DPDK patches and discussions
 help / color / mirror / Atom feed
From: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
To: Huisong Li <lihuisong@huawei.com>, dev@dpdk.org
Cc: ferruh.yigit@intel.com, Thomas Monjalon <thomas@monjalon.net>
Subject: Re: [dpdk-dev] [RFC] lib/ethdev: add dev configured flag
Date: Fri, 2 Jul 2021 13:08:43 +0300	[thread overview]
Message-ID: <256aa58e-7455-e861-ee22-eb87ddd2db67@oktetlabs.ru> (raw)
In-Reply-To: <644f214d-887a-02ae-5476-af8e6d100221@huawei.com>

@Thomas, @Ferruh, I tend to accept it (with minor style fixes),
but I need your opinion on it before doing it.

Thanks,
Andrew.

On 6/29/21 5:27 AM, Huisong Li wrote:
> 
> 在 2021/6/14 23:37, Andrew Rybchenko 写道:
>> Summary should start from "ethdev: "
>>
>> Don't forget to include all maintainers in Cc the next time.
>> Just use --cc-cmd or --to-cmd options.
> ok, thanks!
>>
>> Adding Thomas.
>>
>> On 5/8/21 11:00 AM, Huisong Li wrote:
>>> Currently, if dev_configure is not invoked or fails to be invoked, users
>>> can still invoke dev_start successfully. This patch adds a
>>> "dev_configured"
>>> flag in "rte_eth_dev_data" to control whether dev_start can be invoked.
>>
>> In theory there is an indirect condition. If number of configured Tx
>> *and* Rx queues is 0, device is not configured.
> 
> That's true. If the framework doesn't have this check, each driver needs
> to do this.
> 
> But it's a common thing, and it's probably more reasonable to put it in
> the ethdev layer.
> 
>>
>> I have no strong opinion on the topic. Extra flag requires
>> extra housekeeping. Indirect conditions are not always good
>> and could be a subject to change.
>>
>>> Signed-off-by: Huisong Li <lihuisong@huawei.com>
>>> ---
>>>   lib/ethdev/rte_ethdev.c      | 11 +++++++++++
>>>   lib/ethdev/rte_ethdev_core.h |  6 +++++-
>>>   2 files changed, 16 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/lib/ethdev/rte_ethdev.c b/lib/ethdev/rte_ethdev.c
>>> index a187976..7d74b17 100644
>>> --- a/lib/ethdev/rte_ethdev.c
>>> +++ b/lib/ethdev/rte_ethdev.c
>>> @@ -1604,6 +1604,8 @@ rte_eth_dev_configure(uint16_t port_id,
>>> uint16_t nb_rx_q, uint16_t nb_tx_q,
>>>       }
>>>         rte_ethdev_trace_configure(port_id, nb_rx_q, nb_tx_q,
>>> dev_conf, 0);
>>> +    dev->data->dev_configured = 1;
>>> +
>>>       return 0;
>>>   reset_queues:
>>>       eth_dev_rx_queue_config(dev, 0);
>>> @@ -1614,6 +1616,8 @@ rte_eth_dev_configure(uint16_t port_id,
>>> uint16_t nb_rx_q, uint16_t nb_tx_q,
>>>           dev->data->mtu = old_mtu;
>>>         rte_ethdev_trace_configure(port_id, nb_rx_q, nb_tx_q,
>>> dev_conf, ret);
>>> +    dev->data->dev_configured = 0;
>>> +
>>>       return ret;
>>>   }
>>>   @@ -1749,6 +1753,13 @@ rte_eth_dev_start(uint16_t port_id)
>>>         RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->dev_start, -ENOTSUP);
>>>   +    if (dev->data->dev_configured == 0) {
>>> +        RTE_ETHDEV_LOG(INFO,
>>> +            "Device with port_id=%"PRIu16" is not configured.\n",
>>> +            port_id);
>>> +        return -EINVAL;
>>> +    }
>>> +
>>>       if (dev->data->dev_started != 0) {
>>>           RTE_ETHDEV_LOG(INFO,
>>>               "Device with port_id=%"PRIu16" already started\n",
>>> diff --git a/lib/ethdev/rte_ethdev_core.h b/lib/ethdev/rte_ethdev_core.h
>>> index 4679d94..b508769 100644
>>> --- a/lib/ethdev/rte_ethdev_core.h
>>> +++ b/lib/ethdev/rte_ethdev_core.h
>>> @@ -167,7 +167,11 @@ struct rte_eth_dev_data {
>>>           scattered_rx : 1,  /**< RX of scattered packets is ON(1) /
>>> OFF(0) */
>>>           all_multicast : 1, /**< RX all multicast mode ON(1) /
>>> OFF(0). */
>>>           dev_started : 1,   /**< Device state: STARTED(1) /
>>> STOPPED(0). */
>>> -        lro         : 1;   /**< RX LRO is ON(1) / OFF(0) */
>>> +        lro         : 1,  /**< RX LRO is ON(1) / OFF(0) */
>>> +        dev_configured : 1;
>>> +        /**< Device configuration state:
>>> +         * CONFIGURED(1) / NOT CONFIGURED(0).
>>> +         */
>>>       uint8_t rx_queue_state[RTE_MAX_QUEUES_PER_PORT];
>>>           /**< Queues state: HAIRPIN(2) / STARTED(1) / STOPPED(0). */
>>>       uint8_t tx_queue_state[RTE_MAX_QUEUES_PER_PORT];
>>>
>>
>> .


  reply	other threads:[~2021-07-02 10:08 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-08  8:00 Huisong Li
2021-05-31  8:51 ` Huisong Li
2021-06-14 15:37 ` Andrew Rybchenko
2021-06-29  2:27   ` Huisong Li
2021-07-02 10:08     ` Andrew Rybchenko [this message]
2021-07-02 11:57       ` Ferruh Yigit
2021-07-02 13:23         ` Ananyev, Konstantin
2021-07-03  8:35           ` Huisong Li
2021-07-03 11:04             ` Ananyev, Konstantin
2021-07-05  3:03               ` Huisong Li
2021-07-05  9:50                 ` Andrew Rybchenko
2021-07-05 11:22                   ` Ananyev, Konstantin
2021-07-06  1:47                     ` Huisong Li
2021-07-04 20:05 ` Thomas Monjalon
2021-07-05  3:18   ` Huisong Li
2021-07-05  6:07     ` Thomas Monjalon
2021-07-05  9:50       ` Andrew Rybchenko
2021-07-06  1:48         ` Huisong Li
2021-07-06  3:24 ` [dpdk-dev] [PATCH V1] ethdev: " Huisong Li
2021-07-06  4:10 ` [dpdk-dev] [PATCH V2] " Huisong Li
2021-07-06  8:36   ` Andrew Rybchenko
2021-07-07  2:55     ` Huisong Li
2021-07-07  8:25       ` Andrew Rybchenko
2021-07-07  9:26         ` Huisong Li
2021-07-07  7:39     ` David Marchand
2021-07-07  8:23       ` Andrew Rybchenko
2021-07-07  9:36         ` David Marchand
2021-07-07  9:59           ` Thomas Monjalon
2021-07-07 10:40             ` David Marchand
2021-07-07 10:57               ` Thomas Monjalon
2021-07-06 17:49   ` Ananyev, Konstantin
2021-07-07  9:53 ` [dpdk-dev] [PATCH V3] " Huisong Li
2021-07-08  9:56   ` David Marchand

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=256aa58e-7455-e861-ee22-eb87ddd2db67@oktetlabs.ru \
    --to=andrew.rybchenko@oktetlabs.ru \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@intel.com \
    --cc=lihuisong@huawei.com \
    --cc=thomas@monjalon.net \
    /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).