On Fri, Nov 30, 2012 at 4:27 PM, Jesse Gross <je...@nicira.com> wrote: > On Fri, Nov 30, 2012 at 4:24 PM, Ben Pfaff <b...@nicira.com> wrote: >> On Fri, Nov 30, 2012 at 04:20:33PM -0800, Jesse Gross wrote: >>> On Fri, Nov 30, 2012 at 4:13 PM, Ben Pfaff <b...@nicira.com> wrote: >>> > On Fri, Nov 30, 2012 at 04:07:29PM -0800, Ansis Atteka wrote: >>> >> This patch adds logging support for skb_mark and skb_priority. >>> >> >>> >> Signed-off-by: Ansis Atteka >>> > >>> > skb_priority has two 16-bit subfields, at least with the qdiscs that >>> > I'm familiar with, so it's easier to read when it's printed in >>> > hexadecimal. This means that I'd change "%"PRIu32 to "%#"PRIx32 and >>> > MFS_DECIMAL to MFS_HEXADECIMAL for this field. >>> >>> If we do that (which sounds fine), we probably should also update >>> odp-util.c to print priority in hex. >> >> Good idea. >> >>> It might not be a bad idea to make mark hex as well. >> >> I don't know anything about the semantics of skb_mark. I'll take your >> word for it. > > I think it's pretty much user-defined unstructured data. I don't > think there is a strong reason to chose one or the other but I find > binary blobs easier to read in hex.
Ok, I will change skb_mark to hex format too. Mainly because later on we might want to partially mask out skb_mark. For example, as of now it is possible to install single IPSEC policy that does not match just a specific mark, but rather a masked out range. Also, it seems that it would make more sense to change "priority" to "skb_priority" in odp-util.c for the sake of consistency. _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
