[net] virtio-net: unbreak cusmed packet for small buffer XDP
diff mbox series

Message ID 1498614843-8163-1-git-send-email-jasowang@redhat.com
State New, archived
Headers show
Series
  • [net] virtio-net: unbreak cusmed packet for small buffer XDP
Related show

Commit Message

Jason Wang June 28, 2017, 1:54 a.m. UTC
We should allow csumed packet for small buffer, otherwise XDP_PASS
won't work correctly.

Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
The patch is needed for -stable.
---
 drivers/net/virtio_net.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Michael S. Tsirkin June 28, 2017, 2:02 a.m. UTC | #1
On Wed, Jun 28, 2017 at 09:54:03AM +0800, Jason Wang wrote:
> We should allow csumed packet for small buffer, otherwise XDP_PASS
> won't work correctly.
> 
> Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
> Signed-off-by: Jason Wang <jasowang@redhat.com>

The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set.
What do you think?

> ---
> The patch is needed for -stable.
> ---
>  drivers/net/virtio_net.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 143d8a9..499fcc9 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -413,7 +413,7 @@ static struct sk_buff *receive_small(struct net_device *dev,
>  		void *orig_data;
>  		u32 act;
>  
> -		if (unlikely(hdr->hdr.gso_type || hdr->hdr.flags))
> +		if (unlikely(hdr->hdr.gso_type))
>  			goto err_xdp;
>  
>  		xdp.data_hard_start = buf + VIRTNET_RX_PAD + vi->hdr_len;
> -- 
> 2.7.4
Jason Wang June 28, 2017, 2:14 a.m. UTC | #2
On 2017年06月28日 10:02, Michael S. Tsirkin wrote:
> On Wed, Jun 28, 2017 at 09:54:03AM +0800, Jason Wang wrote:
>> We should allow csumed packet for small buffer, otherwise XDP_PASS
>> won't work correctly.
>>
>> Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
>> Signed-off-by: Jason Wang <jasowang@redhat.com>
> The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set.
> What do you think?

I think it's safe. For XDP_PASS, it work like in the past. For XDP_TX, 
we zero the vnet header. For adjusting header, XDP prog should deal with 
csum.

Thanks

>
>> ---
>> The patch is needed for -stable.
>> ---
>>   drivers/net/virtio_net.c | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
>> index 143d8a9..499fcc9 100644
>> --- a/drivers/net/virtio_net.c
>> +++ b/drivers/net/virtio_net.c
>> @@ -413,7 +413,7 @@ static struct sk_buff *receive_small(struct net_device *dev,
>>   		void *orig_data;
>>   		u32 act;
>>   
>> -		if (unlikely(hdr->hdr.gso_type || hdr->hdr.flags))
>> +		if (unlikely(hdr->hdr.gso_type))
>>   			goto err_xdp;
>>   
>>   		xdp.data_hard_start = buf + VIRTNET_RX_PAD + vi->hdr_len;
>> -- 
>> 2.7.4
Michael S. Tsirkin June 28, 2017, 2:17 a.m. UTC | #3
On Wed, Jun 28, 2017 at 10:14:34AM +0800, Jason Wang wrote:
> 
> 
> On 2017年06月28日 10:02, Michael S. Tsirkin wrote:
> > On Wed, Jun 28, 2017 at 09:54:03AM +0800, Jason Wang wrote:
> > > We should allow csumed packet for small buffer, otherwise XDP_PASS
> > > won't work correctly.
> > > 
> > > Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
> > > Signed-off-by: Jason Wang <jasowang@redhat.com>
> > The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set.
> > What do you think?
> 
> I think it's safe. For XDP_PASS, it work like in the past.

That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP
tools assume it's value.

> For XDP_TX, we
> zero the vnet header.

Again TX offload is disabled, so packets will go out with an invalid
checksum.

> For adjusting header, XDP prog should deal with csum.
> 
> Thanks

That part seems right.

> > 
> > > ---
> > > The patch is needed for -stable.
> > > ---
> > >   drivers/net/virtio_net.c | 2 +-
> > >   1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> > > index 143d8a9..499fcc9 100644
> > > --- a/drivers/net/virtio_net.c
> > > +++ b/drivers/net/virtio_net.c
> > > @@ -413,7 +413,7 @@ static struct sk_buff *receive_small(struct net_device *dev,
> > >   		void *orig_data;
> > >   		u32 act;
> > > -		if (unlikely(hdr->hdr.gso_type || hdr->hdr.flags))
> > > +		if (unlikely(hdr->hdr.gso_type))
> > >   			goto err_xdp;
> > >   		xdp.data_hard_start = buf + VIRTNET_RX_PAD + vi->hdr_len;
> > > -- 
> > > 2.7.4
Jason Wang June 28, 2017, 2:45 a.m. UTC | #4
On 2017年06月28日 10:17, Michael S. Tsirkin wrote:
> On Wed, Jun 28, 2017 at 10:14:34AM +0800, Jason Wang wrote:
>>
>> On 2017年06月28日 10:02, Michael S. Tsirkin wrote:
>>> On Wed, Jun 28, 2017 at 09:54:03AM +0800, Jason Wang wrote:
>>>> We should allow csumed packet for small buffer, otherwise XDP_PASS
>>>> won't work correctly.
>>>>
>>>> Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
>>>> Signed-off-by: Jason Wang <jasowang@redhat.com>
>>> The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set.
>>> What do you think?
>> I think it's safe. For XDP_PASS, it work like in the past.
> That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP
> tools assume it's value.

DATA_VALID is CHECKSUM_UNCESSARY on the host, and according to the 
comment in skbuff.h


"
  *   The hardware you're dealing with doesn't calculate the full checksum
  *   (as in CHECKSUM_COMPLETE), but it does parse headers and verify 
checksums
  *   for specific protocols. For such packets it will set 
CHECKSUM_UNNECESSARY
  *   if their checksums are okay. skb->csum is still undefined in this case
  *   though. A driver or device must never modify the checksum field in the
  *   packet even if checksum is verified.
"

The csum is correct I believe?

Thanks

>
>> For XDP_TX, we
>> zero the vnet header.
> Again TX offload is disabled, so packets will go out with an invalid
> checksum.
>
>> For adjusting header, XDP prog should deal with csum.
>>
>> Thanks
> That part seems right.
>
>>>> ---
>>>> The patch is needed for -stable.
>>>> ---
>>>>    drivers/net/virtio_net.c | 2 +-
>>>>    1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
>>>> index 143d8a9..499fcc9 100644
>>>> --- a/drivers/net/virtio_net.c
>>>> +++ b/drivers/net/virtio_net.c
>>>> @@ -413,7 +413,7 @@ static struct sk_buff *receive_small(struct net_device *dev,
>>>>    		void *orig_data;
>>>>    		u32 act;
>>>> -		if (unlikely(hdr->hdr.gso_type || hdr->hdr.flags))
>>>> +		if (unlikely(hdr->hdr.gso_type))
>>>>    			goto err_xdp;
>>>>    		xdp.data_hard_start = buf + VIRTNET_RX_PAD + vi->hdr_len;
>>>> -- 
>>>> 2.7.4
Michael S. Tsirkin June 28, 2017, 3:31 a.m. UTC | #5
On Wed, Jun 28, 2017 at 10:45:18AM +0800, Jason Wang wrote:
> 
> 
> On 2017年06月28日 10:17, Michael S. Tsirkin wrote:
> > On Wed, Jun 28, 2017 at 10:14:34AM +0800, Jason Wang wrote:
> > > 
> > > On 2017年06月28日 10:02, Michael S. Tsirkin wrote:
> > > > On Wed, Jun 28, 2017 at 09:54:03AM +0800, Jason Wang wrote:
> > > > > We should allow csumed packet for small buffer, otherwise XDP_PASS
> > > > > won't work correctly.
> > > > > 
> > > > > Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
> > > > > Signed-off-by: Jason Wang <jasowang@redhat.com>
> > > > The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set.
> > > > What do you think?
> > > I think it's safe. For XDP_PASS, it work like in the past.
> > That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP
> > tools assume it's value.
> 
> DATA_VALID is CHECKSUM_UNCESSARY on the host, and according to the comment
> in skbuff.h
> 
> 
> "
>  *   The hardware you're dealing with doesn't calculate the full checksum
>  *   (as in CHECKSUM_COMPLETE), but it does parse headers and verify
> checksums
>  *   for specific protocols. For such packets it will set
> CHECKSUM_UNNECESSARY
>  *   if their checksums are okay. skb->csum is still undefined in this case
>  *   though. A driver or device must never modify the checksum field in the
>  *   packet even if checksum is verified.
> "
> 
> The csum is correct I believe?
> 
> Thanks

That's on input. But I think for tun it's output, where that is equivalent
to CHECKSUM_NONE


> > 
> > > For XDP_TX, we
> > > zero the vnet header.
> > Again TX offload is disabled, so packets will go out with an invalid
> > checksum.
> > 
> > > For adjusting header, XDP prog should deal with csum.
> > > 
> > > Thanks
> > That part seems right.
> > 
> > > > > ---
> > > > > The patch is needed for -stable.
> > > > > ---
> > > > >    drivers/net/virtio_net.c | 2 +-
> > > > >    1 file changed, 1 insertion(+), 1 deletion(-)
> > > > > 
> > > > > diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> > > > > index 143d8a9..499fcc9 100644
> > > > > --- a/drivers/net/virtio_net.c
> > > > > +++ b/drivers/net/virtio_net.c
> > > > > @@ -413,7 +413,7 @@ static struct sk_buff *receive_small(struct net_device *dev,
> > > > >    		void *orig_data;
> > > > >    		u32 act;
> > > > > -		if (unlikely(hdr->hdr.gso_type || hdr->hdr.flags))
> > > > > +		if (unlikely(hdr->hdr.gso_type))
> > > > >    			goto err_xdp;
> > > > >    		xdp.data_hard_start = buf + VIRTNET_RX_PAD + vi->hdr_len;
> > > > > -- 
> > > > > 2.7.4
Jason Wang June 28, 2017, 3:40 a.m. UTC | #6
On 2017年06月28日 11:31, Michael S. Tsirkin wrote:
> On Wed, Jun 28, 2017 at 10:45:18AM +0800, Jason Wang wrote:
>> On 2017年06月28日 10:17, Michael S. Tsirkin wrote:
>>> On Wed, Jun 28, 2017 at 10:14:34AM +0800, Jason Wang wrote:
>>>> On 2017年06月28日 10:02, Michael S. Tsirkin wrote:
>>>>> On Wed, Jun 28, 2017 at 09:54:03AM +0800, Jason Wang wrote:
>>>>>> We should allow csumed packet for small buffer, otherwise XDP_PASS
>>>>>> won't work correctly.
>>>>>>
>>>>>> Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
>>>>>> Signed-off-by: Jason Wang<jasowang@redhat.com>
>>>>> The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set.
>>>>> What do you think?
>>>> I think it's safe. For XDP_PASS, it work like in the past.
>>> That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP
>>> tools assume it's value.
>> DATA_VALID is CHECKSUM_UNCESSARY on the host, and according to the comment
>> in skbuff.h
>>
>>
>> "
>>   *   The hardware you're dealing with doesn't calculate the full checksum
>>   *   (as in CHECKSUM_COMPLETE), but it does parse headers and verify
>> checksums
>>   *   for specific protocols. For such packets it will set
>> CHECKSUM_UNNECESSARY
>>   *   if their checksums are okay. skb->csum is still undefined in this case
>>   *   though. A driver or device must never modify the checksum field in the
>>   *   packet even if checksum is verified.
>> "
>>
>> The csum is correct I believe?
>>
>> Thanks
> That's on input. But I think for tun it's output, where that is equivalent
> to CHECKSUM_NONE
>
>

Yes, but the comment said:

"
CKSUM_NONE:
  *
  *   The skb was already checksummed by the protocol, or a checksum is not
  *   required.
  *
  * CHECKSUM_UNNECESSARY:
  *
  *   This has the same meaning on as CHECKSUM_NONE for checksum offload on
  *   output.
  *
"

So still correct I think?

Thanks
Michael S. Tsirkin June 28, 2017, 4:01 a.m. UTC | #7
On Wed, Jun 28, 2017 at 11:40:30AM +0800, Jason Wang wrote:
> 
> 
> On 2017年06月28日 11:31, Michael S. Tsirkin wrote:
> > On Wed, Jun 28, 2017 at 10:45:18AM +0800, Jason Wang wrote:
> > > On 2017年06月28日 10:17, Michael S. Tsirkin wrote:
> > > > On Wed, Jun 28, 2017 at 10:14:34AM +0800, Jason Wang wrote:
> > > > > On 2017年06月28日 10:02, Michael S. Tsirkin wrote:
> > > > > > On Wed, Jun 28, 2017 at 09:54:03AM +0800, Jason Wang wrote:
> > > > > > > We should allow csumed packet for small buffer, otherwise XDP_PASS
> > > > > > > won't work correctly.
> > > > > > > 
> > > > > > > Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
> > > > > > > Signed-off-by: Jason Wang<jasowang@redhat.com>
> > > > > > The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set.
> > > > > > What do you think?
> > > > > I think it's safe. For XDP_PASS, it work like in the past.
> > > > That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP
> > > > tools assume it's value.
> > > DATA_VALID is CHECKSUM_UNCESSARY on the host, and according to the comment
> > > in skbuff.h
> > > 
> > > 
> > > "
> > >   *   The hardware you're dealing with doesn't calculate the full checksum
> > >   *   (as in CHECKSUM_COMPLETE), but it does parse headers and verify
> > > checksums
> > >   *   for specific protocols. For such packets it will set
> > > CHECKSUM_UNNECESSARY
> > >   *   if their checksums are okay. skb->csum is still undefined in this case
> > >   *   though. A driver or device must never modify the checksum field in the
> > >   *   packet even if checksum is verified.
> > > "
> > > 
> > > The csum is correct I believe?
> > > 
> > > Thanks
> > That's on input. But I think for tun it's output, where that is equivalent
> > to CHECKSUM_NONE
> > 
> > 
> 
> Yes, but the comment said:
> 
> "
> CKSUM_NONE:
>  *
>  *   The skb was already checksummed by the protocol, or a checksum is not
>  *   required.
>  *
>  * CHECKSUM_UNNECESSARY:
>  *
>  *   This has the same meaning on as CHECKSUM_NONE for checksum offload on
>  *   output.
>  *
> "
> 
> So still correct I think?
> 
> Thanks

Hmm maybe I mean NEEDS_CHECKSUM actually.

I'll need to re-read the spec.
Jason Wang June 28, 2017, 12:05 p.m. UTC | #8
On 2017年06月28日 12:01, Michael S. Tsirkin wrote:
> On Wed, Jun 28, 2017 at 11:40:30AM +0800, Jason Wang wrote:
>>
>> On 2017年06月28日 11:31, Michael S. Tsirkin wrote:
>>> On Wed, Jun 28, 2017 at 10:45:18AM +0800, Jason Wang wrote:
>>>> On 2017年06月28日 10:17, Michael S. Tsirkin wrote:
>>>>> On Wed, Jun 28, 2017 at 10:14:34AM +0800, Jason Wang wrote:
>>>>>> On 2017年06月28日 10:02, Michael S. Tsirkin wrote:
>>>>>>> On Wed, Jun 28, 2017 at 09:54:03AM +0800, Jason Wang wrote:
>>>>>>>> We should allow csumed packet for small buffer, otherwise XDP_PASS
>>>>>>>> won't work correctly.
>>>>>>>>
>>>>>>>> Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
>>>>>>>> Signed-off-by: Jason Wang<jasowang@redhat.com>
>>>>>>> The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set.
>>>>>>> What do you think?
>>>>>> I think it's safe. For XDP_PASS, it work like in the past.
>>>>> That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP
>>>>> tools assume it's value.
>>>> DATA_VALID is CHECKSUM_UNCESSARY on the host, and according to the comment
>>>> in skbuff.h
>>>>
>>>>
>>>> "
>>>>    *   The hardware you're dealing with doesn't calculate the full checksum
>>>>    *   (as in CHECKSUM_COMPLETE), but it does parse headers and verify
>>>> checksums
>>>>    *   for specific protocols. For such packets it will set
>>>> CHECKSUM_UNNECESSARY
>>>>    *   if their checksums are okay. skb->csum is still undefined in this case
>>>>    *   though. A driver or device must never modify the checksum field in the
>>>>    *   packet even if checksum is verified.
>>>> "
>>>>
>>>> The csum is correct I believe?
>>>>
>>>> Thanks
>>> That's on input. But I think for tun it's output, where that is equivalent
>>> to CHECKSUM_NONE
>>>
>>>
>> Yes, but the comment said:
>>
>> "
>> CKSUM_NONE:
>>   *
>>   *   The skb was already checksummed by the protocol, or a checksum is not
>>   *   required.
>>   *
>>   * CHECKSUM_UNNECESSARY:
>>   *
>>   *   This has the same meaning on as CHECKSUM_NONE for checksum offload on
>>   *   output.
>>   *
>> "
>>
>> So still correct I think?
>>
>> Thanks
> Hmm maybe I mean NEEDS_CHECKSUM actually.
>
> I'll need to re-read the spec.
>

Not sure this is an issue. But if it is, we can probably checksum the 
packet before passing it to XDP. But it would be a little slow.

Thanks
Michael S. Tsirkin July 3, 2017, 5:03 p.m. UTC | #9
On Wed, Jun 28, 2017 at 08:05:06PM +0800, Jason Wang wrote:
> 
> 
> On 2017年06月28日 12:01, Michael S. Tsirkin wrote:
> > On Wed, Jun 28, 2017 at 11:40:30AM +0800, Jason Wang wrote:
> > > 
> > > On 2017年06月28日 11:31, Michael S. Tsirkin wrote:
> > > > On Wed, Jun 28, 2017 at 10:45:18AM +0800, Jason Wang wrote:
> > > > > On 2017年06月28日 10:17, Michael S. Tsirkin wrote:
> > > > > > On Wed, Jun 28, 2017 at 10:14:34AM +0800, Jason Wang wrote:
> > > > > > > On 2017年06月28日 10:02, Michael S. Tsirkin wrote:
> > > > > > > > On Wed, Jun 28, 2017 at 09:54:03AM +0800, Jason Wang wrote:
> > > > > > > > > We should allow csumed packet for small buffer, otherwise XDP_PASS
> > > > > > > > > won't work correctly.
> > > > > > > > > 
> > > > > > > > > Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
> > > > > > > > > Signed-off-by: Jason Wang<jasowang@redhat.com>
> > > > > > > > The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set.
> > > > > > > > What do you think?
> > > > > > > I think it's safe. For XDP_PASS, it work like in the past.
> > > > > > That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP
> > > > > > tools assume it's value.
> > > > > DATA_VALID is CHECKSUM_UNCESSARY on the host, and according to the comment
> > > > > in skbuff.h
> > > > > 
> > > > > 
> > > > > "
> > > > >    *   The hardware you're dealing with doesn't calculate the full checksum
> > > > >    *   (as in CHECKSUM_COMPLETE), but it does parse headers and verify
> > > > > checksums
> > > > >    *   for specific protocols. For such packets it will set
> > > > > CHECKSUM_UNNECESSARY
> > > > >    *   if their checksums are okay. skb->csum is still undefined in this case
> > > > >    *   though. A driver or device must never modify the checksum field in the
> > > > >    *   packet even if checksum is verified.
> > > > > "
> > > > > 
> > > > > The csum is correct I believe?
> > > > > 
> > > > > Thanks
> > > > That's on input. But I think for tun it's output, where that is equivalent
> > > > to CHECKSUM_NONE
> > > > 
> > > > 
> > > Yes, but the comment said:
> > > 
> > > "
> > > CKSUM_NONE:
> > >   *
> > >   *   The skb was already checksummed by the protocol, or a checksum is not
> > >   *   required.
> > >   *
> > >   * CHECKSUM_UNNECESSARY:
> > >   *
> > >   *   This has the same meaning on as CHECKSUM_NONE for checksum offload on
> > >   *   output.
> > >   *
> > > "
> > > 
> > > So still correct I think?
> > > 
> > > Thanks
> > Hmm maybe I mean NEEDS_CHECKSUM actually.
> > 
> > I'll need to re-read the spec.
> > 
> 
> Not sure this is an issue. But if it is, we can probably checksum the packet
> before passing it to XDP. But it would be a little slow.
> 
> Thanks



Right. I confused DATA_VALID with NEEDS_CHECKSUM.

IIUC XDP generally refuses to attach if checksum offload
is enabled.

Could you pls explain how to reproduce the issue you are seeing?
Jason Wang July 4, 2017, 12:20 p.m. UTC | #10
On 2017年07月04日 01:03, Michael S. Tsirkin wrote:
> On Wed, Jun 28, 2017 at 08:05:06PM +0800, Jason Wang wrote:
>>
>> On 2017年06月28日 12:01, Michael S. Tsirkin wrote:
>>> On Wed, Jun 28, 2017 at 11:40:30AM +0800, Jason Wang wrote:
>>>> On 2017年06月28日 11:31, Michael S. Tsirkin wrote:
>>>>> On Wed, Jun 28, 2017 at 10:45:18AM +0800, Jason Wang wrote:
>>>>>> On 2017年06月28日 10:17, Michael S. Tsirkin wrote:
>>>>>>> On Wed, Jun 28, 2017 at 10:14:34AM +0800, Jason Wang wrote:
>>>>>>>> On 2017年06月28日 10:02, Michael S. Tsirkin wrote:
>>>>>>>>> On Wed, Jun 28, 2017 at 09:54:03AM +0800, Jason Wang wrote:
>>>>>>>>>> We should allow csumed packet for small buffer, otherwise XDP_PASS
>>>>>>>>>> won't work correctly.
>>>>>>>>>>
>>>>>>>>>> Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers")
>>>>>>>>>> Signed-off-by: Jason Wang<jasowang@redhat.com>
>>>>>>>>> The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set.
>>>>>>>>> What do you think?
>>>>>>>> I think it's safe. For XDP_PASS, it work like in the past.
>>>>>>> That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP
>>>>>>> tools assume it's value.
>>>>>> DATA_VALID is CHECKSUM_UNCESSARY on the host, and according to the comment
>>>>>> in skbuff.h
>>>>>>
>>>>>>
>>>>>> "
>>>>>>     *   The hardware you're dealing with doesn't calculate the full checksum
>>>>>>     *   (as in CHECKSUM_COMPLETE), but it does parse headers and verify
>>>>>> checksums
>>>>>>     *   for specific protocols. For such packets it will set
>>>>>> CHECKSUM_UNNECESSARY
>>>>>>     *   if their checksums are okay. skb->csum is still undefined in this case
>>>>>>     *   though. A driver or device must never modify the checksum field in the
>>>>>>     *   packet even if checksum is verified.
>>>>>> "
>>>>>>
>>>>>> The csum is correct I believe?
>>>>>>
>>>>>> Thanks
>>>>> That's on input. But I think for tun it's output, where that is equivalent
>>>>> to CHECKSUM_NONE
>>>>>
>>>>>
>>>> Yes, but the comment said:
>>>>
>>>> "
>>>> CKSUM_NONE:
>>>>    *
>>>>    *   The skb was already checksummed by the protocol, or a checksum is not
>>>>    *   required.
>>>>    *
>>>>    * CHECKSUM_UNNECESSARY:
>>>>    *
>>>>    *   This has the same meaning on as CHECKSUM_NONE for checksum offload on
>>>>    *   output.
>>>>    *
>>>> "
>>>>
>>>> So still correct I think?
>>>>
>>>> Thanks
>>> Hmm maybe I mean NEEDS_CHECKSUM actually.
>>>
>>> I'll need to re-read the spec.
>>>
>> Not sure this is an issue. But if it is, we can probably checksum the packet
>> before passing it to XDP. But it would be a little slow.
>>
>> Thanks
>
>
> Right. I confused DATA_VALID with NEEDS_CHECKSUM.
>
> IIUC XDP generally refuses to attach if checksum offload
> is enabled.

Any reason to do this? (Looks like I don't see any code for this)

>
> Could you pls explain how to reproduce the issue you are seeing?
>

Using small buffer, all csumed packets will be dropped.

Thanks
Michael S. Tsirkin July 6, 2017, 12:07 a.m. UTC | #11
On Tue, Jul 04, 2017 at 08:20:00PM +0800, Jason Wang wrote:
> > IIUC XDP generally refuses to attach if checksum offload
> > is enabled.
> 
> Any reason to do this? (Looks like I don't see any code for this)

Some of it was covered here
https://www.mail-archive.com/netdev@vger.kernel.org/msg162577.html

Patch
diff mbox series

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 143d8a9..499fcc9 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -413,7 +413,7 @@  static struct sk_buff *receive_small(struct net_device *dev,
 		void *orig_data;
 		u32 act;
 
-		if (unlikely(hdr->hdr.gso_type || hdr->hdr.flags))
+		if (unlikely(hdr->hdr.gso_type))
 			goto err_xdp;
 
 		xdp.data_hard_start = buf + VIRTNET_RX_PAD + vi->hdr_len;