#qi-hardware IRC log for Monday, 2013-09-09

Acefacecan someone please help me with this pos atheros ar9485 wireless bug, it is driving me insane01:52
wpwrakDocScrutinizer05: btw, does the GTA04 design have some issues that may be a problem if inherited by Neo900 ? http://lists.openmoko.org/pipermail/community/2013-September/068921.html15:24
DocScrutinizer05wpwrak: yep15:29
DocScrutinizer05one problem I suspect is with suspend-to-ram which is not a sensible mode for OMAP15:30
DocScrutinizer05the problem with modem as well as any power consumption issues may well be caused by improper implementation of this suspend state15:31
DocScrutinizer05one of the reasons to port maemo fremantle15:32
DocScrutinizer05it's known to provide proper power management on this platform15:32
DocScrutinizer05largely reduced number of kernel device drivers to check since they are "new"15:33
DocScrutinizer05suspend to ram also might have real hw problems, e.g with dangling/floating lines15:34
DocScrutinizer05OMAP isn't meant to suspend to RAM, it suspends in situ in the static CPU registers15:34
DocScrutinizer05the whole suspend concept been a botch born from a pinch OM been in, between available SoCs and the availablitiy of zero-clock on those 15:36
wpwrakso OMAP's standby is inefficient ... because you can't shut down the power rails ?15:39
wpwrakfloating lines sound more like a sw problem than hw. well, if they only float in "static"standby :)15:39
DocScrutinizer05anyway http://privatepaste.com/42eb06c8fe15:40
wpwraki suppose you already searches for pull-ups and pull-downs ?15:40
DocScrutinizer05this is with wlan and IRC active!15:40
wpwraknice :)15:41
DocScrutinizer05do that with suspend-to-ram!15:44
DocScrutinizer05NB the device yells as soon as my name gets highlighted on IRC15:44
DocScrutinizer05with GTA02 and even usual distros on GTA04 you have a hard time getting 10h standby in this situation15:46
DocScrutinizer05ooh, I forgot: of course GSM modem onine as well15:46
DocScrutinizer05OMAP is particularly inefficient when USB fsckng mentorgraphics musb-core gets involved15:47
DocScrutinizer05ther are at least 3 power domains you need to take care about: CPU<->musb bus, musb-core itself, musb <-ULPI-> PHY15:48
DocScrutinizer05the whole crap eats 60mA just for being activated15:49
DocScrutinizer05and I guess when you mess up with deactivation, some data lines might float in undefined states15:49
DocScrutinizer05somebody did statistics and found a very strange 2mA granularity of suspend-to-RAM quiescent current15:50
DocScrutinizer05I guess 2mA for each data line that happens to be 1 (or 0) while other end is suspended, or sth like that15:51
DocScrutinizer05you recall similar problem with GTA02 display driver? deactivating video data and the video bus ate some 30mA worst case15:52
DocScrutinizer05depending on last pixel that been pending to get transfered to LCD15:53
DocScrutinizer05only cloudy memories here, but we definitely had sth like that15:53
DocScrutinizer05the GTA04 shouldn't eat more than ~10mA without any powerhogs like LEDs and TX and amps enabled. The OS guys need to check that first15:58
DocScrutinizer05unless we reach proper power consumption in idle, we don't even need to look at suspend at all16:00
wpwraknaw, don't remember that. we may have had GPIOs that weren't being driven properly, though. but that's just a sw fix.16:01
DocScrutinizer05wpwrak: anyway if you're interested: http://projects.goldelico.com/p/gta04-main/downloads/48/   schematics16:28
--- Tue Sep 10 201300:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!