gmane.os.ecos.general

LWIP_ASSERT("pbuf_free: p->ref > 0", p->ref > 0);

Elad Yosef — 2013-06-18T05:25:35

Hi all,
I'm getting this error under TCP down-stream traffic.
I'm using ecos-3.0 with fix I added for thread-safety
(SYS_ARCH_UN/PROTECT is spin_lock).
What can be the cause for this assert?

Thanks

Elad

CANbus driver for stm3220g_eval platform

Oleg Peregudov — 2013-06-11T13:51:15

Hi,

are there any CANbus drivers in eCOS, which I can use for STM3220g EVAL board?

Thanks,
Oleg

Wrong character set in Cofiguration Tool

Buehler, Fabian — 2013-06-07T12:12:56

Hi
The console output of the configuration tool prints out strange symbols. It looks like that there is a wrong character set used! Has anybody an idea how to fix that?

I'm running the configuration tool on Windows XP with cygwin!

Regards Fabian

Migrating to flash_v2

Rodolphe Kos — 2013-06-04T15:55:06

Hi, I'm trying to migrate my system to ecos 3.0 because I've read that it 
includes flash_v2. I want to be able to define two different flash families 
(Atmel and AMD) for the same target and select the right one in run-time. 

First, I'm trying to use the Legacy package so I won't modify my application 
code, but it doesn't seem to work. cyg_io_get_config is returning "Unknown 
error".

Any suggestion?

Re: GCC toolchain with HW floating-point enabled

Davide Pippa — 2013-06-04T12:21:56

I had passed the flags "-mcpu=cortex-m4 -mthumb -mfloat-abi=hard
-mfpu=vfpv4-d16" to both compiler and linker (which both are
arm-eabi-g++"), but problem was still there.

I got things compiled this way:
I had to copy all the libraries located in:
gnutools/arm-eabi/lib/gcc/arm-eabi/4.6.3/thumb/thumb2/fpu/fpv4spd16
into the directory:
gnutools/arm-eabi/lib/gcc/arm-eabi/4.6.3/
where the libraries that are linked by gcc are located.

I don't know exactly if it is the same issue, but googling I found
that there is some issue with gcc linker in the selection of the
libraries, suggesting various workarounds...






2013/6/4 Ilija Kocho <ilijak< at >siva.com.mk>:

Re: GCC toolchain with HW floating-point enabled

Ilija Kocho — 2013-06-04T11:17:21

[SNIP]



GCC will select correct libraries on a base of floating point flags.
Therefore you need to provide same floating point flags for compilation
and linking.

HTH

Ilija

Re: GCC toolchain with HW floating-point enabled

Ilija Kocho — 2013-06-04T11:12:49

Hi Davide

Probably you have some files compiled with -mfloat-abi-soft and some
with -mfloat-abi=hard. They are not compatible. Try to compile all files
with same floating point flags.

Ilija

On 04.06.2013 12:41, Davide Pippa wrote:

Re: GCC toolchain with HW floating-point enabled

Davide Pippa — 2013-06-04T10:41:05

Hi!

I tried to update ecos to latest cvs and rebuilding the whole rom, but
I still have that problem.
I was already using the GCC test release that you pointed me.
What it seems to me is that I'm linking to the wrong version of libgcc
anyway (a version not supporting hardware floating point), as the
problems arise in libgcc (linking log points to bpapi, _divdi3,
_udivdi3 functions):

/cygdrive/h/Works/Arm/PyIde/tools/gnutools/arm-eabi/bin/../lib/gcc/arm-eabi/4.6.3/../../../../arm-eabi/bin/ld:
error: demo1.elf uses VFP register arguments,
/cygdrive/h/Works/Arm/PyIde/tools/gnutools/arm-eabi/bin/../lib/gcc/arm-eabi/4.6.3/libgcc.a(bpabi.o)
does not
makefile:59: recipe for target `demo1.elf' failed
/cygdrive/h/Works/Arm/PyIde/tools/gnutools/arm-eabi/bin/../lib/gcc/arm-eabi/4.6.3/../../../../arm-eabi/bin/ld:
failed to merge target specific data of file
/cygdrive/h/Works/Arm/PyIde/tools/gnutools/arm-eabi/bin/../lib/gcc/arm-eabi/4.6.3/libgcc.a(bpabi.o)
/cygdrive/h/Works/Arm/PyIde/tools/gnutools/arm-eabi/bin/../lib/

lwIP sequential Assert cyg_Mbox

Felix R — 2013-06-04T09:10:24

Hi,

the sequential http-test fails with an assert.

ASSERT FAIL: <5>mboxt.inl[79]void Cyg_Mboxt<T, QUEUE_SIZE>::wakeup_waiter()
Bad thread pointer
or:
ASSERT FAIL: <5>mboxt.inl[605]cyg_bool Cyg_Mboxt<T, QUEUE_SIZE>::tryput()
Bad this pointer

I cannot figure out the reason. can you help me?

trace:
Thread [1] (Suspended: Signal 'SIGINT' received. Description: Interrupt.)
13 cyg_assert_fail() buffer.cxx:748 0x00114a78
12 Cyg_Mboxt<void*, 10>::wakeup_waiter() mboxt.inl:79 0x0011e8d4
11 Cyg_Mboxt<void*, 10>::tryput() mboxt.inl:624 0x0011e804
10 Cyg_Mbox::tryput() mbox.cxx:156 0x0011d678
9 cyg_mbox_tryput() kapi.cxx:899 0x00117804
8 sys_mbox_trypost() sys_arch.c:288 0x0013031c
7 tcpip_input() tcpip.c:336 0x0012fca4
6 lwip_eth_drv_input() sequential.c:314 0x00130be8
5 eth_drv_recv() eth_drv.c:431 0x001243e8
4 zynq_eth_poll() xilinx_zynq_ethernet_controller.c:1177 0x00104cc8
3 zynq_eth_deliver() xilinx_zynq_ethernet_controller.c:1071 0x00104770
2 eth_thread() sequential.c:282 0x00130b64

RE: arm-eabi-gdb crashes

Buehler, Fabian — 2013-06-04T07:20:20

Hi John
Thanks for the information! Now the gdb is running!

Regards
Fabian Buehler


-----Original Message-----
From: John Dallaway [mailto:john< at >dallaway.org.uk]
Sent: Wednesday, May 29, 2013 6:30 PM
To: Buehler, Fabian
Cc: eCos Discussion
Subject: Re: arm-eabi-gdb crashes

Hi Fabian

On 29/05/13 15:44, Buehler, Fabian wrote:


The missing DLL is provided by the "libncurses8" Cygwin package. This is normally installed as a dependency of one of the other Cygwin packages required for eCos development. Regardless, you should be able to add the package to your existing Cygwin installation using the Cygwin setup tool.


There is a more recent test release of the eCos arm-eabi GNU tools. Ref:

  http://ecos.sourceware.org/ml/ecos-discuss/2012-06/msg00047.html

I hope this helps...

John Dallaway
eCos maintainer
http://www.dallaway.org.uk/john

Aw: Re: Re: eCos application crashes

Lukas Riezler — 2013-06-03T23:35:03

Hi Ilja,
 
I think you are right!!!
It seems that the situation for FM3 is the same like for kinetis!
==> I forgot to disable the hardware watchdog!
Now it seems to run! :-)

Thank you very very much for this hint!

Lukas
 
 

Gesendet: Dienstag, 04. Juni 2013 um 00:07 Uhr
Von: "Ilija Kocho" <ilijak< at >siva.com.mk>
An: "Lukas Riezler" <lukas.riezler< at >gmx.net>
Cc: ecos-discuss< at >sourceware.org
Betreff: Re: Aw: Re: [ECOS] eCos application crashes
On 03.06.2013 22:14, Lukas Riezler wrote:

Try to imagine... Check whether your code tries access on some illegal
location.

Is your hardware properly initialized? Some devices like Kinetis have
watchdog that is enabled after reset so you must disable it before u do
anything. Does FM3 have one?

Ilija



--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos[http://ecos.sourceware.org/fom/ecos]
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss[http://ecos.sourceware.org/ml/ecos-discuss]

Re: Aw: Re: eCos application crashes

Ilija Kocho — 2013-06-03T22:07:59

Try to imagine... Check whether your code tries access on some illegal
location.

Is your hardware properly initialized? Some devices like Kinetis have
watchdog that is enabled after reset so you must disable it before u do
anything. Does FM3 have one?

Ilija

Aw: Re: eCos application crashes

Lukas Riezler — 2013-06-03T20:14:28

Hi Ilja,

I think it's the port which crashes...because the application contains only a simple counter.
I did the clocking and hal_diag.c - are there any traps causing me stumble?The clocking I did is on the same way like it is described in examples of FM3 (where the clocks also get initialized). So I can't imagine that my fault depends on clocking.

Lukas
 
 

Gesendet: Montag, 03. Juni 2013 um 21:49 Uhr
Von: "Ilija Kocho" <ilijak< at >siva.com.mk>
An: "Lukas Riezler" <lukas.riezler< at >gmx.net>
Cc: ecos-discuss< at >sourceware.org
Betreff: Re: [ECOS] eCos application crashes
On 03.06.2013 21:31, Lukas Riezler wrote:

Is it the application or the port that chrashes? Have you tried the
application on some mature port?

HTH
Ilija


--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss[http://ecos.sourceware.org/ml/ecos-discuss]

Re: eCos application crashes

Ilija Kocho — 2013-06-03T19:49:46

Is it the application or the port that chrashes? Have you tried the
application on some mature port?

HTH
Ilija

eCos application crashes

Lukas Riezler — 2013-06-03T19:31:15

Hi eCos-guys!

I have a problem with my fm3 port.
When I start an eCos application first it seems to run fine...but after ~1 second it crashes and restarts.
Do you have any idea what could be the reason for this behaviour?
I've tried to compile eCos without optimization ==> the first running was okay! The application did run fine!!!
But when I started it a few minutes later it crashed again...

I'll become desperate :/
 
With kind regards, Lukas

Ask WiFi module development driver approach

LiuJett — 2013-05-31T07:51:27

Hi,

I'm going to use the WiFi module (Atheros AR6003G), eCos currently no drive, what should I do develop drivers do?

Thanks.
Jett Liu

Re: arm-eabi-gdb crashes

John Dallaway — 2013-05-29T16:29:43

Hi Fabian

On 29/05/13 15:44, Buehler, Fabian wrote:


The missing DLL is provided by the "libncurses8" Cygwin package. This is
normally installed as a dependency of one of the other Cygwin packages
required for eCos development. Regardless, you should be able to add the
package to your existing Cygwin installation using the Cygwin setup tool.


There is a more recent test release of the eCos arm-eabi GNU tools. Ref:

  http://ecos.sourceware.org/ml/ecos-discuss/2012-06/msg00047.html

I hope this helps...

John Dallaway
eCos maintainer
http://www.dallaway.org.uk/john

Re: RTC doesn't work?

Felix R — 2013-05-29T15:09:05

i found a issue in your code, but i this is only a typo in the demo:

while(1) 
{ 
HAL_READ_UINT32(CYGARC_REG_SYSTICK_BASE+CYGARC_REG_SYSTICK_VALUE,
timerVal0); 
timerVal*2*= cyg_current_time(); 
  
printf("%u\n", timerVal0); //it's okay...i counts down! 
printf("%u\n", timerVal*1*); //always 0 :-( 
} 





--
View this message in context: http://sourceware-org.1504.n7.nabble.com/RTC-doesn-t-work-tp233422p233568.html
Sent from the Sourceware - ecos-discuss mailing list archive at Nabble.com.

Re: arm-eabi-gdb crashes

Felix R — 2013-05-29T15:02:28

Have you installed the gdb-package for cygwin?

Of course you can use newer toolchains. Look if your microcontrollers
producer offers a newer version.



--
View this message in context: http://sourceware-org.1504.n7.nabble.com/arm-eabi-gdb-crashes-tp233562p233566.html
Sent from the Sourceware - ecos-discuss mailing list archive at Nabble.com.

arm-eabi-gdb crashes

Buehler, Fabian — 2013-05-29T14:44:37

Hi,
I tried to use the gdb which is part of the eCos installation within cygwin. So it should be obvious that I'm using Windows (XP). If I tried to execute the gdb by calling it inside the command line or the cygwin shell it crashes with the following message:

$ arm-eabi-gdb
/opt/ecos/gnutools/arm-eabi/bin/arm-eabi-gdb.exe: error while loading shared libraries: cygncurses-8.dll: cannot open shared object file: No such file or directory

So does anybody have an idea about that issue?

Anyway I want to ask if there is a newer gnu toolchain than the one provided with the ecos3.0 installation!


Kind Regards
Fabian Buehler

RTC doesn't work?

Lukas Riezler — 2013-05-28T19:41:33

Hi ecos-guys :)
 
I've a question depending on the real time clock in eCos (I try a port for FM3).
It seems it isn't working!
 
The returned value of cyg_current_time() is always 0!
It also seemy cyg_thread_delay(100) doesn't work...
 
I tried a "hard" read of the value of the systick register, and this isn't 0!
 
while(1)
{
HAL_READ_UINT32(CYGARC_REG_SYSTICK_BASE+CYGARC_REG_SYSTICK_VALUE, timerVal0);
timerVal2= cyg_current_time();
 
printf("%u\n", timerVal0); //it's okay...i counts down!
printf("%u\n", timerVal1); //always 0 :-(
}

Any ideas where my problems are?

With kind regards, Lukas