as pointed out earlier. above is a bug. please add negative test cases to see it. See below... there are few bugs here. 1. it adds 4 byte, so it should have been rejected as misaligned. 2. if you add

Hi, Jiong, Thanks for the patch. Comments inlined. Right. We may need to separate out signed comparison vs. others. Will do that later. Sorry for the confusion caused by "ll". I tested your patch

Hi, I have implemented helper bpf_get_current_ns_info(void* buf, int size) as was proposed. Let me know if something else is needed or if any other change in the code is required, I'm currently

Hi Y Song, Thanks, patch updated, please review. Changes since the initial version: * Addressed all comments on instruction unit tests. - Renamed test file to "insn-unit.s". - Removed

Thank you very much for your comments, after reading them, I realized that my change just added too much unneeded code, when we could just return the values of interest in one call. I'll delete the

Hi, Carlos, Thanks for the prototyping. See comments below. iovisor-dev <iovisor-dev@...> wrote: We already have helper to get current task structure. From there, bpf_probe_read should

Helpers that expect ARG_PTR_TO_MAP_KEY and ARG_PTR_TO_MAP_VALUE can only access stack and packet memory. Allow these helpers to directly access map values by passing registers of type

From what I've tested, the other combinations (imm, reg, var) work with this patch too. I'll send a v2 with a couple additional test cases for these.

Hi All, I was working on this bcc issue https://github.com/iovisor/bcc/issues/1329 (PID filtering issues when running bpf script inside container). The current issue is that

thanks for the patch. Yes. It makes sense to teach verifier to recognize this. since we're at it I think we should allow and test the other combinations: u64 *count = bpf_map_lookup_elem(&counts,

I just push a patch (similar to your suggestion below but without assertion) which still has "ll" suffix for the constant, but no "ll" suffix for symbols. The reason is that we use "ll" in the asm

Thank you very much.

In kernel, you need to add your function proto to kprobe_prog_func_proto in kernel/trace/bpf_trace.c

Helpers that expect ARG_PTR_TO_MAP_KEY and ARG_PTR_TO_MAP_VALUE can only access stack and packet memory. Allow these helpers to directly access map values by passing registers of type

Hi Y Song, Thanks for the review and test. [snip] Personally, I prefer drop "ll". The "ll" suffix was there to tell people it is 64bit constant while the register "r1" is with 64-bit width so I

Hi, I'm trying to add new helpers to obtain a pid namespace, I'm working on kernel 4.13 --- linux/linux-4.13/kernel/bpf/helpers.c 2017-09-03 13:56:17.000000000 -0700 +++

Hi, Jiong, The patch looks great! I tested with a few bpf programs from kernel:samples/bpf/ directory. Your test case with verifier like input looks good. I will have a few suggestions later. I

<alexei.starovoitov@...> wrote: Thank you Alexei for your thoughtful reply and sorry for the delay, but you gave me a lot to think about :) If I'm not mistaken, it won't

Hi, As discussed at threads: https://www.spinics.net/lists/xdp-newbies/msg00320.html https://www.spinics.net/lists/xdp-newbies/msg00323.html This patch adds the initial BPF AsmParser

Hi All, Thanks for attending the call today. Here are the notes from the meeting. Netdev 1.2 reminder: paper submissions deadline is Sept 20 Linux Plumbers microconference (next week) If you have