Hengqi Chen
Hengqi Chen
Misc fix
1. code cleanup(comments, unused codes, indention etc.) 2. sync with latest libbpf, replace deprecated API usage and fix memory leak.
I play around with the `xdp_prog_kern.c` in `experiment01-tailgrow`, I change the `MIN_LEN` to `0` and `1`, but they both failed. Can anyone help answer this, thanks.
Signed-off-by: Hengqi Chen
Allow attach uprobe/uretprobe with symbol addr. ping @schu
The following code crashes, but when the array is wrap in a struct, it works properly. ```c #include BPF_PERF_OUTPUT(trace); inline int capture_args(struct pt_regs *ctx) { u8 buf[32] = {0}; u64*...
* Remove all `__always_inline` from BPF program, this reduces size of `trace_bpfel_x86.o` file * Replace `memcpy`(from ``) with `__builtin_memcpy` * Simplify `get_goroutine_id` implementation, we don't need a extra BPF ringbuf
```shell clang \ -I. \ -I/home/ubuntu/bpf/bpftool/include/uapi/ \ -I/home/ubuntu/bpf/bpftool/src/bootstrap/libbpf/include \ -g -O2 -Wall -target bpf -c skeleton/pid_iter.bpf.c -o pid_iter.bpf.o skeleton/pid_iter.bpf.c:47:14: error: incomplete definition of type 'struct bpf_perf_link' perf_link = container_of(link, struct...
`skb_output` and `xdp_output` are simple wrappers around the existing `output_at_index` method. With these wrappers, we can easily submit the raw data blob held by `struct __sk_buff *`/`struct xdp_md *` to...
We have a bunch of libbpf tools require features like `fentry` / `tp_btf` which only available on kernel v5.5+. Running these tools on old kernels result in EINVAL. Instead of...
```shell $ sudo ./tcpaccept PID COMM IP RADDR RPORT LADDR LPORT 1407454 nc 4 9.1.128.159 39688 127.0.0.1 1215 ```` Signed-off-by: Hengqi Chen