DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ferruh Yigit <ferruh.yigit@intel.com>
To: wangyunjian <wangyunjian@huawei.com>, dev@dpdk.org
Cc: keith.wiles@intel.com, jerry.lilijun@huawei.com,
	xudingke@huawei.com, stable@dpdk.org
Subject: Re: [dpdk-dev] [dpdk-stable] [PATCH v3 3/5] net/tap: fix check for mbuf's nb_segs failure
Date: Tue, 7 Apr 2020 16:58:02 +0100	[thread overview]
Message-ID: <d87bf0f4-e7b0-7bb8-5ffa-7ec6ec35502c@intel.com> (raw)
In-Reply-To: <7da1388e-c4c6-27d5-f038-0526a39d9a8c@intel.com>

On 4/7/2020 4:15 PM, Ferruh Yigit wrote:
> On 4/7/2020 5:23 AM, wangyunjian wrote:
>> From: Yunjian Wang <wangyunjian@huawei.com>
>>
>> Now the rxq->pool is mbuf concatenation, But its nb_segs is 1.
>> When do some sanity checks on the mbuf, it fails.
> 
> +1, 'rxq->pool' seems Rx ring representation as linked mbufs and empty ones has
> 'nb_segs' values as 1.
> 
>>
>> Fixes: 0781f5762cfe ("net/tap: support segmented mbufs")
>> CC: stable@dpdk.org
>>
>> Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
>> ---
>>  drivers/net/tap/rte_eth_tap.c | 27 ++++++++++++++++++++++-----
>>  1 file changed, 22 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/net/tap/rte_eth_tap.c b/drivers/net/tap/rte_eth_tap.c
>> index a9ba0ca68..703fcceb9 100644
>> --- a/drivers/net/tap/rte_eth_tap.c
>> +++ b/drivers/net/tap/rte_eth_tap.c
>> @@ -339,6 +339,23 @@ tap_rx_offload_get_queue_capa(void)
>>  	       DEV_RX_OFFLOAD_TCP_CKSUM;
>>  }
>>  
>> +static void
>> +tap_rxq_pool_free(struct rte_mbuf *pool)
>> +{
>> +	struct rte_mbuf *mbuf = pool;
>> +	uint16_t nb_segs = 1;
>> +
>> +	if (mbuf == NULL)
>> +		return;
>> +
>> +	while (mbuf->next) {
>> +		mbuf = mbuf->next;
>> +		nb_segs++;
>> +	}
>> +	pool->nb_segs = nb_segs;
>> +	rte_pktmbuf_free(pool);
>> +}
> 
> Since you are already iterating the chain, why not free immediately instead of
> calculating the nb_segs and making API go through the chain again, what about
> following:
> 
> tap_rxq_pool_free(struct rte_mbuf *pool)
> {
>     struct rte_mbuf *next;
>     while (pool) {
>          next = pool->next;
>          rte_pktmbuf_free(pool);
>          pool = next;
>     }
> }

Ignore this please, this may be still complaining in mbuf sanity check, so OK to
your usage.

> 
>> +
>>  /* Callback to handle the rx burst of packets to the correct interface and
>>   * file descriptor(s) in a multi-queue setup.
>>   */
>> @@ -389,7 +406,7 @@ pmd_rx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
>>  					goto end;
>>  
>>  				seg->next = NULL;
>> -				rte_pktmbuf_free(mbuf);
>> +				tap_rxq_pool_free(mbuf);
> 
> As far as I can see 'mbuf' should have correct 'nb_segs' value, and it can
> continue to use 'rte_pktmbuf_free()'. If you can observe the problem can you
> please try this?
> 
>>  
>>  				goto end;
>>  			}
>> @@ -1033,7 +1050,7 @@ tap_dev_close(struct rte_eth_dev *dev)
>>  			rxq = &internals->rxq[i];
>>  			close(process_private->rxq_fds[i]);
>>  			process_private->rxq_fds[i] = -1;
>> -			rte_pktmbuf_free(rxq->pool);
>> +			tap_rxq_pool_free(rxq->pool);
>>  			rte_free(rxq->iovecs);
>>  			rxq->pool = NULL;
>>  			rxq->iovecs = NULL;
>> @@ -1072,7 +1089,7 @@ tap_rx_queue_release(void *queue)
>>  	if (process_private->rxq_fds[rxq->queue_id] > 0) {
>>  		close(process_private->rxq_fds[rxq->queue_id]);
>>  		process_private->rxq_fds[rxq->queue_id] = -1;
>> -		rte_pktmbuf_free(rxq->pool);
>> +		tap_rxq_pool_free(rxq->pool);
>>  		rte_free(rxq->iovecs);
>>  		rxq->pool = NULL;
>>  		rxq->iovecs = NULL;
>> @@ -1480,7 +1497,7 @@ tap_rx_queue_setup(struct rte_eth_dev *dev,
>>  	return 0;
>>  
>>  error:
>> -	rte_pktmbuf_free(rxq->pool);
>> +	tap_rxq_pool_free(rxq->pool);
>>  	rxq->pool = NULL;
>>  	rte_free(rxq->iovecs);
>>  	rxq->iovecs = NULL;
>> @@ -2435,7 +2452,7 @@ rte_pmd_tap_remove(struct rte_vdev_device *dev)
>>  			rxq = &internals->rxq[i];
>>  			close(process_private->rxq_fds[i]);
>>  			process_private->rxq_fds[i] = -1;
>> -			rte_pktmbuf_free(rxq->pool);
>> +			tap_rxq_pool_free(rxq->pool);
>>  			rte_free(rxq->iovecs);
>>  			rxq->pool = NULL;
>>  			rxq->iovecs = NULL;
>>
> 


      parent reply	other threads:[~2020-04-07 15:58 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-07  4:23 [dpdk-dev] " wangyunjian
2020-04-07 15:15 ` [dpdk-dev] [dpdk-stable] " Ferruh Yigit
2020-04-07 15:38   ` Stephen Hemminger
2020-04-07 15:45     ` Ferruh Yigit
2020-04-07 15:49       ` Ferruh Yigit
2020-04-07 16:08       ` Stephen Hemminger
2020-04-08  1:10         ` wangyunjian
2020-04-07 15:58   ` Ferruh Yigit [this message]

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=d87bf0f4-e7b0-7bb8-5ffa-7ec6ec35502c@intel.com \
    --to=ferruh.yigit@intel.com \
    --cc=dev@dpdk.org \
    --cc=jerry.lilijun@huawei.com \
    --cc=keith.wiles@intel.com \
    --cc=stable@dpdk.org \
    --cc=wangyunjian@huawei.com \
    --cc=xudingke@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).