Could you elaborate more on this ? Unfortunately, I cannot find the comments you gave me yesterday. (i think I chose a private reply)

Cilium has an idea to accelerate packet forward performance using sockops & sockhash when using istio service mesh, the code is here. But this function is heavily coupling with cilium codebase so I

What is your llvm version? What is your cmake output? Looks like llvm libraries are not really linked?

Before I started to make a small change to bcc, I thought that I ought to verify that 'master' would actually build -- I'm running on Amazon Linux 2 AMI (HVM), SSD Volume Type - ami-0b69ea66ff7391e80.

No agenda means tomorrow conf call is cancelled. And that would be few months in a row. I think we should stop doing this email roll call as well. The conf call was useful for long time, but not any

Hi, All, Brenden is on vacation this week. So I am sending out the notice. Does anybody have any agenda to discuss on our pre-schedule 30min meeting on Sept 4, 11:00am - 11:30am? If you do, please

<arnaldo.melo@...> wrote: You did not miss anything. Currently, there are no counters to count those drops due to nmi or due to bpf program already running on that cpu. There is effort by

Thanks Matt and Yonghong. Info on the preempt_disable()/preempt_enable() region is exactly what I was looking for.

Thanks for the pointer to the code! --matt

Yes. See https://github.com/torvalds/linux/blob/master/kernel/trace/bpf_trace.c#L88-L97

Great, thanks Yonghong! And I get that this is part of the infrastructure that calls into the bpf code (e.g. perf_events), not bcc. > Also we have per cpu counter to prevent when bpf program

bpf program run is wrapped in preempt_disable()/preempt_enable() region. Also we have per cpu counter to prevent when bpf program interrupted then another bpf to run for tracing programs.

When running bpf code from a kprobe / kretprobe, does anything ensure that cpu_id doesn't change while the bpf is running (e.g. due to preemption)? Does anything ensure that no other bpf code runs on

Hi All, We have the bi-weekly phone conference scheduled for two days from now, does anybody have a discussion topic to add to the agenda? As a reminder, we are planning to hold the meeting only if

preemption is a kernel thing, bcc does not disable it. You need to check kernel configuration CONFIG_PREEMPT in your host.

Hi all, I'm trying to verify that there are no concurrency issues with an approach I'm using cpu_id as a key to a HASH_MAP. My understanding is that bcc disables preemption but the details are

Hi All, Since we haven't received any agenda suggestions for the meeting today, please consider it canceled. Thanks, Brenden

Hi all, I attach an XDP prog(SKB mode) to host net device(enp0s9) and the prog will redirect packet to veth device. If there's only veth device, I can redirect packet into netns through veth device in

Hello Song! I'm using sys_enter_execve/sys_exit_execve and sys_enter_execveat/sys_exit_execveat. Bound loops seem really useful! This is going to solve many of the issues I had. I don't have code I

Which tracepoint do you have in mind for your particular use case? The recent kernel (5.3) added bounded loop support up to 1M instructions. You can have a bounded loop like start = ... for (i