Android Froyo on Beagleboard, MMC ext3 filesystem corrupted after first successful boot

GoogleGroups
1

I am using pre-built binaries from Texas Instruments Android Development Kit

Followed the mkmmc script to copy the files onto SD/MMC.

It is successfully booting from SD card and successful reboots as well.

In case of MMC, boots fine at very first time, could see all the folders got created and can execute 3D videos, storage perf apps. On reset and reboot, it could not see ext3 partition that leads to kernel panic. PC host also could not mount ext3 partition as it was corrupted. I need to reflash the MMC to boot again.

Free memory comparison SD vs MMC:

After successful booting,

SD boot: shows about 50MB free space out of total 256MB

MMC boot: shows about 7MB free space

Suspect memory leak in MMC flow path.

Did anyone come across this issue with pre-built binary? Is there any patch ?

Thanks,
Muthu
|

2

No I have done the same but I do not se the issue

Sorry

Victor Rodriguez

3

Thanks for response.

Which make of MMC and is it MMC 4.3 or 4.4 version ? What is the size of the MMC storage ?

Thanks,
Muthu


— On Thu, 12/2/10, Victor Rodriguez vm.rod25@gmail.com wrote:


> From: Victor Rodriguez vm.rod25@gmail.com
> Subject: Re: [beagleboard] Android Froyo on Beagleboard, MMC ext3 filesystem corrupted after first successful boot
> To: beagleboard@googlegroups.com
> Date: Thursday, December 2, 2010, 10:40 PM
>
> On Thu, Dec 2, 2010 at 12:05 PM, Muthukumar karuppiah kmkr3@yahoo.com wrote:
> >
> > I am using pre-built binaries from http://software-dl.ti.com/dsps/dsps_public_sw/sdo_tii/TI_Android_DevKit/02_00_00/index_FDS.html
> >
> > Followed the mkmmc script to copy the files onto SD/MMC.
> >
> > It is successfully booting from SD card and successful reboots as well.
> >
> > In case of MMC, boots fine at very first time, could see all the folders got created and can execute 3D videos, storage perf apps. On reset and reboot, it could not see ext3 partition that leads to kernel panic. PC host also could not mount ext3 partition as it was corrupted. I need to reflash the MMC to boot again.
> >
> > Free memory comparison SD vs MMC:
> >
> > After successful booting,
> >
> > SD boot: shows about 50MB free space out of total 256MB
> >
> > MMC boot: shows about 7MB free space
> >
> > Suspect memory leak in MMC flow path.
> >
> > Did anyone come across this issue with pre-built binary? Is there any patch ?
> >
> > Thanks,
> > Muthu
> >
> > –
> > You received this message because you are subscribed to the Google Groups “Beagle Board” group.
> > To post to this group, send email to beagleboard@googlegroups.com.
> > To unsubscribe from this group, send email to beagleboard+unsubscribe@googlegroups.com.
> > For more options, visit this group at http://groups.google.com/group/beagleboard?hl=en.
>
> No I have done the same but I do not se the issue
>
> Sorry
>
> Victor Rodriguez
>
> –
> You received this message because you are subscribed to the Google Groups “Beagle Board” group.
> To post to this group, send email to beagleboard@googlegroups.com.
> To unsubscribe from this group, send email to beagleboard+unsubscribe@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/beagleboard?hl=en.

|

4

Thanks for response.

Which make of MMC and is it MMC 4.3 or 4.4 version ? What is the size of the MMC storage ?

Thanks,
Muthu

I have a 2gb SD I go to beagleboard-rev-c4/
and execute

sudo mkmmc-android.sh /dev/sd.. and the number of my SD

and thats all

Regards

Victor

5


Thanks, exactly the same I am doing.

Here is my bootargs and bootcmd that I tried,

setenv bootargs ‘console=ttyS2,115200n8 androidboot.console=ttyS2 mem=256M root=/dev/mmcblk0p2 rw rootfstype=ext3 rootdelay=1 init=/init ip=off omap_vout.vid1_static_vrfb_alloc=y’

setenv bootcmd ‘mmc init 0;fatload mmc 0 80000000 uImage; bootm 80000000’

I also tried with bootargs in releasenotes, did not help.

Could you correct me if I have to make any change in bootargs ?

Thanks for helping.

Thanks,
Muthu

— On Thu, 12/2/10, Victor Rodriguez vm.rod25@gmail.com wrote:


> From: Victor Rodriguez vm.rod25@gmail.com
> Subject: Re: [beagleboard] Android Froyo on Beagleboard, MMC ext3 filesystem corrupted after first successful boot
> To: beagleboard@googlegroups.com
> Date: Thursday, December 2, 2010, 10:50 PM
>
> On Thu, Dec 2, 2010 at 12:16 PM, Muthukumar karuppiah kmkr3@yahoo.com wrote:
> >
> > Thanks for response.
> >
> > Which make of MMC and is it MMC 4.3 or 4.4 version ? What is the size of the MMC storage ?
> >
> > Thanks,
> > Muthu
> >
>
> I have a 2gb SD I go to beagleboard-rev-c4/
> and execute
>
> sudo mkmmc-android.sh /dev/sd… and the number of my SD
>
> and thats all
>
> Regards
>
> Victor
>
> > — On Thu, 12/2/10, Victor Rodriguez vm.rod25@gmail.com wrote:
> >
> > From: Victor Rodriguez vm.rod25@gmail.com
> > Subject: Re: [beagleboard] Android Froyo on Beagleboard, MMC ext3 filesystem corrupted after first successful boot
> > To: beagleboard@googlegroups.com
> > Date: Thursday, December 2, 2010, 10:40 PM
> >
> > On Thu, Dec 2, 2010 at 12:05 PM, Muthukumar karuppiah kmkr3@yahoo.com wrote:
> > >
> > > I am using pre-built binaries from http://software-dl.ti.com/dsps/dsps_public_sw/sdo_tii/TI_Android_DevKit/02_00_00/index_FDS.html
> > >
> > > Followed the mkmmc script to copy the files onto SD/MMC.
> > >
> > > It is successfully booting from SD card and successful reboots as well.
> > >
> > > In case of MMC, boots fine at very first time, could see all the folders got created and can execute 3D videos, storage perf apps. On reset and reboot, it could not see ext3 partition that leads to kernel panic. PC host also could not mount ext3 partition as it was corrupted. I need to reflash the MMC to boot again.
> > >
> > > Free memory comparison SD vs MMC:
> > >
> > > After successful booting,
> > >
> > > SD boot: shows about 50MB free space out of total 256MB
> > >
> > > MMC boot: shows about 7MB free space
> > >
> > > Suspect memory leak in MMC flow path.
> > >
> > > Did anyone come across this issue with pre-built binary? Is there any patch ?
> > >
> > > Thanks,
> > > Muthu
> > >
> > > –
> > > You received this message because you are subscribed to the Google Groups “Beagle Board” group.
> > > To post to this group, send email to beagleboard@googlegroups.com.
> > > To unsubscribe from this group, send email to beagleboard+unsubscribe@googlegroups.com.
> > > For more options, visit this group at http://groups.google.com/group/beagleboard?hl=en.
> >
> > No I have done the same but I do not se the issue
> >
> > Sorry
> >
> > Victor Rodriguez
> >
> > –
> > You received this message because you are subscribed to the Google Groups “Beagle Board” group.
> > To post to this group, send email to beagleboard@googlegroups.com.
> > To unsubscribe from this group, send email to beagleboard+unsubscribe@googlegroups.com.
> > For more options, visit this group at http://groups.google.com/group/beagleboard?hl=en.
> >
> >
> > –
> > You received this message because you are subscribed to the Google Groups “Beagle Board” group.
> > To post to this group, send email to beagleboard@googlegroups.com.
> > To unsubscribe from this group, send email to beagleboard+unsubscribe@googlegroups.com.
> > For more options, visit this group at http://groups.google.com/group/beagleboard?hl=en.
>
> –
> You received this message because you are subscribed to the Google Groups “Beagle Board” group.
> To post to this group, send email to beagleboard@googlegroups.com.
> To unsubscribe from this group, send email to beagleboard+unsubscribe@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/beagleboard?hl=en.

|

6

this is a wild guess but try changing rootdelay=1 to rootwait

the kernel docs say:

rootwait: Wait (indefinitely) for root device to show up.
Useful for devices that are detected asynchronously
(e.g. USB and MMC devices).

rootdelay=1 waits exactly one second and then fails if the root device
doesn't exist.

if it hangs after that then maybe there are bad contacts on the card?

7

Thanks, exactly the same I am doing.

Here is my bootargs and bootcmd that I tried,

setenv bootargs 'console=ttyS2,115200n8 androidboot.console=ttyS2 mem=256M root=/dev/mmcblk0p2 rw rootfstype=ext3 rootdelay=1 init=/init ip=off omap_vout.vid1_static_vrfb_alloc=y'

setenv bootcmd 'mmc init 0;fatload mmc 0 80000000 uImage; bootm 80000000'

I also tried with bootargs in releasenotes, did not help.

Could you correct me if I have to make any change in bootargs ?

My boot args

mmc init
fatload mmc 0 0x80300000 uImage
setenv bootargs 'mem=128M androidboot.console=ttyS2 console=tty0
console=ttyS2,115200n8 root=/dev/mmcblk0p2 rw init=/init rootwait
omapfb.video_mode=640x480MR-16@60'
bootm 0x80300000

NY the way I have a problem far from this with my beagle. Does
anyboady has been able to record on the beagle> is necessary a
preamplifier before the audio in ?

I know is not the way to ask this but make me doubt about it

Regards

Victor Rodriguez

8

I'm using the same TI download on my BB B5 board, the kernel stops at:
"Uncompressing Linux..........................................................
done, booting the kernel."

Nothing happened afterwards. I also tried coredruids which boots
kernel fine, but afterwards I got a static screen with "Android" shown
in the middle, nothing else and it's useless.

Anyone seen this? Or Froyo does not work on B5 at all?

Thanks,
Xianghua

9

OK I modified boot.scr to use 128M RAM instead of 256M(B5 only has
128MB), now it boots with a crappy image on screen, from serial port I
saw:

# enabling adb
adb_open
PVR_K: (FAIL) SGXInit: Incompatible HW core rev (10003) and SW core rev (10201).

# warning: `zygote' uses 32-bit capabilities (legacy support in use)
request_suspend_state: wakeup (3->0) at 40152404788 (1970-01-01
00:00:40.152374271 UTC)
init: untracked pid 799 exited
request_suspend_state: wakeup (0->0) at 45376129154 (1970-01-01
00:00:45.376098636 UTC)
init: untracked pid 911 exited
request_suspend_state: wakeup (0->0) at 50620086672 (1970-01-01
00:00:50.620086672 UTC)
init: untracked pid 932 exited

looks like TI's release does not work on B5(SGXInit fail shown above).