Daniel Xu

Too fast to see progress is fine. I'd expect that we don't draw the UI at all in that case. UI should only be drawn past some time threshold.

Just so I can understand the problem better, why are users creating `perf probe` events and then attaching to them with bpftrace? Seems like they could also either do all...

Backing up a bit, what is the use case for this language feature? I'd tend to agree this is useful in the abstract, but is there a user or use...

Chatted offline. We agreed use case makes sense. But it is not a blocker for bcc script conversion work. For context, the script conversion work is not necessarily done for...

> What do you think about shipping the tools and tools with the text files? Yeah that works too. Probably makes more sense - the tools are much more difficult...

My original proposal was that `tools.tar.zst` contains **only** the scripts. And `tools-help.tar.zst` to contain **only** the help files. Alternatively, expanding the comments seems fine to me as well.

There are quite a few folks. The linked issues have some indication. I also use it every now and then. It's really useful to be able to quickly grab a...

Playing with snake some more, a carefully placed loop: ```diff diff ~/scratch/snake.bt ~/scratch/snake.bt.optimized 58c58 < unroll(50) { --- > while($y < 50) { ``` brings insn count down from 400K...

> We should add an early pass that translates regs("reg_name") to ctx.reg_name and allow the standard field analyzer to compute the offset. In the future, when we have appropriate BTF...

I took a look at the code. It seems the code path is: ``` perf_kprobe_event_init perf_kprobe_init create_local_trace_kprobe max_active=0 __register_trace_kprobe ``` https://github.com/torvalds/linux/blob/master/kernel/trace/trace_kprobe.c#L1405 is where maxactive is hardcoded to 0. Since bpftrace...