> Given that you're still doing this in syscall_trace_enter(), I imagine > it could live in secure_computing(). Indeed, We agree with that adding light_syscall_filter in seccomp_computing(). > I wonder if aarch64 has higher overhead for calling into the TIF_WORK > trace stuff? (Or if aarch64's BPF JIT is not as efficient as x86?) We also guess that there are many possible reasons. And we think that placing the bitmap filter the further forward the better. Our test results show that placing the bitmap filter forward can solve single filter total overhead. What is your opinion about that? > Anyway, the functionality here is similar to what I've been working > on for bitmaps (having a global preallocated bitmap isn't going to be > upstreamable, but it's good for PoC). The complications are with handling > differing architecture (for compat systems), tracking/choosing between > the various basic SECCOMP_RET_* behaviors, etc. Firstly, thank you for correction in code, yes, it is just a PoC for performance test. Indeed, your bitmap idea is basicly same with us. And, we are trying to find a solution to improve the seccomp performance for product use. So What is your plan to have it done? Could we do something to help you proceed it?
