On Tue Jun 30, 2026 at 7:59 PM PDT, Leon Hwang wrote: > On 1/7/26 07:12, Alexei Starovoitov wrote: >> On Fri Jun 26, 2026 at 8:43 AM PDT, Leon Hwang wrote: > [...] >> >> I don't think we need such fallback in patch [3]. > > Do you mean we can add the internal BPF_ADDR_PERCPU insn support to the > interpreter?
No. See why CONFIG_BPF_JIT_ALWAYS_ON was introduced. The interpreter is a security hole. It has to be _compiled out_ of the kernel. Hence all patches that improve fallback to interpreter are misguided. The interpreter has to be removed completely... sooner or later.
