| Message ID | 20230330001522.120774-1-andrew@daynix.com |
|---|---|
| Headers | show |
| Series | eBPF RSS through QMP support. | expand |
On Thu, Mar 30, 2023 at 8:33 AM Andrew Melnychenko <andrew@daynix.com> wrote: > > This series of patches provides the ability to retrieve eBPF program > through qmp, so management application may load bpf blob with proper capabilities. > Now, virtio-net devices can accept eBPF programs and maps through properties > as external file descriptors. Access to the eBPF map is direct through mmap() > call, so it should not require additional capabilities to bpf* calls. > eBPF file descriptors can be passed to QEMU from parent process or by unix > socket with sendfd() qmp command. > > Overall, the basic scenario of using the helper looks like this: > * Libvirt checks for ebpf_fds property. > * Libvirt requests eBPF blob through QMP. > * Libvirt loads blob for virtio-net. > * Libvirt launches the QEMU with eBPF fds passed. Is there a libvirt side draft that can be used as a reference? Thanks > > Andrew Melnychenko (4): > ebpf: Added eBPF initialization by fds and map update. > virtio-net: Added property to load eBPF RSS with fds. > ebpf: Added declaration/initialization routines. > qmp: Added new command to retrieve eBPF blob. > > ebpf/ebpf.c | 48 +++++++++++++ > ebpf/ebpf.h | 25 +++++++ > ebpf/ebpf_rss-stub.c | 6 ++ > ebpf/ebpf_rss.c | 124 +++++++++++++++++++++++++++------ > ebpf/ebpf_rss.h | 10 +++ > ebpf/meson.build | 1 + > hw/net/virtio-net.c | 77 ++++++++++++++++++-- > include/hw/virtio/virtio-net.h | 1 + > monitor/qmp-cmds.c | 17 +++++ > qapi/misc.json | 25 +++++++ > 10 files changed, 307 insertions(+), 27 deletions(-) > create mode 100644 ebpf/ebpf.c > create mode 100644 ebpf/ebpf.h > > -- > 2.39.1 >
On Thu, Mar 30, 2023 at 9:57 AM Jason Wang <jasowang@redhat.com> wrote: > > On Thu, Mar 30, 2023 at 8:33 AM Andrew Melnychenko <andrew@daynix.com> wrote: > > > > This series of patches provides the ability to retrieve eBPF program > > through qmp, so management application may load bpf blob with proper capabilities. > > Now, virtio-net devices can accept eBPF programs and maps through properties > > as external file descriptors. Access to the eBPF map is direct through mmap() > > call, so it should not require additional capabilities to bpf* calls. > > eBPF file descriptors can be passed to QEMU from parent process or by unix > > socket with sendfd() qmp command. > > > > Overall, the basic scenario of using the helper looks like this: > > * Libvirt checks for ebpf_fds property. > > * Libvirt requests eBPF blob through QMP. > > * Libvirt loads blob for virtio-net. > > * Libvirt launches the QEMU with eBPF fds passed. > > Is there a libvirt side draft that can be used as a reference? > > Thanks > I'm working on it. This is why it's RFC. I have a small script that checks that eBPF retrieved through QMP can be loaded. Also, please check the eBPF blob declaration/initialization routine and qmp commands. So, this API should be implemented once and right. > > > > Andrew Melnychenko (4): > > ebpf: Added eBPF initialization by fds and map update. > > virtio-net: Added property to load eBPF RSS with fds. > > ebpf: Added declaration/initialization routines. > > qmp: Added new command to retrieve eBPF blob. > > > > ebpf/ebpf.c | 48 +++++++++++++ > > ebpf/ebpf.h | 25 +++++++ > > ebpf/ebpf_rss-stub.c | 6 ++ > > ebpf/ebpf_rss.c | 124 +++++++++++++++++++++++++++------ > > ebpf/ebpf_rss.h | 10 +++ > > ebpf/meson.build | 1 + > > hw/net/virtio-net.c | 77 ++++++++++++++++++-- > > include/hw/virtio/virtio-net.h | 1 + > > monitor/qmp-cmds.c | 17 +++++ > > qapi/misc.json | 25 +++++++ > > 10 files changed, 307 insertions(+), 27 deletions(-) > > create mode 100644 ebpf/ebpf.c > > create mode 100644 ebpf/ebpf.h > > > > -- > > 2.39.1 > > >