Re: [PATCH net-next 0/8] net: The beginning of the end for NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM

[Tom Herbert <tom@herbertland.com>]

On Tue, Dec 15, 2015 at 2:11 PM, Tom Herbert <tom@herbertland.com> wrote:
> On Tue, Dec 15, 2015 at 1:32 PM, David Miller <davem@davemloft.net> wrote:
>> From: Tom Herbert <tom@herbertland.com>
>> Date: Mon, 14 Dec 2015 11:19:39 -0800
>>
>>> This patch set starts to address one front in the battle against
>>> protocol ossification.
>>
>> Looks good series applied, thanks.
>>
>> Two things:
>>
>> 1) I'd ask you to reconsider passing that struct with the bit fields into
>>    the chk routines.  Some cpu ABIs pass that thing as an aggregate on
>>    the stack.  Whereas if you just use a simple bit mask and BIT(x) values
>>    we know it's always going to be passed in as an integer or whatever.
>>
>> 2) As per the TCP check, it is definitely more strict now.  But I seriously
>>    doubt get ever gained anything real with the previous, looser, check.
>>    Besides, if it bites a user segment we will get a report and can adjust
>>    things.
>>
>> Thanks!
>
> Looks like I may have missed a NETIF_F_ALL_CSUM in netronome driver,
> will post fix shortly.

Actually looks okay, did you fix this up Dave?

Thanks,
Tom