On Mon, Apr 24, 2023 at 15:26, Vladimir Oltean <[email protected]> wrote: > On Sun, Apr 23, 2023 at 10:47:15AM +0200, Hans Schultz wrote: >> I do not understand this patch. It seems to me that it basically blocks >> any future use of dynamic fdb entries from userspace towards drivers. >> >> I would have expected that something would be done in the DSA layer, >> where (switchcore) drivers would be able to set some flags to indicate >> which features are supported by the driver, including non-static >> fdb entries. But as the placement here is earlier in the datapath from >> userspace towards drivers it's not possible to do any such thing in the >> DSA layer wrt non-static fdb entries. > > As explained too many times already in the thread here: > https://patchwork.kernel.org/project/netdevbpf/patch/[email protected]/ > the plan is:
Ahh yes thanks, I see the comment you wrote on march the 27th.
