Re: [PATCH 3/6] net/vhost: enable compliant offloading mode

[Maxime Coquelin <maxime.coquelin@redhat.com>]

Hi Chenbo,

On 5/16/22 15:26, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -----Original Message-----
>> From: Maxime Coquelin <maxime.coquelin@redhat.com>
>> Sent: Thursday, May 5, 2022 6:27 PM
>> To: dev@dpdk.org; jasowang@redhat.com; Xia, Chenbo <chenbo.xia@intel.com>;
>> david.marchand@redhat.com; olivier.matz@6wind.com
>> Cc: stable@dpdk.org; Maxime Coquelin <maxime.coquelin@redhat.com>
>> Subject: [PATCH 3/6] net/vhost: enable compliant offloading mode
>>
>> This patch enables the compliant offloading flags mode by
>> default, which prevents the Rx path to set Tx offload flags,
>> which is illegal. A new legacy-ol-flags devarg is introduced
>> to enable the legacy behaviour.
>>
>> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
>> ---
>>   doc/guides/nics/vhost.rst         |  6 ++++++
>>   drivers/net/vhost/rte_eth_vhost.c | 19 ++++++++++++++++---
>>   2 files changed, 22 insertions(+), 3 deletions(-)
>>
>> diff --git a/doc/guides/nics/vhost.rst b/doc/guides/nics/vhost.rst
>> index ee802ec4a8..d7c0e2ade8 100644
>> --- a/doc/guides/nics/vhost.rst
>> +++ b/doc/guides/nics/vhost.rst
>> @@ -64,6 +64,12 @@ The user can specify below arguments in `--vdev` option.
>>       It is used to enable external buffer support in vhost library.
>>       (Default: 0 (disabled))
>>
>> +#.  ``legacy-ol-flags``:
>> +
>> +    It is used to restore legacy behavior for offloading that was not
>> +    compliant with offloading API.
>> +    (Default: 0 (disabled))
>> +
>>   Vhost PMD event handling
>>   ------------------------
>>
>> diff --git a/drivers/net/vhost/rte_eth_vhost.c
>> b/drivers/net/vhost/rte_eth_vhost.c
>> index 070f0e6dfd..0a2e8d9b29 100644
>> --- a/drivers/net/vhost/rte_eth_vhost.c
>> +++ b/drivers/net/vhost/rte_eth_vhost.c
>> @@ -31,9 +31,10 @@ enum {VIRTIO_RXQ, VIRTIO_TXQ, VIRTIO_QNUM};
>>   #define ETH_VHOST_CLIENT_ARG		"client"
>>   #define ETH_VHOST_IOMMU_SUPPORT		"iommu-support"
>>   #define ETH_VHOST_POSTCOPY_SUPPORT	"postcopy-support"
>> -#define ETH_VHOST_VIRTIO_NET_F_HOST_TSO "tso"
>> -#define ETH_VHOST_LINEAR_BUF  "linear-buffer"
>> -#define ETH_VHOST_EXT_BUF  "ext-buffer"
>> +#define ETH_VHOST_VIRTIO_NET_F_HOST_TSO	"tso"
>> +#define ETH_VHOST_LINEAR_BUF		"linear-buffer"
>> +#define ETH_VHOST_EXT_BUF		"ext-buffer"
>> +#define ETH_VHOST_LEGACY_OL_FLAGS	"legacy-ol-flags"
>>   #define VHOST_MAX_PKT_BURST 32
>>
>>   static const char *valid_arguments[] = {
>> @@ -1563,6 +1564,7 @@ rte_pmd_vhost_probe(struct rte_vdev_device *dev)
>>   	int tso = 0;
>>   	int linear_buf = 0;
>>   	int ext_buf = 0;
>> +	int legacy_ol_flags = 0;
>>   	struct rte_eth_dev *eth_dev;
>>   	const char *name = rte_vdev_device_name(dev);
>>
>> @@ -1672,6 +1674,17 @@ rte_pmd_vhost_probe(struct rte_vdev_device *dev)
>>   			flags |= RTE_VHOST_USER_EXTBUF_SUPPORT;
>>   	}
>>
>> +	if (rte_kvargs_count(kvlist, ETH_VHOST_LEGACY_OL_FLAGS) == 1) {
>> +		ret = rte_kvargs_process(kvlist,
>> +				ETH_VHOST_LEGACY_OL_FLAGS,
>> +				&open_int, &legacy_ol_flags);
>> +		if (ret < 0)
>> +			goto out_free;
>> +	}
>> +
>> +	if (legacy_ol_flags == 0)
>> +		flags |= RTE_VHOST_USER_NET_COMPLIANT_OL_FLAGS;
> 
> Putting this check into above '{}' like other option does, will look better.

Actually, it would change the behavior because we want to set the flag
even if legacy devrarg is not passed.

Regards,
Maxime

> Thanks,
> Chenbo
> 
>> +
>>   	if (dev->device.numa_node == SOCKET_ID_ANY)
>>   		dev->device.numa_node = rte_socket_id();
>>
>> --
>> 2.35.1
>