I was just thinking about this and searched in Internet and I found that this is already implemented and part of OVS. Please see the series of patches here:
http://openvswitch.org/pipermail/dev/2015-February/051002.html
Also, if you want to look at the patched OVS source code, check this Linux distribution :
https://github.com/azhou-nicira/net-next
You can check the net/openvswitch directory to understand BPF changes.
Essentially, above patches are addressing one of the concerns of OF critics. OF specifications today define actions. Any new action requirement has to go through the standardization process in ONF, which can take few months, if not years. Many companies, like NICIRA, Freescale defined their own proprietary extensions. But the acceptability is low as these are not part of the OF specifications. Now with BPF integration in the Kernel OVS data path, one can define proprietary actions as BPF set of BPF programs. As long as this generic BPF program action is standardized in OF specifications, then new actions can be implemented as BPF programs without having to go through standardization process.
As of now, Linux provides hooking the BPF programs to three areas - system call level, CLS scheduling level and at the "dev" layer. Once above patches are accepted in Linux Kernel, OVS datapath becomes fourth area where BPF programs can be associated.
I am hoping that, BPF programs can be executed not only in the Kernel, but also in DPDK accelerated OVS as well as smart-NIC (FPGA, Network processors, AIOP etc..).
Thanks
Srini
http://openvswitch.org/pipermail/dev/2015-February/051002.html
Also, if you want to look at the patched OVS source code, check this Linux distribution :
https://github.com/azhou-nicira/net-next
You can check the net/openvswitch directory to understand BPF changes.
Essentially, above patches are addressing one of the concerns of OF critics. OF specifications today define actions. Any new action requirement has to go through the standardization process in ONF, which can take few months, if not years. Many companies, like NICIRA, Freescale defined their own proprietary extensions. But the acceptability is low as these are not part of the OF specifications. Now with BPF integration in the Kernel OVS data path, one can define proprietary actions as BPF set of BPF programs. As long as this generic BPF program action is standardized in OF specifications, then new actions can be implemented as BPF programs without having to go through standardization process.
As of now, Linux provides hooking the BPF programs to three areas - system call level, CLS scheduling level and at the "dev" layer. Once above patches are accepted in Linux Kernel, OVS datapath becomes fourth area where BPF programs can be associated.
I am hoping that, BPF programs can be executed not only in the Kernel, but also in DPDK accelerated OVS as well as smart-NIC (FPGA, Network processors, AIOP etc..).
Thanks
Srini
No comments:
Post a Comment