Re: git: 9e469437993b - main - inet.4: drop "experimetal use" from net0 and net240 descriptions
- Reply: Bjoern A. Zeeb: "Re: git: 9e469437993b - main - inet.4: drop "experimetal use" from net0 and net240 descriptions"
- In reply to: Zhenlei Huang : "Re: git: 9e469437993b - main - inet.4: drop "experimetal use" from net0 and net240 descriptions"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Wed, 13 Nov 2024 20:27:42 UTC
On Tue, 12 Nov 2024 at 08:56, Zhenlei Huang <zlei@freebsd.org> wrote: > > Emm, while I was reviewing this change, I also wondered if it is too early to remove the word *experimental* as the RFC is still in draft status. Well later I think ( maybe I'm wrong ) RFC actually lag behind the reality ( I'm saying FreeBSD as a vendor and move fast action ) so I think it is OK to remove now. My argument for the man page update was that the sysctl controls use of those addresses, regardless of purpose. Note that the sysctl description captures this subtle distinction already: net.inet.ip.allow_net240: Allow use of Experimental addresses, aka Class E (240/4) That is, it's "... use of Experimental addresses ..." not "... experimental use of ..." Unfortunately it's a bit more of a mess than that. Most operating systems are allowing 240/4 by default and as it turns out we (unintentionally?) also allow configuring 240/4 the host's IPv4 address already. This sysctl actually controls only forwarding, and we'll need to update anyway. I'm inclined to indicate in the man page that these sysctls control forwarding, and add a note along the lines of "at time of writing this region is reserved."