Hi All,

Thanks for attending the call today. Below are my notes.

=== Status ===

Alexei:
- 1 patch / day
- burdensome on alexei/daniel
  - how to scale?
  - to be discussed at netconf
- working on calls
  - can load relatively large files
  - rfc to come before netdev
  - memory pressure in the verifier
- First TCP tracepoints accepted
  - 5 more coming soon
  - tcp set state has some corner case where it is missed

Daniel:
- fixes for verifier
  - llvm generates some compare/jumps that aren't covered
  - off by 1 in xdp data_end checks
- lpm needs get next key helper for dumping to userspace
- plan to support more drivers for xdp metadata

Brendan:
- LISA in a couple weeks - 2 bpf talks from Sasha, 1 from Brendan

Jakub:
- How to print 2nd phase bpf verifier output from offloaded program?
  - quirk: it runs at attach time, not load time
  - Maybe try a dry run somehow?
  - Intermixing of driver specific errors?
  - Same problem for maps
  - Include offload check at load time instead
  - New system call for "BPF_PROG_PREPARE", in between load and offload
    - jit a loaded bpf prog for a specific arch/offload

=== Attendees ===
Marco Leogrande
Brenden Blanco
Alexander Duyck
Daniel Borkmann
Jakub Kicinski
Carlos Neira
Quentin Monnet
Alexei Starovoitov
Jiong
Brendan Gregg
Mauricio Vasquez
Nic Viljoen
Gordon Marler
_______________________________________________
iovisor-dev mailing list
iovisor-dev@lists.iovisor.org
https://lists.iovisor.org/mailman/listinfo/iovisor-dev

Reply via email to