On 04/26/2018 10:27 AM, Matthieu Baerts wrote:
>
> Approach to indirect call optimization by Eric Dumazet:
>     that's one example, it is a bit specific to MD5 but the idea is 
> there. That's what we had in mind (but the solution is not generic)
Can someone point me to it.
>
>
>
> Discussion of Oracle's patches:
>     - Rao is asking what to do next with these patches
>     - (Matthieu: I got disconnected, I maybe missed something 
> important, feel free to comment)
>     - One main concerning point raised during the call is that we will 
> still have a big intrusiveness with a lot of "if(mptcp)" in the code
What is a lot ? Our patch has 14 instances...

rshoaib(a)caduceus5:/home/mptcp/upstream/net/ipv4(upstream)$sed -n '{/if 
*(mptcp/p}' tcp_*.c
                 if (mptcp(tp))
                 if (mptcp(tp)) {
                 if (mptcp(tp))
         if (mptcp(tp))
                                 if (mptcp(tp))
         if (mptcp(tp))
                 if (mptcp(tp)) {
                         if (mptcp_init_tw_sock(sk, tcptw)) {
         if (mptcp(tcp_sk(child)))
         if (mptcp_is_data_fin(skb))
         if (mptcp_is_data_fin(skb))
                         if (mptcp(tcp_sk(sk)))
                 if (mptcp(tcp_sk(sk)))
rshoaib(a)caduceus5:/home/mptcp/upstream/net/ipv4(upstream)$cd ../ipv6
rshoaib(a)caduceus5:/home/mptcp/upstream/net/ipv6(upstream)$!grep
rshoaib(a)caduceus5:/home/mptcp/upstream/net/ipv6(upstream)$grep "if 
(mptcp" *.c
tcp_ipv6.c:     if (mptcp(tp))
tcp_ipv6.c:                     if (mptcp(tp))

And all of this code is under #ifdef MPTCP --- Does MD5 have any such code ?
Here is a better comparison of intrusiveness.

rshoaib(a)caduceus5:/home/mptcp/upstream/net/ipv4(upstream)$grep MD5 
tcp_*.c | wc -l
73
rshoaib(a)caduceus5:/home/mptcp/upstream/net/ipv4(upstream)$grep MPTCP 
tcp_*.c | wc -l
39

Once again note, any MPTCP specific code is under #ifdef MPTCP. Our 
patch allows tcp to be built with MPTCP and does not require any mptcp 
file including header files.

>     - Rao would prefer not to rewrite the current MPTCP implementation 
> as much as proposed here during the meetings. Propose this first 
> implementation based on a previous mptcp_trunk version.

The trunk version is irrelevant as we are not going to keep up with what 
mptcp-dev does. Having said that, I believe any changes to MPTCP code 
should just work.

Even though I do not think sentiments towards our patch will change. I 
wanted to point out the facts as the comments were made without truly 
evaluating the changes.

Shoaib