DPDK patches and discussions
 help / color / mirror / Atom feed
From: Jeff Guo <jia.guo@intel.com>
To: "Wei Hu (Xavier)" <xavier.huwei@huawei.com>,
	beilei.xing@intel.com, xiao.w.wang@intel.com,
	qi.z.zhang@intel.com, heinrich.kuhn@netronome.com,
	jerinj@marvell.com, ndabilpuram@marvell.com,
	kirankumark@marvell.com, rmody@marvell.com, shshaikh@marvell.com
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH v4 2/2] ethdev: fix VLAN offloads set if no relative capabilities
Date: Sun, 5 Jul 2020 19:22:56 +0800
Message-ID: <17684471-ef21-65eb-a196-786e423e209d@intel.com> (raw)
In-Reply-To: <e9993f6c-e6cd-1df0-eb44-74fed8eb8524@huawei.com>

hi, xavier

On 7/3/2020 9:28 AM, Wei Hu (Xavier) wrote:
> Hi, Heinrich Kuhn & Jerin Jacob & Nithin Dabilpuram & Kiran Kumar K &
>
>   Rasesh Mody & Shahed Shaikh & Qi Zhang & Xiao Wang & Beilei Xing & 
> Jeff Guo.
>
> Could you please help to review this patch and give 'Acked-by' if 
> there are no objects.
>
> Thanks, Xavier
>
>
> On 2020/6/29 9:34, Wei Hu (Xavier) wrote:
>> Hi, Heinrich Kuhn & Harman Kalra & Jerin Jacob & Nithin Dabilpuram & 
>> Kiran Kumar K &
>>
>> Rasesh Mody & Shahed Shaikh & Qi Zhang & Xiao Wang & Beilei Xing & 
>> Jeff Guo.
>>
>> Could you please help to review this patch and give 'Acked-by' if 
>> there are no objects.
>>
>> Thanks, Xavier
>>
>> On 2020/6/22 16:09, Wei Hu (Xavier) wrote:
>>> Currently, there is a potential problem that calling the API function
>>> rte_eth_dev_set_vlan_offload to start VLAN hardware offloads which the
>>> driver does not support. If the PMD driver does not support certain 
>>> VLAN
>>> hardware offloads and does not check for it, the hardware setting will
>>> not change, but the VLAN offloads in 
>>> dev->data->dev_conf.rxmode.offloads
>>> will be turned on.
>>>
>>> It is supposed to check the hardware capabilities to decide whether the
>>> relative callback needs to be called just like the behavior in the API
>>> function named rte_eth_dev_configure. And it is also needed to cleanup
>>> duplicated checks which are done in some PMDs. Also, note that it is
>>> behaviour change for some PMDs which simply ignore (with 
>>> error/warning log
>>> message) unsupported VLAN offloads, but now it will fail.
>>>
>>> Fixes: a4996bd89c42 ("ethdev: new Rx/Tx offloads API")
>>> Cc: stable@dpdk.org
>>>
>>> Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
>>> Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
>>> Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
>>> Acked-by: Hyong Youb Kim <hyonkim@cisco.com>
>>> ---
>>> v3 -> v4: Delete "next_mask" label and modify the function that when 
>>> the
>>>       offload is not supported the function fail.
>>> v2 -> v3: Add __rte_unused to avoid unused parameter 'dev' and 'mask'
>>>       warning.
>>> v1 -> v2: Cleanup duplicated checks which are done in some PMDs.
>>> ---
>>>   drivers/net/dpaa2/dpaa2_ethdev.c           | 12 +++---------
>>>   drivers/net/enic/enic_ethdev.c             | 12 ------------
>>>   drivers/net/fm10k/fm10k_ethdev.c           | 23 
>>> ++---------------------
>>>   drivers/net/hinic/hinic_pmd_ethdev.c       |  6 ------
>>>   drivers/net/i40e/i40e_ethdev.c             |  5 -----
>>>   drivers/net/nfp/nfp_net.c                  |  5 -----
>>>   drivers/net/octeontx/octeontx_ethdev_ops.c | 10 ----------
>>>   drivers/net/octeontx2/otx2_vlan.c          |  5 -----
>>>   drivers/net/qede/qede_ethdev.c             |  3 ---
>>>   lib/librte_ethdev/rte_ethdev.c             | 21 +++++++++++++++++++++
>>>   10 files changed, 26 insertions(+), 76 deletions(-)
>>>
>>> diff --git a/drivers/net/dpaa2/dpaa2_ethdev.c 
>>> b/drivers/net/dpaa2/dpaa2_ethdev.c
>>> index 2f031ec..17a7a49 100644
>>> --- a/drivers/net/dpaa2/dpaa2_ethdev.c
>>> +++ b/drivers/net/dpaa2/dpaa2_ethdev.c
>>> @@ -147,7 +147,7 @@ dpaa2_vlan_offload_set(struct rte_eth_dev *dev, 
>>> int mask)
>>>   {
>>>       struct dpaa2_dev_priv *priv = dev->data->dev_private;
>>>       struct fsl_mc_io *dpni = dev->process_private;
>>> -    int ret;
>>> +    int ret = 0;
>>>         PMD_INIT_FUNC_TRACE();
>>>   @@ -155,7 +155,7 @@ dpaa2_vlan_offload_set(struct rte_eth_dev 
>>> *dev, int mask)
>>>           /* VLAN Filter not avaialble */
>>>           if (!priv->max_vlan_filters) {
>>>               DPAA2_PMD_INFO("VLAN filter not available");
>>> -            goto next_mask;
>>> +            return -ENOTSUP;
>>>           }
>>>             if (dev->data->dev_conf.rxmode.offloads &
>>> @@ -168,14 +168,8 @@ dpaa2_vlan_offload_set(struct rte_eth_dev *dev, 
>>> int mask)
>>>           if (ret < 0)
>>>               DPAA2_PMD_INFO("Unable to set vlan filter = %d", ret);
>>>       }
>>> -next_mask:
>>> -    if (mask & ETH_VLAN_EXTEND_MASK) {
>>> -        if (dev->data->dev_conf.rxmode.offloads &
>>> -            DEV_RX_OFFLOAD_VLAN_EXTEND)
>>> -            DPAA2_PMD_INFO("VLAN extend offload not supported");
>>> -    }
>>>   -    return 0;
>>> +    return ret;
>>>   }
>>>     static int
>>> diff --git a/drivers/net/enic/enic_ethdev.c 
>>> b/drivers/net/enic/enic_ethdev.c
>>> index 32d5397..ef8900d 100644
>>> --- a/drivers/net/enic/enic_ethdev.c
>>> +++ b/drivers/net/enic/enic_ethdev.c
>>> @@ -374,18 +374,6 @@ static int enicpmd_vlan_offload_set(struct 
>>> rte_eth_dev *eth_dev, int mask)
>>>               enic->ig_vlan_strip_en = 0;
>>>       }
>>>   -    if ((mask & ETH_VLAN_FILTER_MASK) &&
>>> -        (offloads & DEV_RX_OFFLOAD_VLAN_FILTER)) {
>>> -        dev_warning(enic,
>>> -            "Configuration of VLAN filter is not supported\n");
>>> -    }
>>> -
>>> -    if ((mask & ETH_VLAN_EXTEND_MASK) &&
>>> -        (offloads & DEV_RX_OFFLOAD_VLAN_EXTEND)) {
>>> -        dev_warning(enic,
>>> -            "Configuration of extended VLAN is not supported\n");
>>> -    }
>>> -
>>>       return enic_set_vlan_strip(enic);
>>>   }
>>>   diff --git a/drivers/net/fm10k/fm10k_ethdev.c 
>>> b/drivers/net/fm10k/fm10k_ethdev.c
>>> index f537ab2..f5b854e 100644
>>> --- a/drivers/net/fm10k/fm10k_ethdev.c
>>> +++ b/drivers/net/fm10k/fm10k_ethdev.c
>>> @@ -1575,28 +1575,9 @@ fm10k_vlan_filter_set(struct rte_eth_dev 
>>> *dev, uint16_t vlan_id, int on)
>>>   }
>>>     static int
>>> -fm10k_vlan_offload_set(struct rte_eth_dev *dev, int mask)
>>> +fm10k_vlan_offload_set(struct rte_eth_dev *dev __rte_unused,
>>> +               int mask __rte_unused)
>>>   {
>>> -    if (mask & ETH_VLAN_STRIP_MASK) {
>>> -        if (!(dev->data->dev_conf.rxmode.offloads &
>>> -            DEV_RX_OFFLOAD_VLAN_STRIP))
>>> -            PMD_INIT_LOG(ERR, "VLAN stripping is "
>>> -                    "always on in fm10k");
>>> -    }
>>> -
>>> -    if (mask & ETH_VLAN_EXTEND_MASK) {
>>> -        if (dev->data->dev_conf.rxmode.offloads &
>>> -            DEV_RX_OFFLOAD_VLAN_EXTEND)
>>> -            PMD_INIT_LOG(ERR, "VLAN QinQ is not "
>>> -                    "supported in fm10k");
>>> -    }
>>> -
>>> -    if (mask & ETH_VLAN_FILTER_MASK) {
>>> -        if (!(dev->data->dev_conf.rxmode.offloads &
>>> -            DEV_RX_OFFLOAD_VLAN_FILTER))
>>> -            PMD_INIT_LOG(ERR, "VLAN filter is always on in fm10k");
>>> -    }
>>> -
>>>       return 0;
>>>   }
>>>   diff --git a/drivers/net/hinic/hinic_pmd_ethdev.c 
>>> b/drivers/net/hinic/hinic_pmd_ethdev.c
>>> index 0c3e1c0..0009a61 100644
>>> --- a/drivers/net/hinic/hinic_pmd_ethdev.c
>>> +++ b/drivers/net/hinic/hinic_pmd_ethdev.c
>>> @@ -1701,12 +1701,6 @@ static int hinic_vlan_offload_set(struct 
>>> rte_eth_dev *dev, int mask)
>>>                 nic_dev->proc_dev_name, dev->data->port_id);
>>>       }
>>>   -    if (mask & ETH_VLAN_EXTEND_MASK) {
>>> -        PMD_DRV_LOG(ERR, "Don't support vlan qinq, device: %s, 
>>> port_id: %d",
>>> -              nic_dev->proc_dev_name, dev->data->port_id);
>>> -        return -ENOTSUP;
>>> -    }
>>> -
>>>       return 0;
>>>   }
>>>   diff --git a/drivers/net/i40e/i40e_ethdev.c 
>>> b/drivers/net/i40e/i40e_ethdev.c
>>> index 970a31c..9211cf5 100644
>>> --- a/drivers/net/i40e/i40e_ethdev.c
>>> +++ b/drivers/net/i40e/i40e_ethdev.c
>>> @@ -3877,11 +3877,6 @@ i40e_vlan_offload_set(struct rte_eth_dev 
>>> *dev, int mask)
>>>       struct i40e_vsi *vsi = pf->main_vsi;
>>>       struct rte_eth_rxmode *rxmode;
>>>   -    if (mask & ETH_QINQ_STRIP_MASK) {
>>> -        PMD_DRV_LOG(ERR, "Strip qinq is not supported.");
>>> -        return -ENOTSUP;
>>> -    }
>>> -


This part of change looks good if it checking capability by device 
capability info, but please add the related history patch into the Fixes 
commit log if

it is not involved by " a4996bd89c42 ("ethdev: new Rx/Tx offloads API") 
", other is the same i think.


>>>       rxmode = &dev->data->dev_conf.rxmode;
>>>       if (mask & ETH_VLAN_FILTER_MASK) {
>>>           if (rxmode->offloads & DEV_RX_OFFLOAD_VLAN_FILTER)
>>> diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
>>> index 2460ee1..b7a91ae 100644
>>> --- a/drivers/net/nfp/nfp_net.c
>>> +++ b/drivers/net/nfp/nfp_net.c
>>> @@ -2353,11 +2353,6 @@ nfp_net_vlan_offload_set(struct rte_eth_dev 
>>> *dev, int mask)
>>>       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
>>>       new_ctrl = 0;
>>>   -    if ((mask & ETH_VLAN_FILTER_OFFLOAD) ||
>>> -        (mask & ETH_VLAN_EXTEND_OFFLOAD))
>>> -        PMD_DRV_LOG(INFO, "No support for ETH_VLAN_FILTER_OFFLOAD or"
>>> -            " ETH_VLAN_EXTEND_OFFLOAD");
>>> -
>>>       /* Enable vlan strip if it is not configured yet */
>>>       if ((mask & ETH_VLAN_STRIP_OFFLOAD) &&
>>>           !(hw->ctrl & NFP_NET_CFG_CTRL_RXVLAN))
>>> diff --git a/drivers/net/octeontx/octeontx_ethdev_ops.c 
>>> b/drivers/net/octeontx/octeontx_ethdev_ops.c
>>> index ff627a6..dbe13ce 100644
>>> --- a/drivers/net/octeontx/octeontx_ethdev_ops.c
>>> +++ b/drivers/net/octeontx/octeontx_ethdev_ops.c
>>> @@ -43,16 +43,6 @@ octeontx_dev_vlan_offload_set(struct rte_eth_dev 
>>> *dev, int mask)
>>>         rxmode = &dev->data->dev_conf.rxmode;
>>>   -    if (mask & ETH_VLAN_EXTEND_MASK) {
>>> -        octeontx_log_err("Extend offload not supported");
>>> -        return -ENOTSUP;
>>> -    }
>>> -
>>> -    if (mask & ETH_VLAN_STRIP_MASK) {
>>> -        octeontx_log_err("VLAN strip offload not supported");
>>> -        return -ENOTSUP;
>>> -    }
>>> -
>>>       if (mask & ETH_VLAN_FILTER_MASK) {
>>>           if (rxmode->offloads & DEV_RX_OFFLOAD_VLAN_FILTER) {
>>>               rc = octeontx_vlan_hw_filter(nic, true);
>>> diff --git a/drivers/net/octeontx2/otx2_vlan.c 
>>> b/drivers/net/octeontx2/otx2_vlan.c
>>> index 322a565..7357b06 100644
>>> --- a/drivers/net/octeontx2/otx2_vlan.c
>>> +++ b/drivers/net/octeontx2/otx2_vlan.c
>>> @@ -717,11 +717,6 @@ otx2_nix_vlan_offload_set(struct rte_eth_dev 
>>> *eth_dev, int mask)
>>>         rxmode = &eth_dev->data->dev_conf.rxmode;
>>>   -    if (mask & ETH_VLAN_EXTEND_MASK) {
>>> -        otx2_err("Extend offload not supported");
>>> -        return -ENOTSUP;
>>> -    }
>>> -
>>>       if (mask & ETH_VLAN_STRIP_MASK) {
>>>           if (rxmode->offloads & DEV_RX_OFFLOAD_VLAN_STRIP) {
>>>               offloads |= DEV_RX_OFFLOAD_VLAN_STRIP;
>>> diff --git a/drivers/net/qede/qede_ethdev.c 
>>> b/drivers/net/qede/qede_ethdev.c
>>> index c4f8f12..04d17c7 100644
>>> --- a/drivers/net/qede/qede_ethdev.c
>>> +++ b/drivers/net/qede/qede_ethdev.c
>>> @@ -1036,9 +1036,6 @@ static int qede_vlan_offload_set(struct 
>>> rte_eth_dev *eth_dev, int mask)
>>>           }
>>>       }
>>>   -    if (mask & ETH_VLAN_EXTEND_MASK)
>>> -        DP_ERR(edev, "Extend VLAN not supported\n");
>>> -
>>>       qdev->vlan_offload_mask = mask;
>>>         DP_INFO(edev, "VLAN offload mask %d\n", mask);
>>> diff --git a/lib/librte_ethdev/rte_ethdev.c 
>>> b/lib/librte_ethdev/rte_ethdev.c
>>> index b0b0474..bf79749 100644
>>> --- a/lib/librte_ethdev/rte_ethdev.c
>>> +++ b/lib/librte_ethdev/rte_ethdev.c
>>> @@ -3260,12 +3260,14 @@ rte_eth_dev_set_vlan_ether_type(uint16_t 
>>> port_id,
>>>   int
>>>   rte_eth_dev_set_vlan_offload(uint16_t port_id, int offload_mask)
>>>   {
>>> +    struct rte_eth_dev_info dev_info;
>>>       struct rte_eth_dev *dev;
>>>       int ret = 0;
>>>       int mask = 0;
>>>       int cur, org = 0;
>>>       uint64_t orig_offloads;
>>>       uint64_t dev_offloads;
>>> +    uint64_t new_offloads;
>>>         RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV);
>>>       dev = &rte_eth_devices[port_id];
>>> @@ -3319,6 +3321,25 @@ rte_eth_dev_set_vlan_offload(uint16_t 
>>> port_id, int offload_mask)
>>>       if (mask == 0)
>>>           return ret;
>>>   +    ret = rte_eth_dev_info_get(port_id, &dev_info);
>>> +    if (ret != 0)
>>> +        return ret;
>>> +
>>> +    /*
>>> +     * New added Rx VLAN offloading which are not enabled in
>>> +     * rte_eth_dev_configure() must be within its device capabilities
>>> +     */
>>> +    if ((dev_offloads & dev_info.rx_offload_capa) != dev_offloads) {
>>> +        new_offloads = dev_offloads & ~orig_offloads;
>>> +        RTE_ETHDEV_LOG(ERR,
>>> +            "Ethdev port_id=%u requested new added VLAN offloads "
>>> +            "0x%" PRIx64 " must be within Rx offloads capabilities "
>>> +            "0x%" PRIx64 " in %s()\n",
>>> +            port_id, new_offloads, dev_info.rx_offload_capa,
>>> +            __func__);
>>> +        return -EINVAL;
>>> +    }
>>> +
>>> RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->vlan_offload_set, -ENOTSUP);
>>>       dev->data->dev_conf.rxmode.offloads = dev_offloads;
>>>       ret = (*dev->dev_ops->vlan_offload_set)(dev, mask);
>>
>> .
>>
>

  reply	other threads:[~2020-07-05 11:23 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-22  8:08 [dpdk-dev] [PATCH v4 0/2] ethdev: minor bugfixes Wei Hu (Xavier)
2020-06-22  8:08 ` [dpdk-dev] [PATCH v4 1/2] ethdev: fix data room size verification in Rx queue setup Wei Hu (Xavier)
2020-06-22  8:24   ` Sachin Saxena (OSS)
2020-06-22  8:09 ` [dpdk-dev] [PATCH v4 2/2] ethdev: fix VLAN offloads set if no relative capabilities Wei Hu (Xavier)
2020-06-22  8:27   ` Sachin Saxena (OSS)
2020-06-22  9:08   ` Wangxiaoyun (Cloud)
2020-06-29  1:34   ` Wei Hu (Xavier)
2020-06-29 14:47     ` [dpdk-dev] [EXT] " Harman Kalra
2020-07-01  2:32       ` Wei Hu (Xavier)
2020-07-03  1:28     ` [dpdk-dev] " Wei Hu (Xavier)
2020-07-05 11:22       ` Jeff Guo [this message]
2020-07-06  6:32         ` Wei Hu (Xavier)

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=17684471-ef21-65eb-a196-786e423e209d@intel.com \
    --to=jia.guo@intel.com \
    --cc=beilei.xing@intel.com \
    --cc=dev@dpdk.org \
    --cc=heinrich.kuhn@netronome.com \
    --cc=jerinj@marvell.com \
    --cc=kirankumark@marvell.com \
    --cc=ndabilpuram@marvell.com \
    --cc=qi.z.zhang@intel.com \
    --cc=rmody@marvell.com \
    --cc=shshaikh@marvell.com \
    --cc=xavier.huwei@huawei.com \
    --cc=xiao.w.wang@intel.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

DPDK patches and discussions

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://inbox.dpdk.org/dev/0 dev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 dev dev/ https://inbox.dpdk.org/dev \
		dev@dpdk.org
	public-inbox-index dev

Example config snippet for mirrors.
Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.dev


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git