On 20/08/2019 15:44, Pablo Neira Ayuso wrote: > It looks to me this limitation is coming from tc pedit. > > Four actions to mangle an IPv6 address consume more memory when making > the translation, and if you expect a lot of rules. Your change means that now every pedit uses four hw entries, even if it was only meant to be a 32-bit mangle. Host memory used to keep track of the pedit actions is cheap, hw entries in pedit tables are not. Nor is driver implementation complexity. NAK.
- [PATCH net-next 1/2] net: flow_offload: mangle 128-bit p... Pablo Neira Ayuso
- [PATCH net-next 2/2] netfilter: nft_payload: packet... Pablo Neira Ayuso
- Re: [PATCH net-next 1/2] net: flow_offload: mangle ... Edward Cree
- Re: [PATCH net-next 1/2] net: flow_offload: man... Pablo Neira Ayuso
- Re: [PATCH net-next 1/2] net: flow_offload:... Edward Cree
- Re: [PATCH net-next 1/2] net: flow_offl... Pablo Neira Ayuso
- Re: [PATCH net-next 1/2] net: flow... Edward Cree
- Re: [PATCH net-next 1/2] net: ... Pablo Neira Ayuso
- Re: [PATCH net-next 1/2] n... Edward Cree
