH v1 02/13] bus/pci: add device matching field id Gaetan Rivet
2018-08-30 13:41 ` [dpdk-dev] [PATCH v1 03/13] bus/vdev: implement device iteration Gaetan Rivet
2018-08-30 13:41 ` [dpdk-dev] [PATCH v1 04/13] bus/vdev: add device matching field driver Gaetan Rivet
2018-08-30 13:41 ` [dpdk-dev] [PATCH v1 05/13] ethdev: add private generic device iterator Gaetan Rivet
2018-08-31 10:09   ` Andrew Rybchenko
2018-08-31 10:22     ` Gaëtan Rivet
2018-08-30 13:41 ` [dpdk-dev] [PATCH v1 06/13] ethdev: register ether layer as a class Gaetan Rivet
2018-08-31 10:09   ` Andrew Rybchenko
2018-08-30 13:41 ` [dpdk-dev] [PATCH v1 07/13] ethdev: add device matching field name Gaetan Rivet
2018-08-31 10:10   ` Andrew Rybchenko
2018-08-30 13:41 ` [dpdk-dev] [PATCH v1 08/13] app/testpmd: add show device command Gaetan Rivet
2018-08-30 13:42 ` [dpdk-dev] [PATCH v1 09/13] bus/pci: pre-process declarative PCI devargs Gaetan Rivet
2018-08-30 13:42 ` [dpdk-dev] [PATCH v1 10/13] bus/vdev: pre-process declarative vdev devargs Gaetan Rivet
2018-08-30 13:42 ` [dpdk-dev] [PATCH v1 11/13] bus/pci: process declarative PCI devargs Gaetan Rivet
2018-08-30 16:15   ` Stephen Hemminger
2018-08-30 16:37     ` Gaëtan Rivet
2018-08-30 13:42 ` [dpdk-dev] [PATCH v1 12/13] ethdev: process declarative eth devargs Gaetan Rivet
2018-08-31 10:10   ` Andrew Rybchenko
2018-08-31 12:16     ` Gaëtan Rivet [this message]
2018-08-30 13:42 ` [dpdk-dev] [PATCH v1 13/13] eal: add generic dev parameter Gaetan Rivet
2018-08-30 15:42 ` [dpdk-dev] [PATCH v1 00/13] Implement new devargs framework Stephen Hemminger
2018-09-19 16:03 ` [dpdk-dev] [PATCH v2 " Gaetan Rivet
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 01/13] bus/pci: implement device iteration and comparison Gaetan Rivet
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 02/13] bus/pci: add device matching field id Gaetan Rivet
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 03/13] bus/vdev: implement device iteration Gaetan Rivet
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 04/13] bus/vdev: add device matching field driver Gaetan Rivet
2018-09-20 16:11     ` Thomas Monjalon
2018-09-21 11:53       ` Gaëtan Rivet
2018-09-21 12:55         ` Thomas Monjalon
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 05/13] ethdev: add private generic device iterator Gaetan Rivet
2018-09-20 10:02     ` Andrew Rybchenko
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 06/13] ethdev: register ether layer as a class Gaetan Rivet
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 07/13] ethdev: add device matching field name Gaetan Rivet
2018-09-20 16:17     ` Thomas Monjalon
2018-09-21 12:16       ` Gaëtan Rivet
2018-09-21 13:06         ` Thomas Monjalon
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 08/13] app/testpmd: add show device command Gaetan Rivet
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 09/13] bus/pci: pre-process declarative PCI devargs Gaetan Rivet
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 10/13] bus/vdev: pre-process declarative vdev devargs Gaetan Rivet
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 11/13] bus/pci: process declarative PCI devargs Gaetan Rivet
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 12/13] ethdev: process declarative eth devargs Gaetan Rivet
2018-09-20 10:11     ` Andrew Rybchenko
2018-09-19 16:03   ` [dpdk-dev] [PATCH v2 13/13] eal: add generic dev parameter Gaetan Rivet
2018-10-03 12:31   ` [dpdk-dev] [PATCH v2 00/13] Implement new devargs framework Thomas Monjalon
2020-02-19  5:43     ` Pavan Nikhilesh Bhagavatula

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=20180831121646.x46u43rhfyi45rng@bidouze.vm.6wind.com \
    --to=gaetan.rivet@6wind.com \
    --cc=arybchenko@solarflare.com \
    --cc=dev@dpdk.org \
    /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).