On Sun, Jan 23, 2022 at 2:47 AM Gabriele <phoenix1987@xxxxxxxxx> wrote: > > On Fri, 21 Jan 2022 at 02:09, Alexei Starovoitov > <alexei.starovoitov@xxxxxxxxx> wrote: > > > > How would bpf prog know the pid of the python interpreter? > > Then how would it know the pids of the threads? > > I'm not against exposing find_get_task_by_vpid(), but > > we need to understand the real usage first. > > If we do end up exposing find_get_task_by_vpid(), it's probably > > best to do via refcnt-ed kfunc approach (unstable helpers). > > For example: https://lore.kernel.org/all/20220114163953.1455836-7-memxor@xxxxxxxxx/ > > This is a simple but somewhat unrealistic example but hopefully it > will give the idea. Suppose we are tracing sys_kill on entry and that > we have an application that uses it to check if a process exists by > sending the 0 signal to its PID. During the handling of this event, we > might want to read a certain area of the VM (which we would have > identified a priori from user-space) of the process identified by the > PID passed to the syscall. Fair enough. Pls prepare a patch to make find_get_task_by_vpid into refcnted unstable helper for certain prog types. Probably sleepable only.
