tcp/lro: Refactor the free/active list operation.

[sepherosa_gmail.com]

Summary

Ease more work concerning active list, e.g. hash table etc.

[gallatin]

The tcp_lro_entry_get() abstraction adds an extra compare to the critical path (the compare against NULL in the function itself, in addition to the same compare in the main routine). At least it does at the C level. Have you verified that the compiler is smart enough to continue to use a single compare?

If this adds an extra compare to the actual executed code, then I object.

[rrs]

It would be worth checking the assembly output.. if it truly inlines it then
it should probably only do one compare.. but worth checking...

[sepherosa_gmail.com]

In D6137#130603, @gallatin wrote:

The tcp_lro_entry_get() abstraction adds an extra compare to the critical path (the compare against NULL in the function itself, in addition to the same compare in the main routine). At least it does at the C level. Have you verified that the compiler is smart enough to continue to use a single compare?

If this adds an extra compare to the actual executed code, then I object.

OK, I see your concern. I think instead of depending on the compiler, we probably just leave the original code for the free list unchanged :).

[sepherosa_gmail.com]

Skip the free list operation, which may introduce extra cmp on hot code path

[gallatin]

Thanks, I'm good w/this version