On Thu, Dec 19, 2019 at 7:41 AM Daniel Borkmann <daniel@xxxxxxxxxxxxx> wrote: > > On Wed, Dec 18, 2019 at 11:06:57PM -0800, Andrii Nakryiko wrote: > > Convert one of BCC tools (runqslower [0]) to BPF CO-RE + libbpf. It matches > > its BCC-based counterpart 1-to-1, supporting all the same parameters and > > functionality. > > > > runqslower tool utilizes BPF skeleton, auto-generated from BPF object file, > > as well as memory-mapped interface to global (read-only, in this case) data. > > Its makefile also ensures auto-generation of "relocatable" vmlinux.h, which is > > necessary for BTF-typed raw tracepoints with direct memory access. > > > > [0] https://github.com/iovisor/bcc/blob/11bf5d02c895df9646c117c713082eb192825293/tools/runqslower.py > > > > Signed-off-by: Andrii Nakryiko <andriin@xxxxxx> > > --- > > tools/lib/bpf/tools/runqslower/.gitignore | 2 + > > tools/lib/bpf/tools/runqslower/Makefile | 60 ++++++ > > .../lib/bpf/tools/runqslower/runqslower.bpf.c | 101 ++++++++++ > > tools/lib/bpf/tools/runqslower/runqslower.c | 187 ++++++++++++++++++ > > tools/lib/bpf/tools/runqslower/runqslower.h | 13 ++ > > tools/lib/bpf/tools/ is rather weird, please add to tools/bpf/ which is the > more appropriate place we have for small tools. Could also live directly in > there, e.g. tools/bpf/runqslower.{c,h,bpf.c} and then built/run from selftests, > but under libbpf directly is too odd. runqslower is as much as a showcase of how to build a stand-alone tool with libbpf and CO-RE, as a separate tool, which is why I put it under libbpf directory. It's also not really BPF-specific tool, wouldn't that make it weird to put it under tools/bpf along the bpftool? If we added few more such tools using BPF CO-RE + libbpf, would you feel it's still a good idea to put them under tools/bpf? I tried to follow BCC structure, which has tools/ subdirectory with all kinds of examples/tools built with BCC (but not necessarily used by BCC itself). This is the same idea here. > > Thanks, > Daniel
