**** BEGIN LOGGING AT Sat Nov 14 02:59:56 2020
Nov 14 03:49:51 <Grommish> Anyone familar with arm devices?
Nov 14 03:54:50 <mangix> Grommish: what about them"?
Nov 14 03:55:08 <Grommish> armv7 is supported, what other generations are supported?
Nov 14 03:55:17 <mangix> supported by what?
Nov 14 03:55:58 <Grommish> openwrt.. example, the r8000 is a armv7_muslgnueabi (.. yeah.. really), but only reports as a arm in $(ARCH)
Nov 14 03:56:09 <Grommish> are their armv8 devices?
Nov 14 03:56:16 <mangix> of course
Nov 14 03:56:26 <Grommish> Supported by openwrt I mean
Nov 14 03:56:31 <mangix> of course
Nov 14 03:56:43 <Grommish> Do they all report as just arm?
Nov 14 03:57:08 <mangix> no idea
Nov 14 03:57:20 <mangix> i think armv8 devices are 64-bit
Nov 14 03:57:50 <Grommish> I'm trying to figure out the best way to deal with target triples in rust
Nov 14 03:58:01 <mangix> which would mean they report as aarch64
Nov 14 03:58:25 <Grommish> Ah, ok.. I've got those already.. so its probably just the v7's
Nov 14 03:58:35 <Grommish> they report as arm rather than armv7
Nov 14 04:25:29 <aparcar[m]> grift: ping
Nov 14 04:53:49 <Grommish> mangix: Ah ha..I can check CONFIG_arm_v7 :)
Nov 14 06:29:05 >KGB-1< https://tests.reproducible-builds.org/openwrt/openwrt_ath79.html has been updated. (99.1% images and 100.0% packages reproducible in our current test framework.)
Nov 14 07:11:58 <grift> aparcar[m]: whats shaking?
Nov 14 07:15:45 <aparcar[m]> grift: can you send me your presentation?
Nov 14 07:19:53 <grift> sure, whats yout email address?
Nov 14 07:26:06 <grift> aparcar[m]: https://defensec.nl/~kcinimod/misc/SELinuxOWRT.pdf
Nov 14 09:09:12 <ReDaLeRt> Hi, guys.
Nov 14 09:09:24 <ReDaLeRt> The website is giving 502 errors: https://downloads.openwrt.org/snapshots/targets/
Nov 14 09:09:34 <ReDaLeRt> "502 Bad Gateway"
Nov 14 09:09:50 <ReDaLeRt> Could anyone check this, please? :)
Nov 14 09:23:26 >KGB-0< https://tests.reproducible-builds.org/openwrt/openwrt_sunxi.html has been updated. (0% images and 100.0% packages reproducible in our current test framework.)
Nov 14 10:58:32 <aparcar[m]> grift: thank you
Nov 14 13:17:47 >KGB-1< https://tests.reproducible-builds.org/openwrt/openwrt_omap.html has been updated. (0% images and 100.0% packages reproducible in our current test framework.)
Nov 14 14:00:05 <Hauke> Wireless Battlemesh v13 is starting: https://battlemesh.org/BattleMeshV13#Where
Nov 14 15:20:00 <nbd> ldir: fyi, i've installed latest macos on kvm and fixed up the openwrt build issues i encountered
Nov 14 15:25:40 <Hauke> nbd: could you also check the 19.07 branch
Nov 14 16:59:55 <nbd> Hauke: sure. working on it now
Nov 14 17:00:57 <Hauke> nbd: thanks
Nov 14 17:01:18 <Hauke> how do you run macOS in kvm?
Nov 14 17:02:04 <nbd> https://github.com/kholia/OSX-KVM
Nov 14 17:02:23 <nbd> + a custom tweaked config for libvirt/virsh
Nov 14 17:03:30 <nbd> took some trial & error to make it work well with libvirt and vnc as display output
Nov 14 17:03:36 <nbd> but it's working quite well now
Nov 14 17:06:03 <Hauke> nbd: nice
Nov 14 17:10:22 <nbd> are you interested in using it yourself?
Nov 14 17:20:04 <Hauke> nbd: currently I have a freebsd VM to test build u-boot host tool
Nov 14 17:39:04 >KGB-0< https://tests.reproducible-builds.org/openwrt/openwrt_lantiq.html has been updated. (98.2% images and 100.0% packages reproducible in our current test framework.)
Nov 14 17:45:54 <nbd> Hauke: toolchain/gdb only builds if i backport the update to 10.1
Nov 14 18:07:51 <Hauke> nbd: I think we should not update gdb
Nov 14 18:07:55 <Hauke> in 19.07
Nov 14 18:09:00 <nbd> i'll see if it's feasible to make a smaller fix
Nov 14 18:09:25 <nbd> i've ported the other fixes and gdb is the only remaining issue with a small default config
Nov 14 18:09:53 <Hauke> ok nice
Nov 14 18:50:13 <dangole> OSX-KVM worked great for me until recently with HighSierra. revent updates broke it though, guess my CPU is too old to support setting up HPET in the way they do and now wall-clock is about 10 times as fast as it should be, leading to timeouts everywhere (and lots of other issues obviously)
Nov 14 19:21:07 <ldir> nbd: thanks Felix - I'll give it a try.
Nov 14 20:19:27 <xdarklight> Hauke: https://github.com/openwrt/openwrt/pull/3085#issuecomment-727257773 - is there any documentation on how long it takes for the GPHY firmware to fully initialize? in gswip_gphy_fw_load from drivers/net/dsa/lantiq_gswip.c we just deassert the reset line, but don't wait for anything to complete
Nov 14 20:37:05 <Hauke> xdarklight: I do not know
Nov 14 20:37:22 <Hauke> there is a public documentation for the GPHY
Nov 14 20:37:34 <Hauke> but I do not know if this part is documented
Nov 14 20:38:51 <Hauke> xdarklight: are these problems with internal or external phy?
Nov 14 20:39:07 <xdarklight> Hauke: I think the 7412 uses the internal GPHY
Nov 14 20:39:26 <Hauke> I used the 100M PHY FW on my device and it loaded sucecssfully
Nov 14 20:40:19 <xdarklight> Hauke: the user suspects a race condition. I just checked that public datasheet and it states: First MDIO access after reset release (tMDIO), min: 300.0 ms
Nov 14 20:40:36 <Hauke> but timing could be a problem, it probably takes some time till the FW is booted up
Nov 14 20:41:09 <xdarklight> okay. I'll cook a patch and ask the user to test it
Nov 14 20:41:21 <Hauke> xdarklight: thanks
Nov 14 20:45:56 <xdarklight> Hauke: this is what I have in mind: https://gist.github.com/xdarklight/6a204caec318b55bb18ce42e7c6e6337
Nov 14 20:46:19 <xdarklight> (wait 300ms for any number of internal GPHYs, instead of waiting 300ms for each internal GPHY)
Nov 14 20:55:20 <Hauke> xdarklight: you should use msleep()
Nov 14 20:55:37 <xdarklight> Hauke: oops, indeed
Nov 14 20:56:25 <xdarklight> fixed, thanks
Nov 15 00:21:15 <mangix> cool, kmods are broken on the buildbots
Nov 15 01:39:03 <voltagex> is there a way to compile kernel modules with NLS but not the rest of the system? Enabling NLS is breaking the compilation of gammu
Nov 15 01:43:23 <voltagex> nevermind, got it - you don't need NLS globally for NLS in the kernel it seems
Nov 15 01:44:03 <pkgadd> I assume taking out https://github.com/openwrt/packages/blob/master/utils/gammu/Makefile#L27 would already do (untested)
Nov 15 01:44:37 <pkgadd> at worst you add some cmake config options
Nov 15 01:44:56 <voltagex> not sure, and I can't get builds of single packages working to test.
Nov 15 01:45:33 <voltagex> shouldn't that like mean that building with nls works?
Nov 15 01:47:40 <Grommish> ./staging_dir/hostpkg/bin is in the build path, correct?
**** ENDING LOGGING AT Sun Nov 15 02:59:56 2020