Date: Sat, 14 Jul 2018 20:52:47 -0400 From: Rich Felker <dalias@...c.org> To: musl@...ts.openwall.com Subject: Re: getaddrinfo(3) / AI_ADDRCONFIG On Sat, Jul 14, 2018 at 08:19:50PM -0400, Rich Felker wrote: > On Sat, Jul 14, 2018 at 08:07:17PM -0400, Rich Felker wrote: > > On Sat, Jul 14, 2018 at 07:53:53PM -0400, Christopher Friedt wrote: > > > On Fri, Jul 13, 2018 at 10:31 PM Rich Felker <dalias@...c.org> wrote: > > > > Here's a version with the loop. I've tested it now with ::1 removed > > > > from device lo, but the connect to ::1 still succeeds; I suspect > > > > presence of a default route for IPv6 makes it work since ::1 is > > > > "routable" then. Can you confirm that it actually suppresses IPv6 in > > > > your purely-no-IPv6 environment, as intended? > > > > > > Works for me. > > > > Thanks! > > Minor issue: as written the patch left the cancel state disabled if it > exited early via return EAI_NONAME. Un-hoisting the setcancelstate > calls to fix. And a couple other issues: socket() may fail with EAFNOSUPPORT, and in that case, the family needs to be rejected rather than accepted like it was with the continue. For other socket() failures (like EMFILE/ENFILE), the result is indeterminate and we need to return EAI_SYSTEM rather than wrong results. Making these changes and I think it will be ready to commit. Rich
Powered by blists - more mailing lists
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.