mary marchini

It's probably best to not strip the binary. _Maybe_ on the slim image, but even then I'm not sure. Users who need an image smaller than alpine or slim can...

Documentation seems fine to remove IMO? All others can cause some impact to our users. If we're moving forward with this we'll probably also need to remove it from the...

I'm not familiar with yarn enough to know how impactful removing it would be to our users, so I'll defer to others to answer that

I'm not blocking it as it is right no btw :) This is a trade-off between base image download (happens once) vs installation of packages during final image build (happens...

The BPF validator prevents us from passing addresses of known Kernel structures to helpers. I think we'll need to put some thinking on how this operator should work. And since...

Suuper hacky, but there's a workaround for that: ```c // script.bt struct IPv6 { union { u8 v6[16]; struct { char padding[12]; u8 v4[4]; }; }; } tracepoint:tcp:tcp_set_state { $a...

> composes the perf event output struct on the BPF stack @Birch-san I opened #750 which should solve this, you might be interested :) > I wonder if the "probe_read...

That makes sense. Maybe fixing #59 will fix this as well.

USDT needs a lot of usability improvements. We should make them work similarly to [bcc/trace](https://github.com/iovisor/bcc/blob/master/tools/trace.py). Another limitation we have today is that USDT providers must have the same name as...

The test was disabled on [`616cf73` (#1066)](https://github.com/iovisor/bpftrace/pull/1066/commits/616cf73487289dbd101cb01e15b34caf914f10c3), so if anyone wants to take a stab at it they'll need to re-enable the test as it was defined before that commit.