Alastair Robertson

Yep, BPF's maps-in-maps is too limited for what we wanted to do, so range-deletes it is.

Agreed that it would be a useful feature to have. It's just a lack of time/desire that's meant that it hasn't been implemented yet. We'd lock this feature behind bpftrace's...

We could really do with improving arg parsing for scripts and I really like the ideas presented here! How far did you get with the implementation @fbs? One use-case I...

This was done in #1492, although it only works for a very limited set of kernel functions.

cc @lenticularis39

It seems to be the "AFTER_COLON" state which is causing the problem: https://github.com/bpftrace/bpftrace/blob/26e310f6854ed837a4f2468ae114d48660a9fd12/src/lexer.l#L100

I've got a possible fix here: https://github.com/ajor/bpftrace2/commit/113ffab039620f67741fb0046d4996f4e15da889. Not very familiar with flex though. Feel free to take this on if you fancy it, otherwise I'll continue some time next week....

Reported in BCC as well https://github.com/iovisor/bcc/issues/2555

Reopening: we've got to try and fix all our wrong-results bugs. @ttreyer is working on a solution for this one in #3095

Following on from discussions with @arthurshau, it turns out Approach 2 is not possible due to an existing form of `print()` which takes multiple arguments. Updated the description with some...