On 02/15/2014 02:30 PM, Patrick McHardy wrote: > On 15. Februar 2014 13:17:22 GMT+00:00, Nikolay Aleksandrov <nikolay@xxxxxxxxxx> wrote: >> The new "write" expression can be used to manipulate packet data. >> The parameters that it has are source register (source for the bytes >> which are to be written), offset in the packet and length to write. >> It uses a select_ops method to choose between fast ops in the cases >> length is 1,2 or 4 bytes and slow ops (i.e. using memcpy) in other >> cases. >> >> Signed-off-by: Nikolay Aleksandrov <nikolay@xxxxxxxxxx> >> --- >> I needed a way (other than passing the packets to user-space) to alter >> the ToS field via nftables, so I decided to make it a bit more general. >> I >> use it with the immediate expression to load the new ToS and then write >> it. >> If you find this useful I can post the libnftnl patch as well. >> Right now as you can see it continues even if the "write" wasn't >> successful >> which should be probably changed to NFT_BREAK for that case. > > Yes. > >> This patch applies to Dave's net-next tree. > > I think this is a useful addition. However I prefer to put thus into the payload expression and select the proper ops based on the presence of sreg/dreg. > Okay, makes sense. I'll re-write it in such form taking into consideration the other comments and will re-post after some testing. Thank you for the feedback, Nik -- To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
