Jordan Rome

Well this is definitely a strange one. I'm not able to repro this behavior on ubuntu, centos, fedora, or debian (6.1.0-13-arm64, which is after loading the nf_nat kernel module).

Ok, I was finally able to repro. Also asked around and it might be that this patch needs to get backported to fix this issue: https://patchwork.kernel.org/project/netdevbpf/patch/[email protected]/

It was just merged into 6.1 stable: https://lore.kernel.org/stable/[email protected]/T/#u > One more strange thing was if you see latest kernel Linux 6.2.0-1019-gcp even nat symbols are not present in bpftrace. Did...

I use `modprobe` to load kernel modules: https://opensource.com/article/18/5/how-load-or-unload-linux-kernel-module I believe it has to do with kconfig settings that determines which ones get loaded by default. > Can i use tracepoints...

We can use a `const volatile` variable for `nr_cpus` and update it from userspace before program load then use `bpf_for(i, 0, nr_cpus) {` to iterate. Note: Open-coded iterators didn't make...

@janca-ucdavis I also wish there was some internal BPF utility to do something like this but I think the issue for bpftrace (as you alluded to) is the asynchronous part...

@janet-campbell I think I understand what you mean with the "bank" variable but I'd still be curious to see some example scripts if you're willing to share.

+1 to what @danobi said. These are super interesting use cases for bpftrace that we want to become more familiar with and help support.

Closing this as 'sum', 'count', 'min', 'max', and 'avg' all now can be implicitly casted.

> Maybe something similar to what we do with HelperVerifierError might be a good solution - parse the verifier log and if it matches the pattern we expect, provide a...