Michael Schulte | 13 Apr 14:47 2016

ELDK5.8 errors during installation


Although I am no programmer, why not trying to install ELDK5.8 on OpenSuse 13.2.
After downloading and mounting the iso image I got here:

./install.sh -d /home/sc/work/ELDK/eldk5.8 armv7a
Error: SDK image "gmae" is not available
Available SDK images for target 'armv7a':
        'toolchain', 'toolchain-qte', 'toolchain-xenomai-qte'.

The default setting for the SDK "gmae" in the install script is not
available, so I chose one of the available SDK images and got:

./install.sh -d /home/sc/work/ELDK/eldk5.8 -s toolchain-qte armv7a
*** Installing ./targets/armv7a/eldk-glibc-i686-arm-toolchain-qte-5.8.sh
into /home/sc/work/ELDK/eldk5.8/armv7a
NOTICE: superuser priviledges will be needed to install the
root file system; make sure you have sufficient permissions.
Error: Can't read rootfs tarball

First assuming a permission problem I then reviewed the install script and
found that the tarball variable isn't set properly.

		tarball=$(echo ${SRC_DIR}/targets/${target}/\
(Continue reading)

Rama Chandra | 21 Mar 07:05 2016

Fwd: request for information on a reference document of Porting Linux for AMCC 440gx Ref Platform


Recently i am started Embedded Linux, Could you have about any information
or reference document of Porting Linux for AMCC 440gx Ref Platform
Thanks & Regards,
RamaChandra PC


Thanks & Regards,
RamaChandra PC
Albrecht Dreß | 18 Feb 20:53 2016

glibc getaddrinfo() bug (CVE-2015-7547)

Hi all,

short question - is the glibc version (2.21?) coming with ELDK 5.8 affected by the recently published glibc
getaddrinfo() bug CVE-2015-7547 [1]?  If so, will you provide a patched version of '5.8 (like 5.8.1), or do
we have to re-compile glibc with a fix [2] ourselves?

IMO, this bug is a really critical one, much worse than CVE-2015-0235 aka 'GHOST' which strikes the
obsolescent (though still used by some older applications) gethostbyname() function only.

I still use ELDK 5.4 on two PowerPC platforms (MPC5200; P2020) which *is* vulnerable on both according to
the proof-of-concept [3].  This in turn means that *any* system built with ELDK 5.4 (and earlier and later
versions?) is also vulnerable if any application running on it uses getaddrinfo() - which is /very/ likely.

As ELDK 5.8 now comes with gcc 4.9.1 which should have the issue described in [4] fixed, this would be the
perfect time to move to the new ELDK, if CVE-2015-7547 is fixed.

Any insight would be highly appreciated!

Thanks in advance,

[1] <https://googleonlinesecurity.blogspot.de/2016/02/cve-2015-7547-glibc-getaddrinfo-stack.html>
[2] <https://sourceware.org/ml/libc-alpha/2016-02/msg00416.html>
[3] <https://github.com/fjserna/CVE-2015-7547>
[4] <http://lists.denx.de/pipermail/eldk/2014-October/002548.html>
eldk mailing list
(Continue reading)

Rama Chandra | 15 Feb 08:23 2016

Supported Target Architectures

Hai DENX Members,

      I am using IBM PPC440GX it will supports eldk Supported Target
Architectures or not and which ELDK is useful for IBM PPC440GX (which one i
can select ftp://ftp.denx.de/pub/eldk/5.6/targets/powerpc-4xx/)


Thanks & Regards,
RamaChandra PC
Thomas Lange | 12 Feb 15:04 2016

Cannot compile using stdlibs with 5.8

there seems to be a problem with gcc system paths
on my system when running ELDK v5.8 on Debian/Jessie:

   k1:/tmp$ arm-linux-gnueabi-gcc -Wall x.c
   x.c:1:19: fatal error: stdio.h: No such file or directory
    #include <stdio.h>

gcc says it is configured with
which seems weird to me.

I would appreciate any hints, I am kind of stuck here.

Full example:

/* First a patch to make install.sh work with debian */

k1:/tmp/work$ diff -Naur /mnt/work/install.sh install.sh -w
--- /mnt/work/install.sh        2016-02-03 11:17:25.000000000 +0100
+++ install.sh  2016-02-12 07:04:10.000938733 +0100
 <at>  <at>  -288,7 +288,7  <at>  <at> 
                 echo -e "*** Installing ${shar}\n    into ${inst_path}" >&2

                 echo -e "${inst_path}\ny" | \
-               ${DRY_RUN} sh ${shar} >/dev/null || exit 1
+               ${DRY_RUN} bash ${shar} >/dev/null || exit 1
(Continue reading)

Thomas Lange | 12 Feb 09:54 2016

[PATCH] eldk-switch.sh: Handle empty lines in environment script

ELDK v5.8 includes an empty line in environment-setup-*
which caused this:

  k33:~$ eval `eldk-switch.sh -r 5.8 armv7a-hf`
  Setup for armv7a-hf (using ELDK 5.8)
  bash: syntax error near unexpected token `;'

So remove all empty lines before substituting <newline> with ;

Signed-off-by: Thomas Lange <thomas@...>
 eldk-switch.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/eldk-switch.sh b/eldk-switch.sh
index dcdc193..252065a 100755
--- a/eldk-switch.sh
+++ b/eldk-switch.sh
 <at>  <at>  -346,7 +346,7  <at>  <at>  else
 	# Use our pruned path to add the new path in our environment
 	pathcmd=$(cat ${config} | grep " PATH=")
 	eval $pathcmd
-	cmds=$(cat ${config} | grep -v " PATH=" | sed 's/$/ ; /g')
+	cmds=$(cat ${config} | grep -v " PATH=" | sed '/^$/d; s/$/ ; /g')
 	# We want to reference ${TARGET_PREFIX}, so evaluate it
 	eval $(cat ${config} | grep "TARGET_PREFIX=")
 	# Built minimal set of variables, i.e. PATH, CROSS_COMPILE and ARCH

(Continue reading)

Thomas Lange | 11 Feb 17:20 2016

5.8 install fails on Debian

installing 5.8 on Debian (Jessie) fails:

root <at> k33:/mnt/loop# ./install.sh -r lsb-sdk -s toolchain
*** Installing ./targets/armv7a-hf/eldk-glibc-i686-arm-toolchain-5.8.sh
into /opt/eldk-5.8/armv7a-hf
./targets/armv7a-hf/eldk-glibc-i686-arm-toolchain-5.8.sh: 72: read: 
Illegal option -e

The reason seems to be:

1. install.sh uses 'sh' to run the toolchain shell script:

     ${DRY_RUN} sh ${shar} >/dev/null || exit 1

   On Debian, 'sh' means 'dash' (https://wiki.debian.org/Shell).

2. eldk-glibc-i686-arm-toolchain-5.8.sh doesn't work with dash.

Workarounds that seems to work for me:

A) ${DRY_RUN} bash ${shar} >/dev/null || exit 1

B) ${DRY_RUN} ./${shar} >/dev/null || exit 1

(Continue reading)

Wolfgang Denk | 10 Feb 23:30 2016

[STATUS] ELDK v5.8 released


this is to let you know that ELDK v5.8  is available:

- from the git repository, see tag "eldk-5.8" (in branch
- from the Amazon Cloud Drive, see [1]

This release is based on Yocto v1.8.1 ("fido"). ELDK v5.8 introduces
recent tools (like GCC 4.9.2, Binutils 2.24) and software (like U-Boot
v2015.10, Linux v3.18.12 LTS and Qt v4.8.6). 

[1] https://www.amazon.com/clouddrive/share/ZDYTL7dwcLT1lkNoHzM4D7UoTJ4ulmPKFQBSW8IuG0I

Best regards,

Wolfgang Denk


DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@...
Laughter and tears are meant to turn the wheels of the same machinery
of sensibility; one is wind-power, and the other water-power.
                                          - Oliver Wendell Holmes, Sr.
Albrecht Dreß | 19 Nov 20:55 2015

[maybe ot] PPC e500v2 floating-point in eldk 5.4

Hi all,

I use ELDK 5.4 to develop software for a PowerPC e500v2 (QorIQ P2020) system, and have a (hopefully not too
off-topic) question regarding the floating-point implementation in glibc.

Although I did not enable any of the C99 floating-point exceptions, one of my applications throws SIGFPE,
according to the backtrace in exp() (__exp_finite).  Unfortunately, I do not have a core image, so I cannot
check which input value triggered the signal (which is wrong anyway, IMHO, as no exception is enabled).

When I tried to reproduce the effect with a stripped-down version of the code (which didn't work, i.e. I
could /not/ provoke the signal), I noticed that exp(NAN) returns 1.79769e+308.  If I understand IEEE-754
correctly, this is wrong, as any operation involving NAN shall result in NAN.  Actually, running exp(NAN)
on a PC (Ubuntu 14.04 w/ libc 2.19) correctly returns NAN.

I'm not sure if these two effects are linked, but might this be an indication that the exp() implementation
is somehow broken for the e500v2?

Thanks in advance for any insight,
eldk mailing list
Nick Leverton | 17 Nov 10:26 2015

What's loaded in the FPGA on an MCVEVK ?

I'm trying to understand what is loaded in the FPGA on the MCVEVK as it comes 
out-of-the-box.  It doesn't seem to be documented what if any functionality 
this provides to my evaluation kit.

I think that something must be loaded into it because 
/sys/class/fpga/fpga0/status returns "user mode" but what, and at which of the 
several boot stages is it loaded and where from, and where does it fit into 
ELDK and the RFS image generation ?

Hope someone can clarify this,

KwangMin Kim | 20 Oct 13:18 2015

Compile error in eldk(ppc_6xx-g++)

Hi. I'm elementary developer in korea.

I was ever good use "eldk". 

During this time, it was replaced with a new computer, it was supposed to be transferred to the "/ opt" 
        folder of the existing installation that has been "eldk" is also a new computer.

Problems arising By trying to compile transferred to the source code is also new computer programs that 
        are developed using "eldk" from the old computer.

c compiler (ppc_6xx-gcc), which have a normal operation, it is possible to see that compile error occurs in
cpp compiler.

please What I missed part? (need a package, and etc...)

Compile error is here.

/* ...ellipsis...

.rebased /opt/eldk/usr/bin/ppc_6xx-g++ -DHAVE_CONFIG_H -I.
-I../../../../../tools/shared/linux -I ../../..  -I../../../../../tools/shared/linux/../
-I../../../../../wireless/shared -I../../../../../wireless/shared/linux
-I../../../../../tools/shared/3rd_party/iniparser -I../../../../../tools/mttools/shared
-DMTLK_PACKAGE_VERSION="\"\"" -include
-Wall -Wno-multichar -include
(Continue reading)