> To recap, nivedita proposed a TCP_DELACK sockopt to allow users to
> completely disabled delayed acks. Alexey thought this a bad idea. We're
> not convinced it is a bad idea, and it has a lot of attraction for us.
> Alexey's comments are plain, and my comments in <<<>>>
I did not write the phrases in <<<>>>. :-)
Yes, I really insist: disabling delayed ACKs is _never_ good.
Such curcumstances just do not exist in the nature.
I do not understand what happens in your case, you did not provide
tcpdumps. Though I even can suspect that your problem is due to
too frequent acking. :-)