Re:Five years of IPv6: whither the next five?
by cm2187 on Tuesday June 06, 2017 @05:03PM
(#14501305)
What would a 64bit address space solve that a 128bit would break?
[ Parent
| on HN ]
Re:Five years of IPv6: whither the next five?
by scbrg on Tuesday June 06, 2017 @05:06PM
(#14501326)
The possibility of remembering addresses.
Re:Five years of IPv6: whither the next five?
by tomschlick on Tuesday June 06, 2017 @06:01PM
(#14501813)
DNS. You're not supposed to remember ip addresses.
Re:Five years of IPv6: whither the next five?
by nikanj on Wednesday June 07, 2017 @02:04AM
(#14504054)
We are the sorry sods working on your VPNs, LANs and the like. There's no DNS for routable blocks yet.
Re:Five years of IPv6: whither the next five?
by zurn on Thursday June 08, 2017 @02:51PM
(#14516227)
Configuration systems for those typically do support symbolic prefixes for easy renumbering and deployment of sibling networks.
Also, most people dealing with rfc1918 networks would much rather get rid of addressing conflicts and other problems resulting from non-unique addresses, like security problems arising from ambiguous addressing. (Or don't know what they're missing...)
Re:Five years of IPv6: whither the next five?
by paulddraper on Tuesday June 06, 2017 @09:28PM
(#14503038)
I remember 8.8.8.8
Re:Five years of IPv6: whither the next five?
by therein on Wednesday June 07, 2017 @12:18AM
(#14503729)
I remember 208.67.222.222 and 208.67.220.220 being someone who grew up in a country with DNS-level censorship way before Google announced their Public DNS.
Re:Five years of IPv6: whither the next five?
by tomschlick on Tuesday June 06, 2017 @11:07PM
(#14503434)
Longer, but still memorable.
2001:4860:4860::8888
Re:Five years of IPv6: whither the next five?
by nazgul17 on Tuesday June 06, 2017 @07:17PM
(#14502322)
If you're talking about local addresses, can't you just use the short notation and use even-shorter-than-64-bits addresses?
If not, DNS is there for that.
Re:Five years of IPv6: whither the next five?
by dingo_bat on Tuesday June 06, 2017 @10:06PM
(#14503172)
Passing around an IP address in a single register. Today's CPUs are powerful enough to not need hardware forwarding chips to route up to 1Gbps of traffic easily, without much custom software optimisation. However that only holds true for IPv4 traffic. Throw in IPv6 traffic and suddenly each compare takes twice as much time in the CPU. Functions cannot pass addresses and masks in the registers, requiring stack juggling or worse, memory access.
Re:Five years of IPv6: whither the next five?
by tveita on Wednesday June 07, 2017 @06:08AM
(#14504948)
> Throw in IPv6 traffic and suddenly each compare takes twice as much time in the CPU
So an operation responsible for less than 0.1% of the total time spent routing a packet now takes twice as much time... or it would if CPUs weren't superscalar for a long time now. That's paid back in tenfold just by the omission of the header checksum.
Do you have any benchmarks showing a real-world difference? https://www.unix-experience.fr/2013/ipv4ipv6-performances-co... shows performance to be identical on modern computers.
Re:Five years of IPv6: whither the next five?
by legulere on Wednesday June 07, 2017 @03:12AM
(#14504287)
Compile a function that takes a 128 bit struct as an argument and look at the assembly. It is simply passed in 2 registers.
Re:Five years of IPv6: whither the next five?
by DaiPlusPlus on Wednesday June 07, 2017 @04:24AM
(#14504550)
Depends on the ISA and optimization level - and the OS. Win32's x86-only __stdcall convention always passes arguments on the stack instead of in registers.