Viktor Malík

I really like the proposal @BurntBrunch. Setting up a minimal version and vendoring libbpf with no patches is a nice compromise that will make distro packager's life easier and give...

> * IIUC distros are encouraged to package libbpf from the GH mirror: https://github.com/libbpf/libbpf I didn't know that this is the recommended way, thanks for pointing out. > Then it...

> Right, a lot of the refactoring here will involve moving things to use libbpf's equivalent functionality instead. That would be great. AFAIK, there are currently 3 major parts where...

LGTM, but there seems to be a problem with positional parameters which causes incorrect behaviour of the script: ``` $ bpftrace tools/dropwatch.bt Attaching 5 probes... Tracing kernel hw packets drops....

Duplicate of #2372.

Note that if we error out on invalid probes, this may break existing tools. E.g. currently: ``` # bpftrace tools/biostacks.bt tools/biostacks.bt:25-27: WARNING: blk_start_request is not traceable (probably it is inlined...

> So if we call multi-wildcard-less attachpoints a "wildcard", then are we still good for what I proposed in [#1955 (comment)](https://github.com/iovisor/bpftrace/issues/1955#issuecomment-894378731) ? So, if I understand correctly, the current behaviour...

The likely problem here is that the process which makes the probe to trigger (one which links the shared lib) no longer exists at the time you press `^C`. We're...

Hm, not sure what the problem here could be. You could try to print the function name as soon as it is catched: ``` sudo bpftrace -e 'uprobe:/export/content/lid/apps/l1proxy-ats/i001/lib64/libats-lib-lixclient.so:_ZN15AtsLibLixClient18LixClientRefresher12getTreatmentERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKN14AtsPluginUtils13ClientRequestERlS8_RS6_RNS_3Lix6TargetERiSI_SI_ { printf("%s\n",...

Hi @shadyabhi, is the problem still occurring or did you manage to solve it somehow? @xutao323 thanks for the thorough analysis, this is very helpful! It seems to me that...