**** BEGIN LOGGING AT Sun Jul 29 03:00:00 2018
Jul 29 05:45:18 <mangix> anyone have an rt3883 device?
Jul 29 06:02:13 <pkgadd> only in the form of the dreaded RT3883F+RT5392L iNIC, which most probably won't help here
Jul 29 06:08:39 <mangix> it probably will
Jul 29 06:08:47 <mangix> i just need /proc/cpuinfo output
Jul 29 06:14:16 <pkgadd> that's the problem, the router SOC is standard lantiq VRX268, RT3883 iNIC is just a secondary SOC for handling all the wlan stuff (with a tiny ~600 KB rtos 'firmware', only connection to the system SOC is via stacked switches and VLAN)
Jul 29 06:17:14 <mangix> ah ok
Jul 29 06:17:31 <pkgadd> https://openwrt.org/toh/astoria/arcadyan_astoria_easybox_904xdsl_r01
Jul 29 06:18:50 <mangix> I just looked through the git repo and see that there are only two users of the 74kc target
Jul 29 06:56:06 <mangix> ldir: your .59 bump does not build on ipq806x
Jul 29 07:00:07 <gnustomp> mangix: this is an rt3883 - https://paste.debian.net/1035619/
Jul 29 07:09:34 <mangix> interesting. just what i was looking for. thanks.
Jul 29 07:11:14 <mangix> looks like as far as I can tell, MIPS 74kc is the only MIPS processor that supports DSPR2
Jul 29 08:08:07 <ldir> mangix: do you have any more clues?  particular board?  At the moment I can't even get a toolchain to build! :-)
Jul 29 08:10:34 <mangix> oh
Jul 29 08:10:36 <mangix> uhhh
Jul 29 08:10:45 <mangix> it was one of my local patches
Jul 29 08:11:43 <ldir> ah ok :-)  Well I'll try to get a toolchain built anyway
Jul 29 08:30:33 <m4t> hrm for some reason on ath79 my 2 ath9k pci cards are sharing the same interrupt (12) rather than being assigned different interrupts (40, 41) on ar71xx
Jul 29 08:30:46 * m4t has been staring at the pci driver for hours
Jul 29 08:31:02 <m4t> i actually don't know if it's even an issue...
Jul 29 08:43:56 <m4t> ar71xx_pci_irq_init doesn't seem to be taking into consideration ATH79_PCI_IRQ_BASE which is 40
Jul 29 08:55:36 <ultito> gninrom
Jul 29 08:56:22 * ldir holds a mirror up to life
Jul 29 09:19:28 <Sagi> I'm trying to set up a gre ptp link between two openwrt 17.04 devices. I have installed gre and kmod-gre and use the documented options, but no tunnel is brought up. Any pointers on how to debug this?
Jul 29 09:22:31 <m4t> can someone on ath79 using 2 or more pci wifi cards pastebin me the output of /proc/interrupts?
Jul 29 09:31:12 <m4t> i think pci-ar71xx.c might want to use irq_domain_add_simple() rather than irq_domain_add_linear()
Jul 29 09:35:35 <m4t> damn, well it maps them both @ irq 41 now lol. better than 12 i think :|
Jul 29 09:41:38 <rmilecki> jow: SPI regression fixed
Jul 29 09:42:02 <rmilecki> jow: i'm done handling kernel regressions
Jul 29 09:42:34 <rmilecki> jow: two more problem I may manage to fix by Monday: lxc building regression (DonkeyHotei's pull request) and php regression that just needs backporting from master
Jul 29 10:43:49 <Sagi> re: gre setup not working, manual setup appears to work
Jul 29 11:27:55 <Sagi> it's a mystery to me why the uci script don't work
Jul 29 12:09:51 <rmilecki> DonkeyHotei: I think you lxc fix should also go to master, right?
Jul 29 12:12:39 <Sagi> ah, the interface name was too long and I was missing resolveip.
Jul 29 12:47:59 <DonkeyHotei> rmilecki: yes
Jul 29 12:52:21 <DonkeyHotei> i'd hoped https://github.com/openwrt/openwrt/pull/679 could make it in time for the release as well, having been open for 6 months now
Jul 29 13:55:45 <KanjiMonster> rmilecki: dropped the local patches (and linux,part-probe for most devices)
Jul 29 15:08:21 <rmilecki> KanjiMonster: hooray! thanks
Jul 29 15:15:03 <rmilecki> DonkeyHotei: oh, I just compiled lxc without a problem, before applying your patch
Jul 29 15:15:15 <rmilecki> DonkeyHotei: do youknow which package causes these problems?
Jul 29 15:15:21 <rmilecki> I think it's one of lxc tools
Jul 29 15:15:48 <DonkeyHotei> rmilecki: iirc it's lxc itself, when compiled with -O3
Jul 29 15:16:05 <rmilecki> DonkeyHotei: how to trigger -O3?
Jul 29 15:16:24 <rmilecki> make -O3 V=s ?
Jul 29 15:16:59 <DonkeyHotei> CONFIG_TARGET_OPTIMIZATION
Jul 29 15:17:18 <rmilecki> oh, so it's sth different than what I've seen
Jul 29 15:18:05 <DonkeyHotei> there was somebody else here who ran into a problem with one of the tools
Jul 29 15:18:22 <rmilecki> ok, I need some help...
Jul 29 15:18:33 <rmilecki> what's wrong with memcpy(dest, src, 0); ?
Jul 29 15:18:41 <rmilecki> is that illegal call?
Jul 29 15:18:48 <rmilecki> man memcpy doesn't mention that case
Jul 29 15:18:54 <rmilecki> i'd expect it to just do nothing
Jul 29 15:20:17 <rmilecki> ah, it's a buffer overflow? how does check for length help then?
Jul 29 15:24:02 <DonkeyHotei> what's wrong with it is that NULL is explicitly passed
Jul 29 15:24:35 <DonkeyHotei> NULL is not allowed as either src or dst even if len==0
Jul 29 15:25:49 <rmilecki> ok, I see that now
Jul 29 15:25:54 <ldir> there may be an optimisation for non-overlapping copies, so you need src, dst & len to work out if there's an overlap.
Jul 29 15:28:39 <aparcar[m]> hi, the adm5120/rb1xx snapshot builds are still called lede- instead of openwrt, is that intentionally?
Jul 29 15:29:24 <ldir> scrap that, I'm talking bollocks - it's memmove that can cope with overlaps.
Jul 29 15:35:17 <jow> aparcar[m]: that target has been dropped
Jul 29 15:35:26 <jow> what you see are the last snapshot builds of it
Jul 29 15:36:06 <aparcar[m]> jow: thanks
Jul 29 15:36:55 <aparcar[m]> jow: you may want to add one of your filters to keep it from popping up in the json-targets 😉
Jul 29 16:54:45 <xn0r> still can't build 18.06: Patch failed!  Please fix ./patches-2.0/100-musl-compat.patch!
Jul 29 16:56:13 <jow> xn0r: try make ... IGNORE_ERRORS=m
Jul 29 17:00:54 <xn0r> jow: nice, it just skips broken packages
Jul 29 17:38:19 <ultito> gninrom
Jul 29 17:38:43 <Borromini> noonretfa
Jul 29 17:39:17 <salcedo> (: lla, gnineve
Jul 29 17:39:42 <Borromini> xD
Jul 29 18:00:48 <xn0r> jow: odhcpd-ipv6only fails with checK_data_file_clashes with odhcpd
Jul 29 18:57:09 <jow> xn0r: yes, you cannot include both
Jul 29 19:06:04 <guifipedro[m]> how to do a large pool of dhcp, for example from 192.168.98.5 to 192.168.99.250
Jul 29 19:13:25 <jow> guifipedro[m]: just specify a large limit
Jul 29 19:13:39 <jow> and make sure that the underlying interface (lan) has a large enough netmask set
Jul 29 19:16:05 <jow> guifipedro[m]: start 5, limit 501 with ipaddr 192.168.98.1 und netmask 255.255.254.0 will yield 192.168.98.5-192.168.99.250
Jul 29 19:16:15 <jow> check with ipcalc.sh 192.168.98.1 255.255.254.0 5 501
Jul 29 19:17:28 <guifipedro[m]> thanks jow! and I see that ipcalc.sh is included in the router, wow! :)
Jul 29 19:26:53 <swalker> updated openwrt/upstream, https://sdwalker.github.io/uscan/index.html
Jul 29 19:36:36 <xn0r> jow: make menuconfig doesn't allow not to enable both
Jul 29 19:37:42 <xn0r> odhcpd-ipv6only can only be built-in or module
Jul 29 20:01:50 <DonkeyHotei> rmilecki: i think you forgot to bump PKG_RELEASE in master
Jul 29 20:02:11 <rmilecki> DonkeyHotei: i immediately pushed 2nd commit bumping that
Jul 29 20:02:18 <rmilecki> sorry for missing that in the 1st commit
Jul 29 20:03:26 <DonkeyHotei> ah
Jul 29 20:04:10 <rmilecki> 6e1104cc6da9 ("lxc: bump PKG_RELEASE after backporting nl patch")
Jul 29 21:47:51 <rmilecki> jow: i'm not aware of any more blockers for the 18.06
Jul 29 21:59:37 <DonkeyHotei> pr 679 is not a blocker, but awaiting merge for 6 months
**** ENDING LOGGING AT Mon Jul 30 03:00:00 2018