Messages in this thread | ![/](/images/icornerl.gif) | | From | takeru hayasaka <> | Date | Wed, 31 Jan 2024 12:15:41 +0900 | Subject | Re: [Intel-wired-lan] [PATCH net-next RESENT v3] ethtool: ice: Support for RSS settings to GTP from ethtool |
| |
Hi Harald-san
I apologize for the delay in making further progress on this patch.(I'm three months late...
> There are many different GTPv2C messages
Thank you for your assistance in explaining GTPC:)
Thanks, Takeru
2024年1月30日(火) 19:15 Harald Welte <laforge@gnumonks.org>: > > hi Marcin, > > Disclaimer: I have no understanding of the proposed implementation here, just commenting > on this from a 3GPP protocol architecture point of view. > > On Tue, Jan 30, 2024 at 10:59:40AM +0100, Marcin Szycik wrote: > > >> gtpc(4|6) doesn't include TEID, so what is its purpose? > > > In GTPC communication, there is no TEID in the CSR (Create Session Request). > > > Therefore, there are cases of GTPC that do not include TEID. > > > > The way I understand it now, this patch (and the ethtool one) adds hashing on > > TEID field in GTP* headers. So I wanted to ask why do we have a case (gtpc(4|6)) > > that doesn't include TEID? Do we hash on other fields in this header? > > There are many differen GTPv2C messages, most of which contain a TEID. So it does > in general still make sense to be able to use RSS for all those other messages. > > The CSR (Create Session Request) will not be able to benfit from it, but > it's just the first message initiating a dialogue between two elements > (think of it like a TCP SYN). All the follow-up messages in that > dialogue contain TEIDs and hence can benefit from RSS. > > -- > - Harald Welte <laforge@gnumonks.org> https://laforge.gnumonks.org/ > ============================================================================ > "Privacy in residential applications is a desirable marketing option." > (ETSI EN 300 175-7 Ch. A6)
| ![\](/images/icornerr.gif) |