Over the last year there were multiple attempts to let eBPF programs output data into perf events by He Kuang and Wangnan. The last one was: https://lkml.org/lkml/2015/7/20/736 It was almost perfect with exception that all bpf programs would sent data into one global perf_event. This patch set takes different approach by letting user space open independent PERF_COUNT_SW_BPF_OUTPUT events, so that program output won't collide.
Wangnan is working on corresponding perf patches. Alexei Starovoitov (3): perf: pad raw data samples automatically bpf: introduce bpf_perf_event_output() helper samples: bpf: add bpf_perf_event_output example include/uapi/linux/bpf.h | 11 +++ include/uapi/linux/perf_event.h | 1 + kernel/bpf/arraymap.c | 2 + kernel/bpf/verifier.c | 3 +- kernel/events/core.c | 15 ++- kernel/trace/bpf_trace.c | 46 +++++++++ samples/bpf/Makefile | 7 ++ samples/bpf/bpf_helpers.h | 2 + samples/bpf/trace_output_kern.c | 31 +++++++ samples/bpf/trace_output_user.c | 196 +++++++++++++++++++++++++++++++++++++++ 10 files changed, 308 insertions(+), 6 deletions(-) create mode 100644 samples/bpf/trace_output_kern.c create mode 100644 samples/bpf/trace_output_user.c -- 1.7.9.5 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
