**** BEGIN LOGGING AT Mon Oct 27 03:00:01 2014
**** BEGIN LOGGING AT Mon Oct 27 12:03:18 2014
Oct 27 15:41:40 <lumag_> hello ant_work bluelightning
Oct 27 15:41:48 <bluelightning> hi lumag_, ant_work
Oct 27 15:43:22 <lumag_> ant_work, I've sent you the patches for locomo.
Oct 27 15:43:27 <lumag_> Did you have chance to check them?
Oct 27 15:44:53 <ant_work> hi
Oct 27 15:45:19 <ant_work> sort of...but I have to refactor the gpio_power patch (doesn't apply anymore)
Oct 27 15:45:50 <lumag_> Hmm.
Oct 27 15:45:53 <ant_work> bluelightning: I was pinging you about the pending meta-hh patches
Oct 27 15:46:08 <ant_work> lumag: I commented out 3 patches to apply your patchset
Oct 27 15:46:14 <lumag_> upstream kernel is broken for pxa
Oct 27 15:46:15 <ant_work> 3 of the meta-hh recipe
Oct 27 15:46:23 <lumag_> Maybe.
Oct 27 15:46:28 <bluelightning> ant_work: ah ok, I will try to merge those... this weekend was a bit crazy
Oct 27 15:46:29 <ant_work> two for locomo and on efor gpio_charger
Oct 27 15:46:30 <lumag_> Which patches?
Oct 27 15:46:40 <lumag_> ok.
Oct 27 15:47:21 <ant_work> locomo-fix-SCL-SDA.patch
Oct 27 15:47:28 <ant_work> locomolcd-suspend-resume.patch
Oct 27 15:47:43 <ant_work> ^^ clashes with locomo
Oct 27 15:47:45 <ant_work> collie-use-gpio-charger.patch
Oct 27 15:47:51 <ant_work> ^^^ clashes with collie.c
Oct 27 15:48:11 <ant_work> about the latter, I see you have rebased and sent upstream
Oct 27 15:48:17 <lumag_> yep.
Oct 27 15:48:20 <lumag_> No response though.
Oct 27 15:48:44 <ant_work> I need to rebase it on top of the patchset
Oct 27 15:49:15 <ant_work> bluelightning: np
Oct 27 15:49:29 <lumag_> gpio-charger patches should be applicable, two others should be obsolete.
Oct 27 15:49:30 <ant_work> bluelightning: just the queue is getting long
Oct 27 15:49:41 <bluelightning> understood
Oct 27 15:49:48 <lumag_> btw: what is the history behind SCL_SDA patch? I don't remember :((
Oct 27 15:50:06 <ant_work> -	r &=  ~(LOCOMO_DAC_SCLOEB);
Oct 27 15:50:06 <ant_work> +	r &=  ~(LOCOMO_DAC_SDAOEB);
Oct 27 15:50:07 <lumag_> I should double check it.
Oct 27 15:50:10 <lumag_> Yes, I remember.
Oct 27 15:50:14 <ant_work> joking ;)
Oct 27 15:50:18 <lumag_> I just don't remember, why I wrote it.
Oct 27 15:50:35 <lumag_> I have dumped most of my local patches.
Oct 27 15:50:46 <lumag_> Unfortunately sa1100 patches seems to be really ignored :(
Oct 27 15:51:03 <ant_work> no worries, I preserve religiously in meta-hh
Oct 27 15:52:41 <ant_work> lumag_: btw about locomo_kbd_tweak-r2.patch
Oct 27 15:52:47 <lumag_> yes.
Oct 27 15:52:51 <ant_work> Haojian sais 'send upstream'
Oct 27 15:53:04 <ant_work> why is tagged 'unappropriate' ?
Oct 27 15:53:28 <ant_work> most suspect is
Oct 27 15:53:29 <ant_work> -#define SCAN_INTERVAL		(HZ/10)
Oct 27 15:53:29 <ant_work> +#define SCAN_INTERVAL		(HZ/20)
Oct 27 15:54:01 <ant_work> on collie SManual there are all keys mapping afaik
Oct 27 15:54:19 <ant_work> if only collie and poodle are using this driver we could indeed 'fix' it
Oct 27 15:54:32 <ant_work> (there are the codes on collie serv man iirc)
Oct 27 15:55:45 <ant_work> at least in kexecboot menu, poodle is even too rapid to loop over the menu
Oct 27 15:56:03 <ant_work> collie is slower (is just the clock?)
Oct 27 15:56:41 <ant_work> i.e. poodle triggers easily double clicks pressing key down
Oct 27 15:58:06 <ant_work> both are using the same  --enable-evdev-rate=1000,250
Oct 27 16:16:25 <lumag_> ant_work, I don't know.
Oct 27 16:17:04 <lumag_> I'm currently concerned with 195b58add463f697fb802ed55e26759094d40a54
Oct 27 16:17:09 <lumag_> It breaks badly in qemu
Oct 27 16:17:47 <lumag_> on pxa targets
Oct 27 16:21:10 <lumag_> sound on spitz is broken
Oct 27 16:22:15 <ant_work> ARM: Avoid writing to control register on every exception
Oct 27 16:22:18 <ant_work> ?
Oct 27 16:22:23 <lumag_> Yes.
Oct 27 16:22:36 <ant_work> I confess I did miss that one ;)
Oct 27 16:22:43 <ant_work> Is it really that bad?
Oct 27 16:25:57 <lumag_> I can't run pxa targets in qemu with that patch.
Oct 27 16:26:08 <lumag_> Didn't try bare metal, but I suspect it will trigger errors too.
Oct 27 16:26:48 <ant_work> poodle was booting 3.17 with the dlldebug fix
Oct 27 16:27:07 <lumag_> Yes, that is 3.18-rc1/2
Oct 27 16:27:16 <ant_work> I see
Oct 27 16:36:54 <lumag_> 14621c7e5e72200ec021a7580121130ce7f2ff22 broke audio on spitz
Oct 27 16:49:08 <lumag_> ok, I hope sound will be fixed.
Oct 27 16:49:15 <lumag_> pinged broonie.
Oct 27 16:55:10 <lumag_> ant_work, regarding LoCoMo. were you able to actually test patches?
Oct 27 16:55:43 <ant_work> no, just built w/out gpio_charger. I'll test in a few hours
Oct 27 16:56:29 <ant_work> btw, I have only changed the MFD inthe defconfig
Oct 27 16:56:49 <ant_work> any other manual change requested?ù
Oct 27 16:56:54 <lumag_> you will also need several more symbols.
Oct 27 16:57:24 <lumag_> I will send a list in a few minutes.
Oct 27 16:57:37 <ant_work> I did menuconfig and changed the mfs. Verified there is a (new) locomolcd, did not change leds...
Oct 27 16:57:47 <lumag_> also gpio_locomo
Oct 27 16:58:06 <ant_work> ok
Oct 27 16:58:52 <lumag_> Strange.
Oct 27 16:59:01 <lumag_> collie works with 195b58...
Oct 27 16:59:09 <lumag_> tosa and spitz do not
Oct 27 16:59:12 <lumag_> (in qemu).
Oct 27 17:03:08 <ant_work> have you queued  ARM: pxa: fix hang on startup with DEBUG_LL ?
Oct 27 17:04:42 <ant_work> is not yet in -next
Oct 27 17:08:19 <lumag_> Robert said, that it is in pxa/fixes
Oct 27 17:09:43 <lumag_> https://github.com/rjarzmik/linux/commit/cde7fc879969f933614b1256df2625d6ff637bab
Oct 27 17:09:58 <ant_work> ok, missed rc1 & 2
Oct 27 17:10:06 <lumag_> It's up to him to ping arm-soc maintainers
Oct 27 17:10:34 <ant_work> when I tested that one poodle was booting
Oct 27 17:10:52 <lumag_> The issue arrives when DEBUG_LL is selected (and enabled?)
Oct 27 17:11:01 <ant_work> I'll backport for linux-yocto_3.17
Oct 27 17:12:06 <ant_work> btw btw, did you add dummy clock_set_rate() ? There is still that revert
Oct 27 17:14:21 <lumag_> No :(
Oct 27 17:14:29 <lumag_> If you need it, you can add it.
Oct 27 17:14:58 <ant_work> what was needing it? usb phy?
Oct 27 17:15:40 <ant_work> ERROR: "clk_set_rate" [drivers/usb/phy/phy-generic.ko] undefined!
Oct 27 17:16:28 <ant_work> is it really needed or is it a poor driver fault?
Oct 27 17:17:30 <lumag_> It is part of COMMON_CLK API.
Oct 27 17:17:37 <lumag_> <linux/clk.h>
Oct 27 17:17:45 <lumag_> However sa-1100 does not provide full API.
Oct 27 17:18:21 <lumag_> I hope to create COMMON_CLK implementation for sa-1100 (easy) but I'd like rmk first to agree & pick the cpu_clk patches.
Oct 27 17:18:58 <ant_work> I really don't know why that driver is compiled...must be yocto default
Oct 27 17:20:06 <lumag_> usb/phy seems to be a part of usb (pseudo-)otg framework.
Oct 27 17:20:16 <lumag_> Used e.g. for VBUS handling in pxa25x/27x udc drivers
Oct 27 17:32:32 <ant_work> bbl
Oct 27 23:06:23 <lumag> ant__, good evening.
Oct 27 23:06:36 <ant__> hi
Oct 27 23:07:05 <ant__> I'm having hard time with a broken linux-yocto git tree...
Oct 27 23:07:08 <ant__> | error: arch/arm/mach-sa1100/collie.c: patch does not apply
Oct 27 23:07:09 <ant__> | ERROR. could not update git tree
Oct 27 23:07:27 <ant__> applying the last one by hand right now
Oct 27 23:10:24 <lumag> It looks like RMK's patch triggered some bug in qemu.
Oct 27 23:10:56 <lumag> I reflashed my poodle and the kernel works.
Oct 27 23:11:27 <lumag> I need to update my local qemu
Oct 27 23:13:31 <ant__> lumag, btw the mfd driver does not appear alphabetically ordered
Oct 27 23:14:17 <lumag> It looks so
Oct 27 23:23:33 <ant__> this is collie
Oct 27 23:23:34 <ant__> http://pastebin.com/04p7LfZN
Oct 27 23:27:50 <ant__> I'll test mmctomorrow
Oct 27 23:27:53 <ant__> gn
**** ENDING LOGGING AT Tue Oct 28 03:00:00 2014