Daniel Xu

@jordalgo I'm confused - isn't `errno` a libc thing? I'm looking at `__sys_bpf` in the kernel and I see it returning things like `-EFAULT`

The man page had me doubting reality, so I ran a test: ``` $ gcc main.c $ ./a.out raw syscall failed, ret=-1, errno=0 wrapped syscall failed, ret=-1, errno=1 $ sudo...

Yeah, so from userspace side we only get -1 as the retval. Meaning if you were to uprobe a syscall. But if you hook in syscall with kprobe/kfunc, then I...

Also similar to https://github.com/iovisor/bpftrace/issues/2384

Yes, this feature is often requested and would be nice. I recall discussing this at one point (maybe office hours?) - I think we mostly agreed that it would be...

Whoever implements it ought to do a blog post. I think it's a pretty interesting topic (transforming a loop into a callback). Could help attract compiler enthusiasts when they see...

Yeah, use cases seem totally reasonable. Exact syntax should hopefully leverage type annotation work done for user defined functions.

I think the approach makes sense. I'd add on to @viktormalik 's comment in the now-closed PR that we should definitely keep this to compile time. Concatenating literal strings at...

There's some documentation about bpf2bpf calls here: https://docs.cilium.io/en/stable/bpf/architecture/#bpf-to-bpf-calls . I think requiring bpf2bpf calls should be necessary. Inlining functions leads to way too much bloat and unpredictability. And bpf2bpf calls...

> > There's some documentation about bpf2bpf calls here: https://docs.cilium.io/en/stable/bpf/architecture/#bpf-to-bpf-calls . I think requiring bpf2bpf calls should be necessary. Inlining functions leads to way too much bloat and unpredictability. And...