Mario Baldi
Mario Baldi
I wonder if it would make sense to make the behavior for packets experiencing parsing errors configurable. Even though it makes sense that the P4 program running in the Control...
> @mariobaldi I spent a little time editing your 2 example programs so that `p4test` compiles them with 0 error messages using the latest version of p4c code, and the...
> @mariobaldi In your program pna-example-tcp-connection-state-tracking.p4, the ct_tcp_table_miss action assigns an initial value for either the parameter `n2h_seqNo`, or the parameter `h2n_seqNo`, but never for both of those, and never...
> @mariobaldi It would be good to put these in a separate directory for now, until the p4test front end compiles them without errors, to avoid Github CI issues. I...
This looks good to me. I am in favor of approving this.
Maybe I do not understand well what @jnfoster and @mbudiu-vmw have proposed, but I have the impression that the parallel with inout has brought us away from the initial intent....
The initial value is actually supplied through add_entry extern when an entry is created by the data plane, and through the API (TDI/P4Runtime) when an entry is created by the...
The AMD Pensando compiler team expects to start working on the feature at the beginning of 2024.