**** BEGIN LOGGING AT Thu Oct 03 02:59:57 2019 Oct 03 08:19:55 Hi people Oct 03 09:00:51 someone with flyspray access, there seems to be a flurry of spam targetting it today FS#2526, FS#2528-FS#2530 Oct 03 09:16:43 ynezz: btw the TIME_EPOCH commit did its job https://aparcar.gitlab.io/-/rebuild/-/jobs/310680565/artifacts/output/SNAPSHOT/ath79/generic/index.html Oct 03 12:14:44 just saw that Art is not at Wave Computing (MIPS) any more: https://twitter.com/david_schor/status/1179443925515415552 Oct 03 12:14:55 I have only seen this teet and have no other information Oct 03 12:16:35 *tweet Oct 03 12:24:46 aparcar[m]: ask Daniel about Gitorious, he was one of the affected :) Oct 03 15:14:38 I've a favor to ask: If anyone is using ath10k-ct on recent openwrt with 988x (wave-1) card, could you please cat out /sys/kernel/debug/ieee80211/[phyX]/ath10k/fw_regs and see if they are all zeros or not? Hopefully not. Oct 03 15:18:03 greearb_: does this help? https://pastebin.com/RTpsRtnV Oct 03 15:22:04 greearb_: It's all 0s here. :( Oct 03 15:22:10 weird, you see all zeros too. Works fine for me locally on non-openwrt Oct 03 15:26:17 if anyone has time to debug this, maybe add printk in ath10k_refresh_target_regs and ath10k_refresh_peer_stats_t Oct 03 15:30:49 Looks like I'll need to spend some quality time with openwrt code sometime soon....but other things first Oct 03 16:03:16 greearb_: I'm sure it would benefit from your expert eyes - if nothing else to prove /package/kernel/ath10k-ct/Makefile is building the correct driver version & relevant patches Oct 03 16:40:50 Noltari: pingn ? Oct 03 16:42:31 hmmm, looks like dump1090 is broken for rpi4 :/ Oct 03 17:10:08 hmm, target/linux/brcm2708/patches-4.19/950-0699-xhci-add-quirk-for-host-controllers-that-don-t-updat.patch is supposed to fix that Oct 03 17:10:53 but it doesn't match my Via USB hub device ID Oct 03 17:28:58 meh Oct 03 17:36:30 Hauke: saw the iw commit. -flto should be in CFLAGS Oct 03 17:37:23 oh I'm looking at the usb id but I should be looking at the PCI ID, which is correct Oct 03 17:37:31 strange that the quirk doesn't work then :/ Oct 03 17:51:56 oh well, using a rpi3 then Oct 03 17:54:28 mmmm, I want to look at a .so library that (I assume) Fritzbox uses to push firmware to the second SoC, I'm not quite able to read MIPS assembly, so getting it down to C would be nice... Any decompilers that are capable? Oct 03 17:54:56 (or someone that would like to take a look at the file and perhaps be able to tell me how it does it's magic? :P ;) ) Oct 03 17:57:28 decompilers aren't perfect, so you still need to stare at assembly Oct 03 17:59:44 yeh, I'm quite aware :P Oct 03 18:00:16 and i've only used them on x86 :S. Oct 03 18:00:20 though it might make this 'magic' something more like a magician for me (i.e.: something that with a bit of time can be understood) Oct 03 18:01:33 (Fritzbox 5490 contains 2 SoC's, one Lantiq SoC, and one Atheros Q9CA558 connected via switch (MDIO), and it boots via MDIO) Oct 03 18:02:01 if I can figure out how this second linux kernel gets delivered that would be great ;) Oct 03 18:14:17 i'm trying to figure out how to identify the GPIO for Archer C6 (US), gpl source is a bit fuzzy... After uploading the .bin it is brick, i will restore the flash and try again later. may it be related to misconfigured GPIO pins? Oct 03 18:19:34 ok, i found it.. openwrt/qca/src/linux/arch/mips/ath79/mach-archer-a6-v2.c ... now lets fix gpios and try the luck again Oct 03 19:40:40 Anywhere I could find QCA9558 datasheets? Oct 03 19:49:37 meh, same problem on rpi3 Oct 03 19:50:58 must be something openwrt related then, it worked fine on raspbian Oct 03 20:21:31 gpeskens: google for "github atheros datasheet" Oct 03 20:22:03 blocktrron, thanks! Someone did pm me the link a few minutes ago :) Oct 03 20:26:31 :) Oct 03 20:26:46 gpeskens: Ghidra is a nice decompiler which also supports MIPS Oct 03 20:28:24 ty Hauke ! Oct 03 20:28:27 :) Oct 03 20:43:13 evening Oct 03 20:53:39 Done! OpenWRT support for Archer C6 (US) working ;) later i'll submit pull request Oct 03 20:59:08 *OpenWrt Oct 03 21:00:24 yes, sorry ;) Oct 03 21:36:16 mangix: thanks for the info Oct 03 21:36:26 hmm the flags handling in iw looks strange Oct 03 21:36:43 I will have a closer look at this in the next days Oct 03 22:35:31 is firmware partition (kernel/rootfs) generated automatically? Oct 03 22:50:52 gmtandi: usually yes, by a dynamic partition splitter in the kernel Oct 03 22:52:20 ok, i'm having a issue with rootfs, should i fix the os-image and file-system to the same size as stock firmware? Oct 03 23:13:08 it was partition table, fixed ;) now it is working 100% Oct 03 23:13:22 ty @jow **** ENDING LOGGING AT Fri Oct 04 03:02:52 2019