| 1 Jun 2025 |
 magic_rb | I like being right :> | 11:52:02 |
 emily | I feel like "BASE-T" is reasonable. I'm too much of a pedant to say RJ45 but I need something short or I'll go insane. | 11:52:18 |
| emily | though "NBASE-T" refers only to 2.5GBASE-T and 5GBASE-T which, aaagh | 11:52:40 |
| emily | also yeah I'm really hoping for the latter here since it seems like it'll give the kernel more visibility/control | 11:53:01 |
| emily | like I won't have to worry about the SFP+ fucking up jumbo frames | 11:53:13 |
| magic_rb | Lol, btw, if 2.5g isnt real, then how do all those 2.5 motherboards do it? Like the copper version | 11:53:29 |
| emily | do you know what I'd look for in chip datasheets for that? or is it a decision by the transceiver manufacturer rather than the chip? | 11:53:31 |
| magic_rb | Or are just 2.5 sfps not real | 11:53:34 |
| emily | 2.5GBASE-T is real, it's 2500BASE-X that's weird, AIUI. | 11:53:40 |
| emily | I assume the mobos will have chips that turn it into whatever messed up vendor variant of 2500BASE-X the chips support. | 11:54:19 |
| magic_rb | Whats the difference? | 11:54:20 |
| emily | okay that's a good question. my read has been "-X is the raw signals passed through by transceivers etc." but I guess it's actually an umbrella term and 2.5GBASE-T should morally be part of 2500BASE-X? | 11:55:12 |
| emily | I don't know why 2.5GBASE-T would be more interoperable than 2500BASE-X from SFP+s. I don't know anything. K900 is the one who knows :P | 11:55:37 |
 K900 | Hmm actually looking closer at the kernel | 11:57:02 |
| magic_rb | I wonder if one us should do a nixcon talk on the bpi-r4 and nixos routers | 11:57:17 |
| emily | fwiw if anyone knows for sure of an SFP+ that can do all of these and that has power consumption equal to or lower than the BCM84891L chip I will buy it instantly | 11:57:19 |
| K900 | I'm getting ideas for potentially more crimes | 11:57:21 |
| magic_rb | Could be kinda cool | 11:57:22 |
| emily | fwiw there are already hack patches in the various bpi-r4 trees to handle 2500BASE-X | 11:57:37 |
| emily | by disabling "inband"(? no idea what that is) and stuff | 11:57:42 |
| K900 | And also I'm pretty sure the PCS subsystem is just Wrong | 11:57:43 |
| emily | apparently it's broken with autoneg but not if you don't use autoneg? | 11:57:46 |
| K900 | In reply to @emilazy:matrix.org
fwiw there are already hack patches in the various bpi-r4 trees to handle 2500BASE-X Uhh got a link | 11:57:53 |
| emily | people have definitely gotten 2.5G to work | 11:57:56 |
| emily | ugh I'll dig it up :P | 11:58:02 |
| K900 | In reply to @emilazy:matrix.org
apparently it's broken with autoneg but not if you don't use autoneg? It's not broken with autoneg, it does not have the concept of autoneg | 11:58:05 |
| emily | wait what | 11:58:12 |
| emily | it can definitely autoneg for some rates at least | 11:58:17 |
| K900 | But the phylink subsystem as implemented doesn't understand that | 11:58:24 |
| K900 | In reply to @emilazy:matrix.org
it can definitely autoneg for some rates at least 2500base-x is what doesn't have the concept of autoneg | 11:58:46 |
