Comment #16 on issue 28943 by [email protected]: Don't wait for AAAA queries when machine has no IPv6 connectivity http://code.google.com/p/chromium/issues/detail?id=28943
So AI_NUMERICHOST is broken in even as late as 10.4? Wow. I thought they had fixed it by then (they certainly fixed AI_ADDRCONFIG in 10.4). If that's the case, then it's a tradeoff between buggy implementations and suboptimal implementations. How many implementations of getaddrinfo hit the DNS even if AI_NUMERICHOST is specified, how many hit the DNS first without trying to convert an IP address to a string, and how many users does each have? Rolling our own function to decide whether a string is an IP address instead of calling getnameinfo would provide more consistant performance, but it would add cruft that would need to be updated (e.g. if we want to support MS ip6-literal.net names like fe80-0-0-0-a0b6-99c9-2f6c-5759.ipv6-literal.net). Is it worth it? I certainly wouldn't bother adding hacks to Chrome code just because of this ubuntu issue - that's a library bug that needs to be fixed, and working around it in Chrome will only provide less incentive to fix it. getaddrinfo seems not to hit the DNS if I pass it an IP address, at least on ubuntu hardy and karmic. Not sure what other implementations such as windows, mac, or android do. -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings -- Automated mail from issue updates at http://crbug.com/ Subscription options: http://groups.google.com/group/chromium-bugs
