netdev - Re: [PATCH bpf-next 01/13] bpf: xor of a/x in cbpf can be done

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <754bf668-2151-5a5c-c03d-b97c7da012a5@iogearbox.net>
Date:   Sun, 28 Jan 2018 21:51:37 +0100
From:   Daniel Borkmann <daniel@...earbox.net>
To:     "Naveen N. Rao" <naveen.n.rao@...ux.vnet.ibm.com>, ast@...nel.org
Cc:     netdev@...r.kernel.org, Sandipan Das <sandipan@...ux.vnet.ibm.com>
Subject: Re: [PATCH bpf-next 01/13] bpf: xor of a/x in cbpf can be done

On 01/28/2018 07:58 PM, Naveen N. Rao wrote:
> in 32 bit alu
> 
> Daniel Borkmann wrote:
>> Very minor optimization; saves 1 byte per program in x86_64
>> JIT in cBPF prologue.
> 
> ... but increases program size by 4 bytes on ppc64 :(
> In general, this is an area I've been wanting to spend some time on.  Powerpc doesn't have 32-bit sub-registers, so we need to emit an additional instruction to clear the higher 32-bits for all 32-bit operations. I need to look at the performance impact.

Right, I think one way to optimize this could be on JIT level in such
case when CPU doesn't have subregs. There is the bpf_prog_was_classic()
helper that can be used there in order to skip some of the bpf_alu32_trunc
goto cases e.g. for some of the bit ops as an example, since we know that
upper part in cBPF must be zero here anyway, this should definitely be a
low hanging fruit given we use alu32 in the cBPF to eBPF conversion in a
lot of places.

Thanks,
Daniel