This is the mail archive of the ecos-discuss@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Re: Re: AT91EB40 RedBoot install problems - solution !


My problem with installing RedBoot in AT91EB40 has been resolved. I suspect
it is not hardware related. My host is Windows 2000 based. I ended up using
the Multi2000 debugger to talk to the AT91EB40 via Angel. With the Multi2000
debugger, I downloaded the redboot_RAM.elf and ran it. It worked the very
first time !! At this point, I have to suspect that my version of the GDB
was not able to download the redboot_RAM.elf correctly to the AT91EB40 V2.0.
May be it has some problems talking to the Angel monitor.

Now my new problem is in building the library with configtool :-( I loaded
eCOS 1.3.1 and have used CVS Windows client to get the latest eCOS source
code. Also the configuration tool has been updated to "1.3.net Oct 10 2001".


However, the ecos.db has not been updated as there is AT91EB40 when I tried
to select "Build -> Templates -> Hardware". Any idea how to add the AT91EB40
? 

I also got a lot of warning messages when I bring up the eCos Configuration
Tool :

ecos.db, package CYGPKG_DEVICES_WALLCLOCK: warning Version subdirectory
`mips' does not have a CDL script `wallclock.cdl'.

If I select the "ARM development board (PID)" and do a build library, I got
the following errors:

arm-elf-gcc -c  -I/ecos-c/eCOSWork/ARMPID_install/include
-I/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1
-I/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/src
-I/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/tests -I.
-I/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/src/sync/
-mcpu=arm7tdmi -Wall -Wpointer-arith -Wstrict-prototypes -Winline -Wundef
-Woverloaded-virtual -g -O2 -ffunction-sections -fdata-sections -fno-rtti
-fno-exceptions -fvtable-gc -finit-priority -Wp,-MD,src/sync/cnt_sem.tmp -o
src/sync/kernel_cnt_sem.o
/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/src/sync/cnt_sem.cxx
arm-elf-gcc -c  -I/ecos-c/eCOSWork/ARMPID_install/include
-I/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1
-I/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/src
-I/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/tests -I.
-I/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/src/sync/
-mcpu=arm7tdmi -Wall -Wpointer-arith -Wstrict-prototypes -Winline -Wundef
-Woverloaded-virtual -g -O2 -ffunction-sections -fdata-sections -fno-rtti
-fno-exceptions -fvtable-gc -finit-priority -Wp,-MD,src/sync/flag.tmp -o
src/sync/kernel_flag.o
/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/src/sync/flag.cxx
/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/src/sync/flag.cxx:62:
default
   argument given for parameter 1 of `Cyg_Flag::Cyg_Flag(unsigned int = 0)'
/ecos-c/eCOSWork/ARMPID_install/include/cyg/kernel/flag.hxx:102: after
previous 
   specification in `Cyg_Flag::Cyg_Flag(unsigned int = 0)'
/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/src/sync/flag.cxx:125:
default
   argument given for parameter 1 of `void Cyg_Flag::maskbits(unsigned int =
0)
   '
/ecos-c/eCOSWork/ARMPID_install/include/cyg/kernel/flag.hxx:110: after
previous 
   specification in `void Cyg_Flag::maskbits(unsigned int = 0)'
/ecos-c/PROGRA~1/REDHAT~1/eCos/packages/kernel/v1_3_1/src/sync/flag.cxx:149:
default
   argument given for parameter 1 of `void Cyg_Flag::setbits(unsigned int =
-1)
   '
/ecos-c/eCOSWork/ARMPID_install/include/cyg/kernel/flag.hxx:106: after
previous 
   specification in `void Cyg_Flag::setbits(unsigned int = -1)'
make[1]: *** [src/sync/flag.o.d] Error 1
make[1]: Leaving directory `/ecos-c/eCOSWork/ARMPID_build/kernel/v1_3_1'
make: *** [build] Error 2
make: Leaving directory `/ecos-c/eCOSWork/ARMPID_build'




On Tue, 10 Sep 2002, J Sevy wrote:

> Hi Raymond,
> well, the board is really glitchy with regard to getting RedBoot to run 
> from RAM! Once you get the image into ROM, it works flawlessly, but of 
> course you need to get it to run from RAM first to load the image to
flash.


I have 2 EB40's - one "glitchy" one not. I never could get the FLASH
programmed correctly in the "glitchy" one. The good one works fine. Then
one day my good one was "glitchy". The glitchiness was manifested as bogus
data coming from the UART (at least minicom didn't like it). Then I
noticed my power supply was adjusted to 6.5 Volts. I turned it up to 7.2
and it worked just fine. Now, I probably should go back and looked at the
glitch board...



-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]