On Wed, Jan 31, 2024 at 7:56 AM Leon Hwang <hffilwlqm@xxxxxxxxx> wrote: > > This patchset introduces a new generic kfunc bpf_ffs64(). This kfunc > allows bpf to reuse kernel's __ffs64() function to improve ffs > performance in bpf. > The downside of using kfunc for this is that the compiler will assume that R1-R5 have to be spilled/filled, because that's function call convention in BPF. If this was an instruction, though, it would be much more efficient and would avoid this problem. But I see how something like ffs64 is useful. I think it would be good to also have popcnt instruction and a few other fast bit manipulation operations as well. Perhaps we should think about another BPF ISA extension to add fast bit manipulation instructions? > In patch "bpf: Add generic kfunc bpf_ffs64()", there is some data to > confirm that this kfunc is able to save around 10ns for every time on > "Intel(R) Xeon(R) Silver 4116 CPU @ 2.10GHz" CPU server, by comparing > with bpf-implemented __ffs64(). > > However, it will be better when convert this kfunc to "rep bsf" in > JIT on x86, which is able to avoid a call. But, I haven't figure out the > way. > > Leon Hwang (2): > bpf: Add generic kfunc bpf_ffs64() > selftests/bpf: Add testcases for generic kfunc bpf_ffs64() > > kernel/bpf/helpers.c | 7 +++ > .../testing/selftests/bpf/prog_tests/bitops.c | 54 +++++++++++++++++++ > tools/testing/selftests/bpf/progs/bitops.c | 21 ++++++++ > 3 files changed, 82 insertions(+) > create mode 100644 tools/testing/selftests/bpf/prog_tests/bitops.c > create mode 100644 tools/testing/selftests/bpf/progs/bitops.c > > > base-commit: c5809f0c308111adbcdbf95462a72fa79eb267d1 > -- > 2.42.1 >
