Re: [RFC PATCH v2 tip 0/7] 64-bit BPF insn set and tracing filters

[Daniel Borkmann]

On 02/13/2014 11:32 PM, H. Peter Anvin wrote:
> On 02/06/2014 05:20 PM, Alexei Starovoitov wrote:
> > I believe that old BPF outlived itself and BPF64 should
> > replace it in all current use cases plus a lot more.
> > It just cannot happen at once.
> > BPF64 can come in. bpf32->bpf64 converter functioning.
> > JIT from bpf64->aarch64 and may be sparc64 needs to be in place.
> > Then old bpf can fade away.
>
> I don't think that is doable any time soon.  Right now pretty much all
> mobile devices, for example, are 32 bits and they really want to use
> syscall filtering for security.  Performance matters greatly there.

Well, if that would be the case, then seccomp would have had JIT support
long ago. ;-) Right now BPF filters with seccomp are not JIT compiled
for _any_ architecture.

> As such, 32-bit JIT support is going to be very important for a long
> time to come.

True, I think that pretty much depends if we can manage to find a way
to cleanly integrate it into net/core/filter.c while still supporting
the old instructions as I've mentioned earlier.

> 	-hpa
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/