Quantcast

Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

classic Classic list List threaded Threaded
17 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

JamesAng
Hi,

I'm experiencing occasional 'lock-up' during boot if MT9V032 is not detected/connected to the Overo.
This is especially so during the 1st boot after reflashing the system.

[snip]
omap3isp omap3isp: Revision 2.0 found
omap-iommu omap-iommu.0: isp: version 1.1
overo: setting xclk to 25000000 hz
mt9v032 3-005c: MT9V032 power up failed
isp_register_subdev_group: Unable to register subdev mt9v032 <-- 'lock-up' here indefinitely till "Ctrl-C" is send via console port
udev: renamed network interface eth0 to eth1
[snip]

Can someone confirm if they had the same occurrence and any solution to it?

Thanks in adv.

James.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Overo tide not booting on tobi with angstrom

garry.allen
I've created a boot disk which works with an overo fire on the tobi board. This drive does not work on an Overo Tide. From a quick google, it appears that i need to create a boot.scr file on the boot disk. is this correct? Where does it go? And what does it need to contain?
thanks
Garry Allen
------------------------------------------------------------------------------
Cloud Services Checklist: Pricing and Packaging Optimization
This white paper is intended to serve as a reference, checklist and point of
discussion for anyone considering optimizing the pricing and packaging model
of a cloud services business. Read Now!
http://www.accelacomm.com/jaw/sfnl/114/51491232/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Overo tide not booting on tobi with angstrom

ashcharles
Hey Gary,

On Wed, Dec 7, 2011 at 11:04 PM, Garry Allen
<[hidden email]> wrote:
> I've created a boot disk which works with an overo fire on the tobi board. This drive does not work on an Overo Tide. From a quick google, it appears that i need to create a boot.scr file on the boot disk. is this correct? Where does it go? And what does it need to contain?
A Tide shouldn't require a boot.scr---the default u-boot boot commands
should be sufficient to boot the COM into Linux.  At what point in the
boot does the Tide fail?  Does the Fire actually use the MLO and
U-boot.bin (look at the date stamps in the console when it boots) from
the microSD card?

If the MLO wasn't the very first thing written to the SD card after
formatting, the Fire may boot (albeit using the MLO from NAND) where
the Tide would fail (as it doesn't have an MLO in NAND on which to
fall back on).

-Ash

------------------------------------------------------------------------------
Cloud Services Checklist: Pricing and Packaging Optimization
This white paper is intended to serve as a reference, checklist and point of
discussion for anyone considering optimizing the pricing and packaging model
of a cloud services business. Read Now!
http://www.accelacomm.com/jaw/sfnl/114/51491232/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Overo tide not booting on tobi with angstrom

garry.allen
Hello Ash
I followed the directions for creating a flash card on gumstix.org. I had issues with sfdisk and ended up using fdisk to create the boot and rootfs partitions. Once, i successfully created the partitions, I then copied the fiels on.
At the moment, nothing appears in the console. It appears to be failing very early in the process.I am wondering if this is because the default baud rate is set to the wrong value (I,m using 115200, n, 8, 1, with no flow control)
I've tried this card with 2 different tides. i have another 3 I havent touched yet
thanks
Garry
On 09/12/2011, at 10:15 AM, Ash Charles wrote:

> Hey Gary,
>
> On Wed, Dec 7, 2011 at 11:04 PM, Garry Allen
> <[hidden email]> wrote:
>> I've created a boot disk which works with an overo fire on the tobi board. This drive does not work on an Overo Tide. From a quick google, it appears that i need to create a boot.scr file on the boot disk. is this correct? Where does it go? And what does it need to contain?
> A Tide shouldn't require a boot.scr---the default u-boot boot commands
> should be sufficient to boot the COM into Linux.  At what point in the
> boot does the Tide fail?  Does the Fire actually use the MLO and
> U-boot.bin (look at the date stamps in the console when it boots) from
> the microSD card?
>
> If the MLO wasn't the very first thing written to the SD card after
> formatting, the Fire may boot (albeit using the MLO from NAND) where
> the Tide would fail (as it doesn't have an MLO in NAND on which to
> fall back on).
>
> -Ash
>
> ------------------------------------------------------------------------------
> Cloud Services Checklist: Pricing and Packaging Optimization
> This white paper is intended to serve as a reference, checklist and point of
> discussion for anyone considering optimizing the pricing and packaging model
> of a cloud services business. Read Now!
> http://www.accelacomm.com/jaw/sfnl/114/51491232/
> _______________________________________________
> gumstix-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gumstix-users


------------------------------------------------------------------------------
Learn Windows Azure Live!  Tuesday, Dec 13, 2011
Microsoft is holding a special Learn Windows Azure training event for
developers. It will provide a great way to learn Windows Azure and what it
provides. You can attend the event by watching it streamed LIVE online.  
Learn more at http://p.sf.net/sfu/ms-windowsazure
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

JamesAng
In reply to this post by JamesAng
Hi all,

Can some one verify if there is 'lag' after that particular line from console?

Thanks.

James.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

ezekiel247
I just built a 3.0 image this morning. The boot consistently hangs for 2 minutes on:

mt9v032 3-005c: MT9V032 power up failed
isp_register_subdev_group: Unable to register subdev mt9v032

eventually I can login, but the system is overloaded to begin with ( load >1) and then the load decays down to <.05 after a few minutes. I can't see any processes during this time that appear to be hogging resources.


Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

JamesAng
Hi,

Thanks for verifying that I'm not the only one experiencing this for my image.

Occasionally, the next message that appeared will be those related to the Wireless module driver.
Can advise how to turn off the wireless module driver. i.e. bluetooth, wifi. to debug if either is the cause of this?

Many thanks in adv.

James.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

Steven
I have the same problem with a freshly compile linux 3.0 image.  I removed the module from the kernel and it still hang for a long time, but sometimes it gets to the libertas initialization line.  Although it could be hanging because board-overo.c does some check for mt9v032 even without the module compiled in.  I'll try to remove those and see what happens.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

JamesAng
Thanks for helping.

I'm surprise this issue was not picked up for such a long time.
2min of delay is very unusual..

James
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

Steven
I removed MT9V032 from board-overo.c, and removed the kernel module, and it hang again at libertas_sdio init, so I'm assuming MT9V032 wasn't the culprit.

I have noticed in kernel 2.6.39 it boots to the login screen without hanging, but the radio won't actually come up until after 2 minutes or so.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

JamesAng
It seems like that the delay is due to the wifi module.

It need to be fix somehow but in the meantime, how can we disable it then?
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

maik
This post has NOT been accepted by the mailing list yet.
I now ran into the same problem with kernel 3.0.0. Seems that the Wifi hangs. Is there anything new ?
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

raji
Hi All,

I encountered the same problem and found this work around on the mailing list.
Tried it on the linux-omap3.0 image and it seems to fix it.
The Nick Starzicki's solution in that post seems to solve it .  Here is the link
http://gumstix.8.n6.nabble.com/WiFi-problems-td564654.html

Hope it helps,
Raji
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

JamesAng
In reply to this post by Steven
Hi Steven,

Steven wrote
I removed MT9V032 from board-overo.c, and removed the kernel module, and it hang again at libertas_sdio init, so I'm assuming MT9V032 wasn't the culprit.

I have noticed in kernel 2.6.39 it boots to the login screen without hanging, but the radio won't actually come up until after 2 minutes or so.
I tried the same setup but ran Steve's GNOME-R13 image.

There is NO delay experienced in this case!
With or Without the Caspa camera attached, the system boot fine.

The 2 boot messages

libertas_sdio mmc1:0001:1: (unregistered net_device): 00:19:88:23:f9:1d, fw 9.70.3p14, cap 0x00000303
libertas_sdio mmc1:0001:1: wlan0: Marvell WLAN 802.11 adapter

appears immediately after the starting up for the omap3isp subsystem

omap3isp omap3isp: Revision 2.0 found
omap-iommu omap-iommu.0: isp: version 1.1
mt9v032 3-005c: Probing MT9V032 at address 0x5c
overo: setting xclk to 25000000 hz
overo: setting xclk to 0 hz
mt9v032 3-005c: MT9V032 detected at address 0x5c
overo: setting xclk to 25000000 hz
overo: setting xclk to 0 hz

This is weird!!!
Both uses the same kernel recipe but the the only difference is the image recipe used!

James Ang
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

AKS-4
If you suspect a driver is causing problems, you could
make it modular (instead of it being tightly coupled or
integrated into the kernel) may help, but you will have
to load drivers manually by "insmod" at system start-up
after creating that device entry point with "mkdev" - for
character device drivers. My 2 cents!

On 2/9/12, JamesAng <[hidden email]> wrote:

> Hi Steven,
>
>
> Steven wrote
>>
>> I removed MT9V032 from board-overo.c, and removed the kernel module, and
>> it hang again at libertas_sdio init, so I'm assuming MT9V032 wasn't the
>> culprit.
>>
>> I have noticed in kernel 2.6.39 it boots to the login screen without
>> hanging, but the radio won't actually come up until after 2 minutes or so.
>>
>
> I tried the same setup but ran Steve's GNOME-R13 image.
>
> There is NO delay experienced in this case!
> With or Without the Caspa camera attached, the system boot fine.
>
> The 2 boot messages
>
> libertas_sdio mmc1:0001:1: (unregistered net_device): 00:19:88:23:f9:1d, fw
> 9.70.3p14, cap 0x00000303
> libertas_sdio mmc1:0001:1: wlan0: Marvell WLAN 802.11 adapter
>
> appears immediately after the starting up for the omap3isp subsystem
>
> omap3isp omap3isp: Revision 2.0 found
> omap-iommu omap-iommu.0: isp: version 1.1
> mt9v032 3-005c: Probing MT9V032 at address 0x5c
> overo: setting xclk to 25000000 hz
> overo: setting xclk to 0 hz
> mt9v032 3-005c: MT9V032 detected at address 0x5c
> overo: setting xclk to 25000000 hz
> overo: setting xclk to 0 hz
>
> This is weird!!!
> Both uses the same kernel recipe but the the only difference is the image
> recipe used!
>
> James Ang
>
> --
> View this message in context:
> http://gumstix.8.n6.nabble.com/Linux-omap3-v3-0-may-lock-up-if-MT9V032-is-not-detected-tp1319945p4378221.html
> Sent from the Gumstix mailing list archive at Nabble.com.
>
> ------------------------------------------------------------------------------
> Virtualization & Cloud Management Using Capacity Planning
> Cloud computing makes use of virtualization - but cloud computing
> also focuses on allowing computing to be delivered as a service.
> http://www.accelacomm.com/jaw/sfnl/114/51521223/
> _______________________________________________
> gumstix-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gumstix-users
>

------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

srmaddox
In reply to this post by JamesAng
Interesting, I've been having the same issue - but it appears somewhat intermittent.

My Overo AIR Com + RoboVero arrived and on first boot I had CRC errors and it wouldn't boot.

Troubleshooting, I encountered the following:

(1) Booting from MicroSD with pre-built factory image booted up w/o any issues, and wifi was immediately available.
(2) After a number of reboots/some playing with Python client library, the examples (i.e. servo.py) quit working without any actual system changes. In an attempt to troubleshoot, I created a new SD card with the latest development image.
(3) The new development image seems to have resolved the servo/IO issues, however the developmental image now experiences these delays and very odd wifi related issues. Interestingly, every boot now falls into one of two categories:
(a) Bootup hangs as previously described in this thread
(b) Bootup does not hang, however, the wifi module is completely inaccessible and doesn't even appear recognized by the system for about 5 minutes after boot.
(4) Going back to the factory image, I have the servo problems again, but ALSO have the exact same wifi issues, and about 50/50 split between the two boot scenarios outlined above.

At this point, rebooting the device will yield these different results without any changes made to the system - literally, boot, wait for login prompt, shutdown -r now - every other boot exhibits these two boot scenarios.

The issues are now persisting with various images I'm using, so I'm beginning to wonder if this is some intermittent hardware issue which is interacting with the actual image/kernel booted.

At this point in time, my init script for networking simply spawns a script which waits 5 minutes and then goes through my WiFi configuration. It's sort of a pain, but so far this is the first time that every boot will reliably come up with networking, even if the actual delay time varies based on whether or not it hangs during boot or just waits forever before the wifi module is loaded after boot.

I'm still experiencing unresolvable issues with the servo output - at this point, the motor headers are barely functional (the unmodified servo.py will barely move the servo upon changing angle, but never appears to complete the movement - so, for instance, New Angle: 0; New Angle: 180; has about 1 degree of attempted movement, then the servo goes "limp").

All of that noted, I've tried different servos and tested the servo's response to a STAMP controller - the servos are fine.

I mention all of these issues because I'm suspecting an issue either with the COM or the RoboVero module itself, hardware wise - I wonder if there's some hardware issue that we're all experiencing, but it's intermittent and soft/firmware dependent enough to make troubleshooting difficult.

I'll probably just be buying a new Air COM and robovero board and attempting from a fresh start - I'll compare whether or not these issues are seen on a brand new COM/expansion board, that should rule out any hardware defects, though it may not rule out some kind of hardware bug.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Linux-omap3 v3.0 may lock-up if MT9V032 is not detected

DaWaN
I do have similar  issues with the camera.
Like 75% of the time the camera works but booting takes as much as 2 minutes
The other 25% time the board boots within 30 seconds but then my camera does not work

It seems to be a software issue because all the Overo boards I have here suffer from the same issue
Loading...