Re: [RFC PATCH bpf-next 0/5] bpf: tailcalls in BPF subprograms

2020-07-13 Thread Maciej Fijalkowski
On Fri, Jul 10, 2020 at 05:10:08PM -0700, Alexei Starovoitov wrote: > On Thu, Jul 02, 2020 at 03:49:25PM +0200, Maciej Fijalkowski wrote: > > Hello, > > > > today bpf2bpf calls and tailcalls exclude each other. This set is a > > proposal to make them work together. It is still a RFC because we nee

Re: [RFC PATCH bpf-next 0/5] bpf: tailcalls in BPF subprograms

2020-07-10 Thread Alexei Starovoitov
On Thu, Jul 02, 2020 at 03:49:25PM +0200, Maciej Fijalkowski wrote: > Hello, > > today bpf2bpf calls and tailcalls exclude each other. This set is a > proposal to make them work together. It is still a RFC because we need > to decide if the performance impact for BPF programs with tailcalls is > a

[RFC PATCH bpf-next 0/5] bpf: tailcalls in BPF subprograms

2020-07-02 Thread Maciej Fijalkowski
Hello, today bpf2bpf calls and tailcalls exclude each other. This set is a proposal to make them work together. It is still a RFC because we need to decide if the performance impact for BPF programs with tailcalls is acceptable or not. Note that I have been focused only on x86 architecture, I am n