That's a great idea Alan. If we are OK to merge this improvement with support for 127.0.0.1 only, I'll follow up with another patch for the full range of the loopback adapter and some tests for the macro in all 3 modes.
Thanks. Nikola -----Original Message----- From: Alan Bateman <alan.bate...@oracle.com> Sent: July 23, 2020 4:59 AM To: Nikola Grcevski <nikola.grcev...@microsoft.com>; Alex Menkov <alexey.men...@oracle.com>; net-dev@openjdk.java.net Subject: Re: RFR(s): Improving performance of Windows socket connect on the loopback adapter On 23/07/2020 02:52, Nikola Grcevski wrote: > Hi Alex, > > Yes, you are right about that. I didn't think about extending the check to > more than 127.0.0.1. > > The existing macro was only testing for 127.0.0.1 in both IP4 and IP6 > modes, so if the consensus is that we want to extend to all of the > possible loopback addresses, I can modify the macro to add AND > operations to clear the bits before we compare. It would require that we make > our own macro for ::1/128, since the Windows SDK only tests for ::1 in > IN6_IS_ADDR_LOOPBACK. > Alex makes a good point because 127.0.0.0/8 is reserved for the loopback. It's okay to separate this out as a separate improvement if you want. -Alan.
