DPDK patches and discussions
 help / color / mirror / Atom feed
From: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
To: "Wei Hu (Xavier)" <huwei013@chinasoftinc.com>,
	dev@dpdk.org, stephen@networkplumber.org, ferruh.yigit@intel.com,
	kalesh-anakkur.purayil@broadcom.com
Cc: xavier.huwei@huawei.com
Subject: Re: [dpdk-dev] [PATCH v4] ethdev: check if queue setup in queue-related APIs
Date: Tue, 13 Oct 2020 11:08:06 +0300	[thread overview]
Message-ID: <4beba6fb-0ba4-74e3-b585-8056d5b30a93@oktetlabs.ru> (raw)
In-Reply-To: <20201013024126.26492-1-huwei013@chinasoftinc.com>

On 10/13/20 5:41 AM, Wei Hu (Xavier) wrote:
> From: Chengchang Tang <tangchengchang@huawei.com>
> 
> This patch adds checking whether the related Tx or Rx queue has been setup

"This patch adds checking" -> "Add checking"

> in the queue-related API functions to avoid illegal address access. And
> validity check of the queue_id is also added in the API functions
> rte_eth_dev_rx_intr_enable and rte_eth_dev_rx_intr_disable.

Thanks for the patch. LGTM with few minor notes.

Based on the "And ..." in the description I'd consider to split
it into two patches since it has few logical changes.
 * Addition of helper without extra check for queue setup in
    it. I.e. just code restructuring without any changes.
    Definitely harmless.
 * Additoin of queue setup check to helpers
 * Addition of queue ID checks to Rx interrupt control routines

> 
> Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
> Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
> Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
> Acked-by: Stephen Hemminger <stephen@networkplumber.org>
> ---
> v3 -> v4:
> 	 1. dropping the 'rte_' prefix from the funcitons names.
> 	 2. get "struct rte_eth_dev *dev" as parameter and drop the port_id
> 	    validation in the function named eth_dev_validate_rx_queue and
> 	    eth_dev_validate_tx_queue.
> 	 3. replace "setupped" with "setup" in the commit log and titile.
> v2 -> v3:
> 	 don't break lines in format strings.
> v1 -> v2:
> 	 1. replace %"PRIu16" with %u.
> 	 2. extact two common functions which validate RXQ/TXQ ids and
> 	    whether it has been set up or not.
> ---
>  lib/librte_ethdev/rte_ethdev.c | 86 ++++++++++++++++++++++++++++++++++--------
>  lib/librte_ethdev/rte_ethdev.h |  3 +-
>  2 files changed, 72 insertions(+), 17 deletions(-)
> 
> diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
> index 892c246..04d30f9 100644
> --- a/lib/librte_ethdev/rte_ethdev.c
> +++ b/lib/librte_ethdev/rte_ethdev.c
> @@ -877,10 +877,53 @@ rte_eth_dev_rx_queue_config(struct rte_eth_dev *dev, uint16_t nb_queues)
>  	return 0;
>  }
>  
> +static inline int

Pleae, remove inline. The compiler should be wise enought to do
the job. Also it is control path helpers, so there is no point
to blow binaries size.

> +eth_dev_validate_rx_queue(struct rte_eth_dev *dev, uint16_t rx_queue_id)

May be add 'const' to dev to highlight that it changes nothing.

> +{
> +	uint16_t port_id;
> +
> +	if (rx_queue_id >= dev->data->nb_rx_queues) {
> +		RTE_ETHDEV_LOG(ERR, "Invalid RX queue_id=%u\n", rx_queue_id);

RX -> Rx
Please, log port ID as well.
I see that the log is simply copies from previous code,
but let's improve it on the way.

> +		return -EINVAL;
> +	}
> +
> +	if (dev->data->rx_queues[rx_queue_id] == NULL) {
> +		port_id = dev->data->port_id;
> +		RTE_ETHDEV_LOG(ERR,
> +			       "Queue %u of device with port_id=%u has not been setup\n",
> +			       rx_queue_id, port_id);
> +		return -EINVAL;
> +	}
> +
> +	return 0;
> +}
> +
> +static inline int

same as above

> +eth_dev_validate_tx_queue(struct rte_eth_dev *dev, uint16_t tx_queue_id)

same as above

> +{
> +	uint16_t port_id;
> +
> +	if (tx_queue_id >= dev->data->nb_tx_queues) {
> +		RTE_ETHDEV_LOG(ERR, "Invalid TX queue_id=%u\n", tx_queue_id);

TX -> Tx
plus port_id logging

> +		return -EINVAL;
> +	}
> +
> +	if (dev->data->tx_queues[tx_queue_id] == NULL) {
> +		port_id = dev->data->port_id;
> +		RTE_ETHDEV_LOG(ERR,
> +			       "Queue %u of device with port_id=%u has not been setup\n",
> +			       tx_queue_id, port_id);
> +		return -EINVAL;
> +	}
> +
> +	return 0;
> +}
> +
>  int
>  rte_eth_dev_rx_queue_start(uint16_t port_id, uint16_t rx_queue_id)
>  {
>  	struct rte_eth_dev *dev;
> +	int ret;
>  
>  	RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -EINVAL);
>  
> @@ -892,10 +935,9 @@ rte_eth_dev_rx_queue_start(uint16_t port_id, uint16_t rx_queue_id)
>  		return -EINVAL;
>  	}
>  
> -	if (rx_queue_id >= dev->data->nb_rx_queues) {
> -		RTE_ETHDEV_LOG(ERR, "Invalid RX queue_id=%u\n", rx_queue_id);
> -		return -EINVAL;
> -	}
> +	ret = eth_dev_validate_rx_queue(dev, rx_queue_id);
> +	if (ret)

DPDK coding style requires explicit comparion to 0.
Please, add != 0

> +		return ret;
>  
>  	RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->rx_queue_start, -ENOTSUP);
>  
> @@ -922,14 +964,15 @@ int
>  rte_eth_dev_rx_queue_stop(uint16_t port_id, uint16_t rx_queue_id)
>  {
>  	struct rte_eth_dev *dev;
> +	int ret;
>  
>  	RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -EINVAL);
>  
>  	dev = &rte_eth_devices[port_id];
> -	if (rx_queue_id >= dev->data->nb_rx_queues) {
> -		RTE_ETHDEV_LOG(ERR, "Invalid RX queue_id=%u\n", rx_queue_id);
> -		return -EINVAL;
> -	}
> +
> +	ret = eth_dev_validate_rx_queue(dev, rx_queue_id);
> +	if (ret)

same as above, if (ret != 0)

> +		return ret;
>  
>  	RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->rx_queue_stop, -ENOTSUP);
>  
> @@ -955,6 +998,7 @@ int
>  rte_eth_dev_tx_queue_start(uint16_t port_id, uint16_t tx_queue_id)
>  {
>  	struct rte_eth_dev *dev;
> +	int ret;
>  
>  	RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -EINVAL);
>  
> @@ -966,10 +1010,9 @@ rte_eth_dev_tx_queue_start(uint16_t port_id, uint16_t tx_queue_id)
>  		return -EINVAL;
>  	}
>  
> -	if (tx_queue_id >= dev->data->nb_tx_queues) {
> -		RTE_ETHDEV_LOG(ERR, "Invalid TX queue_id=%u\n", tx_queue_id);
> -		return -EINVAL;
> -	}
> +	ret = eth_dev_validate_tx_queue(dev, tx_queue_id);
> +	if (ret)

same as above, if (ret != 0)

> +		return ret;
>  
>  	RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->tx_queue_start, -ENOTSUP);
>  
> @@ -994,14 +1037,15 @@ int
>  rte_eth_dev_tx_queue_stop(uint16_t port_id, uint16_t tx_queue_id)
>  {
>  	struct rte_eth_dev *dev;
> +	int ret;
>  
>  	RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -EINVAL);
>  
>  	dev = &rte_eth_devices[port_id];
> -	if (tx_queue_id >= dev->data->nb_tx_queues) {
> -		RTE_ETHDEV_LOG(ERR, "Invalid TX queue_id=%u\n", tx_queue_id);
> -		return -EINVAL;
> -	}
> +
> +	ret = eth_dev_validate_tx_queue(dev, tx_queue_id);
> +	if (ret)

same as above, if (ret != 0)

> +		return ret;
>  
>  	RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->tx_queue_stop, -ENOTSUP);
>  
> @@ -4458,11 +4502,16 @@ rte_eth_dev_rx_intr_enable(uint16_t port_id,
>  			   uint16_t queue_id)
>  {
>  	struct rte_eth_dev *dev;
> +	int ret;
>  
>  	RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV);
>  
>  	dev = &rte_eth_devices[port_id];
>  
> +	ret = eth_dev_validate_rx_queue(dev, queue_id);
> +	if (ret)

same as above, if (ret != 0)

> +		return ret;
> +
>  	RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->rx_queue_intr_enable, -ENOTSUP);
>  	return eth_err(port_id, (*dev->dev_ops->rx_queue_intr_enable)(dev,
>  								queue_id));
> @@ -4473,11 +4522,16 @@ rte_eth_dev_rx_intr_disable(uint16_t port_id,
>  			    uint16_t queue_id)
>  {
>  	struct rte_eth_dev *dev;
> +	int ret;
>  
>  	RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV);
>  
>  	dev = &rte_eth_devices[port_id];
>  
> +	ret = eth_dev_validate_rx_queue(dev, queue_id);
> +	if (ret)

same as above, if (ret != 0)

> +		return ret;
> +
>  	RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->rx_queue_intr_disable, -ENOTSUP);
>  	return eth_err(port_id, (*dev->dev_ops->rx_queue_intr_disable)(dev,
>  								queue_id));
> diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
> index 5bcfbb8..f4cc591 100644
> --- a/lib/librte_ethdev/rte_ethdev.h
> +++ b/lib/librte_ethdev/rte_ethdev.h
> @@ -4721,7 +4721,8 @@ rte_eth_rx_queue_count(uint16_t port_id, uint16_t queue_id)
>  	RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -EINVAL);
>  	dev = &rte_eth_devices[port_id];
>  	RTE_FUNC_PTR_OR_ERR_RET(*dev->rx_queue_count, -ENOTSUP);
> -	if (queue_id >= dev->data->nb_rx_queues)
> +	if (queue_id >= dev->data->nb_rx_queues ||
> +	    dev->data->rx_queues[queue_id] == NULL)
>  		return -EINVAL;
>  
>  	return (int)(*dev->rx_queue_count)(dev, queue_id);
> 


  parent reply	other threads:[~2020-10-13  8:08 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-10  7:12 [dpdk-dev] [PATCH] ethdev: check if queue setupped " Wei Hu (Xavier)
2020-10-10 15:24 ` Stephen Hemminger
2020-10-12  3:21   ` Wei Hu (Xavier)
2020-10-10 16:38 ` Kalesh Anakkur Purayil
2020-10-12  3:22   ` Wei Hu (Xavier)
2020-10-12  3:19 ` [dpdk-dev] [PATCH v2] " Wei Hu (Xavier)
2020-10-12  6:29   ` Stephen Hemminger
2020-10-12  7:32 ` [dpdk-dev] [PATCH v3] " Wei Hu (Xavier)
2020-10-12 15:12   ` Ferruh Yigit
2020-10-12 15:15   ` Stephen Hemminger
2020-10-13  2:41 ` [dpdk-dev] [PATCH v4] ethdev: check if queue setup " Wei Hu (Xavier)
2020-10-13  4:28   ` Kalesh Anakkur Purayil
2020-10-13  8:08   ` Andrew Rybchenko [this message]
2020-10-13 11:50 ` [dpdk-dev] [PATCH v5 0/3] check queue id " Wei Hu (Xavier)
2020-10-13 11:50   ` [dpdk-dev] [PATCH v5 1/3] ethdev: extract checking queue id into common functions Wei Hu (Xavier)
2020-10-13 12:49     ` Andrew Rybchenko
2020-10-13 11:50   ` [dpdk-dev] [PATCH v5 2/3] ethdev: check if queue setup in queue-related APIs Wei Hu (Xavier)
2020-10-13 11:50   ` [dpdk-dev] [PATCH v5 3/3] ethdev: check queue id in Rx interrupt control routine Wei Hu (Xavier)
2020-10-13 13:17   ` [dpdk-dev] [PATCH v5 0/3] check queue id in queue-related APIs Ferruh Yigit
2020-10-13 14:13     ` Ferruh Yigit

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=4beba6fb-0ba4-74e3-b585-8056d5b30a93@oktetlabs.ru \
    --to=andrew.rybchenko@oktetlabs.ru \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@intel.com \
    --cc=huwei013@chinasoftinc.com \
    --cc=kalesh-anakkur.purayil@broadcom.com \
    --cc=stephen@networkplumber.org \
    --cc=xavier.huwei@huawei.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).