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

Hello Brenden! I’m not sure if it counts as an agenda item, but I’m interested in recording process events using tracepoints, and I would like to know what are the best practices when attempting

Hi Ferenc, I think sock_ops is a good option for me, it seems the cilium project also use this to accelerate networks. Yes, all TCP connection established. There're no connection pressures and I send

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

<cneirabustos@...> wrote: looks good. A couple of comments below. reverse Christmas order. Looks like -EINVAL is returned for all error cases. This is suboptimal. There are cases where other

Hi Yonghong, I had time yesterday to finish the code to avoid using getname_kernel , self-tests run ok with expected results. Here is the current code to avoid the getname_kernel call. diff

Carlos, Any update on this? It would be good to submit the patch earlier for a release cycle so we got a chance it may be accepted in the current release cycle. <cneirabustos@...> wrote:

Hi! Well it seems like a different problem from what I first thought. I have a few idea, maybe some of them helps: 1. If you able to modify the source code of user space receiver program, maybe try to

Hi Kanthi, In the case of OVS AF_XDP netdev, we've used multiple PMD threads. Each thread has its own AF_XDP socket and its own UMEM bind to each rx queue. I think multiple processes should also

Hi Kanthi, Yes, it's possible! Björn

On Thu, Aug 1, 2019 at 07:16 PM, Yonghong Song wrote: I guess you mean value size. here. When you got the values from kernel, the value size is rounded to 8.

For array map, the key type must be 32bit int. ``` int array_map_alloc_check(union bpf_attr *attr) { bool percpu = attr->map_type == BPF_MAP_TYPE_PERCPU_ARRAY; int numa_node =

map defined: struct bpf_map_def SEC("maps/protocount") proto_count = { .type = BPF_MAP_TYPE_PERCPU_ARRAY, .key_size = sizeof(__u32), .value_size = sizeof(__u32), .max_entries =

Could you post how to reproduce the issue so people can help? If you do provide reproducible steps, could you submit an issue to help tracking? Thanks,

Hi all, I define a BPF_MAP_TYPE_PERCPU_ARRAY and use it to count packets in the xdp program. When I read the map from userspace program, I find that the entry number doesn't match local CPU numbers. I

Hi, Is it possible to have 2 processes each with its own AF_XDP socket and its own UMEM? Thanks, Kanthi

Hi Ferenc: The packets are really small, I just send a curl get request. Sometimes I can receive the packets from my program, I think it is because the packets arrive before I set the socket desc so

Thanks Yonghong!. Yes, I was looking at filename_lookup today on how to do it, I'll work on that. Thank you very much.

<alexei.starovoitov@...> wrote: Looks like we can call filename_lookup() directly. We do not need to call getname_kernel(). We know the exact path name. We can do a atomic allocation inside the