Activity
From 06/24/2013 to 07/23/2013
07/19/2013
- MS 10:44 AM PCB Development: RE: EMC Emissions from MityArm 335X MOdule
- Hi Mike,
I am good for now. Thanks for the information.
Regards,
Mike
07/18/2013
- BD 03:48 PM Software Development: RE: NFS configuration
- Great news Leonid! Thanks for the detailed error messages.
-Bob
- LV 01:32 PM Software Development: RE: NFS configuration
- Bob,
That was the problem. It's booting now.
Thanks everybody,
Leonid - BD 12:09 PM Software Development: RE: NFS configuration
- Leonid,
One more suggestion: did you untar the sample filesystem as root? Or did you untar it as a normal user?
For example (you'll need to adjust the filesystem.tgz location and name): -
01:14 PM
Software Development: RE: 3352 Module can't boot. Unknown memory type.
- Tim,
It does boot with a MLO and uboot built from the repository. Just need to put our
SPI changes back in.
Thanks,
John C. -
11:37 AM
Software Development: RE: 3352 Module can't boot. Unknown memory type.
- Ok Tim,
I'll pull that in and give it a try.
Thanks,
John C. - TI 11:35 AM Software Development: RE: 3352 Module can't boot. Unknown memory type.
- John
The error message you see about "Error - Factory Configuration Invalid" is because the u-boot you downloaded is too old (sorry about that)
and it doesn't understand the new factory config block.
We are in the midst of updating "e... -
11:21 AM
Software Development: RE: 3352 Module can't boot. Unknown memory type.
- Tim,
The error is using the latest mlo and uboot from the Critical Link website.
John C. - TI 11:15 AM Software Development: RE: 3352 Module can't boot. Unknown memory type.
- John
All the 3352 based modules use DDR3, so will all require the new u-boot.
cheers
/Tim -
11:05 AM
Software Development: RE: 3352 Module can't boot. Unknown memory type.
- Tim,
Just to add some more info. We tried booting from three different 3352 modules and
they all have the same issue.
John C. -
10:25 AM
Software Development: RE: 3352 Module can't boot. Unknown memory type.
- Tim,
As a test I downloaded the MLO and uboot from your web site and got the following when trying to boot the 3352 module:
-------
U-Boot SPL 2011.09-00000-g045cda5 (Jul 12 2012 - 16:03:00)
Texas Instruments Revision detection u... -
10:14 AM
Software Development: RE: 3352 Module can't boot. Unknown memory type.
- Thanks Tim,
I'll give it a try.
John C. - TI 10:12 AM Software Development: RE: 3352 Module can't boot. Unknown memory type.
- John
You should pull the latest from the AM335XPSP_04.06.00.08 branch on our git repo.
Support for the DDR3 memory configurations came along with a "base update" to the
newer PSP uboot.
cheers
/Tim -
09:27 AM
Software Development: 3352 Module can't boot. Unknown memory type.
- Hi,
We've been running the 3359 modules for some time and have shifted to the 3352. The error
message is:
--------------
U-Boot SPL 2011.09-00000-ge301766-dirty (May 01 2013 - 16:46:12)
Texas Instruments Revision detection unim... - MW 07:32 AM PCB Development: RE: EMC Emissions from MityArm 335X MOdule
- Hi Mike,
OK, those numbers were for the CPU running at 500 MHz. The speed of the CPU may change based on what power management scheme (the CPU governor) that is selected in the kernel configuration.
Our default kernel selects "per...
07/17/2013
- JC 10:47 AM Software Development: RE: NFS configuration
- Lenoid,
It might be helpful to put the filesystem on an SD card and make sure it boots from there.
-Jonathan - LV 10:42 AM Software Development: RE: NFS configuration
- Hi All,
I agree, it looks like the original issue is fixed and now problem is happenning after mounting nfs. Probably, when *init* is running. I tried to play with *init=* commands in u-boot, but it didn't help.
The file system on ... - MW 09:26 AM Software Development: RE: NFS configuration
- On your host, can you type "sudo exportfs" and list the results?
-Mike
- JC 08:41 AM Software Development: RE: NFS configuration
- Leonid,
It looks like nfs may be mounting. I'd expect an error similar to below if it didn't. Is the filesystem on the nfs server the same as the one on the sd card? How was it created? - MW 07:53 AM PCB Development: RE: EMC Emissions from MityArm 335X MOdule
- Hi Mike,
Hold on on that last post. We were looking at the numbers and something does not look correct (there should not be a 2 GHz clock or a 1 GHz clock, and the CPU should be running at 700 MHz).
I may have posted this premat... - MW 07:46 AM PCB Development: RE: EMC Emissions from MityArm 335X MOdule
- Hi Mike,
We needed to enable a kernel flag here. Here is a dump of the PLL derived clock tree from our dev kit. Most of these frequencies don't leave the CPU. The frequencies should be the same for you with the exception of perhaps...
07/16/2013
- LV 06:01 PM Software Development: RE: NFS configuration
- Hi Bob,
The current status is
1. After installing nfs-utils-client I can mount NFS directory when I boot from SD card
2. NFS boot is hanging after the following message:
_Freeing init memory: 236K_
You are right originally t... - BD 05:37 PM Software Development: RE: NFS configuration
- Hello Leonid,
In a previous post you had an error running this command:
07/15/2013
- MS 08:58 AM PCB Development: RE: EMC Emissions from MityArm 335X MOdule
- Hi Mike,
I had our lead software engineer try to dump the clock tree per the information you provided but he could not access it.
I would appreciate if you could set up a rig and help us through it.
Thanks,
Mike
07/12/2013
- MW 05:14 PM PCB Development: RE: EMC Emissions from MityArm 335X MOdule
- Hi Mike,
The module has 2 crystals on it, a 24 MHz (connected to ARM) and a 32 KHz (connected to PMIC).
I don't know off-hand what the switching supply freqencies are on the PMIC, but I would think they aren't above 2 MHz (probably... - Hi,
We have done a number of EMC frequency scans on our product using the MityArm 3352-GX-X27-RC module. There are a couple of frequencies that are out of specification to meet the IEC 61000-6-3 Emissions for Residential Environments.... - LV 04:57 PM Software Development: RE: NFS configuration
- Hi Mike,
I was able to mount NFS from the card after installing nfs-utils-client using opkg
Not sure what the next step can be.
Leonid - LV 01:37 PM Software Development: RE: NFS configuration
- Mounting with *-t nfs* option produces the same error message.
Re-saved file in txt format
Leonid. - MW 01:18 PM Software Development: RE: NFS configuration
- I can't open your word file. (Word says it is corrupt). Can you upload just a plain text file?
I think you need to type: - LV 12:37 PM Software Development: RE: NFS configuration
- Hello, Mike
Our basic configuration is an isolated network with 5-6 devices connected via Gigabit Network Switch (DEll PowerConnect 2708), all with static ip addresses. One of these devices is my Ubuntu PC (server 10.0.0.1), another M... - MW 08:18 AM Software Development: RE: NFS configuration
- Hello,
I am a little confused. In your posts, you show initial success using 192.168.0.144 as your server. But the failures are using 10.0.0.1 as your server. Is your server really 10.0.0.1? Many times that address is reserved for... - MS 03:28 PM PCB Development: RE: MityArm 335x Minimum Voltage Operation
- Mike,
I have made some changes to my design to try to limit the low end of the operating voltage to 3.2VDC but I don't feel comfortable with my test results to date which show the supply goint down to 3.1VDC under max load and input s...
07/11/2013
- LV 04:44 PM Software Development: RE: NFS configuration
- Mike,
Adding 'eth0' as a device in ip=... does stop system from searching for DHCP.
But instead it gives a Kernel Panic with a following information:
_[ 1.828257] CPSW phy found : id is : 0x70421
[ 1.837751] mmcblk0: p1 p... - MW 04:06 PM Software Development: RE: NFS configuration
- Hmmm....
Can you try specifying "eth0"? It looks like you have a USB network gadget compiled in (or our default kernel does), from https://www.kernel.org/doc/Documentation/filesystems/nfs/nfsroot.txt, - LV 03:43 PM Software Development: RE: NFS configuration
- My first post contains command line and fragment of boot sequence. It does match what I expected.
The full boot sequence for the NFS case and static IP is attached as a file:
Thanks
Leonid - MW 03:17 PM Software Development: RE: NFS configuration
- When the unit boots, it prints out the command line it thinks it receives early in the boot sequence. Can you verify that that command line matches what you expect?
Can you capture the boot log and post it here?
-Mike
- LV 03:08 PM Software Development: RE: NFS configuration
- This is the full list of my u-boot environment variables:
_U-Boot# print__autoload=no
baudrate=115200
bootargs=root=/dev/nfs nfsroot=10.0.0.1:/home/likewise-open/SCHICK_NT/lvolpert/Dev/3D/MityARM-rootfs_kit,nolock rw ip=10.0.0.22:10...
07/05/2013
- MC 01:10 PM Software Development: RE: bad timing of RS485 TXE when using omap-serial driver
- Yes, you were right regarding the falling of RS485-TXE line after data ends: in the version 3.2 we have the RS485-TXE code moved from serial_omap_enable_ier_thri() to serial_omap_stop_tx().
It works perfectly now; in our measurements w...
07/04/2013
- MW 07:58 AM Software Development: RE: bad timing of RS485 TXE when using omap-serial driver
- I think you are missing part of the patch, which was original written against 3.1 of the kernel.
Take a look at the serial_omap_stop_tx() call in our "3.2 branch omap_serial.c":http://support.criticallink.com/gitweb/?p=linux-mityarm-... - MW 07:44 AM Software Development: RE: bad timing of RS485 TXE when using omap-serial driver
- Do you see this problem on all the UARTS? Or just UART4?
-Mike
- MC 02:15 AM Software Development: RE: bad timing of RS485 TXE when using omap-serial driver
- sure, here it is.
I also put the patched omap-serial.c to check that your patch has been properly applied and the points I've put "printk"-s.
Cheers,
Mihai
07/03/2013
- MW 03:44 PM Software Development: RE: bad timing of RS485 TXE when using omap-serial driver
- Hello Mihai,
Would you be willing to share your modified send.c code?
-Mike - I've used your nice guidelines describe here: http://support.criticallink.com/redmine/projects/armc8-platforms/wiki/Driving_RS-485_transceivers
to implement the low-latency RS485 handling in the linux kernel (v3.2) - I used the patch2 b... - KW 03:29 PM Software Development: RE: Writing UBIFS rootfs to AM3354 with 512MB NAND
- Tim,
Thanks for testing this. My module PN: *80-000517RC-1B*.
I believe the problem is with the lower MTD NAND driver for NAND part MT29F4G08ABAEAWP. Simple NAND writes are giving me problems.
Kim
- JC 01:34 PM Software Development: RE: Writing UBIFS rootfs to AM3354 with 512MB NAND
- Kim,
Due to the long holiday weekend, i'm not sure this will get looked at until atleast Monday.
Also what is the part number and model number of your module?
Thanks
Jonathan - JC 01:32 PM Software Development: RE: Writing UBIFS rootfs to AM3354 with 512MB NAND
- For our 512mb nand, I believe the format command should be @ubiformat /dev/mtd8 -s 1024 -O 4096@ but it doesn't work.
- JC 01:31 PM Software Development: RE: Writing UBIFS rootfs to AM3354 with 512MB NAND
- Testing on a 256MB nand som, I was able to format mtd8 and mount it. {{collapse(See output:)
- JC 01:17 PM Software Development: RE: Writing UBIFS rootfs to AM3354 with 512MB NAND
- Kim,
I tried to do the ubiformat on a 512MB nand som here and I ran into the same problem. It marked the whole nand as bad.
I was able to reset the bad block count by running the following in u-boot. Note: This will delete all data ... - KW 09:36 AM Software Development: RE: Writing UBIFS rootfs to AM3354 with 512MB NAND
- Hi Tim,
Could someone at Critical Link send me the commands used for creating the ubifs image and writing it to the 512MB NAND flash on your MityARM development board?
Thanks,
Kim
- I am switching from TI Evaluation Board to MityARM-335x Development Kit. I am using Ubuntu 12.04 host machine for development and NFS with client and server having static IP addresses is our preferred configuration.
It worked well with ...
07/02/2013
- TI 08:09 PM Software Development: RE: Writing UBIFS rootfs to AM3354 with 512MB NAND
- Kim
I am not super familiar with ubifs, but it seems it can be a bit tricky to master. I know it can be done, as some of our customers are using it. One thing I have seen is to make sure the subpage size is the same as the page size (4k... - KW 07:14 PM Software Development: RE: Writing UBIFS rootfs to AM3354 with 512MB NAND
- Here some additional information that maybe helpful. After ubiformat the erase blocks are all marked BAD.
- KW 05:01 PM Software Development: RE: Writing UBIFS rootfs to AM3354 with 512MB NAND
- Hi Tim,
I've been following this steps and it looks like the ubi.img is writing to NAND, but I get error 28 (No space left on device) when trying to attach. Here are my steps for writing ubi.img: - TI 02:53 PM Software Development: RE: Writing UBIFS rootfs to AM3354 with 512MB NAND
- Kim
TI has a few pages on this.. check out http://processors.wiki.ti.com/index.php/UBIFS_Support
As long as you program the ubifs from linux there should not be an issue.
There is currently a disconnect between ECC used by u-boot and ... - What's the procedure for creating and writing the Linux UBIFS rootfs to the AM3354 512MB NAND? What ECC is used for the UBIFS partitions?
-Kim
06/27/2013
- We are moving our design away from the Vitesse ethernet chip (VSC8601) included on the Am3359 dev board to a Texas Instruments TLK110.
We know these are used with AM3359 successfully by TI;
http://www.ti.com/tool/tmdxice3359
We ar... - BC 02:10 PM Software Development: RE: AM3359 Sleep / Power Savings
- We are currently using the 3.1 kernel from the critical link git.
Thanks for the info!
06/25/2013
- JC 04:40 PM Software Development: RE: Difficulty in flashing MLO and U-Boot to the AM3354 with 512MB NAND
- Ok I misunderstood. This might be what your looking for. http://stackoverflow.com/a/180368/780194
-Jonathan -
04:35 PM
Software Development: RE: Difficulty in flashing MLO and U-Boot to the AM3354 with 512MB NAND
- It seem the git diif shows me the differences between what I originally cloned from Critical Link and my current changes in the directory.
John C. - JC 04:03 PM Software Development: RE: Difficulty in flashing MLO and U-Boot to the AM3354 with 512MB NAND
- You could try @git diff@
-Jonathan -
03:57 PM
Software Development: RE: Difficulty in flashing MLO and U-Boot to the AM3354 with 512MB NAND
- Hey Jonathan,
I've simply modified the files with-in the u-boot-mityarm-335x directory. Would there be
any way to diff the changes made to the repo?
Thanks,
John C.
- JC 09:57 AM Software Development: RE: Difficulty in flashing MLO and U-Boot to the AM3354 with 512MB NAND
- John,
Depending on how you did your changes you should be able to do a _git pull_ and git will update the repo, merging with your changes where necessary.
-Jonathan - JC 02:59 PM Software Development: RE: AM3359 Sleep / Power Savings
- What version of the kernel are you running?
This might be of some interest.
http://processors.wiki.ti.com/index.php/AM335x_Power_Management_Standby_User%27s_Guide
Our current 3.2 kernel is based off of http://processors.wiki.ti.co... - Is there a way to enter a low power state in the AM3359?
I've been attempting different variations on the following:
echo "mem" > /sys/power/class
...but it does not appear to do anything. The linux configuration seems to have it...
06/24/2013
- TI 09:28 PM Software Development: RE: booting 3359 from NAND only
- I'll add to what Mike said a wee bit.
You can use the mtdparts u-boot command to display (and set) mtdparts in u-boot [ see http://www.denx.de/wiki/view/DULG/UBootCmdGroupFlash#Section_5.9.3.5. ]
These _should_ line up with the linux v... - MW 07:23 PM Software Development: RE: booting 3359 from NAND only
- Hi Keith,
I don't think the NAND is actually physically partitioned. It's more of a "table" as you say.
The Kernel will use a default partitioning as coded in the arch/arm/mach-davinci/board-mityarm335x.c file. However, you can co... - KF 06:32 PM Software Development: RE: booting 3359 from NAND only
- Hi Tim,
I have just been looking at flashing from the mmc, however, I have managed to confuse myself even more!
From the linux running on the SD card if I use "mtdinfo -au" it tells me that the file system is in mtd8, not mtd7 (there ... - KF 06:09 AM Software Development: RE: booting 3359 from NAND only
- Hi Tim,
Thanks for the quick reply. When you say keep the EEC mode consistent between the u-boot and the kernel is this just the EEC mode used when they were written to NAND, or something more sinister such as something in the way they ... -
02:15 PM
Software Development: RE: Difficulty in flashing MLO and U-Boot to the AM3354 with 512MB NAND
- Tim,
We just ran across this exact problem with our newly purchased 3354 module(we had been using the 3359's).
I had followed the original instructions for cloning the repository and building it but then modified portions of
the...