Hi all,

Thanks for attending the meeting today. Below you will find my notes from
the
call. Hear you all in two weeks!

================================================

Brenden:
Overview of conferences over previous month
2 OpenStack presentations from PG and Huawei
 https://www.youtube.com/watch?v=s-8BSTzD91s
 https://www.youtube.com/watch?v=1oAsRzrwAAw
Linux Plumbers - half day track on tracing
 No notes/video to share, but see netdev mailer and possibly upcoming
lwn.net
 article.
 Good progress made on LSM/cgroup hooks, as well as some discussion on
 userspace tracing
Open vSwitch Conference
 Slides available here: http://www.openvswitch.org/support/ovscon2016/

Daniel:
WIP: XDP bug in mlx4/mlx5 ref counting
Also WIP: hashing of bpf programs for visibility
Soon: bpf in kallsyms (address/name/sha)
Adding bpf() tracepoint

Alexei:
Userspace tracing/live patching for bpf?
 discussions currently in private, to share soon
LRU maps (for LB use case) under review
 multiple modes - regular/no common
cgroups work ongoing - patches soon
discussion: improving debuggability
 add pseudo instruction for "comment"

Joe:
libbpf work - pinning support
ovs con (see above link for slides)
William Tu: presented bpf dataplane
 performance issues due to:
  p4->bcc->bpf script inefficiencies
  1 map lookup per action

Alban:
Had 4 topics from IRC (3 were discussed):
1. Golang libraries for kprobe-eBPF and perf-maps
 We agreed to fork the code in github.com/iovisor/iomodules into its own
 go library.
 AI: Brenden to send proposal for name of new repo
2. dependencies on kernel headers for kprobe-eBPF
 There has been some previous discussion of this in
 github.com/iovisor/bcc/issues/421, but no substantial progress yet. Another
 alternative that Brenden proposed was to let the API accept LLVM-IR as the
 input of a BPF() object, such that frontends could "compile" against the
 headers offline.
3. API stability for a limited set of kprobe-eBPF programs
 We didn't discuss this, Alban would you like to give us some more context?
4. size of bcc.so when included in (and used from within) a container image
wanting to build libbcc against alpine go bindings
 No work has been done to compile bcc using alpine base. This would be a
nice
 to have if someone out there has some experience or free time.

Sargun:
While investigating how to implement IPVS-like functionality in BPF, came
across the 32-prog tail call limit, wondering how to work around that?
Can the limit be enforced with a sysctl?
 A: no, probably not
Instead, we proposed that a map/helper-based approach for capturing the use
case, or a new datastructure, would be a better path forward. Sargun may
investigate.

================================================

Attendees:
Brenden Blanco
Alessandro
Alban Crequy
Marco Leogrande
Matteo
Daniel Borkmann
iaguis
Joe Stringer
Alexei
Sargun
_______________________________________________
iovisor-dev mailing list
[email protected]
https://lists.iovisor.org/mailman/listinfo/iovisor-dev

Reply via email to