**** BEGIN LOGGING AT Wed Nov 19 02:59:59 2014 Nov 19 12:57:41 lumag: wrt SPI found this old log around http://pastebin.com/MZAaXQqp Nov 19 14:02:16 ant_work, hello Nov 19 14:02:56 what is the issue with the log? Nov 19 14:02:59 hi Nov 19 14:03:21 is a 2.6 verbose log: locomo lowers the speed Nov 19 14:05:40 yes, that is normal, I think. Nov 19 14:06:04 Either mmc_spi definition contained lower speed, or the card inserted provided some speed limitations (can it?) Nov 19 14:06:15 seems there was no card Nov 19 14:16:58 Hmm. Nov 19 14:33:47 bbl Nov 19 17:46:06 lumag, after reformatting with SDFormatter tool now SanDisk 1G is seen by Qtopia 2.4 Nov 19 17:46:31 oh, so the problem was really in the partition layout. Nov 19 17:46:56 was multipartitioned iirc Nov 19 17:47:21 I hope 3.18 sees that now Nov 19 17:47:51 this collie, the one refusing to kexec since yesterday, runs beautifully Qtopia Nov 19 17:48:00 and recharges the new battery Nov 19 17:48:14 I'll reflash it in a while, after full charge Nov 19 17:51:01 also the reformatted LExar 830->916MB is seen as well Nov 19 17:51:57 heh..the 4G CF freezes all ... Nov 19 17:52:28 no, ok, just undetected Nov 19 17:52:54 strange..I can eject it but it is not mounted... Nov 19 17:53:05 anyway, CF 4G is detected Nov 19 17:53:22 all that stuff should hopefully work with 3.18 Nov 19 17:54:47 * lumag is booting from CF - there should be no surprises ever. Nov 19 17:55:01 strange ext2 partition is nout found/mounted Nov 19 17:55:30 I'm trying to finish locomo rework. The only thing left is the "dac" that is used to set comadj voltage - part of LCD driver. Nov 19 17:55:46 great Nov 19 17:56:01 linusw suggested to use iio framework, I tend to use voltage regulators. Nov 19 17:56:10 But this piece does not fit into my image. Nov 19 17:56:18 s/image/kernel/ Nov 19 17:58:23 s/kernel/brain/ Nov 19 18:00:57 I see sometimes the frameworks do overlap Nov 19 18:03:53 yes Nov 19 18:13:44 The problem is that regulators do really fit the lcd part. But they do not fit quite well the sound part (on collie the second channel of the DAC is used to control "master" volume - both headphones and speaker). Nov 19 18:14:13 I'll probably just fix the "master" volume and hope that individual volume controls will work better. Nov 19 18:14:40 (both headphones and speaker parts have their own volume controls). Nov 19 19:29:46 lumag, after reflashing Sharp rom, charging and reflashing kexecboot+rootfs 3.18 it boots again Nov 19 19:30:47 last thing I did was putting it to suspend Nov 19 19:30:54 and now again... Nov 19 19:31:09 echo mem > /sys/power/state Nov 19 19:31:51 no power led with AC connected..like dead Nov 19 19:32:21 with 3.17 (before new locomo) resume was ok Nov 19 19:32:48 I'm sure...we tested fixing MTD ;) Nov 19 19:33:07 so 2 possibilities: Nov 19 19:33:27 1) I have really bad luck with hardware and even the 2nd battery is defective Nov 19 19:33:46 2) 3.18 + patches does smthg bad to power related stuff Nov 19 19:36:40 hell. it doesn't kexec after that bad suspend Nov 19 21:20:07 ge ant__ bluelightning Nov 19 21:20:52 ant__, this is really strange. I tested suspend/resume with locomo patches. Nov 19 21:21:22 I have old kexecboot flashed, but that shouldn't matter (hopefully). Nov 19 21:22:05 I'm checking the batteries Nov 19 21:22:26 these get to 3.9 - 4.0 after a couple of mins (ac detached) Nov 19 21:22:59 there is a strange reading in D+P menu Nov 19 21:23:09 one batt does 440, the other 1040 Nov 19 21:23:42 it is AIN0 Nov 19 21:23:55 Battery pack presence/absence Nov 19 21:24:08 I'll check tomorrow. Nov 19 21:24:14 with mine batteries. Nov 19 21:24:49 neverthless Sharp rom runs with bot batteries Nov 19 21:25:13 maybe the move to gpio-charger? Nov 19 21:25:27 No. It should not matter. It toggles the same gpios. Nov 19 21:25:39 You can check various statuses/readings in /sys/class/power Nov 19 21:26:00 ah.ok AIN0 is TEMP Nov 19 21:26:13 I see it in Low Battery menu Nov 19 21:26:37 then it's ok they differ Nov 19 22:08:02 lumag_, crazy... Nov 19 22:08:39 after flashing 1) Sharp 2)l-kexecboot + initrd.bin then it boots from SD Nov 19 22:08:43 err.. CF Nov 19 22:08:53 now that it booted I'll test SD ;) Nov 19 22:09:53 same error 38 Nov 19 22:12:19 http://pastebin.com/chhJVYYE Nov 19 22:15:18 root@collie:~# mmc0: new SD card on SPI Nov 19 22:15:18 mmcblk0: mmc0:0000 LEXAR 952 MiB Nov 19 22:15:18 mmcblk0: p1 Nov 19 22:15:18 mmcblk0: error -14 sending stop command, original cmd response 0x0, card status 0x0 Nov 19 22:15:30 ^Lexar is still ok after reformatting Nov 19 22:28:31 ant__, maybe those -38 were of different kinds. Nov 19 22:28:54 to not to kill the flash, don't test my kernels for a few days, please. Nov 19 22:29:19 eh...I still have one spare collie ;) Nov 19 22:29:51 seems I have *not* to suspend Nov 19 22:29:55 It should be possible to reflash collie using JTag, but I won't recommend doing so. Nov 19 22:30:09 did you suspend on-ac or with ac unplugged? Nov 19 22:30:23 plugged on ac (cradle) Nov 19 22:30:24 because I run all my tests with AC being pluged Nov 19 22:30:29 :) Nov 19 22:31:07 2nd reboot also done ;) Nov 19 22:31:17 seems it's suspend Nov 19 22:32:01 now tried poweroff Nov 19 22:32:40 wowo..no reboot? it did really 'poweroff'? Nov 19 22:33:03 pressing ok wakes kexecboot kernel :) Nov 19 22:33:13 or maybe was on/off Nov 19 22:33:56 now it doesn't kexec :/ Nov 19 22:34:53 again... Nov 19 22:35:13 after 'sleep' state it refuses to kexec new kernels Nov 19 22:36:25 how can it keeps settings when I press hw reset? Nov 19 22:36:36 or extract battery? Nov 19 22:36:58 something is written in the firmware? Nov 19 22:37:04 Maybe the cfi is really rewritten Nov 19 22:37:20 Should we lock the cfi before suspending? Nov 19 22:37:31 3.17 was ok Nov 19 22:37:49 Check the stock 3.18-rc* so that we can be sure - where is the problem. Nov 19 22:37:57 ok Nov 19 22:38:22 I'll do saturday Nov 19 22:38:29 more checks Nov 19 22:38:37 tomorrow 700+700km :/ Nov 19 22:38:46 ? Nov 19 22:38:54 going to France and back Nov 19 22:38:59 ah. Nov 19 22:39:44 I'll try now to reflash just kexecboot kernel + initrd Nov 19 22:41:08 kexecboot kernel is based on old kernels with old locomo Nov 19 22:41:17 yep Nov 19 22:41:20 maybe some registers are not reset at boot Nov 19 22:42:47 so. Nov 19 22:43:20 If you kexecboot (old kernel) + 3.18-rc5 + locomo then suspend, Nov 19 22:43:39 after that kexecboot boots, but all other kernels can not be loaded? Nov 19 22:43:42 Is that so? Nov 19 22:43:46 correct Nov 19 22:44:18 same if you type poweroff Nov 19 22:44:34 so poweroff also "hacks" the device. Nov 19 22:44:45 strange Nov 19 22:44:49 usually just reboots Nov 19 22:45:03 strangely this time suspended Nov 19 22:45:18 (like C+D flasher does after finishing) Nov 19 22:47:47 I can not understand, why would kexecboot load, but kexec won't work. Nov 19 22:47:53 I confirm: after reflashing kexecboot (C+D) it does kexec again... Nov 19 22:48:04 no need to reflash initrd.bin Nov 19 22:48:13 just the zImage Nov 19 22:48:32 btw Nov 19 22:48:33 ata1.00: CFA: SanDisk SDCFH-004G, HDX 5.04, max PIO4 Nov 19 22:48:33 ata1.00: 7835184 sectors, multi 0: LBA Nov 19 22:48:33 ata1.00: configured for PIO0 Nov 19 22:48:33 scsi 0:0:0:0: Direct-Access ATA SanDisk SDCFH-00 5.04 PQ: 0 ANSI: 5 Nov 19 22:48:33 sd 0:0:0:0: [sda] 7835184 512-byte logical blocks: (4.01 GB/3.73 GiB) Nov 19 22:48:56 this CF 4G seems ok Nov 19 22:49:28 this was the reboot after reflashing Nov 19 22:50:27 retrying poweroff Nov 19 22:50:44 root@collie:~# poweroff Nov 19 22:50:44 Broadcast message from root@collie (ttySA0) (Tue Nov 18 00:10:10 2014): Nov 19 22:50:44 The system is going down for system halt NOW! Nov 19 22:50:44 root@collie:~# sd 0:0:0:0: [sda] Stopping disk Nov 19 22:50:44 reboot: Power down Nov 19 22:51:09 indeed it is sleeping Nov 19 22:51:16 no charger led even Nov 19 22:51:34 on/off wakes it Nov 19 22:52:05 ...it doesn't kexec :// Nov 19 22:52:34 so strange... Nov 19 22:59:01 boots again after reflashing..for the 3rd time Nov 19 23:02:13 bye Nov 19 23:02:14 gn **** ENDING LOGGING AT Thu Nov 20 02:59:58 2014