Camelia, the Perl 6 bug

IRC log for #motoactv, 2012-04-03

| Channels | #motoactv index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:19 Bugaboo75 joined #motoactv
00:44 Quietlyawesome94 joined #motoactv
00:59 Quietlyawesome94 Hi
01:00 Bugaboo75 howdy
01:00 Bugaboo75 kr1t1c4l, are you on?  I think I was chatting with you last night about backing up the 16 gb boot image.
01:01 Quietlyawesome94 Someone want to try an app out for me?
01:01 CEnnis91 Bugaboo75: hey
01:02 CEnnis91 sorry i wasntt around last night, but i saw you were able to get the boot.img extracted correct?
01:02 Quietlyawesome94 My mom is really addicted to the "story" games (Farm story). She would go nuts if she knew she could gift her main account from a watch.
01:02 Bugaboo75 I would, except need to root first.  Need to back up stock image (was having problems with adb push).
01:02 Bugaboo75 hey CEnnis
01:02 Bugaboo75 yup, got the boot.img
01:02 Bugaboo75 think kr1t1c4l has the image
01:03 CEnnis91 i downloaded it as well
01:03 Bugaboo75 OK, good
01:03 CEnnis91 only thing different from the 8gb at this point is the kernel
01:03 CEnnis91 which is good new
01:03 Bugaboo75 just was having issues with adb push of boot image
01:03 CEnnis91 +s
01:03 Bugaboo75 that's good news
01:04 Bugaboo75 sorry, not boot image, but boot sector(?)
01:04 Bugaboo75 you can see I'm a newb. ;)
01:05 Quietlyawesome94 No chanserv guys? not good if everyone leaves :P
01:08 Bugaboo75 guess folks are busy or not active right now
01:09 CEnnis91 lol sorry about that
01:09 CEnnis91 well first
01:09 Bugaboo75 no prob
01:09 CEnnis91 oh you need to push the modified boot.img
01:09 CEnnis91 well
01:09 CEnnis91 what about getting the system.img first
01:09 Bugaboo75 tell me what I need to do
01:10 Bugaboo75 cr1t1c4l wanted to back up my files before I had any corrupted files
01:11 CEnnis91 yes, otherwise this would be all for nothing
01:11 CEnnis91 [mbm]: are you around now?
01:11 Bugaboo75 righto
01:13 [mbm] .
01:18 Bugaboo75 hey all, I can only stay on for about 40 minutes tonight
01:21 kr1t1c4l Bugaboo75: er that was [mbm] you were talking to
01:22 Bugaboo75 oh, okay
01:24 CEnnis91 Bugaboo75: if you're available i'd prefer if [mbm] would do this not me
01:24 CEnnis91 [mbm]: you had issues with permissions last night i saw?
01:24 CEnnis91 during the pull
01:24 Bugaboo75 sure thing CEnnis
01:25 kr1t1c4l CEnnis91: we think it was Windows 7 weirdness
01:26 [mbm] CEnnis91: yeah, adb acts weird on windows
01:26 CEnnis91 [mbm]: do you still have time for him to pull system.img or no?
01:26 CEnnis91 to guide him through it*
01:27 [mbm] trying to think of the least painful way to do it
01:27 Bugaboo75 LOL
01:27 CEnnis91 haha okay
01:28 [mbm] suppose we could just toss it on the sdcard same as how we grabbed boot.img
01:29 [mbm] http://dl.opticaldelusion.org/boot-5.6.0-root.img
01:30 Bugaboo75 ok, got the DL
01:30 [mbm] there's the rooted boot image; since we already have a backup of the stock boot.img it's safe to just overwrite it
01:30 [mbm] fastboot flash boot boot-5.6.0-root.img
01:30 [mbm] fastboot reboot
01:31 [mbm] should boot up looking perfectly normal but connecting usb gives you an adb shell
01:32 Bugaboo75 okay, rebooting now
01:33 Bugaboo75 ok, rebooted into motoactv
01:33 Bugaboo75 what next?
01:33 [mbm] adb shell
01:34 Bugaboo75 okay, typed in
01:34 [mbm] cp /dev/block/system /sdcard/system.img
01:35 Bugaboo75 ok, done
01:35 [mbm] stop adbd
01:36 Bugaboo75 done
01:36 [mbm] now copy the file off the drive
01:36 Bugaboo75 okay, do you want me to upload to dropbox again?
01:36 [mbm] yep
01:37 Bugaboo75 k, uploading now, will send you link in a jiffy
01:38 Bugaboo75 well, more like 6 minutes from now - slow upload
01:39 Bugaboo75 btw, DL'ed the latest motoactv firmware (included golf I believe) earlier this evening
01:39 Bugaboo75 not sure if this would be useful for you or not
01:41 [mbm] ah, so the system.img is newer than the boot.img
01:44 CEnnis91 [mbm]: is that gonna be an issue?
01:44 CEnnis91 er nvm
01:44 CEnnis91 well just patch
01:47 Bugaboo75 joined #motoactv
01:47 Bugaboo75 sorry, dunno what happened with IRC, but it wasn't letting me type anything
01:47 Bugaboo75 http://dl.dropbox.com/u/6642209/system.img
01:48 [mbm] CEnnis91: right
01:49 CEnnis91 155mb?
01:49 CEnnis91 163.8
01:50 Bugaboo75 164 mb
01:51 Bugaboo75 actually, 156
01:51 Bugaboo75 weird, when checking file properties, it seems to have changed size
01:51 Bugaboo75 ??
01:51 CEnnis91 thats windows
01:52 Bugaboo75 yeah, windows seems to suck
01:52 CEnnis91 [mbm]: did you get a chance to analyze the kernels further?
01:52 CEnnis91 also
01:52 kr1t1c4l Windows is just a bit variable
01:52 CEnnis91 Bugaboo75: what version were you on before and after you updated
01:53 CEnnis91 and
01:53 CEnnis91 [mbm]: can you upload me your 5.6.12 system.img?
01:54 CEnnis91 ill make this version the base for my dual support of the devices if they allow it
01:54 Bugaboo75 sorry CEnnis, dunno what version it was.  I think it was 5.6.?.  I think the new firmware is 6.6 or something
01:54 CEnnis91 they = if they aren't that different
01:54 CEnnis91 Bugaboo75: well whats you're current version
01:55 CEnnis91 well
01:55 CEnnis91 nvm
01:55 CEnnis91 ican figure that out lol
01:55 Bugaboo75 yeah, you can probably figure out faster than me!
01:56 CEnnis91 ro.build.version.full=Blur_Version​.5.6.13.Motorola-F100.Retail.en.US
01:56 CEnnis91 okay and you were definitely on a 5.6.x?
01:56 Bugaboo75 yes, definitely
01:56 Bugaboo75 It was the latest version before 6.6
01:57 CEnnis91 well there was a "hidden" release
01:57 CEnnis91 thats why i am concerned
01:57 Bugaboo75 ??
01:57 CEnnis91 lemme find the release page
01:57 Bugaboo75 I know as of yesterday, I had the latest firmware
01:57 Bugaboo75 today, when I checked motocast, it detected 6.6 and I DL'ed
01:57 [mbm] CEnnis91: should be able to match the bootimg by sha1
01:58 CEnnis91 oh that works
01:59 Bugaboo75 gotta run for now.  Will you guys be on again tomorrow?  If yes, what time?
01:59 CEnnis91 Bugaboo75: im usually around all day
01:59 CEnnis91 here and there
01:59 CEnnis91 usually later around this time thou
02:00 Bugaboo75 okay, I am at work during the day.  After my daughter goes to bed (~7ish), I can jump online.
02:00 Bugaboo75 Will be back on tomorrow night ~8:00.
02:00 [mbm] hmm, patch between 5.6.10 and 5.6.12 is larger than I thought it'd be
02:02 CEnnis91 [mbm]: sha1's arent matching
02:03 CEnnis91 at least in the 2 i checked
02:03 CEnnis91 is this what yours is? 107b0082ad335807d87bfa2fa95fa9326012b7b2
02:04 christefano joined #motoactv
02:10 CEnnis91 yep now im confused
02:23 [mbm] that's because there's more data there than needed
02:23 [mbm] look at the applypatch line
02:23 [mbm] should be a size just before the sha1s
02:24 CEnnis91 yea but none of them match what i posted
02:25 [mbm] the boot.img is probably 5.6.1
02:25 CEnnis91 thats what im guessing
02:25 [mbm] hat's the applypatch line from the 5.6.1-5.6.x ?
02:26 [mbm] *whats
02:26 CEnnis91 apply_patch("MTD:boot:2959360:e4df6cd302b​1a478818d431278df843d2d40dce2:2959360:3d5​4e32a6b09cb91921e97ce5a9d7ef13b1d581e",
02:26 CEnnis91 "-", 3d54e32a6b09cb91921e97ce5a9d7ef13b1d581e, 2959360,
02:26 CEnnis91 e4df6cd302b1a478818d431278df843d2d40dce2, package_extract_file("patch/boot.img.p"));
02:26 CEnnis91 like i said, its not there =(
02:26 [mbm] ok, truncate the boot.img to exactly 2959360 bytes
02:26 CEnnis91 5.6.1-5.6.13
02:27 CEnnis91 one sec
02:27 CEnnis91 heh
02:27 CEnnis91 i hate you
02:28 [mbm] :D
02:28 CEnnis91 e4df6cd302b1a478818d431278df843d2d40dce2
02:28 CEnnis91 so explain why i needed to do that
02:28 CEnnis91 is that a file pull fail?
02:28 CEnnis91 (i know what truncating does)
02:28 [mbm] partition is larger than the amount of useful data in it
02:29 [mbm] the weird MTD: thing above is only hecking the first 2959360 bytes
02:29 kr1t1c4l is that all the FFs at the end?
02:29 CEnnis91 okay
02:29 CEnnis91 yep kr1t1c4l
02:29 [mbm] kr1t1c4l: basically
02:29 CEnnis91 but why wouldnt the pull have known that
02:29 CEnnis91 or you just catted the whole partitons?
02:29 CEnnis91 partition*
02:30 [mbm] pull pretty much just cats the whole partition
02:30 CEnnis91 also, [mbm] in a return to stock, does it matter which i put in there? the full/trunc-ed?
02:31 CEnnis91 (ignoring that a truncated file makes more sense size wise)
02:31 [mbm] kr1t1c4l: yeah, fun thing about flash chips, you erase them by zapping everything to 1's and you get 0xFF and then you selectively program in the 0's until you have valid data .. as a result, you usually end up with FF's as padding at the end
02:31 [mbm] CEnnis91: just put the truncated version there, nothing useful after that
02:31 CEnnis91 okay
02:32 CEnnis91 why would flash chips 1 everything...whats the difference from zeroing instead?
02:32 [mbm] also fastboot will occasionally yell because of some stupid off by one error in moto's size check
02:32 [mbm] meaning that if you dd/cat a partition and then try to use fastboot to write it back sometimes fastboot will tell you the dd is too big
02:33 [mbm] nuke the padding and it's fine
02:33 CEnnis91 hmmm okay
02:33 CEnnis91 and now next step, unless you've already done it, [mbm] ive seen you can (at least partially) unpack zImage
02:33 CEnnis91 well and i need the 5.6.12 system.img too
02:33 [mbm] you can't arbitrarily set a bit in a flash chip to 0/1
02:34 [mbm] you have to do this cycle of erasing a whole page at a time to 0xff
02:34 [mbm] and then writing the 0s
02:34 [mbm] once something is a 0 it can't be a 1 again until you zap the entire page back to 0xff
02:35 CEnnis91 weird
02:35 [mbm] pages are at least 4k
02:36 [mbm] which means that if I want to change something in that 4k I need to copy it to ram, edit the ram copy, erase, write
02:37 CEnnis91 ew
02:37 [mbm] on an SSD you'll have something called a TRIM command
02:37 CEnnis91 thats hardly efficient
02:38 [mbm] the TRIM command is basically "I don't give a damn what was stored here before, so you can skip the read/modify/write step"
02:38 [mbm] and it just goes straight to write
02:38 CEnnis91 and im guessing thats what makes them so much faster
02:39 [mbm] yep
02:39 CEnnis91 now for the zImage, is it actually unpackable?
02:39 [mbm] the read/modify/write only kicks in when you're doing a partial update
02:39 [mbm] if you're overwriting the entire page then it's just a direct write
02:39 [mbm] well, erase/write
02:40 [mbm] zImage .. sure
02:40 christefano joined #motoactv
02:40 [mbm] hexedit and look for 1f8b08
02:40 [mbm] that's the start of a gzip
02:40 CEnnis91 [mbm]: and likely thats the only way to compare the 2 zImages to see if they are the same?
02:40 CEnnis91 or do you have a better way
02:41 [mbm] usually goes something like this: dd if=zImage bs=$((0x....)) skip=1 | gzip -dc > uncompressed
02:41 [mbm] there's a nice hexeditor called vbindiff
02:41 CEnnis91 ive been using bless hex editor, not really impressed but havent used it enough to be bothered
02:42 CEnnis91 oh
02:42 CEnnis91 i misunderstood that
02:42 CEnnis91 [mbm]: but i already hexdumped them and they were different
02:42 CEnnis91 (slightly for what i saw)
02:43 [mbm] right, but were you looking at the compressed or uncompressed kernel?
02:43 CEnnis91 compressed
02:44 [mbm] take zImage, uncompress it into a vmlinux and then diff it
02:44 [mbm] you'll still see diffs, but this time they'll be hex representing asm instructions
02:45 [mbm] vs hex representing compressed data
02:45 CEnnis91 but are diffs in the uncompressed form bad?
02:46 CEnnis91 well, meaning they are incompatible?
02:46 [mbm] well, you'll get a certain amount of change just recompiling the same damn kernel multiple times and looking at the resulting images
02:47 [mbm] important bit is that when functions are added or removed it shifts the whole binary
02:48 [mbm] so it won't be just one or two bytes mismatched, but the whole file mismatched from that point on
02:49 CEnnis91 okay, so random bytes mismatched are okay
02:49 CEnnis91 clumps of bytes are not
02:51 [mbm] if the files differ and never converge again you're pretty much screwed ;)
02:52 CEnnis91 [mbm]: so the uncompressed zImage is known as a vmlinux?
02:52 CEnnis91 or no
02:53 [mbm] yep
02:53 [mbm] well, vmlinux normally has an elf header
02:53 CEnnis91 well the offset is the same
02:53 CEnnis91 thats a good sign
02:53 CEnnis91 yea this does now
02:53 CEnnis91 not*
02:53 [mbm] right, when you unpack a zImage you kind of lose that stuff
02:54 CEnnis91 is there a way to "add it back"
02:55 CEnnis91 because i know ive seen vmlinux's when messing with liveusb
02:56 kr1t1c4l [mbm]: interesting explanation about the flash chip - thanks
02:56 [mbm] there's probably a fancy objcopy command that would do it for you, but not worth the effort
02:57 [mbm] so much of of the vmlinux information simply gets tossed away as it gets converted from one format to another
02:58 [mbm] from an elf file perspective you have several different sections and instructions on where those sections are in the file and where to load them into ram
02:58 CEnnis91 hmmm okay
02:59 [mbm] but when you're dealing with bootloaders, often it's simply "copy this whole damn thing into ram"
02:59 CEnnis91 i was looking more from a linux "hacking" standpoint at that moment
02:59 [mbm] so you strip the elf headers and just arrange everything exatly like it'd be in ram
02:59 CEnnis91 also, so far so good on the hex level
02:59 CEnnis91 initial diff said they were different thou
03:00 CEnnis91 alright this is taking too long
03:00 [mbm] lol
03:00 CEnnis91 time to just hexdump
03:01 [mbm] the kernel has things like the cpio archive for the initrd burried in it, that's dynamically generated and timestamped
03:01 CEnnis91 yep
03:01 CEnnis91 looks like time stamps
03:01 [mbm] you'll also have the kernel version string timestamp
03:02 CEnnis91 theres a few lines i want you to see thou
03:02 CEnnis91 that might be it
03:02 CEnnis91 http://pastebin.com/958TYFiB
03:02 [mbm] then as things shift around every so slightly the pointers to various functions shift
03:02 CEnnis91 specifically the top lines
03:03 CEnnis91 im guessing thats the version string?
03:03 CEnnis91 but thats the entire diff
03:03 [mbm] that's probably the cpio archive
03:04 [mbm] if you do a strings you'll see something like this -
03:04 [mbm] 070701000002D2000041ED0000000000000000​000000024F73F3100000000000000003000000​0100000000000000000000000500000000/dev
03:04 [mbm] ^ note that's all ascii, not hex
03:04 CEnnis91 but
03:04 CEnnis91 is that bad?
03:04 CEnnis91 or is that compatible
03:04 [mbm] and the 0707 marks it as a cpio archive
03:05 [mbm] and the remaining ascii-hex there is the file permissions and timestamp
03:05 [mbm] obviously the one I pasted above is the permissions for /dev
03:06 [mbm] drwxr-xr-x   2 root     root            0 Mar 28 22:28 /dev
03:08 [mbm] drwxr-xr-x   2 root     root            0 Mar 29 01:01 /dev
03:08 CEnnis91 070701000002D1000041ED0000000000000000​000000024F58421E0000000000000003000000​0100000000000000000000000500000000/dev
03:08 CEnnis91 070701000002D2000021800000000000000000000​000014F58421E0000000000000003000000010000​0005000000010000000D00000000/dev/console
03:08 CEnnis91 070701000002D3000041C000000000000000000​00000024F58421E000000000000000300000001​00000000000000000000000600000000/root
03:08 CEnnis91 0707010000000000000000000000000000000000​0000010000000000000000000000000000000000​000000000000000000000B00000000TRAILER!!!
03:08 CEnnis91 root=/dev/nfs nfsroot=192.168.0.1:/home/user/buildroot ip=192.168.0.2:192.168.0.1:192.16​8.0.1:255.255.255.0:tgt:eth0:off rw console=ttyS2,115200n8
03:09 CEnnis91 yep, and theres TRAILER!!! again
03:09 [mbm] right, there you have a cpio archive and a commandline
03:09 CEnnis91 so, [mbm] does this mean compatible or doesnt it?
03:09 [mbm] the TRAILER!! marks the end of the cpio
03:09 [mbm] well, as you can see from the two copies of /dev I posted above, the only difference is the timestamp
03:10 [mbm] so it really doesn't matter
03:10 CEnnis91 okay, thought so, i just really wanted you to say it =P
03:10 [mbm] motorola actually uses the exact same kernel for several devices
03:10 CEnnis91 really?
03:11 [mbm] when configuring the kernel it's called a 'mapphone'
03:11 CEnnis91 i mean i knew stuff could be taken, and used elsewhere, but i didnt think they just moved kernels across devices
03:11 CEnnis91 yea
03:11 [mbm] where map stands for motorola android platform
03:11 CEnnis91 but do they really copy paste? or use the same configs?
03:12 [mbm] well, the way it works is that everything board/device specific is stored somewhere else
03:12 [mbm] and the kernels literally are identical
03:12 [mbm] that "somewhere else" is devtree
03:13 [mbm] if you get a shell on a motorola device you can see the decoded devtree in /proc
03:13 CEnnis91 including or excludes sholes?
03:14 CEnnis91 excluding*
03:14 [mbm] droid1 (sholes) wasn't a mapphone
03:14 CEnnis91 thought so =|
03:14 [mbm] so I don't think you'd find a proper devtree
03:14 CEnnis91 what was the first?
03:14 [mbm] milestone was
03:14 CEnnis91 milestone?
03:14 CEnnis91 yea
03:14 CEnnis91 okay, so now we dont have to worry about separate boot.imgs
03:15 [mbm] for hacking purposes no
03:15 CEnnis91 what do you mean?
03:15 CEnnis91 oh
03:15 CEnnis91 stock images yes?
03:15 [mbm] for some stupid reason moto is too stupid to run the same firmware on both devices and hardcodes the sha1s into the ota patches
03:16 CEnnis91 yea
03:16 CEnnis91 and now what about system
03:16 CEnnis91 for hacking purposes do we just assume they are the same?
03:16 [mbm] same there, no actual changes apart from being recompiled
03:16 [mbm] which changes the sha1
03:16 CEnnis91 fine with me
03:17 CEnnis91 you still havent given me the 6.6 system.img, i guess ill just pull it myself whenever i get teh device
03:19 [mbm] it's actually 5.6.12
03:19 CEnnis91 well yea
03:19 [mbm] and you have my 5.6.0 system image
03:19 CEnnis91 but moto calls it 6.6
03:19 CEnnis91 yea ill just do it myself
03:19 [mbm] and you probably have all the patch files
03:20 [mbm] what I'd like to know is why they pulled 5.6.10 from the weekend
03:20 CEnnis91 side quest: is it possible to repack the zImage?
03:20 bigbob joined #motoactv
03:20 [mbm] yes, but why?
03:20 CEnnis91 the ota delta was fairly small i thought, prob just bugfix
03:20 CEnnis91 [mbm]: knowledge =)
03:21 CEnnis91 im just concerned about the header that was removed when i extracted the gzip
03:21 [mbm] compile a kernel and look at the last steps that go into building a zImage
03:21 [mbm] wasn't a header
03:21 [mbm] it as a gunzip routine ;)
03:23 CEnnis91 that whole thing?
03:23 [mbm] if you want the sources for that stuff look in arch/*/boot where * is whatever platform you're building
03:23 CEnnis91 that's awfully big
03:24 CEnnis91 so how is that attached then?
03:24 CEnnis91 if i was to do it manually?
03:24 [mbm] read the arch/arm/boot/Makefile, or just recompile a kernel and watch the output from that
03:25 [mbm] seems the uncompressed zImage is just called Image ;)
03:36 CEnnis91 [mbm]: i see that, but im not seeing how they are combined
03:37 [mbm] it'll be an objcopy command
03:38 [mbm] and probably burried in too many levels of Make-foo to easily spot
03:38 [mbm] which is why I said to just compile the damn thing
03:39 CEnnis91 heh, on my computer itll take longer to compile then to read through all the makefiles
03:40 CEnnis91 [mbm]: ./boot/.zImage.cmd:cmd_arch/arm/boot/zImage := ../../gingerbread/prebuilt/linux-x86/tool​chain/arm-eabi-4.4.0/bin/arm-eabi-objcopy -O binary -R .note -R .note.gnu.build-id -R .comment -S  arch/arm/boot/compressed/vmlinux arch/arm/boot/zImage
03:40 CEnnis91 that looks about right
03:40 CEnnis91 (now to find out exactly what objcopy does...)
03:41 [mbm] objcopy is more or less a cut and paste different sections from elf files
03:42 [mbm] output binary, remove .note* .comment, strip all symbols ...
03:42 CEnnis91 yep thats what i just read
03:44 [mbm] .comment is where gcc likes to spam it's version number
03:45 [mbm] the -O binary is doing all the hard work of unpacking the elf file into the order it'd appear in ram
03:46 CEnnis91 why isnt it just packed that way in the first place?
03:50 [mbm] you mean why do we have an elf format?
04:21 CEnnis91 i suppose that would be the question
04:42 cubecloudos joined #motoactv
04:43 cubecloudos hey is anyone still here I have a short question
04:44 cubecloudos Can we insert custom shutdown animations and sounds and maybe a boot up sound?
04:44 cubecloudos If so where?
04:45 CEnnis91 cubecloudos: definitely a boot sound
04:46 CEnnis91 shutdown animation, not easily
04:46 cubecloudos so where do I push the boot sound?
04:47 CEnnis91 you need to set it up, its not exactly a 2 step thing
04:47 cubecloudos ok
04:47 CEnnis91 i was going to take care of that in a custom rom
04:47 cubecloudos -.- can I get it now? pleeeeaaaseeee
04:49 CEnnis91 nah, i can give you a hint, but thats all for now
04:49 cubecloudos ok
04:50 CEnnis91 =P i actually added boot sounds back in cm6, and i later found out my script started circling around the community
04:51 [mbm] sound? no speaker.
04:51 CEnnis91 cubecloudos: add init.d support in the boot.img, add a script to run a music file via stagefright, include the file in /system/media/audio/ui, make sure it runs at the same time
04:51 CEnnis91 s/same/right/
04:51 CEnnis91 [mbm]: lololol seriously?1
04:52 CEnnis91 you can only listen to music over headphones?
04:52 CEnnis91 and this cant be an alarm clock?
04:52 [mbm] right, no vibrator either, afaik
04:52 ds2 does it work on A2DP headphones?
04:52 CEnnis91 an android watch without one of the core functionality of a digital watch?!
04:52 CEnnis91 gg moto
04:53 [mbm] yep, although it needs to be booted before bluetooth ill connect
04:53 CEnnis91 ds2: even so, a2dp wont be up on system boot
04:53 CEnnis91 so still no boot sound
04:53 cubecloudos ok got it!!! Thanks CEnnis91!
04:53 ds2 ah right
04:53 CEnnis91 [mbm]: that makes me mad
04:54 cubecloudos wait...
04:54 cubecloudos there are no speakers...
04:54 cubecloudos wow... I'm dumb
04:55 CEnnis91 *sigh*
04:55 CEnnis91 seriously that pissed me off
04:55 CEnnis91 it's less of a watch now
04:56 CEnnis91 [mbm]: im putting it in there anywaty
04:56 [mbm] CEnnis91: also a bit to bulky to be a watch
04:56 ds2 donno about spkrs... vibrate would have been nice
04:56 CEnnis91 [mbm]: if it functioned as an actual watch, i may not mind
04:57 CEnnis91 even fitness wise
04:57 CEnnis91 run for 20 minutes, alert me when its done
04:57 CEnnis91 but nope!
04:57 CEnnis91 sorry we didnt include that
04:57 ds2 for the work out stuff, bluetooth headphones
04:57 [mbm] it's a pedometer with a built in mp3 player
04:58 CEnnis91 does this have a headphone jack?
04:58 [mbm] yep
04:58 CEnnis91 well at least it has that
04:58 * CEnnis91 checks if the ipod nano has a speaker
04:59 CEnnis91 guess not
04:59 [mbm] suppose you could wire up a piezo and superglue it to the back
04:59 CEnnis91 well i cant blame them then
05:00 CEnnis91 [mbm]: ds2: theres supposedly a bunch of sensors that can work with this thing right?
05:00 [mbm] yeah, anything ANT+
05:00 [mbm] which I guess is some sort of standard in the fitness world
05:01 ds2 BLE too
05:01 ds2 and normal bluetooth
05:01 CEnnis91 yea looks to be a standard
05:01 ds2 BLE should replace ANT+
05:01 CEnnis91 ds2: whats BLE?
05:02 CEnnis91 initial search wasnt as helpful as "ANT+"
05:02 * [mbm] noticed that having bluetooth paired with the phone 24x7 killed the battery of both the phone and the motoactv even though neither was sending anything :/
05:02 ds2 Bluetooth 4.0 stuff... Bluetooth Low Energy
05:02 CEnnis91 oh
05:02 CEnnis91 yea i knew that then
05:02 CEnnis91 [mbm]: how bad?
05:03 [mbm] been a while since I ran the test
05:03 CEnnis91 how much battery life do you get in a normal day?
05:03 CEnnis91 motoactv not the phone
05:03 [mbm] but I can tell you that the motoactv while idle (screen off, wifi off, bluetooth off) will last at least 2.5 days
05:03 CEnnis91 not bad
05:04 [mbm] but when I was doing the tethering thing I had to recharge it by the end of the day
05:04 CEnnis91 have you messed with the penetration stuff yet?
05:04 [mbm] not yet, too many other things on my todo list
05:04 CEnnis91 i'd like to try that at least once when i have the device =)
05:04 [mbm] yeah
05:05 ds2 I get about 3 days of use out of it
05:05 CEnnis91 i suppose i could ask bdaman80, not sure if he'll know how thou
05:05 ds2 regular time checking, about an hour of walking with GPS on
05:05 CEnnis91 [mbm]: ds2: does this have a battery stats screen?
05:05 ds2 precentage gage
05:05 CEnnis91 no
05:06 [mbm] it does, but the menu to get to it seems to have been nuked
05:06 [mbm] you can get it with an adb shell
05:06 CEnnis91 ds2: i meant more like this http://dl.dropbox.com/u/5849​675/galaxy_nexus_battery.jpg
05:06 CEnnis91 [mbm]: yea im still thinking of my options for revealing all the shit they hid
05:07 ds2 oh that.. don't think so
05:08 CEnnis91 [mbm]: you think it would be easy to re-enable them via smali?
05:08 CEnnis91 via was a horrible word choice
05:10 [mbm] probably, but I really hate digging through java
05:10 CEnnis91 i wasnt asking you to do it =P
05:10 CEnnis91 ive dealt with smali a little before
05:11 CEnnis91 and yea, didn't find it fun
05:13 CEnnis91 [mbm]: oh, another random question
05:14 CEnnis91 [mbm]: back when i make the boot sound, i looked at making a power down animation, just for fun, like cubecloudos mentioned, shouldn't running the bootanimation binary give a power down animation?
05:15 [mbm] adb shell am start -a android.intent.action.POWER_USAGE_SUMMARY
05:15 [mbm] well, it'll give you a boot animation ;)
05:16 CEnnis91 0.o
05:16 CEnnis91 fuck all my usb ports are being used
05:17 CEnnis91 [mbm]: is that's what is called on boot?
05:18 CEnnis91 wait
05:18 [mbm] yep
05:18 CEnnis91 i clearly didnt read what i typed
05:18 cubecloudos I used anycut and found the battery gauge seems to me like the screen takes up more battery than the gps and bluetooth combined 0.o
05:19 CEnnis91 [mbm]: that doesnt do any animation
05:19 [mbm] hmm.. does the razr have nfc? seeing they left some nf stuff in the rom whih I don't understand
05:21 [mbm] CEnnis91: look at logcat; think bootanim requires surfaeflinger
05:21 CEnnis91 ummm wow
05:21 CEnnis91 literally "bootanimation" calls it
05:22 CEnnis91 that definitely didnt happen pre-4
05:22 CEnnis91 looks like the OS is still active underneathe it too
05:22 CEnnis91 soooo i now have a screen saver that i can disable and drains my battery =)
05:23 [mbm] heh
05:23 CEnnis91 ^C but yea
05:23 [mbm] this looks like fun - com.motorola.internal.intent.action​.DISPLAY_THERMAL_PROTECTION_WARNING
05:23 CEnnis91 [mbm]: does /system/bin/bootanimation (or bootanim i think on 2.3) call anything?
05:23 CEnnis91 no params
05:23 [mbm] "The display has been dimmed to help prevent burn-in ..."
05:24 [mbm] don't think so; sources are in aosp .. I seem to remember it just sits there in a loop
05:25 CEnnis91 may have to build a custom binary to make a boot/shutdown anim
05:25 CEnnis91 sholdnt be hard
05:25 CEnnis91 meh wtf is it
05:27 CEnnis91 oh so its /system/bin/bootanimation with service name bootanim
05:34 [mbm] yeah, start bootanim
05:36 CEnnis91 oh its in framework
05:36 CEnnis91 yea i knew that, but i want it to take params
05:37 CEnnis91 mAndroidAnimation = false;
05:37 CEnnis91 status_t err = mZip.open("/data/local/bootanimation.zip");
05:37 CEnnis91 if (err != NO_ERROR) {
05:37 CEnnis91 err = mZip.open("/system/media/bootanimation.zip");
05:37 CEnnis91 if (err != NO_ERROR) {
05:37 CEnnis91 mAndroidAnimation = true;
05:37 CEnnis91 }
05:37 CEnnis91 }
05:37 [mbm] yeah, you can override it
05:37 CEnnis91 good job hardcoding that
05:37 CEnnis91 initTexture(&mAndroid[0], mAssets, "images/android-logo-mask.png");
05:37 CEnnis91 initTexture(&mAndroid[1], mAssets, "images/android-logo-shine.png");
05:37 CEnnis91 hmmm
05:38 [mbm] yeah, that's the fallback animation
05:38 CEnnis91 yea, guess its time i need to brush up on a little more of my c
05:38 [mbm] displays the ANDROID logo/text and it just shimmers
05:39 [mbm] that's if it can't find one of the above animation zip files with the png flipbook
05:39 CEnnis91 yea
05:39 CEnnis91 i guess alternatively i can just make it take parameters via env props
05:43 CEnnis91 meh
05:43 CEnnis91 [mbm]: can i accept parameters using this: int main(int argc, char** argv) <--- in original source
05:43 CEnnis91 or do i need to make it a pointer to an array
05:44 [mbm] that's fine
06:04 CEnnis91 7697 cannot locate '_ZN7android21SurfaceComposerClient14g​etDisplayInfoEiPNS_11DisplayInfoE'...
06:04 CEnnis91 CANNOT LINK EXECUTABLE
06:04 CEnnis91 [mbm]: does it need to be built with framework?...
06:04 CEnnis91 considering it uses surfaceflinger
06:05 CEnnis91 granted i tried using the gingerbread version on my nexus...but still
06:07 CEnnis91 meh, yea it builds alot of shit with it
06:07 CEnnis91 this isnt worth it in that case
06:11 christefano joined #motoactv
06:31 [mbm] hmm
06:31 [mbm] devtree lists a vibtrator
06:33 * kr1t1c4l wonders what one can do with a vibtrator
06:46 CEnnis91 alarm
06:47 CEnnis91 alert
06:50 [mbm] hmm doesn't appear to work
06:50 [mbm] least not from android
06:50 CEnnis91 [mbm]: is there a way to turn on the vibrator from command line?
06:50 CEnnis91 ive never found that
06:50 dproldan joined #motoactv
06:51 [mbm] figured out how to do it from assembly once, but that's a long way from the commandline
06:52 CEnnis91 okay, at least i didnt miss it =)
06:52 AndroUser joined #motoactv
06:52 AndroUser morning
06:54 [mbm] hmm, they've just copied this devtree from a phone, it also says there's a speakerphone
06:55 dproldan joined #motoactv
06:55 [mbm] and a proximity sensor?
06:56 CEnnis91 hahahah
06:56 CEnnis91 nfc?
06:57 [mbm] doesn't claim to have that
06:57 [mbm] guessing that someone just cut and pasted code from the milestone to make this thing
07:01 dproldan with some bits of "begonia"
07:02 dproldan begonia=t-mobile cliq 2
07:08 [mbm] looks like they patched something in the framework since the golf build on the weekend
07:08 [mbm] 420K./patch/system/framework/framework.jar.p
07:11 JohnDoe_ joined #motoactv
07:13 JohnDoe_ After installing the Goo.inside.me zip from 110703, clearing Dalvic, still FCs AND market tries to update to Play which of course doesn't work anymore...
07:13 JohnDoe_ Just a heads up...
07:14 JohnDoe_ I am going to droidwall market, but then that defeats the purpose... anyone know how to stop it from trying to update?  Also I have been looking long and hard for some widget to search market too since we cannot search directly
07:15 dproldan my market updated to play just fine..
07:15 JohnDoe_ Really?
07:16 dproldan yep,  the interface is much better on the MA display
07:16 JohnDoe_ Maybe I should redo the whole process then and let it update again, can you search now on Play?
07:16 dproldan nope
07:17 [mbm] lol, yay, vpn support is still there
07:17 JohnDoe_ There is a widget (free) that lets you search market, I have yet to try it.
07:17 dproldan Ianas released his program to do just that
07:17 JohnDoe_ VPN?  Awesome... I haven't tried it, how to access the system settings for VPN?
07:18 JohnDoe_ I haven't seen Ianas' program on the wiki or forum
07:18 dproldan http://forum.xda-developers.com/sh​owthread.php?t=1437013&amp;page=9
07:18 dproldan last posts
07:18 [mbm] adb shell am start -a android.net.vpn.CENTRALSETTINGS
07:20 dproldan bbl
07:24 JohnDoe__ joined #motoactv
07:24 JohnDoe__ "adb shell am start -a android.net.vpn.CENTRALSETTINGS" - Thanks
07:26 [mbm] adb shell am start -n com.motorola.android.test.wsensor/.WSensorTest
07:27 christefano joined #motoactv
07:27 [mbm] think that's for the ANT sensors
07:27 JohnDoe__ MBM: Do you happen to have the command to access the full android system settings instead of the Moto Settings?
07:31 [mbm] there really isn't; moto took the normal menus and messed with them
07:31 [mbm] best you can do is find fragments of things that they've removed the menu item for
07:32 [mbm] eg, Wifi Hotspot - adb shell am start -n com.android.settings/.wifi.WifiApSettings
07:32 [mbm] although since there's no internet connection that one doesn't do anything :/
07:36 JohnDoe__ MBM: Cool, thanks... I am going to try this market thing again.  I found Ianas' apk, so I'll give this another go.
07:38 JohnDoe__ We cannot modify the GoogleServiceFramework to stop FCing?  Maybe a small dpi issue?  This has probably been covered before, if so my appologies
07:41 [mbm] lmao; com.motorola.blur.service.​blur/.CloudPickerActivity
07:46 JohnDoe__ Could Service?  wtf?  Moto really hacked this system, no?
07:46 JohnDoe__ Cloud...
07:50 [mbm] reason I'm laughing is that's the same update service moto uses on their phones to check for updates
07:51 [mbm] and with a few changes you can have it check for updates on their test and qa servers
07:51 [mbm] google "project cheesecake" for all of that
09:51 cmw joined #motoactv
11:07 JohnDoe_ joined #motoactv
11:18 JohnDoe_ Market has updated to Play, using the search tool by Ianas: working well, still FC sometimes but working...
11:24 dproldan joined #motoactv
11:32 JohnDoe_ dprodan: thanks for the help before, Play seems to be OK now.
11:38 dproldan good,  np
11:40 dproldan joined #motoactv
13:20 cmw joined #motoactv
13:30 cmw_ joined #motoactv
13:38 CEnnis91|Mobile joined #motoactv
13:44 bigbob joined #motoactv
13:54 CEnnis91|Mobile joined #motoactv
14:05 bigbob joined #motoactv
14:07 bigbob So once we are rooted and have a custom recovery on the motoactv, are we still able to go back to stock and update to an official update or are we stuck updating through custom roms that are built off the new updates?
14:08 CEnnis91|Mobile Yes you can go back
14:09 CEnnis91|Mobile 5.6.0 is stock
14:10 bigbob Cool, I saw that there is a new update being rolled out.  I just wanted to make sure I would be able to try it if i would like.  Since it claims to be compatible with other android phones with the new update.
14:12 bigbob I get my watch sometime today, so I am sure once I start flahsing stuff I will be bugging the hell out of you guys.
14:23 CEnnis91|Mobile Heh
14:23 CEnnis91|Mobile Once I get the device I will fix everything
14:23 CEnnis91|Mobile I'll make a universal update.zip that'll root any version
14:24 CEnnis91|Mobile Mostly because I'm lazy =)
14:29 bigbob lol...Well I have no complaints, when do you think you will get one.  That's awesome that you have done as much as you have with out it.
14:31 bigbob Hell if you have a paypal account I will donate some funds to help you out.
14:32 CEnnis91|Mobile I'm only going to be borrowing on
14:32 CEnnis91|Mobile One*
14:32 CEnnis91|Mobile I should get it either this week or next
14:33 CEnnis91|Mobile And I should have access to it as long as I continue to develop on it
14:35 bigbob Well if you decide that you want to buy one let me know, I will donate some money to help you out so you don't stop...of if any other dev wants some help getting one I don't mind donating $50 or something in that range
14:36 CEnnis91|Mobile Thanks I appreciate it
14:37 bigbob no prob man
14:52 dproldan joined #motoactv
14:56 dproldan hehe,  there's something going on behind the scenes : http://www.engadget.com/2012/04/0​3/garmin-approach-s3-golf-watch/
15:21 utkanos joined #motoactv
15:49 [mbm] dproldan: doesn't look like that one plays angry birds
15:50 dproldan definitely not
15:53 [mbm] can't imagine why anyone would want to play actual games while wearing a smart watch; seems too physically exhausting
15:54 dproldan I agree,  apart from the occasional distance hint,  I don't see much use
16:15 CEnnis91 [mbm]: maybe solitare or minesweeper
16:16 CEnnis91 simple clicking games
16:16 utkanos someone should do marble madness
16:16 utkanos and you hold your wrist out
16:16 utkanos to balance the ball
16:16 CEnnis91 if they actually scaled correctly, bejeweled perhaps?
16:16 CEnnis91 utkanos: trolol
16:17 [mbm] utkanos: there's a nice android port of enigma; might be fun on the motoactv
16:17 utkanos nice
16:17 utkanos i brought that unit in
16:17 utkanos sending it back to TEG
16:17 utkanos today
16:19 CEnnis91 okay cool
16:19 CEnnis91 i messaged him on xda, hopefully hell send his out today
16:19 CEnnis91 wheres he shipping from
16:19 utkanos im going to pack it a lot better than it came to me :\
16:20 utkanos one piece of bubble wrap in a manilla envelope
16:20 CEnnis91 ...lol
16:20 utkanos I'm surprised it arrived with an intact screen
16:20 CEnnis91 lol, just taught my gf how to do public links in dropbox, god what an ordeal
16:33 [mbm] aw.. can't play enigma on the motoactv - the acceleration sensors aren't mapped to android
16:33 utkanos shock, awe
16:35 CEnnis91 [mbm]: can they be, or does that need to be done in framework?
16:36 utkanos kernel
16:36 utkanos and I didn't have much success changing the USB gadget to android standard from moto
16:36 [mbm] hardware -> kernel -> libhardware -> framework -> android app
16:37 [mbm] iirc, several other moto devices also use an msp430 for sensors
17:00 dproldan is the motoactv boot loader sending the kernel cmd_line?  looks like this happens in a lot of moto phones
17:01 utkanos the kernel command line is part of the image file the bootloader loads
17:03 dproldan yes,  but looks like in some moto phones,  that is hardcoded in the boot loader
17:03 dproldan son ,  no matter what you put in the boardconfig,  it doesn't make a difference
17:03 utkanos ah, perhaps it is
17:04 utkanos but not the case at least on this
17:04 utkanos lol dproldan
17:04 dproldan :-)  -n
17:11 [mbm] dproldan: comandline comes from either the cdt or the boot/recovery image, later is an override
17:11 [mbm] and yeah, gets bounced by the bootloader
17:13 [mbm] for x in $(seq 1 100); do { msp430 tread $x 4 | tr -d '\n' ; echo; }; done
17:14 [mbm] around 97 seems to be a timer
17:15 dproldan that's what makes the msp340 messages appear in dmesg every few seconds?
17:15 dproldan or logcat
17:15 [mbm] hmm? no, that's just a test read command
17:17 [mbm] not seeing anything in logcat
17:17 dproldan liveDataHandler...
17:19 [mbm] not seeing it
17:19 dproldan one sec
17:20 dproldan pastebin.com/h9egnvXh
17:21 dproldan maybe changing the SET_EQUIPMENT_TYPE will make it behave like a normal phone?
17:22 [mbm] think it has more to do with the libhardware aand framework
17:23 [mbm] hmm,seems to be to msp firmwares
17:23 [mbm] mspfirmware.bin mspfirmwarefactory.bin
17:23 [mbm] aka "msp430 boot -f" and "msp430 bootfactory -f"
17:34 CEnnis91 [mbm]: from BugaBoo75: Hey CEnnis, are there additional images that you need or I should be backing up beyond boot & system? I was wondering if I needed to join IRC tonight.
17:34 CEnnis91 we dont need him anymore right
17:35 [mbm] for most purposes boot and system will suffice
17:36 [mbm] there are dozens of other partitions, but probably not stuff that anyone will touch
17:36 CEnnis91 yea, but are they worth to backup at all?
17:36 CEnnis91 er i should probably make him extract preinstall
17:36 CEnnis91 because i plan to use it
17:37 [mbm] use it how?
17:37 CEnnis91 possibly for backup
17:37 CEnnis91 not sure yet, but there is extra space in there
17:37 CEnnis91 and i want to try to give users all of that limited /data space
17:38 CEnnis91 it all depends on what i see when i get this thing
17:38 [mbm] you can tell him to back it up, but explaining how to do it (particularly from windows) is painful
17:38 [mbm] so I'll leave the details of that to you
17:39 CEnnis91 9:33:57 PM <[mbm]> adb shell
17:39 CEnnis91 9:34:23 PM <Bugaboo75> okay, typed in
17:39 CEnnis91 9:34:53 PM <[mbm]> cp /dev/block/system /sdcard/system.img
17:39 CEnnis91 9:35:36 PM <Bugaboo75> ok, done
17:39 CEnnis91 9:35:44 PM <[mbm]> stop adbd
17:39 CEnnis91 9:36:01 PM <Bugaboo75> done
17:39 CEnnis91 9:36:11 PM <[mbm]> now copy the file off the drive
17:39 CEnnis91 sed s/system/preinstall/g?
17:39 CEnnis91 or is it a different block name
17:40 [mbm] ...
17:40 CEnnis91 what?
17:41 [mbm] I just said I wasn't going to provide instructions, so obviously the first thing you do is ask me about the instructions; can't possibly answer that
17:42 CEnnis91 heh
17:42 CEnnis91 fine
17:56 dproldan is it normal/right that the kernel cmdline has a serial number in it?
17:56 dproldan console=/dev/null rw mem=254M@0x80000000 init=/init ip=off brdrev=P1 omap3_die_id androidboot.bootloader=0x0000 mmcparts=mmcblk1:p6(pds),p12(boot),p13(r​ecovery),p14(cdrom),p15(misc),p16(cid),p​17(kpanic),p18(system),p19(cache),p20(pr​einstall),p21(userdata),p22(emstorage) androidboot.bootloader=0313 androidboot.serialno=0A3AA66405035020 androidboot.mode=reboot
17:56 dproldan also,  two bootloader
17:56 dproldan ?
17:57 [mbm] some of that is populated by the bootloader and lbl
17:58 [mbm] if you look about halfway through the cdt you'll find the default commandline and some atags
17:58 [mbm] bootloader will add a few more atags
17:59 [mbm] and then hand off to lbl
17:59 [mbm] lbl does some atag parsing and then hands off to the linux kernel
17:59 dproldan ok
17:59 * [mbm] should really just put up a wiki somewhere with all of this crap ;)
18:00 CEnnis91 motoactv.wikispa....oh nvm
18:00 CEnnis91 nothing useful on there
18:00 CEnnis91 =)
18:00 dproldan it's a wiki,  you create the content...
19:44 bigbob joined #motoactv
20:25 dproldan [mbm]:  I want to check if the system.img I'm building is being mounted correctly and see why this is not booting,  I think I can do that from recovery,  but I can't access recovery without  a working system.   can I use that misc.img you made?
20:25 dproldan fastboot flash misc misc.img?
20:26 dproldan will I be able to get to fast boot again?
20:39 utkanos yes you can get back to fastboot mode dproldan
20:39 utkanos and I never tried his misc, but he said it worked so I trust it
20:40 dproldan how does it get back to fast boot?  the misc.img is still flashed, isn't it'
20:52 lownox joined #motoactv
20:52 dproldan when I reboot from recovery and try to enter fastboot,  won't it just go back into recovery because of the misc commands
20:54 utkanos you'll have to ask mbm I guess
20:55 dproldan k
20:56 [mbm] sup?
20:57 dproldan will your misc.img put the device in a recovery loop?   :-)
20:57 [mbm] yep
20:57 [mbm] well, not a loop
20:57 [mbm] misc gets wiped each bootup
20:58 dproldan oh,  that's good
21:00 dproldan so I should be able to check if my system.img is getting mounted correctly using it,  then after rebooting,  I can get to fast boot normally, yes
21:00 dproldan ?
21:05 dproldan thanks,  I'm trying it
21:31 bigbob joined #motoactv
21:33 bigbob I am getting ready to run the motoACTV Root Tool, but out of the versions to choose from mine seems to be an older one than what root tool says..Its says versions either 4.55.78/4.55.97 or 5.6.0/1.6.0 where mine is acutally 4.53.50...does this make a difference?
21:34 dproldan why don't you try and upgrade the motoactv with the mococast software?
21:35 bigbob Thats what I was thinking but If I upgrade to the newest version will I still be able to root?
21:37 [mbm] you can root any version
21:37 ds2 fastboot is wonderful, isn't it? :D
21:37 [mbm] to be perfectly blunt, the bootloader is unlocked and has absolutely no checks over what software you install, so you can always just send it a rooted image
21:38 bigbob Ok cool, well then I will upgrade first and then run the tool....thanks..
21:38 [mbm] there aren't any exploits involved in rooting
21:39 bigbob Well that makes me feel better, I am just used to my phone being locked down and having to be careful when upgrading so root doesn't get blocked.
21:39 utkanos [mbm]: I wonder if they'll release an update to prevent flashing eventually
21:39 utkanos or if they dont care that much
21:39 [mbm] they probably don't care
21:39 utkanos in the OTAs there are diff bootloaders
21:39 ds2 this really smells like it is a lab project that they decided to sell and see if ther eis money to be made
21:40 ds2 OTH, this is, IMO, better then the other 2 Android watches out there
21:40 [mbm] I think someone at moto played with an ipod nano and decided to make a cleap knockoff using old moto hardware
21:41 utkanos pretty much
21:41 ds2 yep... everything in there looks like it it recycled
21:41 utkanos though cheap is relative
21:41 utkanos :P
21:42 utkanos i find it interesting the board files and other references to mapphone, I wonder if they had intended to add a radio to this
21:42 [mbm] utkanos: well, it's not a top of the line omap4, this is hardware from 2+ years ago
21:42 utkanos yeah
21:42 utkanos its omap 3 isn't it?
21:43 ds2 [mbm]: top of the line omap4 won't last long with the tiny battery
21:43 [mbm] right
21:43 [mbm] utkanos: every android device moto makes is a "mapphone"
21:43 [mbm] it's just how they've named their kernel
21:43 utkanos ah
21:44 [mbm] they use the exact same kernel on all their android devices and just use a devicetree to handle the board specific stuff
21:44 utkanos hmm
21:44 utkanos they use the same defconfig?
21:44 [mbm] yep
21:44 utkanos guess that's smart
21:44 [mbm] there's a /proc/device-tree with the real "config"
21:45 utkanos at least from a centralized location
21:45 [mbm] which is stored in the devtree partition
21:46 [mbm] although it looks like they recycled the devtree; references hardware that doesn't exist like vibrator and loudspeaker
21:46 [mbm] think I even saw a keypad reference in there
21:49 utkanos yup
21:50 utkanos bunch of sholes files :P
21:55 CEnnis91|Mobile joined #motoactv
21:57 CEnnis91|Mobile joined #motoactv
22:15 [mbm] dproldan: fix your /system?
22:31 dproldan what would be the good commas to mount /system in this recovery
22:31 dproldan command
22:36 [mbm] mount /dev/block/system /system
22:37 dproldan invalid argument
22:39 dproldan there is already a /system in recovery,  guess I have to mkdir a new one?
22:43 [mbm] you can mount ontop of the existing system
22:43 dproldan ok,  i'll try with mmcblk0p18?
22:44 [mbm] cat /proc/partitions
22:44 [mbm] I think it as mmcblk1p18
22:44 dproldan yeah
22:45 CEnnis91|Mobile what'd you do
22:45 dproldan nah,  invalid argument
22:45 [mbm] look at dmesg
22:46 [mbm] sounds like system isn't a valid filesystem
22:46 dproldan [  976.829071] yaffs: dev is 187695154 name is "mmcblk1p18"
22:46 dproldan [  976.829223] yaffs: passed flags ""
22:46 dproldan [  976.829467] yaffs: Attempting MTD mount on 179.50, "mmcblk1p18"
22:47 dproldan this is not yaffs...
22:47 [mbm] mount -t ext3 ...
22:47 dproldan yeah,  i tried that before
22:47 [mbm] usually it's smart enough to figure out the filesystem type so I skip that
22:47 dproldan still invalid argument
22:47 [mbm] different dmesg error with -t ext3?
22:48 dproldan no,  same error
22:49 dproldan isn't there a mount_ext3.sh in the stock initramfs
22:49 [mbm] there is, but if you can't manually mount it with the above commands there's something weird going on
22:50 dproldan back to the drawing board for me then
22:50 [mbm] there's a hexdump on that recovery if you want to poke around /dev/block/system
22:53 dproldan so,  [  976.829467] yaffs: Attempting MTD mount on 179.50, "mmcblk1p18"
22:54 dproldan is it trying to mount MTD,  isn't this mmc?
22:57 [mbm] it is
22:58 [mbm] if you don't specify a filesystem it will try various filesystems from /proc/filesystems
22:58 [mbm] which can result in strange errors like the above
22:58 dproldan ok
22:59 dproldan the mount command did work with the userdata partition,  looks like it's my system.img that's bad
22:59 [mbm] yep, what'd you do to cause it?
23:00 [mbm] oh, is this your cm port?
23:00 dproldan heh,  that's the problem,  I didn't do anything to make it good
23:00 dproldan yeah
23:00 [mbm] you didn't build system as an ext3 image
23:00 dproldan yeah,  looks like
23:01 [mbm] probably built it as yaffs2 by mistake
23:01 [mbm] and that doesn't work on anything other than an mtd
23:01 dproldan now I have to find where in that environment you tell it to build a ext3
23:01 [mbm] dunno, but you'll override it in the board file
23:01 dproldan any other moto device using ext3 for system,  that I can take a look at
23:03 dproldan_ joined #motoactv
23:04 dproldan joined #motoactv
23:04 utkanos joined #motoactv
23:05 dproldan_ joined #motoactv
23:07 dproldan_ joined #motoactv
23:08 dproldan joined #motoactv
23:10 dproldan really nice the trick to get into recovery,  btw
23:11 [mbm] :)
23:13 dproldan the adb shell works much better,  with tab autocompletion, etc ?
23:15 dproldan brb,  the ubuntu vm is acting up,  rebooting
23:19 dproldan joined #motoactv
23:31 [mbm] I used busybox in recovery, colors and tab completion ;)
23:32 [mbm] actually the shell has history too, so it doesn't need to rely on adb for shell history
23:32 [mbm] normally adb intercepts up/down and retypes commands
23:32 [mbm] kind of annoying if you're trying to use vi
23:33 ds2 is it adb or adbd that does that?
23:34 [mbm] adb (client)
23:34 [mbm] there's a define in the source for shell history
23:35 [mbm] I always recompile adb without that crap
23:35 ds2 I see... the reason I ask is I get different behaviors with the same adb client
23:35 [mbm] different how?
23:36 ds2 up/down arrows work in some devices and not on others
23:37 [mbm] honeycomb/ics includes a newer shell that works better than the old toolbox version
23:37 [mbm] I still tend to use busybox
23:37 ds2 ah...that would explain it
23:38 ds2 the GB devices are the ones that it doesn't work on
23:40 [mbm] yeah, the old shell didn't have shell history so they implemented it in adb

| Channels | #motoactv index | Today | | Search | Google Search | Plain-Text | summary