On 2015/11/13 13:03, Rich Lane wrote:
>> + if (rte_kvargs_count(kvlist, ETH_VHOST_IFACE_ARG) == 1) {
>> + ret = rte_kvargs_process(kvlist, ETH_VHOST_IFACE_ARG,
>> +&open_iface, &iface_name);
>> + if (ret < 0)
>> +
On 2015/11/12 21:52, Wang, Zhihong wrote:
> Hi Tetsuya,
>
> In my test I created 2 vdev using "--vdev
> 'eth_vhost0,iface=/tmp/sock0,queues=1' --vdev
> 'eth_vhost1,iface=/tmp/sock1,queues=1'", and the qemu message got handled in
> wrong order.
> The reason is that: 2 threads are created to handl
> -Original Message-
> From: Tetsuya Mukawa [mailto:mukawa at igel.co.jp]
> Sent: Friday, November 13, 2015 11:10 AM
> To: Wang, Zhihong ; dev at dpdk.org; Liu, Yuanhan
>
> Cc: ann.zhuangyanying at huawei.com
> Subject: Re: [dpdk-dev] [PATCH v3 2/2] vhost: Add VHOS
>
> + if (rte_kvargs_count(kvlist, ETH_VHOST_IFACE_ARG) == 1) {
> + ret = rte_kvargs_process(kvlist, ETH_VHOST_IFACE_ARG,
> +&open_iface, &iface_name);
> + if (ret < 0)
> + goto out_free;
> + }
>
Hi Tetsuya,
In my test I created 2 vdev using "--vdev
'eth_vhost0,iface=/tmp/sock0,queues=1' --vdev
'eth_vhost1,iface=/tmp/sock1,queues=1'", and the qemu message got handled in
wrong order.
The reason is that: 2 threads are created to handle message from 2 sockets, but
their fds are SHARED, so
On 2015/11/10 7:22, Stephen Hemminger wrote:
> On Mon, 9 Nov 2015 14:17:01 +0900
> Tetsuya Mukawa wrote:
>
>> +rte_atomic32_set(&r->while_queuing, 1);
>> +
>> +if (unlikely(rte_atomic32_read(&r->allow_queuing) == 0))
>> +goto out;
> You special 2 variable custom locking here i
Hi Liu,
Thank you so much for your reviewing.
I will fix them, then submit again in this week.
Thanks,
Tetsuya
On 2015/11/09 15:21, Yuanhan Liu wrote:
> Hi Tetsuya,
>
> Here I just got some minor nits after a very rough glimpse.
>
> On Mon, Nov 09, 2015 at 02:17:01PM +0900, Tetsuya Mukawa wrote
On Mon, 9 Nov 2015 14:17:01 +0900
Tetsuya Mukawa wrote:
> + rte_atomic32_set(&r->while_queuing, 1);
> +
> + if (unlikely(rte_atomic32_read(&r->allow_queuing) == 0))
> + goto out;
You special 2 variable custom locking here is buggy.
If you hit second atomic test, you will lea
Hi Tetsuya,
Here I just got some minor nits after a very rough glimpse.
On Mon, Nov 09, 2015 at 02:17:01PM +0900, Tetsuya Mukawa wrote:
...
> +static uint16_t
> +eth_vhost_rx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs)
> +{
> + struct vhost_queue *r = q;
> + uint16_t nb_rx = 0;
> +
The patch introduces a new PMD. This PMD is implemented as thin wrapper
of librte_vhost. It means librte_vhost is also needed to compile the PMD.
The vhost messages will be handled only when a port is started. So start
a port first, then invoke QEMU.
The PMD has 2 parameters.
- iface: The parame
10 matches
Mail list logo
