Activity
From 02/13/2013 to 03/14/2013
03/11/2013
- MK 09:02 PM Software Development: RE: Yocto in the future?
- Mike,
I *think* the kernel source were coming from Critical Link's git repo - this line is in the linux-mityarm-335x_3.1.bb file:
@SRC_URI = "git://support.criticallink.com/home/git/linux-mityarm-335x.git;protocol=git;branch=${BRAN... - MW 08:26 PM Software Development: RE: Yocto in the future?
- Right,
If I read you correctly, you will need to modify the Yocto recipe to use our kernel URL and not Yocto's. This is because the Yocto kernel sources (or anyone else's at the moment, until we upstream stuff) does not have a MACHIN... - MK 04:45 PM Software Development: RE: Yocto in the future?
- Mike,
Thanks again for your quick response.
In Yocto, I've been setting following environment variables in my local.conf:
GCCVERSION = "4.5%"
BINUVERSION = "2.20.1"
I have used pretty much the same BSP in Yocto as for Angstrom... - MW 07:28 AM Software Development: RE: Yocto in the future?
- I don't know why the SDK toolchain isn't working. Are you sure that the .config file was not altered, specifically, the MACHINE type? That would definitely cause problems. Are you using the kernel and u-boot code from our site, or fro...
- We are building a system with extended low temperature range (below 0C) and see that the VSC8601 PHY only goes to 0C.
Vitesse also has an 8641 part with extended temperature range, which appears to be quite similar to the VSC8601, exc...
03/08/2013
- Up until a week or so ago, we were developing MityARM-335x code under Angstrom as a holdover from work done on Beaglebone w/ Angstrom. We built the filesystem, kernel, and uboot using recipes posted on the Critical Link Wiki, and - once...
-
02:39 PM
Software Development: RE: MityArm in u-booting from SPI flash
- Thanks Tim,
How hard to you think it would be to merge the MityArm changes for u-boot into the latest mainline u-boot?
John C. - TI 02:28 PM Software Development: RE: MityArm in u-booting from SPI flash
- John
While the omap l138 based mity SOMs do boot from SPI (NOR) flash, we have not tried to do so on the AM335X based SOMs.
In fact, the current AM335X based SOMs can not boot from on-board flash due to pin mux constraints. You should ... - Does Critical Link plan to support booting from spi flash? We would like to begin the process of migrating the booting
over to spi flash and were hoping you guys might have it ready to go already (fingers crossed...) We understand that
...
02/27/2013
- MK 07:57 PM Software Development: RE: Critical Link Kernel integration w/ Angstrom/OpenEmbedded
- I take back the comment about console-image build not working. I can build it just fine, and the issue is most certainly user error on my part.
- MK 07:15 PM Software Development: RE: Critical Link Kernel integration w/ Angstrom/OpenEmbedded
- After the latest angstrom update, building the console-image program no longer works, but the systemd-image works fine.
Also, I still can't figure out how to get the MLO to build. Using the one shipped with the dev kit seems to work ...
02/20/2013
- MW 08:54 AM Software Development: RE: Where is the real spi_xfer() function located for MityArm in u-boot?
- Cheat:
Look for .o files.
-Mike
-
08:53 AM
Software Development: RE: Where is the real spi_xfer() function located for MityArm in u-boot?
- Mike,
Thanks for the info. It is difficult to trace which files are included in the build using
the am335x_evm.h file's #defines.
-John Conover - MW 08:36 AM Software Development: RE: Where is the real spi_xfer() function located for MityArm in u-boot?
- John,
For the MityARM-335x family, the file is:
drivers/spi/omap3_spi.c
-Mike
- MW 07:50 AM Software Development: RE: Typo in rs485_test code
- Hi Ian,
It depends on the version of the compiler you are using and where it is getting it's kernel headers from.
The TI tools from their 05.03.02.00 SDK, what we used to run the test, have a definition of the their serial_rs485 st...
02/19/2013
- Thank you for putting up the great resource at:
http://support.criticallink.com/redmine/projects/armc8-platforms/wiki/Driving_RS-485_transceivers
But there is a typo in rs485_test.tar. I believe "#define gpio_pin padding [1]" should... - Hi
I've been unable to build u-boot on my Ubuntu 12.04 or 11.xx Virtual machines so I was wondering which u-boot file contains the spi_xfer() function that is compiled in the MityArm build of u-boot? I've searched looking at file...
02/14/2013
-
11:06 AM
Software Development: RE: Booting StarterWare image from NAND
- Tim,
Now it makes sense. I never did get the u-boot to build so I did not know the process of making
the MLO. Now that you say it I remember seeing that command used to build an image with ECC. So if
we had made a plain MLO ... - TI 10:56 AM Software Development: RE: Booting StarterWare image from NAND
- John
My understanding (which may be confused!) is that it doesn't matter so much if you use SW or HW ECC, but that you flash the image using the same ECC mode that you use to read it. So if you are using HW ECC when you flash the image ... -
10:42 AM
Software Development: RE: Booting StarterWare image from NAND
- Hey Tim,
We have referenced that page before, the problem is that there is no mention of ECC being "Required" for any
MLO image past a certain revision to boot. As with a lot of TI documentation on the 3359, certain things ar... - TI 10:30 AM Software Development: RE: Booting StarterWare image from NAND
- John
check out http://processors.wiki.ti.com/index.php/AM335x_U-Boot_User's_Guide
cheers
/Tim -
10:07 AM
Software Development: RE: Booting StarterWare image from NAND
- Mike,
Here is our update. We have our board booting from NAND with no changes to the MLO image. We changed the SYSBOOT[9] to ECC
and our system boots with no issues. Is there any TI documentation/notes that would explain exactl...
02/13/2013
-
09:25 AM
Software Development: RE: Booting StarterWare image from NAND
- Hey Mike,
Are jumper mode is 00100b so it is serial first, MMC and then NAND. Experimenting with your board we got your board to
boot from NAND but we had to remove the header from the MLO image (first 0x200 bytes.) We also ...