| 2 Feb 2025 |
 motiejus | Are you referring to https://lore.kernel.org/all/20240612-6-10-rocket-v1-0-060e48eea250@tomeuvizoso.net/ ?
Looking at a state of rk3588 today, seems like many new things have indeed landed in v6.11. Though HW Crypto engine and h.264 decoders seem to be not in mainline yet (things I still quite care about).
The scariest and most annoying thing, when I tried this about a year ago, was booting it. :/ though this seems quite reasonable today.
| 06:31:51 |
| motiejus | * Are you referring to https://lore.kernel.org/all/20240612-6-10-rocket-v1-0-060e48eea250@tomeuvizoso.net/ ?
Looking at a state of rk3588 today, seems like many new things have indeed landed in v6.11. Though HW Crypto engine and h.264 decoders seem to be not in mainline yet (things I still quite care about).
The scariest and most annoying thing, when I tried this about a year ago, was booting it. :/ though this seems quite reasonable today.
I am in the market for one -- either rk3588 or something from beelink.
| 06:32:24 |
 K900 | In reply to @motiejus:jakstys.lt
Are you referring to https://lore.kernel.org/all/20240612-6-10-rocket-v1-0-060e48eea250@tomeuvizoso.net/ ?
Looking at a state of rk3588 today, seems like many new things have indeed landed in v6.11. Though HW Crypto engine and h.264 decoders seem to be not in mainline yet (things I still quite care about).
The scariest and most annoying thing, when I tried this about a year ago, was booting it. :/ though this seems quite reasonable today.
I am in the market for one -- either rk3588 or something from beelink.
There's a bunch of bits | 07:18:36 |
| K900 | The crypto engine is probably not something you're going to care about much | 07:18:53 |
| K900 | Video decode/encode is being worked on but it's a really funny setup | 07:19:04 |
| motiejus | It'd be great to have at least gigabyte-order-of-manigude-per-second for something that dm-crypt supports (it was a few hundred MB/s last I checked on rk3588) | 07:33:00 |
| K900 | Do you really need disk encryption on a server board that has debug headers sticking out of every single possible hole | 07:38:29 |
 macaroniskengz | In reply to @k900:0upti.me
Do you really need disk encryption on a server board that has debug headers sticking out of every single possible hole Yes because law enforcement can't do anything with debug headers when the device is off | 13:44:30 |
| macaroniskengz | * Yes because law enforcement can't magically read encrypted with debug headers when the device is off | 13:44:44 |
| macaroniskengz | * Yes because law enforcement can't magically read encrypted data with debug headers when the device is off | 13:44:52 |
 @hexa:lossy.network | motiejus: wiki.nixos.org is the current wiki fwiw | 13:46:44 |
|  @pederbs:pvv.ntnu.no changed their display name from pbsds to pbsds (FOSDEM). | 16:04:57 |
|  raboof changed their display name from raboof@FOSDEM to raboof. | 22:18:29 |
| 3 Feb 2025 |
 dotlambda | Does Home Assistant benefit a lot from replacing SQLite with PostgreSQL? | 03:19:10 |
 uep | i haven't tried nor felt the need, and I'm a postgresql fan in general | 07:14:31 |
| uep | but it works fine as sqlite, and it means i just zfs snapshot the dataset for the home-assistant nixos container and replicate that as the backup strategy with no further fuss required | 07:15:38 |
 CRTified | My impression was a noticeable performance boost, although I can't explain it | 07:47:21 |
| CRTified | (keeping 2y of history) | 07:47:44 |
|  @lorenzleutgeb:matrix.org changed their display name from Lorenz Leutgeb 📞6343 to Lorenz Leutgeb. | 08:53:02 |
|  SomeoneSerge (utc+3) changed their display name from SomeoneSerge (utc+3) to SomeoneSerge (Gand St. Pieters). | 13:39:02 |
| @hexa:lossy.network | I personally run with sqlite, but I also cull history quickly | 16:15:17 |
| @hexa:lossy.network | ok, nvm. apparently I don't purge anymore 😄 | 16:21:09 |
| @hexa:lossy.network | i have room temperature values going back to december 2023 | 16:21:25 |
| @hexa:lossy.network | 
Download image.png | 16:21:46 |
| @hexa:lossy.network | but you can clearly see the color indication that older values are thinned out | 16:21:56 |
| @pederbs:pvv.ntnu.no changed their display name from pbsds (FOSDEM) to pbsds. | 16:25:34 |
| @hexa:lossy.network | 
Download image.png | 16:30:32 |
 @adam:robins.wtf | that's quite the downsampling :) | 16:45:03 |
| @hexa:lossy.network | after ~10 days apparently | 16:45:52 |
| 4 Feb 2025 |
|  @qfe6mwof7:matrix.org left the room. | 14:28:41 |
