| Wizzup | freemangordon: what about it, I don't see what you'd like me to check per se | 00:29 |
|---|---|---|
| uvos | freemangordon: upps | 00:29 |
| uvos | freemangordon: thanls | 00:29 |
| uvos | thanks | 00:29 |
| Wizzup | freemangordon: maybe try this in /etc/network/interfaces: https://wiki.debian.org/NetworkConfiguration#The_resolvconf_program | 00:30 |
| freemangordon | Wizzup: there is an option to ignore resolvconf | 01:12 |
| freemangordon | maybe it should be set for VMs | 01:12 |
| freemangordon | hmm, ok | 01:13 |
| Wizzup | freemangordon: well, I added the code not to ignore it and make it work since it is the upstream way | 01:14 |
| freemangordon | ok | 01:15 |
| freemangordon | I will add nameserver in interfaces filke | 01:15 |
| Wizzup | but sure why dhclient would not write resolvconf | 01:15 |
| Wizzup | maybe we need to check | 01:15 |
| Wizzup | hm: | 01:16 |
| Wizzup | Specifically, resolvconf includes a file which modifies the make_resolv_conf shell function used by dhclient-script(8). | 01:16 |
| Wizzup | https://wiki.debian.org/resolv.conf | 01:16 |
| Wizzup | E_TOO_MANY_OPTIONS :) | 01:17 |
| Wizzup | sicelo: wow they accepted the wl1251 patch | 10:36 |
| Wizzup | nice | 10:36 |
| Wizzup | tmlind: would you like me to submit some omap2plus_defconfig changes as patches, or just suggest them here? | 11:57 |
| Wizzup | tmlind: I think disabling CONFIG_COMPACT per defconfig makes sense, and adding CONFIG_SND_SOC_TLV320AIC3X=m and CONFIG_SND_SOC_TLV320AIC3X_I2C=m makes n900 audio work against after a96d2ba2d8248d5e8170f2f44f98d4a33329b08a | 11:58 |
| Wizzup | I haven't seen a reply from the guy who authored CONFIG_COMPACT about making it behave on non-server platforms, so I'm surprised it's enabled by default tbh | 11:59 |
| Wizzup | work again after* | 11:59 |
| Wizzup | happy to send a patch, just not sure if it's normal for me to submit patches for omap2plus_defconfig | 11:59 |
| lel | MerlijnWajer opened an issue: https://github.com/maemo-leste/bugtracker/issues/596 (Investigate support for GAN/UMA) | 12:30 |
| uvos | no way gan can work without the modem co-operating for gan->gms/umts handoff | 12:38 |
| Wizzup | yeah probably. just responding to a request from someone on twitter | 12:38 |
| Wizzup | https://twitter.com/jkepler/status/1471422192944615429 | 12:39 |
| Wizzup | and pavel pulled the led patches | 12:44 |
| Wizzup | :) | 12:44 |
| tmlind | Wizzup: i suggest you resend the CONFIG_COMPACT regression report with linux-mm in cc rather than trying to work around it | 14:25 |
| tmlind | Wizzup: eayh adding the audio stuff as loadable modules makes sense for sure :) | 14:26 |
| Wizzup | ok | 14:35 |
| Wizzup | tmlind: shall I send a patch for the audio modules then? | 14:53 |
| tmlind | Wizzup: sure | 15:54 |
| Wizzup | tmlind: btw that's linux-mm@kvack.org then? | 16:10 |
| * Wizzup assumes yes | 16:11 | |
| uvos | do cc me | 16:15 |
| Wizzup | you're in CC already | 16:17 |
| Wizzup | unless devnull@ doesn't reach you :P | 16:17 |
| uvos | no i get it | 16:23 |
| uvos | that just has a very agressvie spam filter | 16:23 |
| uvos | but you are on the whitelist | 16:23 |
| Wizzup | ok | 16:23 |
| Wizzup | btw, it seemed like you figured out how to users tracers to get more pm debug info yesterday with bencoh's help? | 16:23 |
| uvos | echo 1 > /sys/kernel/debug/tracing/events/workqueue/enable; sleep 10; echo 0 > /sys/kernel/debug/tracing/events/workqueue/enable; cat /sys/kernel/debug/tracing/trace > /home/user/trace.log | 16:24 |
| uvos | that | 16:24 |
| uvos | but it wasent too helpfull, besides the wl12xx events | 16:24 |
| uvos | (that are sane) | 16:24 |
| uvos | it also just lists pretty generic functions | 16:25 |
| uvos | nothing that would help iding what causes the ipis for instance | 16:25 |
| uvos | but tracing can also list the interrupts too | 16:25 |
| uvos | i just havent looked at how | 16:25 |
| uvos | also figured out is very optimisitic i just did what he told me to do :P | 16:26 |
| uvos | so besides wifi dbs_work_handler | 16:27 |
| uvos | is the main culprit | 16:27 |
| uvos | and vmstat_update | 16:27 |
| uvos | i gues increase stat_interval in sysctl would help with the latter | 16:31 |
| uvos | maybe its not so useless on second look | 16:32 |
| Wizzup | uvos: right, maybe we add the vmstat thing to droid4-pm ? | 22:01 |
| Wizzup | ok so the main off mode problem atm is gpio_keys | 22:43 |
| Wizzup | (on n900) | 22:43 |
| Wizzup | is really misbehaves | 22:43 |
| Wizzup | uvos: interesting it's also mostly wl1251 for me | 22:46 |
| Wizzup | I think some trace points are probably just missing | 22:46 |
| Wizzup | basically all CONFIG_*_TRACER is turned off in our config | 22:47 |
| Wizzup | also ke-recv goes nuts when gpio_keys is removed | 22:48 |
| Wizzup | uvos: oh I guess we just need enable more events tracing | 23:03 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!