uClinux development discussion

nunya si | 13 Sep 02:23

iptables not working in 2.6.37?

I have uClinux version 2.6.37-uc0 running on coldfire 5275.

I try to use iptables by enabling iptables in userspace (which report
as iptables v1.4.3.2) as well as:
netfilter, advanced netfilter configuration and packet filtering in
kernel config.

I even go crazy and select every option under netfilter and then

If I enter simple command:
iptables -P INPUT DROP

I get response: iptables: Invalid argument. Run `dmesg' for more information.

or if I run more complicated command:
iptables -A INPUT -i eth0 -d 192.168.1.1 -j ACCEPT
and get:
iptables: No chain/target/match by that name

Only command that work is iptables -L (shows empty IMPUT, FORWARD,
OUTPUT tables)

no matter what command I give iptables it does not work... What
special has to be done to make it work in new kernel?

Is there compatibility problem with userspace and new kernel?

Thank You,
Nunya Si
_______________________________________________

(Continue reading)

headers

Ingrid GUE | 12 Sep 09:15

smc91x

Hi all,

I am looking for drivers for Linux Lan91c1x SMSC version 2.6 for my Ethernet card.

I use a new version (2.6) distribution uclinux on this site: http://www.uclinux.org/pub/uClinux/dist/ card for my type of Freescale ColdFire M5307C3 is already working on an older version (the 2.4).
After compiling my new distribution, I loaded the kernel image (image.elf) SDRAM on my card. Except that I can not ping other user on the same network. So I checked the config Ethernet with the ifconfig command, I found that eth0 was not active. I told him assigned a static address because I have no dhcp server but its not working.
So I reloaded the image. Elf from the old version (2.4). Everything is working properly and the LED at the Ethernet controller is turned on during charging while when I load the new version, the LED remains off.
I do not know just where the problem apparently because the controller is not faulty as it works with the older version.
My controller is Ethernet-type SMSC LAN91C111

Any help is really appreciated.

regards,

Ingrid

_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

headers

Angelo Dureghello | 10 Sep 17:50

Re: [HELP] coldfire, bss.variable offset greater than __bss_end

Hi all,

i reply just to close the thread and to help each other that can have similar issues, this since i have seen
other boards that can have the same issue.

The issue was related to my custom board, and more properly was that the linker added some other .bss.*
variables after __bss_end__ , that of course u-boot was not relocating into sdram. Error was in my
board-related u-boot.lds:

WRONG

  __bss_start = .;
  .bss (NOLOAD)       :
  {
   _sbss = .;
   *(.sbss)
   *(.bss)  <----
   *(COMMON)
   . = ALIGN(4);
   _ebss = .;
  }

CORRECT

  __bss_start = .;
  .bss (NOLOAD)       :
  {
   _sbss = .;
   *(.sbss*)
   *(.bss*)
   *(COMMON)
   . = ALIGN(4);
   _ebss = .;
  }

regards,
angelo

-------- Original Message --------
Subject: [HELP] coldfire, bss.variable offset greater than __bss_end
Date: Sat, 10 Sep 2011 00:41:58 +0200
From: Angelo Dureghello <angelo70 <at> gmail.com>
To: U-Boot <at> lists.denx.de <U-Boot <at> lists.denx.de>

Hi all,

after a recent update and compiling u-boot, bootloader get locked trying to set a global variable, just
after sdram relocation.

Exactly, the program lock here:

1529  void mem_malloc_init(ulong start, ulong size)
1530  {
1531	mem_malloc_start = start;   <<<---program paralize here

Tracing the variable offset in memory, i have seen this variable is located a little bit outside (higher
address) of the total 16M SDRAM memory size.

Investigating further, in the map file, the issue seems to be that u-boot assign a memory space that is:

len = (ulong)&__bss_end__ - CONFIG_SYS_MONITOR_BASE;

But from the map file the variable seems to have an offset greater than __bss_end :

.bss.mem_malloc_start
                0xffc159c8        0x4
 .bss.mem_malloc_start
                0xffc159c8        0x4 common/libcommon.o
                0xffc159c8                mem_malloc_start

.....

.bss            0xffc13e00     0x1544
                0xffc13e00                _sbss = .
 *(.sbss)
 *(.bss)
 *(COMMON)
 COMMON         0xffc13e00        0x4 arch/m68k/lib/libm68k.o
                0xffc13e00                monitor_flash_len

u-boot.lds has been checked many times, i don't see any issue on it.

I am using this compiler:

m68k-elf-gcc --version
m68k-elf-gcc (GCC) 4.2.4
Copyright (C) 2007 Free Software Foundation, Inc.

Any help is really appreciated.

regards,
angelo
_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

headers

Quiliro Ordóñez | 6 Sep 21:12

Kyocera 7135 and 100% libre software

Hi.

I found this distro and would like to thank the developers.

I would like to use it to install only the free (as in freedom) parts to 
a Kyocera 7135 smart phone. It has a Motorola 68VZ328 processor that 
works without MMU. I understand that uCLinux works with this hardware.

Will you please give me guidance about how to connect to this device? I 
suspect I should connect with minicom through the USB or serial 
interface supplied by the phone. Is this correct? Is there a howto about 
that?

If you can give me some pointers about the nonfree parts that you know 
exist in uCLinux it would help me free the OS also.

Thank you very much for your help and keep trucking! 
-- 
Quiliro Ordóñez
09 821 8696
02 340 1517

"No se puede sacrificar la libertad por ningún bien, por ninguna promesa 
de pan o de paz o de justicia, porque ese pan tendría amargura de 
veneno, esa paz sería de muerte, y esa justicia no sería justicia humana 
ni tendría sentido." Alfredo Pérez Guerrero

"Não se pode sacrificar a liberdade por nenhum bem, por nenhuma promessa 
de pan ou de paz ou de justiça, porque esse pan teria amargura de 
veneno, essa paz seria de morte, e essa justiça não seria justiça humana 
nem faria sentido." Alfredo Pérez Guerrero
_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

headers

dhlm83 | 5 Sep 17:01

telnetd: All network ports in use

Hello.

I am using kernel 2.4.34.5-uco, uclinux distro 20110603.

I build successful an Image to NetBurner MOD5272.

But i cont run the telnetd.

I have it in inetd.conf:
         telnet  stream tcp nowait root /bin/telnetd
in start-up telnetd don't start...

If I try start manually i have the following error message:
...
# telnetd
telnetd: getpeername: Socket operation on non-socket
telnetd: All network ports in use.
...

Someone had this problem or similar?

Thanks.
Domingos Goncalves
_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

headers

Angelo Dureghello | 4 Sep 21:53

Coldfire GPIO question

Dear all,

i am running uClinux but with a main line kernel 2.6.36.2 and trying to get 13 GPIO lines of my MCF5307, to do
some tests driving an LCD display, as below:

static struct gpio ks0108_gpios[] = {
{  0, GPIOF_OUT_INIT_LOW, "D0" },
{  1, GPIOF_OUT_INIT_LOW, "D1" },
{  2, GPIOF_OUT_INIT_LOW, "D2" },
{  3, GPIOF_OUT_INIT_LOW, "D3" },
{  4, GPIOF_OUT_INIT_LOW, "D4" },
{  5, GPIOF_OUT_INIT_LOW, "D5" },Linux version 
{  6, GPIOF_OUT_INIT_LOW, "D6" },
{  7, GPIOF_OUT_INIT_LOW, "D7" },
{ 10, GPIOF_OUT_INIT_LOW, "CS1" },
{ 11, GPIOF_OUT_INIT_LOW, "CS2" },
{ 12, GPIOF_OUT_INIT_LOW, "R_W" },
{ 13, GPIOF_OUT_INIT_LOW, "D_I" },
{ 14, GPIOF_OUT_INIT_LOW, "E" },
};

gpio_request_array(ks0108_gpios, 13);

The gpio_request_array() return without any error, but seems i can set correctly levels only lines PP0 to
PP3, not 4to7, 10,11,12,13,14.

Am i doing something wrong ?

Thanks,
angelo
_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

headers

Brent Weatherall | 2 Sep 20:17

At91 and skyeye config files

Hello again. I have just built the uClinux 2.4.x kernel for the Atmel AT91 (using Snapgear toolchain arm-linux-tools-20061213). I want to run this kernel with skyeye. I can successfully run the skye-testsuite AT91 (uclinux_cs8900a) with wonderful results. Warm fuzzy feeling and all that. The reason I am trying to rebuild the kernel is that the current testsuite example (uclinux_cs8900a) only has about 1.5MB RAM and my tests need around 13 MB RAM. When I switch to the kernel I just built, the memory address alignment no longer matches the config file values. I tried matching some values from skyeye.conf (below) to values in the 'make menuconfig' options for DRAM offset and size etc..., but I am not easily seeing the mapping between these values and what is present in the config file:

mem_bank: map=M, type=RW, addr=0x00000000, size=0x00004000

mem_bank: map=M, type=RW, addr=0x01000000, size=0x00400000

mem_bank: map=M, type=R, addr=0x01400000, size=0x00400000, file=./romfs.img

mem_bank: map=M, type=RW, addr=0x02000000, size=0x00400000

mem_bank: map=M, type=RW, addr=0x02400000, size=0x00008000

mem_bank: map=M, type=RW, addr=0x04000000, size=0x00400000

mem_bank: map=I, type=RW, addr=0xf0000000, size=0x10000000

I am currently looking through the skyeye source (skyeye-1.3.2_rc1) for how the config file is parsed and what the mappings are, but this is looking to be an arduous task atm. If anyone has any good high level pointers on how to map kernel settings to match up with config file values or more directly how to increase the RAM avail to the skyeye VM it would greatly help out. I am a bit new to skyeye so there may something glaringly obvious I am as yet not aware of. Thanks in advance if anyone can provide some tips or documentation which clears this up.

Looking around the web it appears there are multiple suggestions on toolchain and uclinux-dist versions from various sites, blogs and forums. I will probably try these out as well.

Regards,

Brent Weatherall

_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

headers

Brent Weatherall | 1 Sep 01:55

Cross compiling uClinux 2.4.x for arm-nommu arch

Hello. I am trying to cross-compile uClinux kernel 2.4.x for the Atmel AT91 board. I AM able to compile uClinux 2.6.x successfully for the Atmel board; however, I need 2.4.x to compile.

Compiling 2.6.x I simply set CROSS_COMPILE in the menuconfig and went about my way.

It appears less straightforward in the 2.4.x build. To override the hard-coded "CROSS_COMPILE" and "KERNEL_CROSS_COMPILE" I modified uClinux-dist/vendors/config/armnommu.arch so I could override the cross compilation variables.

Then I ran:

CROSS_COMPILE=/usr/local/arm-uclinux-gcc-3.4.3/bin/arm-uclinux-elf- KERNEL_CROSS_COMPILE=/usr/local/arm-uclinux-gcc-3.4.3/bin/arm-uclinux-elf- make linux

Where:

ls /usr/local/arm-uclinux-gcc-3.4.3/bin/

arm-elf-addr2line arm-elf-flthdr arm-elf-objcopy arm-uclinux-elf-addr2line arm-uclinux-elf-flthdr arm-uclinux-elf-ld arm-uclinux-elf-run

arm-elf-ar arm-elf-g++ arm-elf-objdump arm-uclinux-elf-ar arm-uclinux-elf-g++ arm-uclinux-elf-ld.real arm-uclinux-elf-size

arm-elf-as arm-elf-gcc arm-elf-ranlib arm-uclinux-elf-as arm-uclinux-elf-gcc arm-uclinux-elf-nm arm-uclinux-elf-strings

arm-elf-c++ arm-elf-gcov arm-elf-readelf arm-uclinux-elf-c++ arm-uclinux-elf-gcc-3.4.3 arm-uclinux-elf-objcopy arm-uclinux-elf-strip

arm-elf-c++filt arm-elf-ld arm-elf-size arm-uclinux-elf-c++filt arm-uclinux-elf-gccbug arm-uclinux-elf-objdump genromfs

arm-elf-cpp arm-elf-ld.real arm-elf-strings arm-uclinux-elf-cpp arm-uclinux-elf-gcov arm-uclinux-elf-ranlib

arm-elf-elf2flt arm-elf-nm arm-elf-strip arm-uclinux-elf-elf2flt arm-uclinux-elf-gdb arm-uclinux-elf-readelf

Am I using the wrong toolchain version? So far I haven't found anything for sure specifying a specific version of the gnu arm-uclinux compiler toolset for uClinux kernel 2.4.x. I may just be missing it.

The error I see appears to start with the compilation of assembly units:

/usr/local/arm-uclinux-gcc-3.4.3//bin/arm-uclinux-elf-gcc -D__ASSEMBLY__ -D__KERNEL__ -I/home/Virtual_Machines/linux_kernels/uClinux-dist/linux-2.4.x/include -DNO_MM -mapcs-32 -march=armv4 -msoft-float -c -o entry-armv.o entry-armv.S

entry-armv.S: Assembler messages:

entry-armv.S:1495: Warning: destination register same as write-back base

entry-armv.S:1721: Error: undefined symbol TSS_FPESAVE used as an immediate value

The above text is the first error. There are many more assembly errors. Any help would be appreciated.

_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

headers

Angelo Dureghello | 28 Aug 11:26

samsung lcd support

Hi all,

is there a driver useable for Coldfire, for Samsung S6B0107 controller ? I tried to see the menuconfig
options but didn't seen it (almost not for Coldfire).

thanks,
angelo
_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

headers

Viet Nguyen | 25 Aug 18:37

Re: mcf5307, i2c rtc

On Wed, Aug 24, 2011 at 11:24 PM, angelo <angelo70 <at> gmail.com> wrote:

1- The device rtc0 must be created from genromfs ? I see some boards that have CONFIG_RTC_HCTOSYS_DEVICE="rtc0" but create a "rtc" instade, with any number.

2 - Do i hae to add some static structure in my board.c to enable rtc ds1307 driver ? Do you have a sample ?

1. rtc0 device node should be created in order to use hwclock, you don't need to create i2c device note by the way.
2. You must register your platform devices:

static struct platform_device rtc_dev_ = {
    .name = "rtc-ds1307",
    .id = -1,
};

static struct resource i2c_res_[] = {
    {
        .start      = MCFI2C_IOBASE,
        .end        = MCFI2C_IOBASE + MCFI2C_IOSIZE - 1,
        .flags      = IORESOURCE_MEM,
    },
    {
        .start      = MCFINT_VECBASE + MCFINT_I2C,
        .end        = MCFINT_VECBASE + MCFINT_I2C,
        .flags      = IORESOURCE_IRQ,
    },
};

static struct platform_device i2c_dev_ = {
   .name = "mcfi2c",
   .id = 0,
   .num_resources = ARRAY_SIZE(i2c_res_),
   .resource = i2c_res_,
};

static struct i2c_board_info __initdata i2c_info_[] = {
    {
        I2C_BOARD_INFO("ds1338", 0x68),
    },
};

static struct platform_device *devices_[] __initdata = {
    &rtc_dev_,
    &i2c_dev_,
};

...
in your arch_init:
i2c_register_board_info(0, i2c_info_, ARRAY_SIZE(i2c_info_));
platform_add_devices(devices_, ARRAY_SIZE(devices_));

PS: I used i2c driver of Steven King

Regrads,
--
Nguyen Quoc Viet

_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

headers

kylix liu | 24 Aug 12:41

Re: uClinux-dev Digest, Vol 100, Issue 14

unsubscribe

2011/8/20 <uclinux-dev-request <at> uclinux.org>

Send uClinux-dev mailing list submissions to
uclinux-dev <at> uclinux.org

To subscribe or unsubscribe via the World Wide Web, visit
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
or, via email, send a message with subject or body 'help' to
uclinux-dev-request <at> uclinux.org

You can reach the person managing the list at
uclinux-dev-owner <at> uclinux.org

When replying, please edit your Subject line so it is more specific
than "Re: Contents of uClinux-dev digest..."

Today's Topics:

1. Re: FEC and brctl after unplug in/out Ethernet cable (nunya si)

----------------------------------------------------------------------

Message: 1
Date: Thu, 18 Aug 2011 16:26:11 -0600
From: nunya si <nunya034144 <at> gmail.com>
Subject: Re: [uClinux-dev] FEC and brctl after unplug in/out Ethernet
cable
To: uclinux-dev <at> uclinux.org
Message-ID:
<CAJ6eg7A3=ZtD3juhSCWELzbnN9sPMQZ9kGyMBMoF5V6pEw2xGA <at> mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

On Wed, Aug 17, 2011 at 6:50 PM, Greg Ungerer <gerg <at> snapgear.com> wrote:
> Hi Nunya,
>
> On 17/08/11 02:22, nunya si wrote:
>>
>> I have project where I take new uClinux on a coldfire 5275 dev board
>> and bridge both Ethernet interfaces together with brctl.
>>
>> If Ethernet plugged in when I turn on the board everything work just
>> fine BUT if I turn on board and THEN plug in Ethernet after it has
>> booted then I can talk to the local address but nothing passes in/out
>> of the bridge. Also if I power board up with Ethernet and un-plug and
>> re-plug in Ethernet then same problem, talk locally but not across
>> bridge.
>>
>> Doing "ifconfig eth0 up" every time I plug in cable will cause
>> everything to work.
>>
>> I play with:
>>
>> fec_enet_adjust_link(struct net_device *dev) function in fec.c and if
>> run only once:
>>
>> static char rerun=0;
>> if(rerun++>0) return;
>>
>> ácan solve up/down/up issue, but not plug in after boot issue. Not
>> know how to run ifconfig from driver automatic.
>>
>> What can do to solve issue to make eth0... Work in bridge after plug
>> in/out Ethernet all by itself?
>
> Do you have either mii-tool or ethtool on your target?
> It would be good to know what they report as the link
> status on eth0.
>
> Regards
> Greg
>
>
>

Thank You Greg,

I install ethtool and report is as follows:

/> ethtool eth0
Settings for eth0:
Supported ports: [ TP MII ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
Advertised auto-negotiation: Yes
Speed: 100Mb/s
Duplex: Full
Port: MII
PHYAD: 1
Transceiver: external
Auto-negotiation: on
Link detected: yes

Also /proc/kmsg announces: "PHY: 1:01 - Link is Up - 100/Full

Also bridge reports after link up: br0: port 1(eth0) entering forwarding state

but still bridge is broken...

I have bad fix like this in fec.c last lines of function static void
fec_enet_adjust_link(struct net_device *dev):

if (status_change)
{
phy_print_status(phy_dev);
dev_ioctl(dev_net(dev), SIOCSIFFLAGS, "eth0"); // hack make eth0 work
}

And this fixes everything so far... no right way to fix but working.

Ideas?

Regards (and thank You),
Nunya Si

------------------------------

_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

End of uClinux-dev Digest, Vol 100, Issue 14
********************************************

_______________________________________________
uClinux-dev mailing list
uClinux-dev <at> uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev <at> uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

15	September 2011
78	August 2011
18	July 2011
22	June 2011
23	May 2011
33	April 2011
45	March 2011
51	February 2011
30	January 2011
33	December 2010
62	November 2010
67	October 2010
49	September 2010
56	August 2010
34	July 2010
106	June 2010
100	May 2010
149	April 2010
135	March 2010
76	February 2010
129	January 2010
142	December 2009
124	November 2009
87	October 2009
95	September 2009
215	August 2009
129	July 2009
117	June 2009
187	May 2009
86	April 2009
508	March 2009
244	February 2009
235	January 2009
130	December 2008
180	November 2008
192	October 2008
192	September 2008
85	August 2008
130	July 2008
176	June 2008
213	May 2008
501	April 2008
245	March 2008
416	February 2008
205	January 2008
287	December 2007
286	November 2007
281	October 2007
254	September 2007
278	August 2007
274	July 2007
257	June 2007
348	May 2007
287	April 2007
287	March 2007
354	February 2007
373	January 2007
301	December 2006
387	November 2006
298	October 2006
414	September 2006
407	August 2006
431	July 2006
515	June 2006
532	May 2006
370	April 2006
589	March 2006
553	February 2006
473	January 2006
399	December 2005
537	November 2005
376	October 2005
320	September 2005
18	August 2005
1	July 2005
15	June 2005
49	May 2005
5	April 2005
6	March 2005
7	February 2005
4	January 2005
23	December 2004
132	November 2004
14	October 2004
12	September 2004
4	August 2004
3	July 2004
10	June 2004
4	May 2004
1	April 2004
18	March 2004
15	February 2004
5	January 2004
17	December 2003
9	November 2003
11	October 2003
17	September 2003
49	August 2003
66	July 2003
22	June 2003
3	May 2003
5	April 2003
11	March 2003
9	February 2003
8	January 2003
5	December 2002
2	November 2002
313	October 2002
765	September 2002
656	August 2002
588	July 2002
1	January 2002

Mon	Tue	Wed	Thu	Fri	Sat	Sun

		1	2	3	4	5
6	7	8	9	10	11	12
13	14	15	16	17	18	19
20	21	22	23	24	25	26
27	28	29