On Fri, Feb 7, 2025 at 11:29 AM David Schinazi <[email protected]> wrote:
> To me the tradeoff is very easy: let's disable the link-layer reordering. > Anyone who cares enough to notice a performance regression will be able to > upgrade their TCP. And of course, this isn't a concern at all for QUIC > since there's no pre-RACK legacy device problem. > Totally agree with the sentiment here. We have real world data that QUIC's out-of-order delivery mitigates the effects of head of line blocking and leads to read world performance gains. With RACK the situation is similar. Let's not hobble the performance of modern protocols in order to *potentially* provide minimal improvements to the performance of obsolete implementations.
