**** BEGIN LOGGING AT Fri Nov 13 02:59:59 2015
Nov 13 03:17:35 <lukesaber> ums? im not sure what that is
Nov 13 03:17:45 <zmatt> usb mass storage mode
Nov 13 03:18:03 <lukesaber> aah. yeah it wont go into that
Nov 13 03:18:04 <zmatt> it's an u-boot command that makes the BBB appear like a USB stick
Nov 13 03:18:12 <lukesaber> im pretty sure that since my uuid in the fstab is wrong
Nov 13 03:18:16 <lukesaber> it cant find the filesystem
Nov 13 03:18:30 <lukesaber> so even though it wants an emergency console, theres nothing there to serve a tty
Nov 13 03:19:02 <lukesaber> is there a hardware io that forces ums on?
Nov 13 03:19:07 <lukesaber> jump a pin high?
Nov 13 03:21:26 <lukesaber> hey.
Nov 13 03:21:30 <zmatt-> it seems my connection at home dropped
Nov 13 03:21:39 <lukesaber> it happens :)
Nov 13 03:21:55 <lukesaber> is there a hardware io that i can bring high to force ums on/
Nov 13 03:21:57 <lukesaber> ?
Nov 13 03:21:59 <zmatt-> anyway, as I was saying, ums is great for fixing small problems like an fstab edit
Nov 13 03:22:02 <zmatt-> but
Nov 13 03:22:10 <zmatt-> you're using windows, which doesn't understand ext4 filesystems
Nov 13 03:22:17 <lukesaber> i do have a linux box
Nov 13 03:22:19 <zmatt-> oh
Nov 13 03:22:24 <zmatt-> but either way
Nov 13 03:22:31 <zmatt-> that still requires a working console
Nov 13 03:22:43 <zmatt-> and then you may as well fix it using initramfs
Nov 13 03:23:06 <lukesaber> im all for fixing it. how would i use initramfs?
Nov 13 03:23:14 <lukesaber> new sd?
Nov 13 03:23:17 <zmatt-> have you checked yet whether hardware handshake (aka hardware flow control) is disabled
Nov 13 03:23:25 <zmatt-> no sd involved, your serial cable needs to work
Nov 13 03:23:29 <lukesaber> gotcha
Nov 13 03:23:31 <lukesaber> let me check
Nov 13 03:23:51 <zmatt-> and hw flow control is suspect #1 for a non-responsive console
Nov 13 03:24:21 <lukesaber> flow control is set to none
Nov 13 03:24:40 <lukesaber> is that how it should be?
Nov 13 03:24:43 <zmatt-> yes
Nov 13 03:24:57 <zmatt-> but it doesn't respond at all to you typing
Nov 13 03:24:58 <zmatt-> ?
Nov 13 03:25:00 <lukesaber> hah, darn.
Nov 13 03:25:19 <lukesaber> no... but if i have a usb keyboard plugged in i can hit ctrl alt del and it reboots
Nov 13 03:25:38 <lukesaber> it wont respont to hit any keep to abort auto booting
Nov 13 03:25:45 <lukesaber> key*
Nov 13 03:25:48 <zmatt-> usb keyboard won't work in u-boot
Nov 13 03:25:55 <zmatt-> but you're right it does work in initramfs
Nov 13 03:26:11 <lukesaber> im using putty as my terminal client
Nov 13 03:26:28 <lukesaber> anything that might bork up the serial experience with putty?
Nov 13 03:26:34 <zmatt-> no idea
Nov 13 03:26:46 <uavcam> try teraterm
Nov 13 03:26:50 <lukesaber> k
Nov 13 03:26:51 <zmatt-> usually if settings are wrong then you also don't see anything (or just garbage)
Nov 13 03:26:55 <zmatt-> but
Nov 13 03:26:57 <zmatt-> you have output
Nov 13 03:26:59 <zmatt-> and a keyboard
Nov 13 03:28:32 <zmatt-> hmm, wait, what exactly is the output you get from initramfs?  can you put it in pastebin ?
Nov 13 03:29:23 <lukesaber> sure if you can tell me what pastebin is
Nov 13 03:29:33 <lukesaber> sorry. total noob with pastebin
Nov 13 03:29:40 <zmatt-> pastebin.com
Nov 13 03:29:50 <lukesaber> k
Nov 13 03:30:03 <zmatt-> a way to show people chunks of text
Nov 13 03:30:10 <zmatt-> there are many similar services, any will do
Nov 13 03:30:32 <zmatt-> (never copy/paste anything more than one or two lines straight into the irc channel)
Nov 13 03:31:43 <lukesaber> gotcha. slamming is frowned upon :)
Nov 13 03:33:11 <lukesaber> http://pastebin.com/x1ngc7SQ
Nov 13 03:34:13 <zmatt-> hey, my home connection hasn't ping-timeouted...
Nov 13 03:34:20 <zmatt> woo, it's up again
Nov 13 03:34:46 <lukesaber> lol
Nov 13 03:34:47 <lukesaber> http://pastebin.com/x1ngc7SQ
Nov 13 03:34:52 <lukesaber> hope your still here
Nov 13 03:34:58 <zmatt> I never left
Nov 13 03:35:13 <lukesaber> ah now i noticed the -
Nov 13 03:35:34 <zmatt> so it seems it actually manages to find the rootfs with that UUID
Nov 13 03:36:37 <zmatt> wait, is that SD card still inserted?  if so, what happens if you try to boot without it ?
Nov 13 03:36:50 <lukesaber> i shall try
Nov 13 03:37:21 <lukesaber> same uuid
Nov 13 03:37:44 <lukesaber> and this uuid differs from my backup file's uuid
Nov 13 03:37:47 <zmatt> still systemd-fsck saying "rootfs: clean" ?
Nov 13 03:37:49 <zmatt> ok
Nov 13 03:37:55 <lukesaber> yes clean
Nov 13 03:38:12 <zmatt> so it can find your rootfs just fine, I also see no error (other than some unimportant noise)
Nov 13 03:38:31 <zmatt> most likely it's the "ro" in the cmdline that's making it enter emergency mode
Nov 13 03:38:48 <zmatt> you tried pressing control-D ?
Nov 13 03:39:23 <lukesaber> i did
Nov 13 03:39:45 <zmatt> what happens then?
Nov 13 03:39:53 <lukesaber> nothing at all
Nov 13 03:40:07 <lukesaber> ive tried entereing the sysctl default and hitting enter
Nov 13 03:40:25 <lukesaber> i see my terminal client echo my keypresses, but no responses from the bb
Nov 13 03:40:38 <zmatt> hmm
Nov 13 03:40:49 <zmatt> we really need to get into u-boot
Nov 13 03:40:54 <zmatt> what serial cable are you using?
Nov 13 03:41:01 <lukesaber> the one for the raspberry pi
Nov 13 03:41:04 <uavcam> have led's on the serial/usb cable?
Nov 13 03:41:14 <lukesaber> (red lead unconnected)
Nov 13 03:41:25 <zmatt> you sure you connected the leads right?
Nov 13 03:41:25 <lukesaber> nope... no leds
Nov 13 03:41:32 <lukesaber> i am sure, yes
Nov 13 03:41:42 <lukesaber> i can try reversing my tx rx lines if you like
Nov 13 03:41:47 <zmatt> no, since you have output
Nov 13 03:41:55 <lukesaber> my thoughts as well
Nov 13 03:41:57 <zmatt> so that one is connected right
Nov 13 03:42:08 <zmatt> still leaves enough options for wrong connections though
Nov 13 03:42:12 <uavcam> so the ground is connected right too.
Nov 13 03:42:17 <lukesaber> yep
Nov 13 03:42:27 <lukesaber> only one unconnected is the 5v supply
Nov 13 03:42:34 <zmatt> rts/cts/5v are not connected on the BBB side anyway
Nov 13 03:42:49 <lukesaber> thats a good thing. ive let enough smoke out of my last one.
Nov 13 03:42:57 <uavcam> using  pin 1/4/5 ?
Nov 13 03:42:57 <lukesaber> like to keep it inside this one...
Nov 13 03:43:08 <lukesaber> correct uavcam
Nov 13 03:43:47 <uavcam> cables works at other devices?
Nov 13 03:43:47 <zmatt> btw instead of putty you can use 'screen' on linux
Nov 13 03:44:09 <lukesaber> this is true...
Nov 13 03:44:15 <lukesaber> ok, another possibility
Nov 13 03:44:21 <zmatt> e.g.   screen /dev/ttyUSB0 115200,clocal,-crtscts,-ixon,-ixany
Nov 13 03:44:43 <lukesaber> if i connect this setup to my linux box, would the uart flas instructions floating around the internet (possibly) work?
Nov 13 03:45:00 <lukesaber> flash*
Nov 13 03:45:32 <zmatt> no idea, but if you only borked one file then reflashing it seems overkill
Nov 13 03:45:56 <zmatt> if you have a working serial connection, you can use ums
Nov 13 03:46:05 <lukesaber> ah whats 20 minutes? :) id be happy if it just worked
Nov 13 03:46:23 <zmatt> 20 minutes? via uart? I doubt that very much
Nov 13 03:47:54 <lukesaber> hm. seems like im dummied.
Nov 13 03:48:16 <zmatt> but try to get into u-boot by pressing any key during the brief window where it says so
Nov 13 03:48:25 <lukesaber> i will try again
Nov 13 03:48:48 <lukesaber> no effect
Nov 13 03:49:45 <zmatt> you're really, *really* sure you connected it right?  since really the only way this can't work is if your cable is broken or you plugged it in wrong
Nov 13 03:50:11 <lukesaber> no effect on the any key
Nov 13 03:50:24 <lukesaber> tried reversing the tx rx lines, hah
Nov 13 03:50:27 <lukesaber> that was some output
Nov 13 03:50:36 <zmatt> (btw, I quickly did the math, the max theoretical transfer rate via uart is 360 KB/s)
Nov 13 03:50:46 <lukesaber> dear god
Nov 13 03:51:16 <zmatt> except your serial cable probably wouldn't support it anyway, if it worked at all
Nov 13 03:51:22 <uavcam> reversing tx an rx --> to same output??
Nov 13 03:51:26 <lukesaber> im still stymied that the board wont detect the flasher on the SD
Nov 13 03:51:36 <lukesaber> uvacam no, many many symbols.
Nov 13 03:51:38 <zmatt> lukesaber: sounds like a misflashed SD
Nov 13 03:52:24 <zmatt> uavcam: pull-up on cable + pull-down on BBB = recipe for randomness
Nov 13 03:52:27 <lukesaber> then im going to go buy a new SD. this one is from radio shack, and id be happier telling myself that at least i ruled out the sd card
Nov 13 03:52:38 <zmatt> lukesaber: have you tried reflashing the card first?
Nov 13 03:52:44 <lukesaber> absolutely
Nov 13 03:52:54 <lukesaber> with multiple images, flasher, and non flasher
Nov 13 03:53:14 <lukesaber> tried dropping the image, and using the flash tool reccomended
Nov 13 03:53:32 <uavcam> yeah. i read same utput not some output :D
Nov 13 03:53:33 <zmatt> BBBlfs is also an option
Nov 13 03:53:44 <lukesaber> lfs?
Nov 13 03:54:23 <zmatt> https://github.com/ungureanuvladvictor/BBBlfs
Nov 13 03:54:43 <lukesaber> linux from scratch
Nov 13 03:54:44 <zmatt> don't use the included scripts though, they're garbage
Nov 13 03:54:46 <zmatt> no
Nov 13 03:55:00 <zmatt> read the description
Nov 13 03:55:32 <zmatt> remember I mentioned usb in the list of boot devices
Nov 13 03:55:50 <lukesaber> you did
Nov 13 03:56:00 <lukesaber> that actually looks like a great option!!
Nov 13 03:56:30 <zmatt> if you connect the BBB via usb to a linux system in that mode, you'll see it appears as a network device (network manager will try to connect to it. it will fail, but it's actually important it tries)
Nov 13 03:57:00 <zmatt> BBBlfs can then boot a tiny linux system on it that makes the eMMC appear as a block device (like an usb stick)
Nov 13 03:57:18 <lukesaber> and then i can check out the goofs i goofed.
Nov 13 03:57:19 <zmatt> same as u-boot's ums, except  1. you don't need console access (though the output is funny to see)
Nov 13 03:57:38 <zmatt> 2. it's fast enough to do complete reflashing, unlike u-boot's ums which is pig slow
Nov 13 03:57:40 * GenTooMan hides from the stick of BBB.
Nov 13 03:57:49 <zmatt> and yes you can
Nov 13 03:57:58 <lukesaber> excellent.
Nov 13 03:58:05 <zmatt> so, only follow the build instructions from BBBlfws
Nov 13 03:58:05 <lukesaber> thanks zmatt!
Nov 13 03:58:16 <zmatt> never run its script
Nov 13 03:58:27 <lukesaber> oh
Nov 13 03:58:36 <zmatt> instead run the tool it compiled called "usb_flasher" (yes, its name is badly chosen)
Nov 13 03:58:42 <lukesaber> another good point lol
Nov 13 03:59:04 <zmatt> that's the core part of BBBlfs which boots the BBB into UMS mode
Nov 13 03:59:10 <zmatt> that's all you need really
Nov 13 03:59:29 <zmatt> the flash script contains wrong assumptions and absolutely zero error-checking
Nov 13 03:59:46 <lukesaber> so i build my dependencies, and run the usb_flasher and hopefully im in
Nov 13 04:00:00 <lukesaber> gotcha. wrong ttys etc
Nov 13 04:00:07 <zmatt> more like wrong partition
Nov 13 04:00:15 <lukesaber> :) sounds familiar.
Nov 13 04:00:36 <zmatt> it assumes the image you want to flash has 2 partitions, one FAT boot partition and one ext4 rootfs
Nov 13 04:00:39 <zmatt> that used to be true
Nov 13 04:00:51 <zmatt> current images no longer have a separate boot partition
Nov 13 04:01:14 <zmatt> but the script will still try to act on partition 2 and it doesn't notice all the failures since, no error checking
Nov 13 04:01:39 <lukesaber> interestingly enough, the sdcard flashers still have two partitions. lol
Nov 13 04:01:44 <zmatt> they have to
Nov 13 04:01:55 <lukesaber> for the usb boot?
Nov 13 04:01:56 <zmatt> the no-FAT boot method only works for eMMC
Nov 13 04:02:01 <zmatt> not for SD cards
Nov 13 04:02:05 <lukesaber> gotcha
Nov 13 04:02:35 <zmatt> but try to fix first, saves you a reflash
Nov 13 04:02:41 <GenTooMan> so use the lo FAT method. ... I hate the fact that the FAT still exists after decades of better options oh well.
Nov 13 04:02:59 <zmatt> GenTooMan: it's gone in current eMMC images
Nov 13 04:03:04 <lukesaber> agree. if its a uuid problem then i can fix that up, otherwise... flash.
Nov 13 04:03:09 <lukesaber> again, thank you very much!
Nov 13 04:03:14 <zmatt> lukesaber: the "ro" doesn't belong in the cmdline either
Nov 13 04:03:15 <GenTooMan> It's not gone in the SD cards though.
Nov 13 04:03:26 <zmatt> GenTooMan: it *can't* .. bootROM requires is
Nov 13 04:03:26 <lukesaber> zmatt ill have to figure out that one as well
Nov 13 04:03:34 <zmatt> *it
Nov 13 04:03:52 <zmatt> GenTooMan: it will not even attempt the no-FAT method if it detects an SD card
Nov 13 04:03:58 <lukesaber> have a good night! im off to bed
Nov 13 04:04:03 <lukesaber> thanks for your help zmatt
Nov 13 04:04:05 <GenTooMan> zmatt Le sigh
Nov 13 04:04:20 <lukesaber> ill let you know how it worked out. :)
Nov 13 04:04:41 <zmatt> GenTooMan: worse still, eMMC has a dedicated boot mode that's reeaaaly easy to use
Nov 13 04:04:48 <zmatt> but bootROM doesn't use that either
Nov 13 04:05:14 <zmatt> (that would be fetched from one of the mmcblk0boot[01] partitions)
Nov 13 04:06:52 <zmatt> the protocol:  pull CMD low, start clocking, when the card is done with internal init it starts sending data blocks from the boot partition until you let CMD go high again to indicate you've had enough (or you hit the end of the boot partition)
Nov 13 04:07:12 <zmatt> how's that for easy
Nov 13 04:08:19 <zmatt> but no, it goes through the motions of normal card initialization and probes four fixed locations (not in the boot partitions) for their custom header
Nov 13 04:11:37 <GenTooMan> Yeah they have some new boot thing mode for NOR flash. Still the biggest available nor flash is 2G
Nov 13 04:12:00 <zmatt> nor flash is a very different story
Nov 13 04:12:06 <zmatt> that supports xip
Nov 13 04:12:11 <zmatt> always has, nothing new about it
Nov 13 04:13:15 <GenTooMan> the most annoying thing is spansion removing the 4K sector mode on their Nor flash. Makes it useless to use it as part of a FAT system.
Nov 13 04:16:41 <GenTooMan> I am not sure why they made the XIP mode. I am also not certain why they went with 24 bit addressing instead of 32bit address modes. That's the same stupid idea that gave us the Y2K bug.
Nov 13 04:19:30 <GenTooMan> Enough whining from me. Sleep time :D
Nov 13 04:20:49 <zmatt> I don't immediately see anything about 24-bit
Nov 13 04:21:09 <zmatt> they say max size is 128 MB... that's bigger than 24-bit addressing
Nov 13 04:23:03 <zmatt> 32-bit addressing is impossible anyhow, max is 28-bit
Nov 13 08:57:49 <HHE> Hi, I want to communicate other board with rs-232. Can I using rx,tx and gnd to other boards RS232
Nov 13 08:57:51 <HHE> _
Nov 13 08:58:35 <_av500_> yes and no
Nov 13 08:58:40 <_av500_> you have to take care about levels
Nov 13 08:58:47 <HHE> levels?
Nov 13 08:58:51 <_av500_> voltage levels
Nov 13 08:59:01 <_av500_> BBB has 0/3.3V on the uarts
Nov 13 08:59:03 <HHE> BBB 3.3V I think
Nov 13 08:59:14 <_av500_> real RS"§" is up to +/-12V
Nov 13 08:59:17 <_av500_> RS232
Nov 13 08:59:35 <_av500_> so you need to know at what voltages the other board operates
Nov 13 09:00:13 <HHE> other board using 3.3V
Nov 13 09:02:49 <HHE> my friend add RS232 port for communication
Nov 13 09:04:01 <HHE> If I buy TTL to RS232. It can be solving my communication problem right ?
Nov 13 09:04:31 <_av500_> if both are using 3.3V TTL you can connect directly
Nov 13 09:04:43 <HHE> can you check link please http://www.robotistan.com/rs232den-ttle-cevirici-modul
Nov 13 09:05:40 <HHE> we did not know can be direct. Than added on board rs232 port
Nov 13 09:06:21 <_av500_> if you have two of these, its fine
Nov 13 09:06:25 <_av500_> and more robust
Nov 13 09:06:29 <_av500_> over longer cables
Nov 13 09:19:26 <HHE> ok thanks @av500
Nov 13 09:29:09 <zmatt> _av500_: direct connection is also asking for trouble due to power domain crossing
Nov 13 09:29:45 <_av500_> "same desk domain"
Nov 13 09:29:50 <_av500_> should work
Nov 13 09:29:56 <zmatt> no it won't
Nov 13 09:30:07 <zmatt> they'd need to share the same power plane
Nov 13 09:30:39 <zmatt> or you need to take care to disable the output drivers before the power down the other side
Nov 13 09:31:01 <zmatt> which is tedious, and you'll forget about it sooner or later
Nov 13 09:32:20 <zmatt> in which case, since serial is high by default, when one device is powered down the other one starts ramming current down the protection diode of its input
Nov 13 09:32:37 <zmatt> which is generally a good way to fry a chip
Nov 13 09:58:00 <KotH> so, they call a level shifter "çevrici" in erdoganistan
Nov 13 09:58:02 <KotH> interesting
Nov 13 09:58:42 <KotH> and expensive too
Nov 13 09:59:00 <KotH> HHE: http://www.aliexpress.com/item/Free-shipping-MAX3232-RS232-to-TTL-Serial-Port-Converter-Module-DB9-Connector-MAX232/32379970719.html?spm=2114.01020208.3.84.kJWyOJ&ws_ab_test=searchweb201556_2_21_79_78_77_92_91_22_80,searchweb201644_5,searchweb201560_9
Nov 13 10:00:06 <Sandhya> zmatt, nerienna here, I tested the script and it doesn't work on my kernel
Nov 13 10:00:09 <KotH> zmatt: you have that problem with RS-232 level converters as well
Nov 13 10:00:36 <Sandhya> maybe I first have to install a package and I don't know which
Nov 13 10:00:41 <KotH> zmatt: it just gets usually ignored becaues there is no way you can parasitically power a whole PC from the data lines over the protection diodes
Nov 13 10:01:12 <Sandhya> Can't locate experimental.pm in @INC (@INC contains: /etc/perl /usr/local/lib/perl/5.14.2 /usr/local/share/perl/5.14.2 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.14 /usr/share/perl/5.14 /usr/local/lib/site_perl .) at ./show-pins-v2.pl line 8.
Nov 13 10:01:45 <zmatt> KotH: you don't have that problem with RS-232
Nov 13 10:02:07 <zmatt> Sandhya: oh, perl is too old...
Nov 13 10:02:20 <zmatt> Sandhya: that's not really related to the kernel at all though
Nov 13 10:02:46 <Sandhya> so I could fix it by installing a newer version?
Nov 13 10:02:55 <KotH> zmatt: you have it. the same as everywhere. It is just that the level shifter sinks all the power flowing into it
Nov 13 10:03:44 <zmatt> KotH: except RS-232 is specifically protected against it
Nov 13 10:03:47 <zmatt> "RS-232 drivers and receivers must be able to withstand indefinite short circuit to ground or to any voltage level up to ±25 volts."
Nov 13 10:04:39 <KotH> withstand short circuit != does not parasitically power
Nov 13 10:05:49 <zmatt> true, but if the chip is designed for it that's usually not a problem
Nov 13 10:06:28 <KotH> you mean the circuit. it's not a problem of the chip per-se
Nov 13 10:06:31 <zmatt> the issue usually isn't that the power rail is being raised slightly (that's dissipated via many routes anyway) but that it's happening through an I/O
Nov 13 10:06:39 <KotH> lol
Nov 13 10:06:47 <KotH> i see, you have never done low power designs :)
Nov 13 10:07:19 <Sandhya> apt tells me pearl is already the newest version
Nov 13 10:07:48 <zmatt> Sandhya: yeah I get it, you have an ancient distro, I'll tweak the script to remove the dependency on 'experimental'
Nov 13 10:08:11 <KotH> it's quite "funny" to try to debug why your circuit misbehaves from time to time, only to realize you forgot to connect power and the 2 GPIO pins you have are powering it
Nov 13 10:08:30 <zmatt> KotH: yeah I've met the phenomenon
Nov 13 10:08:43 <zmatt> however it's not really relevant in this context
Nov 13 10:09:43 <KotH> true that. only chocolate matters
Nov 13 10:11:25 <zmatt> Sandhya: uploaded a new version, try again
Nov 13 10:11:35 <zmatt> (same url)
Nov 13 10:12:43 <zmatt> most likely it'll still fail due to difference in kernel, but who knows, it might accidently work
Nov 13 10:13:57 <Sandhya> have to type it - 2 different PCs and 2 different networks ^^
Nov 13 10:17:02 <Sandhya> and try to remember which ip is my beaglebone...
Nov 13 16:02:56 <bose> Hi
Nov 13 16:03:35 <bose> do I have to boot from external sd , not eMMC , in order to free the eMMC pins on P8 header?
Nov 13 16:03:59 <nerienna> yes
Nov 13 16:04:27 <bose> thanks
Nov 13 16:04:31 <tbr> that's kind of the meaning of not using the eMMC, yes
Nov 13 19:26:44 <aeeee> noob!
Nov 14 01:42:00 * GenTooMan wakes zmatt by making a nonwitty remark.
Nov 14 02:52:55 <lukesaber> zmatt you there?
**** ENDING LOGGING AT Sat Nov 14 02:59:59 2015