**** BEGIN LOGGING AT Sat Dec 26 03:00:13 2020 Dec 26 06:41:23 dorf: intel radios have a history of not being very good at ap-mode (it might be different these days) Dec 26 06:43:21 russell--: ah, ok, noted. and realtek? Dec 26 06:44:36 all mine are either qualcomm/atheros or mediatek Dec 26 06:48:27 the odroid pkgadd mentioned is realtek I think, no? Dec 26 07:27:32 the odroid h2/ h2+ is wired-only (realtek ethernet, not wireless) - and intel wireless is indeed still unusable for AP mode (and realtek /wireless/ is even worse) Dec 26 07:28:25 ah, right, yeah, mixed up eth and wireless. now you mention it, I recall it saying dual intel radios. Dec 26 07:28:44 except that's the hb2. Dec 26 07:35:39 I'm still holding out hope nxp do the right thing and open source their wrt firmware blob :) Dec 26 07:36:03 lol no way Dec 26 07:36:15 If you're listening NXP, I'm sure you'd get max love in here :) Dec 26 07:36:33 the last company that open sourced firmware is atheros I believe Dec 26 07:51:10 mangix: the optimist in me things NXP just need a nudge to put their firmware up on their github site.. they're obviously not entirely ignorant of the concept of opensource :) https://github.com/NXP/mwifiex-firmware Dec 26 07:51:27 "hundreds have tried, all have failed" Dec 26 07:52:54 at least, they know how to use github. Dec 26 07:54:04 mediatek has the biggest chance IMO Dec 26 08:20:55 iw info says the intel 01:00.0 Network controller: Intel Corporation Wireless 7260 (rev 6b) Dec 26 08:21:03 supports ap mode Dec 26 08:21:26 (that's the radio in one of my laptops) Dec 26 08:25:12 Most are unhappy with how AP mode is supported on the 5 GHz band with Intel devices, see "iw list" for the channels it's allowed on. Dec 26 08:25:37 Channels and power. Dec 26 09:58:39 russell--: that's news to me Dec 26 10:18:32 russell--: please share iw list for that so that one could judge how useful the facilities offered for different usecases. Dec 26 10:19:08 Something tells me mangix won't be fascinated :) Dec 26 10:21:27 how so? Dec 26 10:22:52 I expect very limited power and channel selection on the 5 Ghz band. Dec 26 10:25:07 ah of course. that card is meant to be used in laptops Dec 26 10:52:16 PaulFertser: here you go http://sprunge.us/4X1s96 Dec 26 10:55:26 russell--: hm, looks like no-IR on every 5 GHz channel, so no AP at all there? Dec 26 10:56:11 it is in my laptop, so i wouldn't know ;-) Dec 26 11:54:08 >KGB-0< https://tests.reproducible-builds.org/openwrt/openwrt_sunxi.html has been updated. (0% images and 92.0% packages reproducible in our current test framework.) Dec 26 12:30:37 PaulFertser: hi!! Dec 26 12:33:54 PaulFertser: thanks for helping me out in coordinating with blocktrron, seems we are able to be present only alternately to each other :D Dec 26 12:34:01 but I did read the logs Dec 26 12:34:27 mrkiko: welcome. Dec 26 12:35:14 blocktrron: yyes, I would have to test it somehow, don't exactly know how I might do that, except for substituting the ISP's modem with a computer or something Dec 26 12:37:57 no, won't have an easy way of doing that except use another devicei n the meantime to connect all the rest. Let's see... Dec 26 12:40:22 blocktrron: thanks for your port to that device, it's very nice. Dec 26 13:48:09 mrkiko: ar40xx should support port mirroring, otherwise pass the traffic thru a PC Dec 26 13:48:36 in the end, i try to avoid this mess altogether as best as is can Dec 26 14:34:10 blocktrron: thanks! I hope I'll be able to do that. Dec 26 14:34:52 blocktrron: I mean, I guess I'll have to get another 7530 to do it peacefully from this desk **** ENDING LOGGING AT Sat Dec 26 20:43:00 2020 **** BEGIN LOGGING AT Sat Dec 26 20:50:42 2020 **** ENDING LOGGING AT Sat Dec 26 21:02:01 2020 **** BEGIN LOGGING AT Sat Dec 26 21:11:18 2020 Dec 26 22:09:48 hi Dec 26 22:10:15 i'm modifying pci-rt3883.c to support rt6855a but the weird dts structure in rt3883.dtsi gives me headaches Dec 26 22:11:47 https://github.com/openwrt/openwrt/blob/master/target/linux/ramips/dts/rt3883.dtsi#L355 : there are 3 levels : "pci: pci@10140000" (without a device_type), "pci@0", "pci1: pci@1" Dec 26 22:12:17 the PCI ranges (IO/MEM) are loaded from "the PCI host bridge child node" : https://github.com/torvalds/linux/blob/master/arch/mips/pci/pci-rt3883.c#L443 Dec 26 22:13:07 I can't figure out how to write my .dts to work with this strange structure Dec 26 22:14:29 I wonder if I should abandon modifying pci-rt3883.c and just start a new file Dec 26 22:14:38 maybe this is a question for Linux devs though Dec 27 00:06:49 Hey guys. I was bricking my dlink DWR 960 with installing the latest sysupgrade .. jboot recovery even broke. Dec 27 00:07:10 Reset button had been pressed Dec 27 00:07:28 that's the last thing it outputs Dec 27 00:07:49 any generic advice as to what to do? Dec 27 00:34:29 if the 'standard' ways don't work, serial console access remains Dec 27 01:14:02 it is getting more weird. The router is booting fine after releasing the reset button right after the first "Reset button has been pressed". Dec 27 01:14:53 without pressing the reset button it doesn't get out of the boot loader messages. Keeping the reset button pressed until the message shows up twice als doesn't let it boot. Dec 27 01:15:34 to me it looks as if the sysupgrade was overwriting the jboot/jrecovery partition. Dec 27 01:18:28 is anyone familiar with ramips/mt7620/dlink dwr 960? Dec 27 01:21:47 Is there a way to download the sysupgrade.bin from August 28 2020? Dec 27 01:25:10 or have the buildbots build a snapshot for me? Dec 27 01:25:14 if you didn't back that up yourself, no - snapshots are replaced daily and not backed up Dec 27 01:26:41 I was downloading the factory.bin that day, not the sysupgrade.bin Dec 27 01:44:41 sysupgrade -F factory.bin got me back to the original jboot/recovery .. it doesn't make any sense ... Dec 27 01:51:55 >KGB-0< https://tests.reproducible-builds.org/openwrt/openwrt_lantiq.html has been updated. (98.2% images and 97.4% packages reproducible in our current test framework.) **** ENDING LOGGING AT Sun Dec 27 02:59:57 2020