[dpdk-dev] [PATCH] vhost: zero_copy incompatible with client mode

[Maxime Coquelin]

Hi Yinan,

On 4/29/20 10:33 AM, Wang, Yinan wrote:
> Tested-by: Yinan Wang <yinan.wang at intel.com>

Thanks for running tests on Vhost patches, that's much appreciated.

Could you provide more information on the test run?
Does it consist in reproducing the issue the patch is aiming to fix,
apply patch and check issue is fixed?
Or does it consist in running a generic Vhost/Virtio test suite?

Regards,
Maxime

>> -----Original Message-----
>> From: dev <dev-bounces at dpdk.org> On Behalf Of Xuan Ding
>> Sent: 2020年4月29日 11:00
>> To: maxime.coquelin at redhat.com; Wang, Zhihong <zhihong.wang at intel.com>;
>> Ye, Xiaolong <xiaolong.ye at intel.com>
>> Cc: dev at dpdk.org; Ding, Xuan <xuan.ding at intel.com>
>> Subject: [dpdk-dev] [PATCH] vhost: zero_copy incompatible with client mode
>>
>> In server mode, virtio-user inits under the assumption that vhost-user supports a
>> list of features. However, this could be problematic when in_order feature is
>> negotiated but not supported by vhost-user when enables dequeue_zero_copy
>> later.
>>
>> Add handling when vhost-user enables dequeue_zero_copy as client.
>>
>> Signed-off-by: Xuan Ding <xuan.ding at intel.com>
>> ---
>>  lib/librte_vhost/socket.c | 6 ++++++
>>  1 file changed, 6 insertions(+)
>>
>> diff --git a/lib/librte_vhost/socket.c b/lib/librte_vhost/socket.c index
>> 7c8012179..bb8d0d780 100644
>> --- a/lib/librte_vhost/socket.c
>> +++ b/lib/librte_vhost/socket.c
>> @@ -926,6 +926,12 @@ rte_vhost_driver_register(const char *path, uint64_t
>> flags)
>>  			ret = -1;
>>  			goto out_mutex;
>>  		}
>> +		if (!vsocket->is_server) {
>> +			VHOST_LOG_CONFIG(ERR,
>> +			"error: zero copy is incompatible with vhost client
>> mode\n");
>> +			ret = -1;
>> +			goto out_mutex;
>> +		}
>>  		vsocket->supported_features &= ~(1ULL <<
>> VIRTIO_F_IN_ORDER);
>>  		vsocket->features &= ~(1ULL << VIRTIO_F_IN_ORDER);
>>
>> --
>> 2.17.1
>