Here's the basic idea: r1 = -8 r2 = -1 JGT r1, r2, end JSGT r1, 1, end ptr += r1 *(u8 *)ptr = 0 After the JGT, we're in the false branch so r1->min_value = 0 and r1->max_value

Hi, Tetsuo, You are right. The bug is actually introduced by my last patch. I just focused one aspect of issue and inadvertently introduced another *blocker* bug. I just pushed a patch. Thanks a lot

Hello. I changed the hook as below and confirmed that cp == NULL at bpf_probe_read(). That is, it is bpf_usdt_readarg() which got broken. ---------- int do_start(struct pt_regs *ctx) { char *cp

Well the min_value is a s64, but yeah anything negative is supposed to be rejected, so it essentially acts as the range of unsigned absolute values it can hold. I tried to hand craft a way to exploit

Tetsuo, I cannot reproduce the issue. The transformation from: dest = ctx->bx; to: dest = *(volatile uint64_t *)&ctx->bx; should be safe. I launched a psql console and did a query: ... yhs=>

Tetsuo, I cannot reproduce the issue. The transformation from: dest = ctx->bx; to: dest = *(volatile uint64_t *)&ctx->bx; should be safe. I launched a psql console and did a query: ... yhs=>

I think the way Josef intended it to behave is min/max_value are absolute values that 64-bits can hold. In that sense unsigned (JGT) comparison and the false branch are implying that min_value =

I think that I solved it. I changed the code of xdp_ddos01_blacklist_kern.c: case ETH_P_ARP: return XDP_PASS instead of break. Works! :-)

Hi Jesper, First of all, thanks for updating the prototype-kernel packet. It works! I tried to switch to white list and it worked. Though, as a result, there is no way to ping the machine and get a

Hello. I noticed that b0f891d129a9b372 ("Force udst ctx->#reg load to be volatile") broke an USDT probe example shown below. ---------- #!/usr/bin/python import sys import ctypes as ct def

You are right. In this case, r0 = bpf_map_lookup r2 = r0 r3 = r0 r4 = r0 r5 = r0 if (r0 != 0) <=== condition 1 r1 = 1 if (r0 != 0) r1 = 2 if (r3 != 0) *r3 = 0 ... If (r0 != 0) if false, the

A couple of the tests in tools/testing/selftests/bpf/test_verifier.c seem to be bogus: Test "multiple registers share map_lookup_elem bad reg type" is supposed to error with "R3 invalid mem access

Jesper Dangaard Brouer <brouer@...> wrote: Done, updated the documentation in the prototype-kernel github repo. See three top commits in this link:

The prototype-kernel github repo[1] is actually originally meant for developing real kernel module. I guess, I should/could have placed the eBPF/XDP examples[3] in another github repo. [1]

<iovisor-dev@...> wrote: Just looked at the code. This seems a control plane application. You should not compile with "-target bpf". You should just use the default host one if you are

Hi Jesper, Thank you so much for your help. I need to run an eBPF programs that filters packets according to IP address or port. I'm trying to run the prototype-kernel package but encounter

Adel Fuchs <adelfuchs@...> wrote: I assume you are doing: sudo cat /sys/kernel/debug/tracing/trace_pipe The problem could be that the kernel need to be compiled with the right trace config

Hi Jesper, I tried adding your solution, bpf_debug, and I'm now able to run the program with no errors but the trace_pipe file stays empty. I just added this to my program: #ifdef DEBUG /* Only use

Hi, I'm trying to run this program: https://github.com/idosch/iproute2/blob/master/examples/bpf/bpf_prog.c $tc filter add dev enx00e11100329b parent 1: bpf obj bpf.o exp / /tmp/bpf-uds flowid 1:1

Hi All, Thanks for joining the meeting today. Here are the notes! === Conferences === Reminder: CFP for Linux Plumbers Please submit a topic for the Tracing