**** BEGIN LOGGING AT Sun Jan 31 02:59:57 2021 Jan 31 07:38:50 aparcar[m]: pong Jan 31 07:48:23 Merged the zstd thing. Worked for me Jan 31 07:48:33 What about the gettext patch Jan 31 07:48:44 How do I test it again? Jan 31 07:48:49 mangix: -^ Jan 31 07:48:57 ehm Jan 31 07:49:55 gettext is a provider of libintl.so Jan 31 07:50:12 let's see if it has any utilities Jan 31 07:51:47 ah OK. so for target builds, no utilities get installed. just the library. however, gettext-full/host installs all of them Jan 31 07:52:40 https://gist.github.com/neheb/ad570e4ba83ca38bc7a6c087a68640cc is all of them. Jan 31 09:31:44 build #792 of at91/sam9x is complete: Success [build successful] Build details are at http://buildbot.openwrt.org/master/images/builders/at91%2Fsam9x/builds/792 Jan 31 14:24:02 does uci have an option to _insert_ blocks? Jan 31 14:24:23 i.e. I want the last section to stay last, and want to insert a block _before_ that last section Jan 31 14:24:51 or will I have to delete the last and recreate it? Jan 31 17:52:33 updated openwrt/upstream, https://sdwalker.github.io/uscan/index.html Jan 31 19:53:01 Borromini: ping Jan 31 19:53:57 rsalvaterra: pong Jan 31 19:54:25 So, about a year ago, someone opened this topic on the forum… https://forum.openwrt.org/t/ram-size-detection-wrong-xiaomi-ac2100-mt7621/56141 Jan 31 19:54:34 … which means, what I'm seeing is a regression. Jan 31 19:55:27 I'm a bit wary of hacking the device tree itself, but the fact is the RAM detection isn't working properly, it seems… Jan 31 19:57:19 ok Jan 31 19:58:50 I don't know who are the ramips/mt7621 specialists, though. :P Jan 31 20:14:07 :( Jan 31 20:14:11 me neither, sorry. Jan 31 20:14:27 check who authored the autodetection commit maybe? Jan 31 22:30:20 rsalvaterra: as a stopgap you could hardcode the ram into your dts, and roll your own (i assume you already do) Jan 31 22:43:48 rsalvaterra: gch is sometimes here **** ENDING LOGGING AT Mon Feb 01 02:59:57 2021