Re: [iovisor-dev] bpf invalid stack off=-528

Thanks for the reply. > could you increase the max to 1M or so > and see what the actual number of insns it needs? it takes 112,762 insns. And the kernel runs a while (~6 seconds) to verify it. > If the program is short, as you say, can you debug > the state prunning to see why it's not

Re: [iovisor-dev] BPF unrolled loop doesn't work and its workaround

Thanks Yonghong, I'd like to know more about the BPF compiler backend, let me know if there is anything I can help to do or test. William On Tue, Apr 18, 2017 at 10:09 PM, Y Song wrote: > Hi, William, > > Your analysis is correct. In this case, the error message could be

[iovisor-dev] Is BPF map thread-safe?

Hi Experts, I'm working on a project leveraging BPF. I need a global hash table for some per-packet information. Looks like BPF_HASH meets my requirement. The last concern in my mind is that: is BPF_HASH thread-safe? If not, how to introduce lock for bpf maps? Thanks, Wenxian