**** BEGIN LOGGING AT Sun Mar 15 03:02:57 2020 Mar 15 08:20:28 * russell-- is being successful kernel panic'ing a beagle bone black when putting an ath9k-htc radio into monitor mode Mar 15 08:27:33 or, even in station mode Mar 15 08:33:24 Is the panic inside the driver or some other place? Mar 15 08:35:05 http://paste.debian.net/1134997/ Mar 15 08:36:36 undefined instruction Mar 15 08:51:16 undefined instruction is how BUG() is implemented on arm: https://elixir.bootlin.com/linux/latest/source/arch/arm/include/asm/bug.h#L15 so it doesn't mean there's any memory corruption in this case. Mar 15 08:59:25 if (unlikely(skb->data < skb->head)) Mar 15 08:59:25 skb_under_panic(skb, len, __builtin_return_address(0)); Mar 15 08:59:56 in skb_push() in net/core/skbuff.c Mar 15 09:03:16 [ 18.258411] skbuff: skb_under_panic: text:a87a076f len:4 put:4 head:0a6f14f7 data:71182cbe tail:0xdcdbc000 end:0xdcdbc040 dev: Mar 15 09:05:14 not certain about the pointer arithmetic here, but data looks like it's greater than head Mar 15 09:19:11 russell--: I passed the paste to Oleksij Mar 15 09:21:48 He might take a look later Mar 15 09:23:28 Says he was seeing something similar a week ago but hasn't investigated yet. Mar 15 09:43:02 russell--: these head/data values look totally bogus, they probably should look similar to tail/end (0xd*) Mar 15 10:15:41 russell--: yupp, it should be data <= head <= tail < end Mar 15 10:27:48 russell--: could you try this: https://git.kernel.org/linus/cd486e627e67ee9ab66914d36d3127ef057cc010 Mar 15 10:28:16 russell--: or this update to 5.4.24: https://git.openwrt.org/?p=openwrt/staging/hauke.git;a=shortlog;h=refs/heads/mac80211-5.6 Mar 15 10:28:20 it contains this fix Mar 15 10:36:07 Hauke: russell--'s panic is happing at ath9k_wmi_cmd, this fix likely won't help Mar 15 13:07:57 build #121 of at91/sam9x is complete: Failure [failed images] Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/at91%2Fsam9x/builds/121 blamelist: Jun Su , Sungbo Eo , Rafa? Mi?ecki , Adrian Schmutzler , Felix Fietkau , Mar 15 13:07:58 Catrinel Catrinescu Mar 15 16:29:10 jow: is it normal that config.seed doesn't appear on downloads.openwrt.org since 18.06.4 (last version to provide it)? Mar 15 16:45:57 f00b4r0: config.buildinfo Mar 15 16:46:41 pkgadd: sorry, i forgot to mention my comment is for 18.06.x Mar 15 16:46:55 the config seed disappears after 18.06.4 Mar 15 16:47:28 ah, sorry (I'm not following the stable branches) Mar 15 16:47:35 np Mar 15 16:52:36 ynezz: did we ever check ucert after the libubox changes? There's a report in the forum that it is broken Mar 15 16:52:46 f00b4r0: no idea, need to investigate later Mar 15 16:53:01 ok Mar 15 19:18:27 Hi... Anyone got a good pointer to working with tc and sch_htb? I’m trying to set up a guest network that has 20% of my uplink bandwitdth allocated to it, but if it isn’t used then the excess capacity is used by my primary network… both networks are on separate interfaces. all of the examples I can find are about traffic classes or host addresses on the same interface… not about sharing bandwidth across two interfaces. Mar 15 20:01:13 has anyone seen what appears to be a race between procd/ubus/logd when booting from initramfs? it doesn't happen every time, but when it does, logd doesn't seem to start, and procd/ubusd loop very quickly like so: https://paste.ee/p/4lQg5#gQ6KHsLEl4QWtM3xuBWdg5u9kDW6TapB Mar 15 20:02:08 trying to manually run logd can lead to 'Failed to connect to ubus' Mar 15 20:02:16 maybe because its in such a tight loop Mar 15 20:02:43 then eventually i will get a watchdog timeout Mar 15 20:03:43 i thought it was just weirdness when building from an unclean dir (even with all libubox deps cleaned, staging stamps removed), but it persisted occasionally even after a complete rebuild of toolchain and all Mar 15 21:03:18 updated openwrt/upstream, https://sdwalker.github.io/uscan/index.html Mar 15 22:42:31 Hi devs! I purchased this router https://www.aliexpress.com/item/32922557750.html?spm=a2g0s.12269583.0.0.4e10629d3lJ29k ; flashed it with the firmware for https://openwrt.org/toh/hwdata/zbt/zbt_wg3526_16m and everything appears to work out of the box though i'm yet to fully investigate - Should I make some commits to have it fully supported/ make some documentation ? Mar 15 22:42:44 I'm happy to contribute if someone can point me in the right direction :) Mar 15 22:44:12 bcm Hi I am not a dev, but it's the rong time around about now for people to be active. If you don't get a ancer ask again in the morning. Mar 15 22:44:29 A lot of the devs are in the EU Mar 15 22:44:50 Tapper: understood Mar 15 22:45:35 We like to get new people involved with OpenWrt tho. Mar 15 22:47:17 So thanks for your offer of time/effort **** ENDING LOGGING AT Mon Mar 16 02:59:58 2020