Juergen Borleis | 1 Oct 09:01 2014

Re: u-boot_hd.config

Hi Ian,

please keep the mailing list at least on CC. Thanks.

On Wednesday 01 October 2014 07:19:08 ian c wrote:
> thanks for the detailed description. However, I still have question,
> though.
> >> partition spare {
> >> size = 256M
> >> partition-type = 0x83
> >> }
> >> size=1024M
> >> }
> What is mean for size=1024M? If I add partition (e.g. I name its partition
> duplicate), should I reduce that size value?

If you take a closer look, the "size=1024M" is inside the "image" block. So it 
defines the size of the whole image file to be created.



Pengutronix e.K.                              | Juergen Borleis             |
Industrial Linux Solutions                    | http://www.pengutronix.de/  |

Girish Desai | 28 Sep 21:04 2014

UART Communication FriendlyARM-mini2440


I am facing a strange issue.. in case any one has faced this kind of issue and have any suggestions, please let me know...

if I would like to send and array of 8 bytes such as below through UART such as ttySAC1, ttySAC2 etc,  the byte containing 'zero' is not transmitted.. below is the example...

I have an array of unsigned char alph[] = {'a', 'b', 'c', 'd', 'e', 0x00, 'g', 'h'};
when this is transmitted over UART,  the output would be "abcdegh" only seven characters and the zero is missing... 

in the same way, one more.. try sending unsigned char reset[] = {0x01, 0x06, 0x00, 0x01, 0x00, 0x01, 0x19, 0xCA};

I would only see "01 06 01 01 19 CA"  only 6 bytes.. byte containing zeros are missing again..

in case any one have had this kind of issue and have suggestions, please let me know...

Juergen Borleis | 25 Sep 15:41 2014

Re: Changing MD5 of QT4

Hi Girish,

please keep the mailing list at least on CC.

On Thursday 25 September 2014 14:27:54 Girish Desai wrote:
> Thanks for the quick response.. I recently had cleaned up my system and had
> re-installed ubuntu.. at that time, I had backed-up only the un-tar'd
> sources where qt4... is also one of them.. while building bsp(rootfs), all
> the archives were getting downloaded and I just allowed it to as they were
> not taking much time.. but for QT it is around 230MB and due to my slow
> connection it is taking longer.. so I thought to tar the existing un-tar'd
> qt(the same version of what is being downloaded) and put in .../src path so
> that build can proceed.. it is that time where ptxdist says, wrong
> checksum...

Sure. That is why the checksum exists: to ensure the archive you are using is
the same than we are using.

> now i checked, i could not find qt4.make in the path you 
> mentioned.. I think this would be the reason for the error...

[jbe <at> thebe]> ls -l /usr/local/lib/ptxdist-2014.08.0/rules/qt4.make
-rw-r--r-- 1 root staff 18279 Jul 31 14:20 /usr/local/lib/ptxdist-2014.08.0/rules/qt4.make

Where did you install your PTXdist?



Pengutronix e.K.                              | Juergen Borleis             |
Industrial Linux Solutions                    | http://www.pengutronix.de/  |

Girish Desai | 25 Sep 13:49 2014

Changing MD5 of QT4


Could some one let me know where(which file) can I change the MD5sum of QT4 so that ptxdist does not crib saying that MD5SUM error for QT4.  just like as we give 
"PTXCONF_BAREBOX_MD5="d66774abcf9fa6c9530e13543cc707bc" for barebox in "platformconfig-NAND-128M" file, I would like to set md5sum for QT4

Cindy Gold | 18 Sep 00:36 2014

TQMa6x BSP Rev.0104 AHCI configuration

Hello again,

Thanks to all of you for your help so far. I hope you can provide me with some guidance on this AHCI (possibly configuration?) issue.

With Linux TQMa6x BSP Rev.0101, my AHCI device (a Plextor solid state drive) that I had connected to the TQ system was detected. (I show output from various command in the section below labeled "Previous BSP (0101) Version.")

In the newest BSP (Rev. 0104) this same AHCI device is NOT detected. Can you tell if I should expect AHCI to work on Rev 0104 and if so, what changes I must make to get it working?

Here is the result of several commands on the BSP 0104 system (OSELAS(R)-TQC-ARM-BSP-QT-20140718 / MBa6x-20140718 ptxdist-2013.12.0/2014-09-16T16:48:08-0600)       

"uname -a" reports:

Linux MBa6x 3.16.0-rc5-20140718 #4 SMP Wed Sep 10 18:44:53 MDT 2014 armv7l GNU/x

"lspci -m" and "lspci -k" produce no output -- nothing at all.

In /dev, I see no sd* devices.

After installing a BSP 0104 image on the TQ system, these are the files with "ahci" in the name that are installed:

root <at> MBa6x:/ find . -name "*ahci*" -print

This is in contrast to what I see when I install the image I built with BSP version 0101. Here is its output:

--------------------- Previous BSP (0101) Version --------------------------
"uname -a" reports:

Linux MBa6x 3.13.0-ptx20140211-1+tq20140224 #2 SMP Thu Aug 21 16:23:01 MDT 2014x

In /dev, I see:

root <at> MBa6x:/dev file sda
sda: block special 

root <at> MBa6x:/dev file sda1 
sda1: block special

"lspci" reports:

root <at> MBa6x:/dev lspci -m
"lspci" reports:
00:00.0 "Class 0604" "16c3" "abcd" "0000" "0000" 
01:00.0 "Class 0106" "1c28" "0122" "1b4b" "9183" 

root <at> MBa6x:/dev lspci -k
00:00.0 Class 0604: 16c3:abcd pcieport
01:00.0 Class 0106: 1c28:0122 ahci 

Files with "ahci" in the name:

root <at> MBa6x:/ find . -name "ahci*" -print

"hdparm /dev/sda" reports:
hdparm: HDIO_GET_MULTCOUNT: Inappropriate ioctl for device
 IO_support  =2129103361 (???) 

hdparm: HDIO_GET_UNMASKINTR: Inappropriate ioctl for device
hdparm: HDIO_GET_DMA: Inappropriate ioctl for device
hdparm: HDIO_GET_KEEPSETTINGS: Inappropriate ioctl for device
 readonly =  0 (off)
 readahead= 256 (on)
 geometry = 15566/255/63, sectors = 250069680, start = 0 

"parted -l /dev/sda" reports:

Model: ATA PLEXTOR PX-G128M (scsi)
Disk /dev/sda: 128GB
Sector size (logical/physical): 512B/512B   
Partition Table: msdos  

Number  Start   EndSize    TypeFile system  Flags
 1 1049kB  68.7GB  68.7GB  primary  ext4

Error: /dev/mmcblk1rpmb: unrecognised disk label 
Error: /dev/mmcblk1boot0: unrecognised disk label
Error: /dev/mmcblk1boot1: unrecognised disk label

Model: SD  (sd/mmc)
Disk /dev/mmcblk0: 8010MB
Sector size (logical/physical): 512B/512B   
Partition Table: msdos  

Number  Start   EndSize    TypeFile system  Flags
 1 4194kB  8389kB  4194kB  primary
 2 8389kB  277MB   268MB   primary  ext3
 3 277MB   545MB   268MB   primary

Model: MMC MMC04G (sd/mmc)   
Disk /dev/mmcblk1: 3842MB
Sector size (logical/physical): 512B/512B   
Partition Table: msdos  

Number  Start   EndSize    TypeFile system  Flags
 1 4194kB  8389kB  4194kB  primary
 2 8389kB  277MB   268MB   primary  ext3
 3 277MB   545MB   268MB   primary

Smart Monitoring tools (smartctl) reports:

root <at> MBa6x:/dev smartctl -a /dev/sda 
smartctl 5.42 2011-10-20 r3458 [armv7l-linux-3.13.0-ptx20140211-1+tq20140224] ()
Copyright (C) 2002-11 by Bruce Allen, http://smartmontools.sourceforge.net
Device Model:  PLEXTOR PX-G128M6e 
Serial Number: P02426105455 
LU WWN Device Id: 5 002303 10023e8ab 
Firmware Version: 1.03
User Capacity: 128,035,676,160 bytes [128 GB] 
Sector Size:512 bytes logical/physical
Device is:  Not in smartctl database [for details use: -P showall] 
ATA Version is:7
ATA Standard is:  ATA/ATAPI-7 T13 1532D revision 4a 
Local Time is: Thu Jan  1 02:11:07 1970 UTC
SMART support is: Available - device has SMART capability.
SMART support is: Enabled
SMART overall-health self-assessment test result: PASSED
General SMART Values: 
Offline data collection status:  (0x00) Offline data collection activity 
 was never started.
 Auto Offline Data Collection: Disabled.
Self-test execution status:(0) The previous self-test routine completed
 without error or no self-test has ever 
 been run.
Total time to complete Offline 
data collection: (10) seconds. 
Offline data collection
capabilities:  (0x15) SMART execute Offline immediate. 
 No Auto Offline data collection support.
 Abort Offline collection upon new
 No Offline surface scan supported.
 Self-test supported. 
 No Conveyance Self-test supported.
 No Selective Self-test supported.
SMART capabilities:(0x0002) Does not save SMART data before
 entering power-saving mode.
 Supports SMART auto save timer.
Error logging capability:  (0x01) Error logging supported.
 General Purpose Logging supported.
Short self-test routine
recommended polling time:  (1) minutes. 
Extended self-test routine
recommended polling time:  (  10) minutes. 
SCT capabilities:  (0x003d) SCT Status supported.
 SCT Error Recovery Control supported.
 SCT Feature Control supported.
 SCT Data Table supported.
SMART Attributes Data Structure revision number: 1
Vendor Specific SMART Attributes with Thresholds:
  1 Raw_Read_Error_Rate  0x0003100100070 Pre-fail  Always 0
  5 Reallocated_Sector_Ct0x0003100100000 Pre-fail  Always 0
  9 Power_On_Hours 0x0002100100000 Old_ageAlways 0


Thanks for any advice you can provide.

-- Cindy

Cindy Gold | 10 Sep 01:05 2014

Need fstrim program installed


I have built the TQ board support package TQMa6x-BSP-REV.0104 using:

   --- ptxdist-2013.12.0
   --- OSELAS.Toolchain-2013.12.2

And "uname" now reports:

Linux MBa6x 3.16.0-rc5-20140718 #2 SMP Tue Sep 9 11:56:23 MDT 2014 armv7l GNU/Linux

I need the program "fstrim" for use with solid state drives. Please forgive me if I am posting to the wrong list, but I am unsure about where the trouble lies.

After building, I see these files on my host system under the directory TQMa6x-BSP-REV.0104:


After I burn the image (uboot_tqma6s_hd.img) and install it on my TQ system, the fstrim command is apparently not installed. Can you tell me why it is not getting installed on the target TQ system and how I get it installed?

Thank you!

Nick Wernicke | 5 Sep 05:33 2014

Has anyone ever installed PTXDist on Zynq7020

Going to build ptxdist for a zynq7020 dev kit.

Before I attempt this, I am curious if anyone has ever installed PTXDist on the Zynq7020? Xilinx has their own xilinx-gnu toolchain. I'm going to try the Xilinx toolchain first, but if it doesn't work I will try the arm-softfloat-linux-gnu- one. 

Most importantly: is there any possibility that I will break my silicon doing this? Did someone try it and blow up a Zynq or find something else bad that makes this a waste of time?

Guillermo Rodriguez Garcia | 1 Sep 13:11 2014

[mini2440] Replacing udev with mdev

Hello all,

I am playing around to see if I can replace udev with mdev for the mini2440.

Most of it is working but as soon as I remove udev, I also lose the
/dev/fd, /dev/stdin, /dev/stdout, /dev/stderr entries

I have looked through the udev rules (default and mini2440 specific)
and also through the init scripts, and I don't see where these are
defined. There is one place at the /etc/init.d/udev script where these
links are created manually, but this code is not run if devtmpfs is
mounted (which is the case here)

Can anyone help?

Thank you,

Guillermo Rodriguez Garcia

Guillermo Rodriguez Garcia | 26 Aug 17:25 2014

mini2440: NOR flash access from barebox

Hello all,

I am playing with the NOR flash of the mini2440 from barebox and
noticed that the protection status seems to be ignored.

This can be seen with the following test:

mini2440:/ addpart /dev/nor0 512k(dummy0),512k(dummy1)
mini2440:/ protect /dev/nor0.dummy1
cfi_protect: protect 0x00080000 (size 524288)

mini2440:/ devinfo cfi_flash0
base  : 0x00000000
size  : 0x00200000
driver: cfi_flash

CFI conformant FLASH (16 x 16)  Size: 2 MB in 35 Sectors
  AMD Standard command set, Manufacturer ID: 0x1C, Device ID: 0x49
  Erase timeout: 16384 ms, write timeout: 512 us

  Sector Start Addresses:
  00000000        00004000        00006000        00008000        00010000
  00020000        00030000        00040000        00050000        00060000
  00070000        00080000   RO   00090000   RO   000A0000   RO   000B0000   RO
  000C0000   RO   000D0000   RO   000E0000   RO   000F0000   RO   00100000
  00110000        00120000        00130000        00140000        00150000
  00160000        00170000        00180000        00190000        001A0000
  001B0000        001C0000        001D0000        001E0000        001F0000
no parameters available

[ The protected sectors have been properly marked as "RO", however.... ]

mini2440:/ md -s /dev/nor0.dummy1 0+32
00000000: ea0000d8 e59ff014 e59ff014 e59ff014                ................
00000010: e59ff014 e59ff014 e59ff014 e59ff014                ................
mini2440:/ erase /dev/nor0.dummy1
mini2440:/ md -s /dev/nor0.dummy1 0+32
00000000: ffffffff ffffffff ffffffff ffffffff                ................
00000010: ffffffff ffffffff ffffffff ffffffff                ................

Not sure if I have done something wrong, but before I dive further
into the code, is this expected?

I am using OSELAS.BSP-Pengutronix-Mini2440-2013.10.0

Thank you,

Guillermo Rodriguez Garcia

Juergen Borleis | 19 Aug 09:11 2014

Re: Could you assist pleas - one wire ts

Hi MC,

please let us continue our discussion at our public mailing list.

On Tuesday 19 August 2014 00:14:46 MC Potgieter wrote:
> Only saw your email now. Did try newpackage target earlier today but it was
> conflicting with the existing tslib.

After changing one package (e.g. replacing the rule file coming with PTXdist by 
your own BSP-local one) it needs to re-create the whole rootfs (simple way) or 
to re-build all enabled packages which depends on the (replaced) tslib.

> For experimental purposes, how do I remove a package once I've added it?

Just disable it in the menu, run a "ptxdist clean root; ptxdist go" (note 
the 'clean root', not(!) only 'clean'). This will re-assemble the rootfs from 
the already built packages but will omit the now disabled package.

Or if you want to revert the replacement of a package, just remove the 
BSP-local rule file.

If you do your development via NFS root, handling a workflow like this is very 



Pengutronix e.K.                              | Juergen Borleis             |
Industrial Linux Solutions                    | http://www.pengutronix.de/  |

Cindy Gold | 14 Aug 01:51 2014

PTXDist Application Note Kernel Driver Development out of date?


I would like to experiment with building an "out-of-tree" driver and have browsed the PTXDist Application Note "Kernel Driver Development". My version of the board support package (TQMa6x-BSP-REV.0101) doesn't seem to have any of the .../rules/templates files that are mentioned in the document.

So, is the Application Note out of date or is the Application Note not applicable to the TQMa6x?

Any help would be appreciated!

Thank you!