Re: [oselas] Kernel panic after electricity was away.

2015-02-27 Thread Juergen Borleis
Hi Rudi,

On Thursday 26 February 2015 21:42:53 Lappies wrote:
> Like I said I have setup my SDCard with new Partition and formated it new.
> Copy my Backups back on the SDCard.
>
> Here are the output of your command.
> testbox:/home/rudi/kernel/linux-2.6.39.1# addr2line -e vmlinux 0xc00b139c
> /home/rudi/kernel/linux-2.6.39.1/fs/block_dev.c:1275

Hm, hardware issue?

> Doesn´t say me much. :-(
>
> Just to make sure I unterstand where things broke.
>
> The System boot from my uImage, get to the point to where it should load
> the root FS and then things broke up, correct?

Can you send us the whole boot log (from power on, until it fails)?

Regards,
Juergen

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



[oselas] Undefined symbols SOCK_NONBLOCK & SOCK_CLOEXEC

2015-02-27 Thread Doug Abbott
Still trying to build the Mini2440 BSP. ptxdist go fails trying to build 
udev-182/src/libudev-monitor.c saying the above symbols are undefined. 
I'm using the pre-built toolchain OSELAS.Toolchain-2014.12.0. Oddly, I 
find very few header files in that tree and I don't find sys/socket.h at 
all. Where is it getting the header files and why aren't those symbols 
declared?


Thanks, Doug

--
Doug Abbott
Principal Consultant, Intellimetrix
Linux for the Real World
575-590-2788

www.intellimetrix.us




[oselas] Undefined symbols SOCK_NONBLOCK & SOCK_CLOEXEC

2015-02-27 Thread Doug Abbott
OK, so I tracked down the header files to 
sysroot-arm-v4t-linux-gnueabi/usr/include. sys/socket.h includes 
bits/socket.h, which in turn includes bits/socket_type.h, which does in 
fact declare SOCK_NONBLOCK and SOCK_CLOEXEC. So what am I missing here?


--
Doug Abbott
Principal Consultant, Intellimetrix
Linux for the Real World
575-590-2788

www.intellimetrix.us




Re: [oselas] On Flash access timings calculation.

2015-02-27 Thread mind entropy
On Sat, Feb 21, 2015 at 12:21 PM, mind entropy  wrote:
> Hi,
>   I am trying to understand the flash access timing calculations in
> the barebox code and I have some doubts. ( I am a newbie in NAND flash
> ).
>
> In the code the comment says (in 
> arch/arm/boards/friendlyarm-mini2440/config.h)
>
> /*
>  * Flash access timings
>  * Tacls  = 0ns (but 20ns data setup time)
>  * Twrph0 = 25ns (write) 35ns (read)
>  * Twrph1 = 10ns (10ns data hold time)
>  * Read cycle time = 50ns
>  *
>  * Assumed HCLK is 100MHz
>  * Tacls = 1 (-> 20ns)
>  * Twrph0 = 3 (-> 40ns)
>  * Twrph1 = 1 (-> 20ns)
>  * Cycle time = 80ns
>  */
>
> I have a K9F2G08U0B 256MB NAND flash(
> http://115.28.165.193/down/datasheet/Peripheral/NandFlash/K9F2G08x0B.pdf
> ) . I know the HCLK cycle time is 10ns. (1/100MHz). Could you please
> help me out with the calculation of Tacls,Twrph0 and Twph1? What are
> the values/timing diagrams I should look in the NAND datasheet related
> to this?
>
> Thanks,
> Gautam

Can somebody please give some help here.

-Gautam.



[oselas] Latest mini2440BSP bash pidof problems

2015-02-27 Thread Dave Festing
Upgraded a system that has been working nicely with the previous BSP and 3.7
and having a few problems.

My main PHP script starts with:

#!/usr/bin/php5