409.B.GAMESS US with GPU support on debian wheezy --the ACML edition. This works.

Update 27/6/2013:
Please note that Kirill Berezovsky has published a series of posts on GAMESS US, including how to compile it for both CPU and GPU use. See
http://biochemicalmatters.blogspot.com.au/2013/06/gamess-us-frequently-asked-questions_26.html
http://biochemicalmatters.blogspot.ru/2013/06/gamess-us-frequently-asked-questions_1687.html
http://biochemicalmatters.blogspot.ru/2013/06/gamess-us-frequently-asked-questions_1447.html
http://biochemicalmatters.blogspot.com.au/2013/06/gamess-us-frequently-asked-questions.html


Update 21 May 2013: See the comments below this post. This approach most likely works -- what has been confusing me is the lack of reports of GPU timings in the output, but this doesn't necessarily mean that the GPU isn't being used. The poster below, using nvidia-smi, observed GPU usage, although the speed-up was not major.

Blogspot needs versioning.
I lost the entire post when it was almost complete. Screw this.

Everything compiles fine, but no GPU output during calculation.

I see no evidence of the GPU being used at any stage.  Otherwise all is good -- the calcs run fine on the CPU.

Maybe someone else will have a better idea.

I looked at libcchem/aaa.readme.1st and http://combichem.blogspot.com.au/2011/02/compiling-gamess-with-cuda-gpu-support.html to get as far as I did.

---

Setting up gamess
Get gamess (see e.g. http://verahill.blogspot.com.au/2012/09/compiling-and-testing-gamess-us-on.html). Put gamess-current.tar.gz in ~/tmp

sudo apt-get install libboost-all-dev build-essential g++ gfortran automake nvidia-cuda-toolkit python-cheetah openmpi-bin libopenmpi-dev zlib1g-dev checkinstall
mkdir ~/tmp
cd ~/tmp
tar xvf gamess-current.tar.gz
sudo mv gamess /opt/gamess_cuda
sudo chown $USER:$USER /opt/gamess_cuda -R

---

ACML
Download both the 'regular' and the int64 gfortran packages from AMD:
http://developer.amd.com/tools-and-sdks/cpu-development/amd-core-math-library-acml/acml-downloads-resources/#download

tar xvf acml-5-3-1-gfortran-64bit-int64.tgz
tar xvf acml-5-3-1-gfortran-64bit.tgz
sh install-acml-5-3-1-gfortran-64bit-int64.sh

Where do you want to install ACML?  Press return to use
the default location (/opt/acml5.3.1), or enter an alternative path.
The directory will be created if it does not already exist.
> /opt/acml/acml5.3.1

sh install-acml-5-3-1-gfortran-64bit.sh

Where do you want to install ACML?  Press return to use
the default location (/opt/acml5.3.1), or enter an alternative path.
The directory will be created if it does not already exist.
> /opt/acml/acml5.3.1

You'll get something like this:

/opt/acml/acml5.3.1
|-- Doc
|-- gfortran64
|-- gfortran64_fma4
|-- gfortran64_fma4_int64
|-- gfortran64_fma4_mp
|-- gfortran64_fma4_mp_int64
|-- gfortran64_int64
|-- gfortran64_mp
|-- gfortran64_mp_int64
`-- util

where
*  fma4 is for cpus with FMA4 support (use util/cpuid to check)
*  int64 is for double-precision float (integer*8) I think
*  mp is for openmp. For MPI do not use the _mp_ libraries!

Pick your library/ies and add them to the LD_LIBRARY_PATH, e.g.:

echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/acml/acml5.3.1/gfortran64_int64/lib' >> ~/.bashrc
source ~/.bashrc

---

CBLAS

cd /opt/netlib/
wget http://www.netlib.org/blas/blast-forum/cblas.tgz
tar xvf cblas.tgz
cd CBLAS/

Edit Makefile.LINUX

24   
25  BLLIB = /opt/acml/acml5.3.1/gfortran64_int64/lib/libacml.a
26  CBLIB = ../lib/cblas_$(PLAT).a
27  

cp Makefile.LINUX Makefile.in
make

---

patching libboost

sudo su
cd /usr/include/boost
patch -p1 < /opt/gamess_cuda/libcchem/boot/
exit

Make the following changes by hand if the patch didn't work:

/usr/include/boost/mpl/aux_/integral_wrapper.hpp

47 // other compilers (e.g. MSVC) are not particulary happy about it
48 #if BOOST_WORKAROUND(__EDG_VERSION__, <= 238) || defined(__CUDACC__)
49     typedef struct AUX_WRAPPER_NAME type;

/usr/include/boost/mpl/size_t_fwd.hpp

20 
21 BOOST_MPL_AUX_ADL_BARRIER_NAMESPACE_OPEN
22 #if defined(__CUDACC__)
23    typedef std::size_t std_size_t;
24    template< std_size_t N > struct size_t;
25 #else
 26    template< std::size_t N > struct size_t;
 27 #endif
28 
29 BOOST_MPL_AUX_ADL_BARRIER_NAMESPACE_CLOSE

/usr/include/boost/mpl/size_t.hpp

19 #if defined(__CUDACC__)
20   #define AUX_WRAPPER_VALUE_TYPE std_size_t
21   #define AUX_WRAPPER_NAME size_t    
22   #define AUX_WRAPPER_PARAMS(N) std_size_t N
23 #else 
24   #define AUX_WRAPPER_VALUE_TYPE std::size_t
25   #define AUX_WRAPPER_NAME size_t
26   #define AUX_WRAPPER_PARAMS(N) std::size_t N
27 #endif
28 

---

HDF5

mkdir ~/tmp
cd ~/tmp
wget http://www.hdfgroup.org/ftp/HDF5/current/src/hdf5-1.8.10-patch1.tar.gz
tar xvf hdf5-1.8.10-patch1.tar.gz
cd hdf5-1.8.10-patch1/
export CC=/usr/bin/gcc-4.6 && export CXX=/usr/bin/g++-4.6
./configure --prefix=/opt/gamess_cuda/hdf5 --with-pthread --enable-cxx --enable-threadsafe --enable-unsupported
make
mkdir /opt/gamess_cuda/hdf5/lib -p
mkdir /opt/gamess_cuda/hdf5/include -p
sudo checkinstall

This package will be built according to these values: 

0 -  Maintainer: [ root@neon ]
1 -  Summary: [ hdf5-cxx]
2 -  Name:    [ hdf5-1.8.10 ]
3 -  Version: [ 1.8.10-1 ]
4 -  Release: [ 1 ]
5 -  License: [ GPL ]
6 -  Group:   [ checkinstall ]
7 -  Architecture: [ amd64 ]
8 -  Source location: [ hdf5-1.8.10-patch1 ]
9 -  Alternate source location: [  ]
10 - Requires: [  ]
11 - Provides: [ hdf5-1.8.10 ]
12 - Conflicts: [  ]
13 - Replaces: [  ]

Make sure to edit the Version field since Patch-1 leads to an error (must start with digit).

---

LIBCCHEM
Edit /opt/gamess_cuda/libcchem/src/externals/boost/cuda/device_ptr.hpp and /opt/gamess_cuda/libcchem/rysq/src/externals/boost/cuda/device_ptr.hpp. Insert

#include <stddef.h>

somewhere at the beginning of each file.

./configure --with-gamess --with-hdf5=/opt/gamess_cuda/hdf5 CPPFLAGS="-I/opt/gamess_cuda/hdf5/include" --with-cuda=/usr --disable-openmp --prefix=/opt/gamess_cuda/libcchem --with-gpu=fermi --with-integer8 --with-cublas
make
make install

---

Configure GAMESS US

cd /opt/gamess_cuda
./config

please enter your target machine name: linux64
GAMESS directory? [/opt/gamess_cuda]
GAMESS build directory? [/opt/gamess_cuda]
Version? [00] 12
Please enter your choice of FORTRAN: gfortran
Please enter only the first decimal place, such as 4.1 or 4.6:  
4.6
Enter your choice of 'mkl' or 'atlas' or 'acml' or 'none': acml
enter this full pathname: /opt/acml/acml5.3.1
communication library ('sockets' or 'mpi')? mpi
Enter MPI library (impi, mvapich2, mpt, sockets): openmpi
Please enter your openmpi's location: /opt/openmpi/1.6

---

Compile

cd ddi/
./compddi
cd ..

Edit comp

872 #          see ~/gamess/libcchem/aaa.readme.1st for more information
873 set GPUCODE=true
874 if ($GPUCODE == true) then

and

1663 #           -fno-whole-file suppresses argument's data type checking
1664       set OPT='-O0'
1665       if (".$GMS_DEBUG_FLAGS" != .) set OPT="$GMS_DEBUG_FLAGS"

./compall

Edit lked

69 #
70 set GPUCODE=true
71 #
72 #   5. optional MPQC interface

and

958             case openmpi:
959                set MPILIBS="-L$GMS_MPI_PATH/lib"
960                set MPILIBS="$MPILIBS -lmpi -lpthread"
961                breaksw

and

1214 if ($GPUCODE == true) then
1215    echo "   Using 'libcchem' add-in C++ codes for Nvidia/CUDA GPUs."
1216    set GPU_LIBS="-L/opt/gamess_cuda/libcchem/lib -lcchem_gamess -lcchem -lrysq"
1217    set GPU_LIBS="$GPU_LIBS -lcudart -lcublas"
1218    ### GPU_LIBS="$GPU_LIBS -lcudart -lcublas"
1219    set GPU_LIBS="$GPU_LIBS /usr/lib/libboost_thread.a"
1220    set GPU_LIBS="$GPU_LIBS /opt/gamess_cuda/hdf5/lib/libhdf5.a"
1221    set GPU_LIBS="$GPU_LIBS /opt/gamess_cuda/hdf5/lib/libhdf5_cpp.a"
1222    set GPU_LIBS="$GPU_LIBS /opt/gamess_cuda/hdf5/lib/libhdf5_hl.a"
1223    set GPU_LIBS="$GPU_LIBS /opt/gamess_cuda/hdf5/lib/libhdf5.a"
1224    set GPU_LIBS="$GPU_LIBS /opt/acml/acml5.3.1/gfortran64_int64/lib/libacml.a /opt/netlib/CBLAS/lib/cblas_LINUX.a"
1225    set GPU_LIBS="$GPU_LIBS -lz"
1226    set GPU_LIBS="$GPU_LIBS -lstdc++"
1227    ### GPU_LIBS="$GPU_LIBS -lgomp"
1228    set GPU_LIBS="$GPU_LIBS -lpthread"
1229    echo "   libcchem GPU code's libraries are"
1230    echo "$GPU_LIBS"
1231 else

./lked gamess gpu.12

---

Run script
Create rungpu:

#!/bin/csh -v
set TARGET=mpi
set SCR=$HOME/scratch
set USERSCR=/scratch
set GMSPATH=/opt/gamess_cuda
set JOB=$1
set VERNO=$2
set NCPUS=$3
set PPN=$3
   @ NUMGPU=1
   if ($NUMGPU > 0) then
      @ NUMCPU = $NCPUS - 1
      echo libcchem kernels will use $NUMCPU cores and $NUMGPU GPUs per node...
      set echo
      setenv CCHEM_PROFILE 1
      setenv NUM_THREADS $NCPUS
      setenv GPU_DEVICES 0
      #--if ($NUMGPU == 0) setenv GPU_DEVICES -1
      #--if ($NUMGPU == 2) setenv GPU_DEVICES 0,1
      #--if ($NUMGPU == 4) setenv GPU_DEVICES 0,1,2,3
      #setenv LD_LIBRARY_PATH /share/apps/cuda/lib64:$LD_LIBRARY_PATH
      ###### LD_LIBRARY_PATH /usr/local/cuda/lib64:$LD_LIBRARY_PATH
      unset echo
   else
      echo NO GPU
      setenv GPU_DEVICES -1
   endif


if ( $JOB:r.inp == $JOB ) set JOB=$JOB:r
echo "Copying input file $JOB.inp to your run's scratch directory..."
cp $JOB.inp $SCR/$JOB.F05

setenv TRAJECT $USERSCR/$JOB.trj
setenv RESTART $USERSCR/$JOB.rst
setenv INPUT $SCR/$JOB.F05
setenv PUNCH $USERSCR/$JOB.dat
if ( -e $TRAJECT ) rm $TRAJECT
if ( -e  $PUNCH ) rm $PUNCH
if ( -e  $RESTART ) rm $RESTART
source $GMSPATH/gms-files.csh

setenv LD_LIBRARY_PATH /opt/openmpi/1.6/lib:/opt/netlib/CBLAS/lib:/opt/acml/acml5.3.1/gfortran64_int64/lib
set path= ( /opt/openmpi/1.6/bin $path )
/opt/openmpi/1.6/bin/mpiexec -n $NCPUS $GMSPATH/gamess.gpu.$VERNO.x|tee $JOB.out
cp $PUNCH .

chmod +x it to make it executable.

Add /opt/gamess_cuda to path:

echo 'export PATH=$PATH:/opt/gamess_cuda'
source ~/.bashrc

---

Testing

cd /opt/gamess_cuda/tests/standard
gpurun exam44 12 2

409.A.GAMESS US with GPU support on debian wheezy. This works (probably).

Update 27/6/2013:
Please note that Kirill Berezovsky has published a series of posts on GAMESS US, including how to compile it for both CPU and GPU use. See
http://biochemicalmatters.blogspot.com.au/2013/06/gamess-us-frequently-asked-questions_26.html
http://biochemicalmatters.blogspot.ru/2013/06/gamess-us-frequently-asked-questions_1687.html
http://biochemicalmatters.blogspot.ru/2013/06/gamess-us-frequently-asked-questions_1447.html
http://biochemicalmatters.blogspot.com.au/2013/06/gamess-us-frequently-asked-questions.html


Update 21 May 2013: See the comments below this post. This approach most likely works -- what has been confusing me is the lack of reports of GPU timings in the output, but this doesn't necessarily mean that the GPU isn't being used. The poster below this post, using nvidia-smi, observed GPU usage, although the speed-up was not major.

Update 10/05/2013: fixed libcchem compile.

Everything compiles fine and computations run fine and fast. To date there's only one other detailed step-by-step example of successful compilation of GAMESS with GPU support out there. At least based on google.

For various reasons I'm beginning to suspect that ATLAS isn't working out for me -- I've had issues getting things to converge with ATLAS, but which work fine with ACML (see post B).

I was in part following http://combichem.blogspot.com.au/2011/02/compiling-gamess-with-cuda-gpu-support.html and ./libcchem/aaa.readme.1st

This took a while to hammer out, so the write-up is a bit messy.

---

Set up

sudo apt-get install libboost-all-dev build-essential g++ gfortran automake nvidia-cuda-toolkit python-cheetah openmpi-bin libopenmpi-dev zlib1g-dev checkinstall
mkdir ~/tmp

Get gamess (see e.g. http://verahill.blogspot.com.au/2012/09/compiling-and-testing-gamess-us-on.html).

Put gamess-current.tar.gz in  ~/tmp

cd ~/tmp
tar xvf gamess-current.tar.gz
sudo mv gamess /opt/gamess_cuda
sudo chown $USER:$USER /opt/gamess_cuda -R


Preparing Boost

Edit /usr/include/boost/mpl/aux_/integral_wrapper.hpp

47 // other compilers (e.g. MSVC) are not particulary happy about it
 48 #if BOOST_WORKAROUND(__EDG_VERSION__, <= 238) || defined(__CUDACC__)
 49     typedef struct AUX_WRAPPER_NAME type;

Edit /usr/include/boost/mpl/size_t_fwd.hpp

20 
 21 BOOST_MPL_AUX_ADL_BARRIER_NAMESPACE_OPEN
 22 #if defined(__CUDACC__)
 23    typedef std::size_t std_size_t;
 24    template< std_size_t N > struct size_t;
 25 #else
 26    template< std::size_t N > struct size_t;
 27 #endif
 28 
 29 BOOST_MPL_AUX_ADL_BARRIER_NAMESPACE_CLOSE

Edit /usr/include/boost/mpl/size_t.hpp

 19 #if defined(__CUDACC__)
 20   #define AUX_WRAPPER_VALUE_TYPE std_size_t
 21   #define AUX_WRAPPER_NAME size_t    
 22   #define AUX_WRAPPER_PARAMS(N) std_size_t N
 23 #else 
 24   #define AUX_WRAPPER_VALUE_TYPE std::size_t
 25   #define AUX_WRAPPER_NAME size_t
 26   #define AUX_WRAPPER_PARAMS(N) std::size_t N
 27 #endif
 28 

---

HDF5
You'll have to compile that yourself for now since H5Cpp.h missing in the debian packages.(i.e. cxx support)

mkdir ~/tmp
cd ~/tmp
wget http://www.hdfgroup.org/ftp/HDF5/current/src/hdf5-1.8.10-patch1.tar.gz
tar xvf hdf5-1.8.10-patch1.tar.gz
cd hdf5-1.8.10-patch1/
export CC=/usr/bin/gcc-4.6 && export CXX=/usr/bin/g++-4.6
./configure --prefix=/opt/gamess_cuda/hdf5 --with-pthread --enable-cxx --enable-threadsafe --enable-unsupported
make
mkdir /opt/gamess_cuda/hdf5/lib -p
mkdir /opt/gamess_cuda/hdf5/include -p
sudo checkinstall

This package will be built according to these values: 

0 -  Maintainer: [ root@neon ]
1 -  Summary: [ hdf5-cxx]
2 -  Name:    [ hdf5-1.8.10 ]
3 -  Version: [ 1.8.10-1 ]
4 -  Release: [ 1 ]
5 -  License: [ GPL ]
6 -  Group:   [ checkinstall ]
7 -  Architecture: [ amd64 ]
8 -  Source location: [ hdf5-1.8.10-patch1 ]
9 -  Alternate source location: [  ]
10 - Requires: [  ]
11 - Provides: [ hdf5-1.8.10 ]
12 - Conflicts: [  ]
13 - Replaces: [  ]


Make sure to edit the Version field since Patch-1 leads to an error (must start with digit).


Openmpi 1.6

Can't remember why I ended up compiling it myself instead of using the stock debian version. From here.

sudo apt-get install build-essential gfortran
wget http://www.open-mpi.org/software/ompi/v1.6/downloads/openmpi-1.6.tar.bz2
tar xvf openmpi-1.6.tar.bz2
cd openmpi-1.6/

sudo mkdir /opt/openmpi/
sudo chown ${USER} /opt/openmpi/
./configure --prefix=/opt/openmpi/1.6/ --with-sge

make
make install



compiling libcchem

cd /opt/gamess_cuda/libcchem

edit /opt/gamess_cuda/libcchem/rysq/src/externals/boost/cuda/device_ptr.hpp

  4 #include <cstdlib>
  5 #include <iterator>
  6 #include <stddef.h>
  7 
  8 namespace boost {

Edit /opt/gamess_cuda/libcchem/src/externals/boost/cuda/device_ptr.hpp

  4 #include <cstdlib>
  5 #include <iterator>
  6 #include <stddef.h>
  7 
  8 namespace boost {
  9 namespace cuda {

./configure --with-gamess --with-hdf5=/opt/gamess_cuda/hdf5 CPPFLAGS="-I/opt/gamess_cuda/hdf5/include" --with-cuda=/usr --disable-openmp --prefix=/opt/gamess_cuda/libcchem --with-gpu=fermi --with-integer8 --with-cublas
make
make install

Configure Gamess US

Mainly follow this: http://verahill.blogspot.com.au/2012/09/compiling-and-testing-gamess-us-on.html

cd /opt/gamess_cuda
./config


please enter your target machine name: linux64

GAMESS directory? [/opt/gamess_cuda] /opt/gamess_cuda

Setting up GAMESS compile and link for GMS_TARGET=linux64
GAMESS software is located at GMS_PATH=/opt/gamess_cuda
 
Please provide the name of the build locaation.
This may be the same location as the GAMESS directory.
 
GAMESS build directory? [/home/me/tmp/gamess] 

Please provide a version number for the GAMESS executable.
This will be used as the middle part of the binary's name,
for example: gamess.00.x

Version? [00] 12r2

Please enter your choice of FORTRAN: gfortran

gfortran is very robust, so this is a wise choice.

Please type 'gfortran -dumpversion' or else 'gfortran -v' to
detect the version number of your gfortran.
This reply should be a string with at least two decimal points,
such as 4.1.2 or 4.6.1, or maybe even 4.4.2-12.
The reply may be labeled as a 'gcc' version,
but it is really your gfortran version.
Please enter only the first decimal place, such as 4.1 or 4.6:  
4.6


Enter your choice of 'mkl' or 'atlas' or 'acml' or 'none': atlas

Please enter the Atlas subdirectory on your system: /opt/ATLAS/lib

Math library 'atlas' will be taken from /opt/ATLAS
 

If you have an expensive but fast network like Infiniband (IB), and
if you have an MPI library correctly installed,
     choose 'mpi'.
 
communication library ('sockets' or 'mpi')? mpi

Enter MPI library (impi, mvapich2, mpt, sockets): openmpi



Please enter your openmpi's location: /opt/openmpi/1.6

---

Build Gamess US

cd /opt/gamess_cuda/ddi/
./compddi
cd ../

Edit comp

872 #          see ~/gamess/libcchem/aaa.readme.1st for more information
 873 set GPUCODE=true
 874 if ($GPUCODE == true) then

and

1663 #           -fno-whole-file suppresses argument's data type checking
1664       set OPT='-O0'
1665       if (".$GMS_DEBUG_FLAGS" != .) set OPT="$GMS_DEBUG_FLAGS"

./compall

Edit lked

69 #
  70 set GPUCODE=true
  71 #
  72 #   5. optional MPQC interface

and

958             case openmpi:
 959                set MPILIBS="-L$GMS_MPI_PATH/lib"
 960                set MPILIBS="$MPILIBS -lmpi -lpthread"
 961                breaksw

and

1214 if ($GPUCODE == true) then
1215    echo "   Using 'libcchem' add-in C++ codes for Nvidia/CUDA GPUs."
1216    set GPU_LIBS="-L/opt/gamess_cuda/libcchem/lib -lcchem_gamess -lcchem -lrysq"
1217    set GPU_LIBS="$GPU_LIBS -lcudart -lcublas"
1218    ### GPU_LIBS="$GPU_LIBS -lcudart -lcublas"
1219    set GPU_LIBS="$GPU_LIBS /usr/lib/libboost_thread.a"
1220    set GPU_LIBS="$GPU_LIBS /opt/gamess_cuda/hdf5/lib/libhdf5.a"
1221    set GPU_LIBS="$GPU_LIBS /opt/gamess_cuda/hdf5/lib/libhdf5_cpp.a"
1222    set GPU_LIBS="$GPU_LIBS /opt/gamess_cuda/hdf5/lib/libhdf5_hl.a"
1223    set GPU_LIBS="$GPU_LIBS /opt/gamess_cuda/hdf5/lib/libhdf5.a"
1224    set GPU_LIBS="$GPU_LIBS /opt/ATLAS/lib/libcblas.a"
1225    set GPU_LIBS="$GPU_LIBS -lz"
1226    set GPU_LIBS="$GPU_LIBS -lstdc++"
1227    ### GPU_LIBS="$GPU_LIBS -lgomp"
1228    set GPU_LIBS="$GPU_LIBS -lpthread"
1229    echo "   libcchem GPU code's libraries are"
1230    echo "$GPU_LIBS"
1231 else

./lked gamess gpu.12

Create gpurun

#!/bin/csh
set TARGET=mpi
set SCR=$HOME/scratch
set USERSCR=/scratch
set GMSPATH=/opt/gamess_cuda
set JOB=$1
set VERNO=$2
set NCPUS=$3

   @ NUMGPU=1
   if ($NUMGPU > 0) then
      @ NUMCPU = $NCPUS - 1
      echo libcchem kernels will use $NUMCPU cores and $NUMGPU GPUs per node...
      set echo
      setenv CCHEM_PROFILE 1
      setenv NUM_THREADS $NCPUS
      #--if ($NUMGPU == 0) setenv GPU_DEVICES -1
      #--if ($NUMGPU == 2) setenv GPU_DEVICES 0,1
      #--if ($NUMGPU == 4) setenv GPU_DEVICES 0,1,2,3
      #setenv LD_LIBRARY_PATH /share/apps/cuda/lib64:$LD_LIBRARY_PATH
      ###### LD_LIBRARY_PATH /usr/local/cuda/lib64:$LD_LIBRARY_PATH
      unset echo
   else
      setenv GPU_DEVICES -1
   endif


if ( $JOB:r.inp == $JOB ) set JOB=$JOB:r
echo "Copying input file $JOB.inp to your run's scratch directory..."
cp $JOB.inp $SCR/$JOB.F05

setenv TRAJECT $USERSCR/$JOB.trj
setenv RESTART $USERSCR/$JOB.rst
setenv INPUT $SCR/$JOB.F05
setenv PUNCH $USERSCR/$JOB.dat
if ( -e $TRAJECT ) rm $TRAJECT
if ( -e  $PUNCH ) rm $PUNCH
if ( -e  $RESTART ) rm $RESTART
source $GMSPATH/gms-files.csh

setenv LD_LIBRARY_PATH /opt/openmpi/lib:$LD_LIBRARY_PATH
set path= ( /opt/openmpi/bin $path )
mpiexec -n $NCPUS $GMSPATH/gamess.gpu.$VERNO.x|tee $JOB.out
cp $PUNCH .

echo 'export PATH=$PATH:/opt/gamess_cuda' >> ~/.bashrc
source ~/.bashrc
chmod +x gpurun
cd test/standard/
 gpurun exam44 12 2

The only evidence of GPU usage in the output is e.g. in exam44.out:

388           -----------------------
389           MP2 CONTROL INFORMATION
390           -----------------------
391           NACORE =        6  NBCORE =        6
392           LMOMP2 =        F  AOINTS = DUP
393           METHOD =        2  NWORD  =               0
394           MP2PRP =        F  OSPT   = NONE
395           CUTOFF = 1.00E-09  CPHFBS = BASISAO
396           CODE   = GPU
397 
398           NUMBER OF CORE -A-  ORBITALS =     6
399           NUMBER OF CORE -B-  ORBITALS =     6

but in the summary only CPU utilisation is mentioned.

---

I modified rungms:

me@neon:/opt/gamess_cuda/tests/standard$ diff /opt/gamess_cuda/gpurungms /opt/gamess/rungms 
59,62c59,62
< set TARGET=mpi
< set SCR=$HOME/scratch
< set USERSCR=/scratch
< set GMSPATH=/opt/gamess_cuda
---
> set TARGET=sockets
> set SCR=/scr/$USER
> set USERSCR=~$USER/scr
> set GMSPATH=/u1/mike/gamess
67d66
< set NNODES=1
513c512
< set PPN=$3
---
>    set PPN=$4
601c600
<          @ PPN2 = $PPN
---
>          @ PPN2 = $PPN + $PPN
742c741
<    @ NUMGPU=1
---
>    @ NUMGPU=0
752c751
< #      setenv LD_LIBRARY_PATH /share/apps/cuda/lib64:$LD_LIBRARY_PATH
---
>       setenv LD_LIBRARY_PATH /share/apps/cuda/lib64:$LD_LIBRARY_PATH
793c792,793
<       /opt/openmpi/1.6/bin/mpiexec -n $NPROCS $GMSPATH/gamess.$VERNO.x < /dev/null
---
>       mpiexec.hydra -f $PROCFILE -n $NPROCS \
>             /home/mike/gamess/gamess.$VERNO.x < /dev/null

408. Briefly: Tor on Debian -- the quick option

Tor can -- under the right conditions -- be used to anonymize your connection. Encryption, anonymity etc. is a minefield is you want to do it right, and I won't pretend to be an expert, so do your own reading.

Anyway.

In the process of looking at manually setting up Tor on Debian I came across the Tor browser bundle. Using it is pretty straightforward, but given that linux users are at varying skill-levels, a step by step guide with pictures can't hurt (and another post for me...).

sudo mkdir /opt/torbundle
sudo chown $USER:$USER /opt/torbundle
cd /opt/torbundle
wget https://www.torproject.org/dist/torbrowser/linux/tor-browser-gnu-linux-x86_64-2.3.25-6-dev-en-US.tar.gz
tar xvf tor-browser-gnu-linux-x86_64-2.3.25-6-dev-en-US.tar.gz
echo "alias torbrowser='/opt/torbundle/tor-browser_en-US/./start-tor-browser'" >> ~/.bashrc
source ~/.bashrc

Start by typing

torbrowser

Vidalia will open, and once you're connected to the tor network a browser session will automatically open.

Vidalia

407. Building less (458) as a temporary solution on Debian Jessie

Currently less conflicts with man-db/yelp/gnome-core/gnome on debian jessie. There are probably ways of overriding the conflict, but I prefer to simply compile my own less and install it.

Note that this doesn't take into account WHY less and man-db are listed as conflicting for versions of less below 4.5.6. I simply want less and the way to do it is to compile an approved version of less.

sudo apt-get install build-essential checkinstall
wget http://www.greenwoodsoftware.com/less/less-458.tar.gz
tar xvf less-458.tar.gz
cd less-458/
./configure
make
sudo checkinstall

0 -  Maintainer: [ root@niobium ]
1 -  Summary: [ less 4.5.8 ]
2 -  Name:    [ less ]
3 -  Version: [ 458 ]
4 -  Release: [ 1 ]
5 -  License: [ GPL ]
6 -  Group:   [ checkinstall ]
7 -  Architecture: [ amd64 ]
8 -  Source location: [ less-458 ]
9 -  Alternate source location: [  ]
10 - Requires: [  ]
11 - Provides: [ less ]
12 - Conflicts: [  ]
13 - Replaces: [  ]

406. Briefly: Missing package in debian wheezy -- forgot apt-pinning settings

In the off-chance that someone for some reason has made the same mistake as I have...

Wheezy is now stable and while that's all fine and dandy, when trying to install tor I kept on getting errors along the lines of

Reading package lists... Done
Building dependency tree       
Reading state information... Done
Package tor is not available, but is referred to by another package.
This may mean that the package is missing, has been obsoleted, or
is only available from another source

aptitude show tor

gave

No current or candidate version found for tor
Package: tor
State: not installed
Version: 0.2.3.25-1
Priority: optional

and

apt-cache policy tor

tor:
  Installed: (none)
  Candidate: (none)
  Version table:
     0.2.3.25-1 0
        -10 http://ftp.iinet.net.au/debian/debian/ wheezy/main amd64 Packages

I got similar errors for e.g. wine and virtualbox.
The solution is in the output of apt-cache -- I set up apt-pinning a long time ago (for mpich2?) and forgot about it.

cat /etc/apt/preferences

Package: *
Pin: release a=testing
Pin-Priority: 990

Package: *
Pin: release a=unstable
Pin-Priority: -10

Package: *
Pin: release a=stable
Pin-Priority: -10

Well, wheezy is now stable (and I am tracking wheezy only in my sources.list now) so the problem was quickly solved by simply deleting /etc/apt/preferences.

405. First breakage in Debian Jessie? less vs man-db

Update 12/5/13: less 456 is in the debian repos now so the breakage is resolved:
http://packages.debian.org/search?keywords=less

For some reason less (<456) and man-db conflict. man-db in turn is a requirement for yelp, which is a requirement for gnome-core which is a requirement for gnome. In other words, you currently have a choice between less or gnome.

More here: http://www.mail-archive.com/debian-bugs-dist@lists.debian.org/msg1117622.html

verahill@debianstd:~/tmp/poppler_build$ sudo apt-get install less

Reading package lists... Done
Building dependency tree       
Reading state information... Done
The following packages will be REMOVED:
  gnome gnome-core gnome-user-guide man-db yelp
The following NEW packages will be installed:
  less
0 upgraded, 1 newly installed, 5 to remove and 0 not upgraded.
Need to get 0 B/135 kB of archives.
After this operation, 29.4 MB disk space will be freed.
Do you want to continue [Y/n]?

404. Briefly: Debian Jessie now out (sort of, and in places)

update 6/5/13: The first upgrades and dist-upgrades are now in jessie. Nothing particularly exciting, beyond a new lsb_release package. As far as I can tell the gnome desktop background hasn't been touched either, but if memory serves me right the themes for the new stable is decided around the time of the freeze of testing, so there're another 2-3 years to go.

update: jessie is now at ftp.au.debian.org too

Original post
ftp.us.debian.org now has a copy of jessie, even though ftp.au.debian.org still doesn't.

This means that you can switch to the new testing (jessie) by editing your /etc/apt/sources.list:

deb http://ftp.us.debian.org/debian/ jessie main contrib non-free
deb http://www.deb-multimedia.org jessie main non-fre

At this point jessie is simply a copy of the 'old' testing, wheezy, so if you've got an up-to-date wheezy there are currently no updates involved in switching to jessie.

If I've understood things right sid was frozen at the same time as wheezy, so that it will take a little while before changes will occur in jessie since they first need to be introduced to sid, and then filter through.

Note that once updates start flowing into jessie the odd breakage might occur, so make sure to install apt-listbugs to get warnings about known bugs.

---

Upgrading from Squeeze to Wheezy
Upgrading from the old stable to the new stable is simple enough.

First make sure that Squeeze is up to date

sudo apt-get update && sudo apt-get upgrade && sudo apt-get dist-upgrade

Edit /etc/apt/sources.list and replace all instances of squeeze with wheezy. Then update, and download all updates before upgrading (-d).

sudo apt-get update && sudo apt-get dist-upgrade -d
sudo apt-get upgrade && sudo apt-get dist-upgrade

If you get an error about default-jre you can uninstall openjdk-6-jre and then run dist-upgrade again. It should work.

403. Kernel 3.9 on Debian Wheezy/Testing

Kernel 3.9 is out now -- here's how to build it on debian wheezy. Nothing odd in comparison to earlier versions and it barely warrants a separate post.

* To compile a kernel under Arch linux, see here: http://verahill.blogspot.com.au/2013/03/355-kernel-382-on-arch-linux-exploration.html

* To compile a kernel without kernel-package on debian, see here: http://verahill.blogspot.com.au/2013/02/344-compile-kernel-38-without-using-kpkg.html

So it begins

sudo apt-get install kernel-package fakeroot build-essential ncurses-dev
mkdir ~/tmp
cd ~/tmp
wget http://www.kernel.org/pub/linux/kernel/v3.0/linux-3.9.tar.bz2
tar xvf linux-3.9.tar.bz2
cd linux-3.9/
cat /boot/config-`uname -r`>.config
make oldconfig

You will be asked a lot of questions -- how many depends on what version you upgrade from. If in doubt, pick the default answer (i.e. hit enter). If really in doubt, use google.

Then continue:

make-kpkg clean

Do

make menuconfig

if you want to make any specific changes to the kernel (e.g. add support for certain devices)

Then continue:

time fakeroot make-kpkg -j4 --initrd kernel_image kernel_headers

As usual 4 is the number of threads you wish to launch -- make it equal to the number of cores that you have for optimum performance during compilation (more about that here).

Install:

sudo dpkg -i ../linux-image-3.9.0_3.9.0-10.00.Custom_amd64.deb ../linux-headers-3.9.0_3.9.0-10.00.Custom_amd64.deb

---

The new stuff
I know it's a bit lazy to simply post the questions as I do below, but...well, I don't have much of an excuse other than you having to figure out for yourself what you want to enable, and what you don't.:

  2. Full dynticks CPU time accounting (VIRT_CPU_ACCOUNTING_GEN) (NEW)
Intel Low Power Subsystem Support (X86_INTEL_LPSS) [N/y/?] (NEW) 
Early load microcode (MICROCODE_INTEL_EARLY) [Y/n/?] (NEW) 
  PCI slot detection driver (ACPI_PCI_SLOT) [N/y/?] (NEW) 
  Container and Module Devices (ACPI_CONTAINER) [Y/?] (NEW) y
Intel P state control (X86_INTEL_PSTATE) [N/y/?] (NEW) 
  "bpf" match support (NETFILTER_XT_MATCH_BPF) [N/m/?] (NEW) 
  "connlabel" match support (NETFILTER_XT_MATCH_CONNLABEL) [N/m/?] (NEW) 
  VLAN filtering (BRIDGE_VLAN_FILTERING) [N/y/?] (NEW) 
  MVRP (Multiple VLAN Registration Protocol) support (VLAN_8021Q_MVRP) [N/y/?] (NEW) 
Virtual Socket protocol (VSOCKETS) [N/m/y/?] (NEW) 
  Enable LED triggers for Netlink based drivers (CAN_LEDS) [N/y/?] (NEW) 
  8 devices USB2CAN interface (CAN_8DEV_USB) [N/m/?] (NEW) 
Fallback user-helper invocation for firmware loading (FW_LOADER_USER_HELPER) [Y/n/?] (NEW) 
  Command line partition table parsing (MTD_CMDLINE_PARTS) [N/m/?] (NEW) 
  IBM FlashSystem 70/80 PCIe SSD Device Driver (BLK_DEV_RSXX) [N/m/y/?] (NEW) 
Device driver for Atmel SSC peripheral (ATMEL_SSC) [N/m/y/?] (NEW) 
Lattice ECP3 FPGA bitstream configuration via SPI (LATTICE_ECP3_CONFIG) [N/m/y/?] (NEW) 
VMware VMCI Driver (VMWARE_VMCI) [N/m/y/?] (NEW) 
    SATA Zero Power Optical Disc Drive (ZPODD) support (SATA_ZPODD) [N/y/?] (NEW) 
    Cache target (EXPERIMENTAL) (DM_CACHE) [N/m/?] (NEW) 
      Broadcom 578xx and 57712 SR-IOV support (BNX2X_SRIOV) [Y/n/?] (NEW) 
      Intel(R) PCI-Express Gigabit adapters HWMON support (IGB_HWMON) [Y/n/?] (NEW) 
  ASIX AX88179/178A USB 3.0/2.0 to Gigabit Ethernet (USB_NET_AX88179_178A) [M/n/?] (NEW) 
    Intel Wireless WiFi MVM Firmware support (IWLMVM) [N/m/?] (NEW) 
  Cypress APA I2C Trackpad support (MOUSE_CYAPA) [N/m/?] (NEW) 
  Support 8250_core.* kernel options (DEPRECATED) (SERIAL_8250_DEPRECATED_OPTIONS) [Y/n/?] (NEW) 
Support for Synopsys DesignWare 8250 quirks (SERIAL_8250_DW) [N/m/y/?] (NEW) 
Comtrol RocketPort EXPRESS/INFINITY support (SERIAL_RP2) [N/m/y/?] (NEW) 
  STMicroelectronics ST33 I2C TPM (TCG_ST33_I2C) [N/m/?] (NEW) 
Intel iSMT SMBus Controller (I2C_ISMT) [N/m/?] (NEW) 
  PXA2xx SSP SPI master (SPI_PXA2XX) [N/m/y/?] (NEW) 
  Intel Lynxpoint GPIO support (GPIO_LYNXPOINT) [N/y/?] (NEW) 
Dual Channel Addressable Switch 0x3a family support (DS2413) (W1_SLAVE_DS2413) [N/m/?] (NEW) 
  Goldfish battery driver (BATTERY_GOLDFISH) [N/m/y/?] (NEW) 
  Maxim MAX6697 and compatibles (SENSORS_MAX6697) [N/m/?] (NEW) 
  TI / Burr Brown INA209 (SENSORS_INA209) [N/m/?] (NEW) 
  Fair-share thermal governor (THERMAL_GOV_FAIR_SHARE) [N/y/?] (NEW) 
  Step_wise thermal governor (THERMAL_GOV_STEP_WISE) [Y/?] (NEW) y
  User_space thermal governor (THERMAL_GOV_USER_SPACE) [N/y/?] (NEW) 
  Thermal emulation mode support (THERMAL_EMULATION) [N/y/?] (NEW) 
  Intel PowerClamp idle injection driver (INTEL_POWERCLAMP) [N/m/?] (NEW) 
  TI LP8755 High Performance PMU driver (REGULATOR_LP8755) [N/m/?] (NEW) 
  V4L2 int device (DEPRECATED) (VIDEO_V4L2_INT_DEVICE) [N/m/?] (NEW) 
    Support for various USB DVB devices v2 (DVB_USB_V2) [N/m/?] (NEW) 
      Cypress firmware helper routines (DVB_USB_CYPRESS_FIRMWARE) [N/m] (NEW) 
      Afatech AF9015 DVB-T USB2.0 support (DVB_USB_AF9015) [N/m/?] (NEW) 
      Afatech AF9035 DVB-T USB2.0 support (DVB_USB_AF9035) [N/m/?] (NEW) 
      Anysee DVB-T/C USB2.0 support (DVB_USB_ANYSEE) [N/m/?] (NEW) 
      Alcor Micro AU6610 USB2.0 support (DVB_USB_AU6610) [N/m/?] (NEW) 
      AzureWave 6007 and clones DVB-T/C USB2.0 support (DVB_USB_AZ6007) [N/m/?] (NEW) 
      Intel CE6230 DVB-T USB2.0 support (DVB_USB_CE6230) [N/m/?] (NEW) 
      E3C EC168 DVB-T USB2.0 support (DVB_USB_EC168) [N/m/?] (NEW) 
      Genesys Logic GL861 USB2.0 support (DVB_USB_GL861) [N/m/?] (NEW) 
      ITE IT913X DVB-T USB2.0 support (DVB_USB_IT913X) [N/m/?] (NEW) 
      MxL111SF DTV USB2.0 support (DVB_USB_MXL111SF) [N/m/?] (NEW) 
      Realtek RTL28xxU DVB USB support (DVB_USB_RTL28XXU) [N/m/?] (NEW) 
NXP Semiconductors TDA998X HDMI encoder (DRM_I2C_NXP_TDA998X) [N/m/?] (NEW) 
  Enable userspace modesetting on radeon (DEPRECATED) (DRM_RADEON_UMS) [N/y/?] (NEW) 
  Goldfish Framebuffer (FB_GOLDFISH) [N/m/y/?] (NEW) 
    Support new DSP code for CA0132 codec (SND_HDA_CODEC_CA0132_DSP) [N/y/?] (NEW) 
Steelseries SRW-S1 steering wheel support (HID_STEELSERIES) [N/m/?] (NEW) 
ThingM blink(1) USB RGB LED (HID_THINGM) [N/m/?] (NEW) 
  Xsens motion tracker serial interface driver (USB_SERIAL_XSENS_MT) [N/m/?] (NEW) 
  USB3503 HSIC to USB20 Driver (USB_HSIC_USB3503) [N/m/?] (NEW) 
  OMAP USB3 PHY Driver (OMAP_USB3) [N/m/y/?] (NEW) 
  OMAP CONTROL USB Driver (OMAP_CONTROL_USB) [N/m/y/?] (NEW) 
  Epson RX-4581 (RTC_DRV_RX4581) [N/m/y/?] (NEW) 
  HID Sensor Time (RTC_DRV_HID_SENSOR_TIME) [N/m/?] (NEW) 
  Synopsys DesignWare AHB DMA support (DW_DMAC) [N/m/y/?] (NEW) 
  Chrome OS Laptop (CHROMEOS_LAPTOP) [N/m/?] (NEW) 
Mailbox Hardware Support (MAILBOX) [N/y/?] (NEW) 
  Step_wise thermal governor (THERMAL_GOV_STEP_WISE) [Y/?] (NEW) y
Intel Non-Transparent Bridge support (NTB) [N/m/y/?] (NEW) 
  Register efivars backend for pstore (EFI_VARS_PSTORE) [Y/n/?] (NEW) 
    Disable using efivars as a pstore backend by default (EFI_VARS_PSTORE_DEFAULT_DISABLE) [N/y/?] (NEW) 
    Enable notifications for userspace key wrap/unwrap (ECRYPT_FS_MESSAGING) [N/y/?] (NEW) 
  Create a snapshot trace buffer (TRACER_SNAPSHOT) [N/y/?] (NEW) 
  CRC32 CRC algorithm (CRYPTO_CRC32) [N/m/y/?] (NEW) 
  CRC32 PCLMULQDQ hardware acceleration (CRYPTO_CRC32_PCLMUL) [N/m/y/?] (NEW) 

---

Links to this post:
http://www.itnews.com.au/News/342158,debian-70-debuts-with-private-cloud-deployment-tools.aspx
http://www.neowin.net/forum/topic/1158614-ubuntu-or-linux-mint/page__st__15

402. Very briefly: what I forgot about gnome 3...a short rant

Update 30 May 2013: frippery extensions are out now for gnome 3.8. Phew! http://intgat.tigress.co.uk/rmy/extensions/index.html

Original post:
After an initial bout of swearing, and the odd persistent nuisance, I took to gnome 3/gnome-shell pretty well -- I use it every day and it works well for me. It hasn't changed the way I work and it's prettier than gnome 2, so it's alright.

In other words: whenever I've read posts about how awful gnome 3 is and how the poster is going to switch to KDE/XFCE/LXDE/Xmonad I've considered it as whiny hyperbole. After all, even I had got used to gnome 3.

Sure, things like

* people like Allan Day and Jon McCann saying things like this and this
* feature deprecation in everything from gnome-screenshot to nautilus -- gnome-screenshot is now unusable without patching.
* incessant renaming* of 'gnome' applications

all sure don't help in convincing anyone that using gnome is a sane long-term strategy. But gnome 3 has worked ok for me.

*[open Image Viewer in gnome. Click on About in Help. Any indication that the program is called eye-of-gnome and that the package is called eog? Epiphany is now web. Palimpsest brings up disk utility but there's no package with that name anymore]

Or so I thought. What I had forgotten about was all the gnome shell extensions that I had installed to make gnome 3 usable.

This was driven home to me when Arch linux upgraded to gnome 3.8 -- facing the usual unusable default interface of gnome 3 I quickly set to rectify it by going to http://intgat.tigress.co.uk/rmy/extensions/index.html and downlo...wait...no extensions for gnome 3.8? http://extensions.gnome.org didn't yield much either. There were a handful of extensions, but it was missing one of the most important ones -- the bottom panel.

For the moment I'm stuck with vanilla gnome 3 -- and I don't like it a bit.

I haven't actually used gnome 3 -- I've been using gnome 2 with a gnome-shell engine. And that's why I've been happy...

401. AMD FX 8150:issues building kernel -- random failures.

Update 4:  I found the receipts for one pair of sticks and took it to MSY in Melbourne -- they were replaced on the spot without any questions asked. Very happy.

Update 3: The errors were all due to 3 bad ram sticks. Using the only good stick everything works fine. That's 24 Gb of bad ram...this won't be cheap if I can't find the receipts...

Update 2:
Running memtest86 I caught lots of errors (51 in 50 minutes) before I killed the test. I'm currently testing each stick one by one. I'm hoping that what is seemingly RAM errors can be caused by inapproriate BIOS settings, because 32 Gb bios is not cheap to replace...

While I'm swapping RAM sticks I'm also testing a separate set of stick on a different box. If they are error free it will be interesting to see if they trigger errors on the troublesome node. I'm still hoping for BIOS as being the culprit...

So far three out of four tested sticks have shown errors -- they all happen during test #6. The fourth stick has passed all tests seven times.

Update 1: dmesg also shows the same message as the OP here sees: https://bugzilla.redhat.com/show_bug.cgi?id=909702

The OP puts it down to a misconfigured bios, so the quest continues.
Searching for 990FX and FX8150 I get a number of hits:

Here's a newegg review for 990FX:

 I purchased This MB to run with the AMD FX 8150. I have built computers from high end to low end and know the ones in the middle last the longest and are the most stable.
[..]
At this point the fun of the build is gone, and I have too many hours dealing with problems. 

And that's not the only negative FX8?50 + 990FX review.

The worst part of it is that I've been thinking about building another, identical node (good value for money) as well as recommending my build to a student whom is about to do calcs.

Mind you, I've only ever had issues when it comes to compiling the kernel -- it's been solid when it comes to running calculations.

Original post:

NOTE: this is NOT a solution. Just observations.

My AMD FX 8150 is a great CPU -- it makes up the heart of the fastest of my computational nodes, and is eminently affordable. It does, however, cause me grief in one respect -- I can't compile the linux kernel.

The system
The box that's causing me trouble has
* AMD FX 8150 cpu
* gigabyte 990FXA-D3 motherboard
* nvidia GeForce 210 video card
* Corsair GS 800 PSU
* 4x8 Gb patriot viper PV316G186C0K RAM
While not top of the range, the components should be of reasonable quality.

In terms of software and OS, it's an up-to-date wheezy install (gcc 4.7), running kernel 3.7.2 (compiled on a different machine).

Compiling the kernel
I'm compiling the kernel as shown here: http://verahill.blogspot.com.au/2013/02/342-compiling-kernel-38-on-debian.html

The errors are shown at the end of the post

The fact that the errors keep changing might also be pointing towards there being a hardware fault with my CPU, rather than with FX 8150 in general.

3.8 built fine twice, and crashed the third time. 3.8.10 crashed twice, then built fine the third time.

It all sounds like I'm having hardware issues...but they only seem to be triggered during kernel builds. During 'normal use (i.e. using 100% cpu for weeks at a time) it is perfectly stable. Compiling e.g. nwchem (another pretty heavy compile) also goes absolutely fine.

Troubleshooting something like this also wouldn't be easy. See the end of the post for a list over various errors that I was getting during compilation of different kernel versions.

Anyway, I hit google...

---

BIOS
That Windows has issues with 8150 might seem unrelated, but it appeared that my errors could be solved by a bios update to my 990 fxa-d3 mobo:
 http://scalibq.wordpress.com/2011/10/19/amd-bulldozer-can-it-get-even-worse/

"The actual reported error is quite random, it just depends on where the CPU fails first. So you generally get a different error code with every BSOD."

and

"AMD’s KB article focuses solely on some boards with the 990FX chipset."

Well, I do have a 990FXA-D3 gigabyte motherboard.

My bios is shown by lshw as

*-firmware
          description: BIOS
          vendor: Award Software International, Inc.
          physical id: 0
          version: F7
          date: 05/30/2012
          size: 128KiB
          capacity: 4032KiB

So the obvious solution was to flash the bios.

Turns out, flashing the BIOS is a headache on Gigabyte motherboards (not buying anything from them again). What happened with simply burning a CD and booting with it in the drive?

Flashing the bios

I downloaded the bios (version F8): http://download.gigabyte.eu/FileList/BIOS/mb_bios_ga-990fxa-d3_f8.exe.

I unzipped it with 7z, giving me 990FXAD3.F8 -- I then put that file in the root of a USB stick..

I've tried with a number of USB sticks, including a blank stick formatted with W95 Fat32 and keeping the stick plugged in before rebooting.

In Q-flash, I always ended up with a prompt saying Floppy A <Drive>, and when I hit enter it says '..    <dir>'. 0 Files found. Yet it also said Total size 7.48G, Free Size: 7.44 G, which matched the size of the USB stick.

Finally I managed to get it to work:
*  in fdisk I only created a 1 gb partition on the USB stick, set type (t) to 6 (Fat16), made it bootable, and wrote changes to disk.
* I then ran mkdosfs -F 16 /dev/sdb1 (my usb stick was /dev/sdb).
*  I then copied the 990FXD3.F8 file to the usb stick root (after mounting it of course) and THAT worked.

Memtest86
Because RAM has traditionally been a major culprit behind hardware errors (especially the random, difficult-to-diagnose type) it's always a good idea to run a memtest. To do that, install memtest86+ (sudo apt-get install memtest86+) and reboot. There should be a new menu item (scroll down) in grub. Memtest takes quite a while, especially if you have a lot of RAM (32 Gb...).

Lo and behold, there are errors:

Tst  Pass  Failing Address              Good        Bad        Err-Bits  Count Chan
------------------------------------------------------------------------------------
6     0     0007383b4f4  -  1848.2MB   fffffbff     ffffffff   00000400    1
6     0     00039c1f294  -   924.1MB   fffffbff     ffffffff   00000004    2
6     0     00120203034  -  4610.0MB   00000004     00000000   00000004    3
6     0     001ca16c464  -  7329.4MB   00020004     00000000   00020000    4
[..]

I counted 51 errors before killing the test (time to identify the bad stick). Many of these occurred in a more limited address space than those shown above. Sigh...the RAM was the most expensive part of this build...

According to this there's a slight chance that the RAM might be ok, but it's still not a good sign.

I've tested each stick by itself -- so far 3 out of 4 sticks have yielded errors during test 6. I did seven passes on the fourth stick and no errors.

The outcome
However, even with the new bios the kernel compiles still fail -- it takes longer for it to fail, but it fails.

I do see the odd thing in dmesg though:

[ 4260.342268] as[29370]: segfault at 4541b5e ip 0000000000410306 sp 00007fff40ec4420 error 4 in as[400000+51000]

So either FX 8150 is still not properly supported by the BIOS, or I've bought a lemon.

The question remains: why do I only see failure during kernel compiles and no other conditions?

---

After bios flash:

Kernel 3.9-rc8

  CC      drivers/base/dd.o
In file included from /home/me/tmp/linux-3.9-rc8/arch/x86/include/asm/processor.h:23:0,
                 from /home/me/tmp/linux-3.9-rc8/arch/x86/include/asm/atomic.h:6,
                 from include/linux/atomic.h:4,
                 from include/linux/sysfs.h:20,
                 from include/linux/kobject.h:21,
                 from include/linux/device.h:17,
                 from drivers/base/dd.c:20:
/home/me/tmp/linux-3.9-rc8/arch/x86/include/asm/special_insns.h: In function 'native_read_cr0':
/home/me/tmp/linux-3.9-rc8/arch/x86/include/asm/special_insns.h:24:2: internal compiler error: in build_int_cst_wide, at tree.c:1238
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
The bug is not reproducible, so it is likely a hardware or OS problem.
make[3]: *** [drivers/base/dd.o] Error 1
make[2]: *** [drivers/base] Error 2
make[1]: *** [drivers] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.9-rc8'
make: *** [debian/stamp/build/kernel] Error 2

Kernel 3.8.10

  UPD     include/generated/compile.h
  CC      init/version.o
  LD      init/built-in.o
ipc/built-in.o:(.debug_info+0x1ed81): undefined reference to `.LASF108'
make[1]: *** [vmlinux] Error 1
make[1]: Leaving directory `/home/me/tmp/linux-3.8.10'
make: *** [debian/stamp/build/kernel] Error 2

Kernel 3.7.6

  CC [M]  fs/gfs2/super.o
  CC [M]  fs/gfs2/sys.o
In file included from /home/me/tmp/linux-3.7.6/arch/x86/include/asm/smp.h:13:0,
                 from include/linux/smp.h:38,
                 from include/linux/sched.h:30,
                 from fs/gfs2/sys.c:10:
/home/me/tmp/linux-3.7.6/arch/x86/include/asm/apic.h:394:1: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
The bug is not reproducible, so it is likely a hardware or OS problem.
make[3]: *** [fs/gfs2/sys.o] Error 1
make[2]: *** [fs/gfs2] Error 2
make[1]: *** [fs] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.7.6'
make: *** [debian/stamp/build/kernel] Error 2

Kernel 3.5

CC [M]  drivers/scsi/lpfc/lpfc_els.o
  CC [M]  drivers/scsi/lpfc/lpfc_hbadisc.o
  CC [M]  drivers/scsi/lpfc/lpfc_init.o
In file included from /home/me/tmp/linux-3.5/arch/x86/include/asm/msr.h:139:0,
                 from /home/me/tmp/linux-3.5/arch/x86/include/asm/processor.h:20,
                 from /home/me/tmp/linux-3.5/arch/x86/include/asm/thread_info.h:22,
                 from include/linux/thread_info.h:54,
                 from include/linux/preempt.h:9,
                 from include/linux/spinlock.h:50,
                 from include/linux/seqlock.h:29,
                 from include/linux/time.h:8,
                 from include/linux/timex.h:56,
                 from include/linux/sched.h:57,
                 from include/linux/blkdev.h:4,
                 from drivers/scsi/lpfc/lpfc_init.c:22:
/home/me/tmp/linux-3.5/arch/x86/include/asm/paravirt.h: In function 'store_gdt':
/home/me/tmp/linux-3.5/arch/x86/include/asm/paravirt.h:304:2: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
The bug is not reproducible, so it is likely a hardware or OS problem.
make[4]: *** [drivers/scsi/lpfc/lpfc_init.o] Error 1
make[3]: *** [drivers/scsi/lpfc] Error 2
make[2]: *** [drivers/scsi] Error 2
make[1]: *** [drivers] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.5'
make: *** [debian/stamp/build/kernel] Error 2

Kernel 3.4.42
Second crash:

  CC [M]  fs/coda/psdev.o
  CC [M]  fs/coda/cache.o
In file included from include/linux/mm.h:256:0,
                 from fs/coda/coda_linux.h:17,
                 from fs/coda/cache.c:24:
include/linux/page-flags.h:232:1: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
The bug is not reproducible, so it is likely a hardware or OS problem.
make[3]: *** [fs/coda/cache.o] Error 1
make[2]: *** [fs/coda] Error 2
make[1]: *** [fs] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.4.42'
make: *** [debian/stamp/build/kernel] Error 2

First crash:

  CC      kernel/signal.o
gcc: internal compiler error: Segmentation fault (program as)
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
make[2]: *** [kernel/signal.o] Error 4
make[1]: *** [kernel] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.4.42'
make: *** [debian/stamp/build/kernel] Error 2

---

Kernels that won't build and the errors -- before bios flash:
3.9-rc8

CC [M]  fs/nfs/nfs4client.o
  CC [M]  fs/nfs/nfs4sysctl.o
  CC [M]  fs/nfs/nfs4session.o
  CC [M]  fs/nfs/pnfs.o
fs/nfs/pnfs.c: In function 'read_seqcount_retry':
fs/nfs/pnfs.c:1951:1: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
The bug is not reproducible, so it is likely a hardware or OS problem.
make[3]: *** [fs/nfs/pnfs.o] Error 1
make[2]: *** [fs/nfs] Error 2
make[1]: *** [fs] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.9-rc8'
make: *** [debian/stamp/build/kernel] Error 2

3.8.10

  CC [M]  fs/nfs/inode.o
In file included from include/net/scm.h:6:0,
                 from include/linux/netlink.h:8,
                 from /home/me/tmp/linux-3.8.10/include/uapi/linux/neighbour.h:5,
                 from include/linux/netdevice.h:51,
                 from include/linux/icmpv6.h:12,
                 from include/linux/ipv6.h:59,
                 from include/net/ipv6.h:16,
                 from include/linux/sunrpc/clnt.h:26,
                 from fs/nfs/inode.c:26:
include/linux/security.h:2581:1: internal compiler error: Segmentation fault
Please submit a full bug report,

3.8.6

CC [M]  drivers/hid/hid-lg.o
  CC [M]  drivers/hid/hid-lgff.o
  CC [M]  drivers/hid/hid-lg2ff.o
  CC [M]  drivers/hid/hid-lg3ff.o
  CC [M]  drivers/hid/hid-lg4ff.o
  CC [M]  drivers/hid/hid-picolcd_core.o
  CC [M]  drivers/hid/hid-picolcd_fb.o
  CC [M]  drivers/hid/hid-picolcd_backlight.o
drivers/hid/hid-picolcd_backlight.c:120:1: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
The bug is not reproducible, so it is likely a hardware or OS problem.
make[3]: *** [drivers/hid/hid-picolcd_backlight.o] Error 1
make[2]: *** [drivers/hid] Error 2
make[1]: *** [drivers] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.8.6'
make: *** [debian/stamp/build/kernel] Error 2

3.8

CC      mm/dmapool.o
  CC      mm/hugetlb.o
/bin/sh: line 1: 25153 Done(2)                 gcc -E -D__GENKSYMS__ -Wp,-MD,mm/.hugetlb.o.d -nostdinc -isystem /usr/lib/gcc/x86_64-linux-gnu/4.7/include -I/home/me/tmp/linux-3.8/arch/x86/include -Iarch/x86/include/generated -Iinclude -I/home/me/tmp/linux-3.8/arch/x86/include/uapi -Iarch/x86/include/generated/uapi -I/home/me/tmp/linux-3.8/include/uapi -Iinclude/generated/uapi -include /home/me/tmp/linux-3.8/include/linux/kconfig.h -D__KERNEL__ -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Wno-format-security -fno-delete-null-pointer-checks -Os -m64 -mtune=generic -mno-red-zone -mcmodel=kernel -funit-at-a-time -maccumulate-outgoing-args -fstack-protector -DCONFIG_AS_CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -DCONFIG_AS_CFI_SECTIONS=1 -DCONFIG_AS_FXSAVEQ=1 -DCONFIG_AS_AVX=1 -DCONFIG_AS_AVX2=1 -pipe -Wno-sign-compare -fno-asynchronous-unwind-tables -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -mno-avx -Wframe-larger-than=2048 -Wno-unused-but-set-variable -fomit-frame-pointer -g -Wdeclaration-after-statement -Wno-pointer-sign -fno-strict-overflow -fconserve-stack -DCC_HAVE_ASM_GOTO -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(hugetlb)" -D"KBUILD_MODNAME=KBUILD_STR(hugetlb)" mm/hugetlb.c
     25154 Segmentation fault      | scripts/genksyms/genksyms -a x86_64 -r /dev/null > mm/.tmp_hugetlb.ver
make[2]: *** [mm/hugetlb.o] Error 139
make[1]: *** [mm] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.8'
make: *** [debian/stamp/build/kernel] Error 2

3.7.6

The errors differ each time:

Second run:

  CC [M]  fs/ext2/namei.o
  CC [M]  fs/ext2/super.o
fs/ext2/super.c: In function 'ext2_fill_super':
fs/ext2/super.c:762:12: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
The bug is not reproducible, so it is likely a hardware or OS problem.
make[3]: *** [fs/ext2/super.o] Error 1
make[2]: *** [fs/ext2] Error 2
make[1]: *** [fs] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.7.6'
make: *** [debian/stamp/build/kernel] Error 2

First run:

  CC      drivers/base/power/main.o
/bin/sh: line 1: 12317 Done                    gcc -E -D__GENKSYMS__ -Wp,-MD,drivers/base/power/.main.o.d -nostdinc -isystem /usr/lib/gcc/x86_64-linux-gnu/4.7/include -I/home/me/tmp/linux-3.7.6/arch/x86/include -Iarch/x86/include/generated -Iinclude -I/home/me/tmp/linux-3.7.6/arch/x86/include/uapi -Iarch/x86/include/generated/uapi -I/home/me/tmp/linux-3.7.6/include/uapi -Iinclude/generated/uapi -include /home/me/tmp/linux-3.7.6/include/linux/kconfig.h -D__KERNEL__ -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Wno-format-security -fno-delete-null-pointer-checks -Os -m64 -mtune=generic -mno-red-zone -mcmodel=kernel -funit-at-a-time -maccumulate-outgoing-args -fstack-protector -DCONFIG_AS_CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -DCONFIG_AS_CFI_SECTIONS=1 -DCONFIG_AS_FXSAVEQ=1 -DCONFIG_AS_AVX=1 -pipe -Wno-sign-compare -fno-asynchronous-unwind-tables -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -mno-avx -Wframe-larger-than=2048 -Wno-unused-but-set-variable -fomit-frame-pointer -g -Wdeclaration-after-statement -Wno-pointer-sign -fno-strict-overflow -fconserve-stack -DCC_HAVE_ASM_GOTO -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(main)" -D"KBUILD_MODNAME=KBUILD_STR(main)" drivers/base/power/main.c
     12318 Segmentation fault      | scripts/genksyms/genksyms -a x86_64 -r /dev/null > drivers/base/power/.tmp_main.ver
make[4]: *** [drivers/base/power/main.o] Error 139
make[3]: *** [drivers/base/power] Error 2
make[2]: *** [drivers/base] Error 2
make[1]: *** [drivers] Error 2

3.7.2
The errors differ every time:

Second run:

CC [M]  drivers/hwmon/tmp102.o
======= Backtrace: =========
/lib/x86_64-linux-gnu/libc.so.6(+0x76d76)[0x2b2bb07f6d76]
/lib/x86_64-linux-gnu/libc.so.6(+0x7a658)[0x2b2bb07fa658]
/lib/x86_64-linux-gnu/libc.so.6(__libc_malloc+0x70)[0x2b2bb07fbb90]
scripts/genksyms/genksyms[0x4075fa]
scripts/genksyms/genksyms[0x4037c0]
scripts/genksyms/genksyms[0x402de6]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xfd)[0x2b2bb079eead]
scripts/genksyms/genksyms[0x400f59]
======= Memory map: ========
00400000-0040e000 r-xp 00000000 08:05 36440720                           /home/me/tmp/linux-3.7.2/scripts/genksyms/genksyms
0060d000-0060e000 rw-p 0000d000 08:05 36440720                           /home/me/tmp/linux-3.7.2/scripts/genksyms/genksyms
0060e000-00616000 rw-p 00000000 00:00 0
0089b000-00c58000 rw-p 00000000 00:00 0                                  [heap]
2b2bb0330000-2b2bb0350000 r-xp 00000000 08:01 11802457                   /lib/x86_64-linux-gnu/ld-2.13.so
2b2bb0350000-2b2bb0352000 rw-p 00000000 00:00 0
2b2bb054f000-2b2bb0550000 r--p 0001f000 08:01 11802457                   /lib/x86_64-linux-gnu/ld-2.13.so
2b2bb0550000-2b2bb0551000 rw-p 00020000 08:01 11802457                   /lib/x86_64-linux-gnu/ld-2.13.so
2b2bb0551000-2b2bb0552000 rw-p 00000000 00:00 0
2b2bb0558000-2b2bb0561000 r-xp 00000000 08:01 2233201                    /usr/lib/x86_64-linux-gnu/libfakeroot/libfakeroot-sysv.so
2b2bb0561000-2b2bb0761000 ---p 00009000 08:01 2233201                    /usr/lib/x86_64-linux-gnu/libfakeroot/libfakeroot-sysv.so
2b2bb0761000-2b2bb0762000 rw-p 00009000 08:01 2233201                    /usr/lib/x86_64-linux-gnu/libfakeroot/libfakeroot-sysv.so
2b2bb0762000-2b2bb0763000 rw-p 00000000 00:00 0
2b2bb077c000-2b2bb077d000 rw-p 00000000 00:00 0
2b2bb0780000-2b2bb0900000 r-xp 00000000 08:01 11802454                   /lib/x86_64-linux-gnu/libc-2.13.so
2b2bb0900000-2b2bb0b00000 ---p 00180000 08:01 11802454                   /lib/x86_64-linux-gnu/libc-2.13.so
2b2bb0b00000-2b2bb0b04000 r--p 00180000 08:01 11802454                   /lib/x86_64-linux-gnu/libc-2.13.so
2b2bb0b04000-2b2bb0b05000 rw-p 00184000 08:01 11802454                   /lib/x86_64-linux-gnu/libc-2.13.so
2b2bb0b05000-2b2bb0b0a000 rw-p 00000000 00:00 0
2b2bb0b10000-2b2bb0b12000 r-xp 00000000 08:01 11802447                   /lib/x86_64-linux-gnu/libdl-2.13.so
2b2bb0b12000-2b2bb0d12000 ---p 00002000 08:01 11802447                   /lib/x86_64-linux-gnu/libdl-2.13.so
2b2bb0d12000-2b2bb0d13000 r--p 00002000 08:01 11802447                   /lib/x86_64-linux-gnu/libdl-2.13.so
2b2bb0d13000-2b2bb0d14000 rw-p 00003000 08:01 11802447                   /lib/x86_64-linux-gnu/libdl-2.13.so
2b2bb0d14000-2b2bb0d16000 rw-p 00000000 00:00 0
2b2bb0d30000-2b2bb0d45000 r-xp 00000000 08:01 11796731                   /lib/x86_64-linux-gnu/libgcc_s.so.1
2b2bb0d45000-2b2bb0f45000 ---p 00015000 08:01 11796731                   /lib/x86_64-linux-gnu/libgcc_s.so.1
2b2bb0f45000-2b2bb0f46000 rw-p 00015000 08:01 11796731                   /lib/x86_64-linux-gnu/libgcc_s.so.1
2b2bb4000000-2b2bb4021000 rw-p 00000000 00:00 0
2b2bb4021000-2b2bb8000000 ---p 00000000 00:00 0
7fff5cd00000-7fff5cd23000 rw-p 00000000 00:00 0                          [stack]
7fff5cdd8000-7fff5cdd9000 r-xp 00000000 00:00 0                          [vdso]
ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0                  [vsyscall]
/bin/sh: line 1: 18863 Done(2)                 gcc -E -D__GENKSYMS__ -Wp,-MD,drivers/acpi/.video.o.d -nostdinc -isystem /usr/lib/gcc/x86_64-linux-gnu/4.7/include -I/home/me/tmp/linux-3.7.2/arch/x86/include -Iarch/x86/include/generated -Iinclude -I/home/me/tmp/linux-3.7.2/arch/x86/include/uapi -Iarch/x86/include/generated/uapi -I/home/me/tmp/linux-3.7.2/include/uapi -Iinclude/generated/uapi -include /home/me/tmp/linux-3.7.2/include/linux/kconfig.h -D__KERNEL__ -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Wno-format-security -fno-delete-null-pointer-checks -Os -m64 -mtune=generic -mno-red-zone -mcmodel=kernel -funit-at-a-time -maccumulate-outgoing-args -fstack-protector -DCONFIG_AS_CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -DCONFIG_AS_CFI_SECTIONS=1 -DCONFIG_AS_FXSAVEQ=1 -DCONFIG_AS_AVX=1 -pipe -Wno-sign-compare -fno-asynchronous-unwind-tables -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -mno-avx -Wframe-larger-than=2048 -Wno-unused-but-set-variable -fomit-frame-pointer -g -Wdeclaration-after-statement -Wno-pointer-sign -fno-strict-overflow -fconserve-stack -DCC_HAVE_ASM_GOTO -Os -DMODULE -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(video)" -D"KBUILD_MODNAME=KBUILD_STR(video)" drivers/acpi/video.c
     18864 Aborted                 | scripts/genksyms/genksyms -a x86_64 -r /dev/null > drivers/acpi/.tmp_video.ver
make[3]: *** [drivers/acpi/video.o] Error 134
make[2]: *** [drivers/acpi] Error 2
make[1]: *** [drivers] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.7.2'
make: *** [debian/stamp/build/kernel] Error 2

First run:

  CHK     include/generated/uapi/linux/version.h
  CHK     include/generated/utsrelease.h
  CALL    scripts/checksyscalls.sh
  Building modules, stage 2.
  MODPOST 2369 modules
ERROR: "ieee80211_get_hdrlen" [drivers/staging/rtl8192u/r8192u_usb.ko] undefined!
ERROR: "ieee80211_is_empty_essid" [drivers/staging/rtl8192u/r8192u_usb.ko] undefined!
make[2]: *** [__modpost] Error 1
make[1]: *** [modules] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.7.2'
make: *** [debian/stamp/build/kernel] Error 2

3.6.3

LD [M]  drivers/input/misc/pcf50633-input.ko
  CC      drivers/input/misc/pcspkr.mod.o
In file included from drivers/input/misc/pcspkr.mod.c:1:0:
include/linux/module.h:299:9: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
The bug is not reproducible, so it is likely a hardware or OS problem.
make[2]: *** [drivers/input/misc/pcspkr.mod.o] Error 1
make[1]: *** [modules] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.6.3'
make: *** [debian/stamp/build/kernel] Error 2

3.5.0
The errors keep changing.

Second run:

  CC [M]  drivers/gpu/drm/via/via_map.o
  CC [M]  drivers/gpu/drm/via/via_mm.o
  CC [M]  drivers/gpu/drm/via/via_dma.o
drivers/gpu/drm/via/via_dma.c:741:21: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
The bug is not reproducible, so it is likely a hardware or OS problem.
make[5]: *** [drivers/gpu/drm/via/via_dma.o] Error 1
make[4]: *** [drivers/gpu/drm/via] Error 2
make[3]: *** [drivers/gpu/drm] Error 2
make[2]: *** [drivers/gpu] Error 2
make[1]: *** [drivers] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.5'
make: *** [debian/stamp/build/kernel] Error 2

First run:

  CC      drivers/hid/hid-sony.mod.o
drivers/hid/hid-sony.mod.c:46:1: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
The bug is not reproducible, so it is likely a hardware or OS problem.
make[2]: *** [drivers/hid/hid-sony.mod.o] Error 1
make[1]: *** [modules] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.5'
make: *** [debian/stamp/build/kernel] Error 2

3.4.42

CC [M]  fs/quota/quota_tree.o
  CC [M]  fs/reiserfs/bitmap.o
fs/reiserfs/bitmap.c: In function 'scan_bitmap_block.constprop.9':
fs/reiserfs/bitmap.c:236:9: warning: 'next' may be used uninitialized in this function [-Wmaybe-uninitialized]
  CC [M]  fs/reiserfs/do_balan.o
  CC [M]  fs/reiserfs/namei.o
gcc: internal compiler error: Segmentation fault (program as)
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
make[3]: *** [fs/reiserfs/namei.o] Error 4
make[2]: *** [fs/reiserfs] Error 2
make[1]: *** [fs] Error 2
make[1]: Leaving directory `/home/me/tmp/linux-3.4.42'
make: *** [debian/stamp/build/kernel] Error 2

---

Another dmesg error:

400. XpressConnect on Debian, Arch: step by step

Here's a step-by-step write-up of this post: http://verahill.blogspot.com.au/2013/04/393-not-fix-xpressconnect-on-ubuntu-vs.html

The 'problem' with running Xpress Connect on non-Ubuntu linux distributions is entirely artificial -- XpressConnect checks whether you are using Ubuntu, and if you're not, it refuses to run.

So the solution is simply to pretend that you are using ubuntu, however annoying that is. I wish universities would take this into account and end their association with Cloudpath, or to force them to support other distributions.

Note: XpressConnect is completely superfluous -- it doesn't do anything other than set up your wireless connection, which is something you could easily do by hand. See e.g. here for eduroam: http://verahill.blogspot.com.au/2013/04/394-eduroam-using-wicd-and-network.html

---

How-to get XpressConnect running
1. Create the file /etc/lsb-release and put the following in it

DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=10.04
DISTRIB_CODENAME=lucid
DISTRIB_DESCRIPTION="Ubuntu 10.04.4 LTS"

If you are completely new to linux, one way of creating the file is to run

gksu gedit /etc/lsb-release

Alternatively, if you're not using gnome, try

sudo nano /etc/lsb-release 

2. Install lshw and iwlist

On debian (and clones like mint, ubuntu etc.):

sudo apt-get install lshw wireless-tools

On arch linux

sudo pacman -S lshw wireless_tools

3. Run XpressConnect
This is the vanilla version -- replace http://hosted.cloudpath.net/Xavier/Production/tools/XpressConnect-Linux.tar with the link to your universities version.

cd ~/Downloads
wget http://hosted.cloudpath.net/Xavier/Production/tools/XpressConnect-Linux.tar
tar xvf XpressConnect-Linux.tar
./XpressConnect-DoubleClickToRun

That's it. Simple as that.

399. Looking at speeding up (re)boot on debian wheezy.

I'd be interested in getting my beowulf cluster nodes to boot a little bit faster -- (re)boots of the nodes very are infrequent, but the front node doubles as my work desktop and is normally rebooted at least once per month (kernel upgrades etc.) -- rebooting the front node makes me nervous, however, and the faster it boots, the better it is.

I should probably build a low-powered front node specifically for my cluster though...but that takes money, and money takes time.

Anyway, boot. In spite of the impetus for this post I'm testing this on my laptop which has wheezy, gnome 3.4 and an SSD -- it's not that representative of the target system and I'll have to repeat this on a normal desktop with a spinning hdd at a later stage.

I'm more or less following http://wiki.debian.org/BootProcessSpeedup. Note that insserv seems to be set up and enabled by default in Wheezy.

---

Timing it -- Setting up bootchart2
I first tried to define boot times arbitrarily as the time from me hitting enter in GRUB, to the visual appearance of the log-in prompt in GDM3, but it was too imprecise (up to +- 2) relative to the time a boot took (ca 9-10s).

I ended up installing bootchart and bootchart-view instead.

sudo apt-get install bootchart2

Then edit /etc/default/grub as shown here:

GRUB_CMDLINE_LINUX_DEFAULT="quiet initcall_debug printk.time=y init=/sbin/bootchartd"

and run

sudo update-grub

After a boot, run

pybootchartgui
eog bootchart.png

You'll get something like this:

Look at the top, right above the first chart -- it says 'time: 6.61s'. I'll use that as the metric.

Most of the time bootchart2 worked fine, but for the odd boot the /var/log/bootchart.tgz wasn't accepted by pybootchartgui.

---

Normal boot, pre-optimisation: 
'Cold' reboots: 6.61, 5.77 seconds
Warm* reboots: 6.46, 5.79, 5.97 seconds

*using shutdown -r now

The variability is very high -- there's almost a second between the fastest and slowest boots. Keep that in mind when looking at the numbers later on.

---

Using readahead-fedora to pre-load files

sudo apt-get install readahead-fedora

After install, readhead-early, -late and stop were enabled in rcconf.

The first boot took over 7 seconds, but later boots were typically around 6 seconds or faster. Note that readahead is solving an issue which isn't really present when using high bandwidth SSDs, and may even slow things down under conditions where you use an SSD or a spinning disk with a high rpm (e.g. >7200 rpm)

First run

'normal' run

Not exactly an improvement. Looking at /etc/readahead.d/custom.early shows that the wrong kernel files are loaded -- I'm using a custom kernel (3.8.5-ck1) but the stock kernel files are loaded (3.2.0-4). I edited custom.early to point towards my current kernel, and then did a warm reboot.

---

Speeding up reboots -Kexec

sudo apt-get install kexec-tools

Shutdown your computer once, then boot up. After that first time you can do warm reboots (sudo shutdown -r now) without going through the BIOS and grub stages. The only -- visible -- downside is that your screen will go crazy for a few seconds as the running kernel is being overwritten by the new kernel (I presume). Doesn't look pretty, but reboot is fast.

I couldn't get bootchart to time the hot reboots, but they look 'fast'.

---

I'll be repeating this on a system with a spinning disk at a later stage.

398. Securing your Dropbox with encfs

Quite some time ago I made a post about how to use truecrypt and dropbox together. It wasn't a very elegant solution as it consisted of putting a truecrypt container in the dropbox folder.

 Apart from being pretty obvious, it suffers from drawbacks, such as the fixed size of the truecrypt container taking up space whether you're using it all or not.

[Note that mounting the dropbox folder inside a truecrypt container isn't a real solution either since it won't actually encrypt anything.]

Playing around with encfs made me realise that THIS is the perfect solution for secure Dropbox.

We'll make the assumption that you don't want to encrypt everything in your dropbox, but only things which is kept in a specific folder.

I'm presuming that you've got dropbox set up and working. If not, get the deb files from https://www.dropbox.com/

If you're having issues, make sure that fuse is installed, and that you belong to the group fuse (too add, do sudo usermod -a -G fuse $USER . To check do cat /etc/group|grep fuse)

Getting started:

sudo apt-get install encfs
mkdir ~/Dropbox/encrypted
encfs ~/Dropbox/encrypted ~/decrypted

The directory "/home/me/decrypted" does not exist. Should it be created? (y,n) Y
Creating new encrypted volume.
Please choose from one of the following options:
 enter "x" for expert configuration mode,
 enter "p" for pre-configured paranoia mode,
 anything else, or an empty line will select standard mode.
?> p

Paranoia configuration selected.

Configuration finished.  The filesystem to be created has
the following properties:
Filesystem cipher: "ssl/aes", version 3:0:2
Filename encoding: "nameio/block", version 3:0:1
Key Size: 256 bits
Block Size: 1024 bytes, including 8 byte MAC header
Each file contains 8 byte header with unique IV data.
Filenames encoded using IV chaining mode.
File data IV is chained to filename IV.
File holes passed through to ciphertext.

-------------------------- WARNING --------------------------
The external initialization-vector chaining option has been
enabled.  This option disables the use of hard links on the
filesystem. Without hard links, some programs may not work.
The programs 'mutt' and 'procmail' are known to fail.  For
more information, please see the encfs mailing list.
If you would like to choose another configuration setting,
please press CTRL-C now to abort and start over.

Now you will need to enter a password for your filesystem.
You will need to remember this password, as there is absolutely
no recovery mechanism.  However, the password can be changed
later using encfsctl.

New Encfs Password: 
Verify Encfs Password:

Put the files you want to be encrypted in ~/decrypted -- NOT ~/Dropbox/encrypted. Once you're done, unmount ~/decrypted:

sudo umount ~/decrypted

---

When you want to access your encrypted files or add more files to the encrypted file, just mount ~/Dropbox/encrypted using encfs:

encfs ~/Dropbox/encrypted ~/decrypted/

397. Briefly: compiling ck (Con Kolivas) kernel 3.8.9 in arch linux

mkdir ~/tmp
cd ~/tmp 
wget http://www.kernel.org/pub/linux/kernel/v3.0/linux-3.8.9.tar.bz2
tar xvf linux-3.8.9.tar.bz2
cd linux-3.8.9/
wget http://ck.kolivas.org/patches/3.0/3.8/3.8-ck1/patch-3.8-ck1.bz2
bunzip2 patch-3.8-ck1.bz2
patch -p1 < patch-3.8-ck1
cp /proc/config.gz .
gunzip config.gz
mv config .config
make oldconfig
make -j2
make -j2 modules
sudo make modules_install
sudo make headers_install INSTALL_HDR_PATH=/usr/src/linux-3.8.9-ck1-ARCH
sudo cp arch/x86_64/boot/bzImage /boot/vmlinuz-3.8.9-ck1-ARCH
sudo cp System.map /boot/System-3.8.9-ck1-ARCH.map
sudo mkinitcpio -k 3.8.9-ck1-ARCH -c /etc/mkinitcpio.conf -g /boot/initramfs-3.8.9-ck1-ARCH.img
sudo grub-mkconfig -o /boot/grub/grub.cfg

396. Compiling gromacs 4.6 with gpu support, openblas and fftw3 on debian wheezy

NOTE: with ACML my performance on my FX8150 and FX8350 nodes is only 25% of that with Openblas (double precision). Yes, for some reason gromacs is four times faster with openblas than with the machine vendor libraries in my tests.

Here are the release notes: http://www.gromacs.org/About_Gromacs/Release_Notes/Versions_4.6.x
As far as I understand you don't have to rely on openmm anymore for CUDA. Yes, the PITA of compiling openmm is gone!

Note that GPU calcs only speed things up under certain, specific conditions  -- and not all nvidia cards are supported (or equal). My own set-up, using statically cooled graphics cards, is definitely not appropriate for a GPU cluster. Once nwchem comes out with GPU support I might upgrade to fancier $200 graphics cards (maybe COSMO in NWChem will finally become more reasonable in terms of computational cost), but there's little reason for that at the moment.

Not all cards are created equal either -- e.g. GT210, which has GPU compute capability 1.2, is too poor to run with gromacs. GT430 (compute cap GT430) works. Both are obviously not viable for professional work.

Also note that it seems that you still need to use OPENMM if you want GPU support for implicit solvation.

Gromacs used to be easy to install. It's become a fair bit more complicated between 4.5.5 and 4.6. See here for gromacs 4.5.5: http://verahill.blogspot.com.au/2012/05/gromacs-with-external-fftw3-and-blas-on.html

CUDA: If you want to build with cuda you need gcc-4.6, which is still available in the wheezy repos. 4.7 won't work. Luckily, you can have both on your system, but you'll need to specify CC and CXX as shown below.

---

Openblas
Note that the links to the openblas file tends to die after a while, so you might have to download it manually.

sudo mkdir /opt/openblas
sudo chown ${USER} /opt/openblas
cd ~/tmp
wget http://github.com/xianyi/OpenBLAS/tarball/v0.2.6
tar xvf v0.2.6
cd xianyi-OpenBLAS-87b4d0c/
wget http://www.netlib.org/lapack/lapack-3.4.1.tgz
make all BINARY=64 CC=/usr/bin/gcc FC=/usr/bin/gfortran USE_THREAD=0 INTERFACE64=1 1> make.log 2>make.err
make PREFIX=/opt/openblas install
cp lib*.*  /opt/openblas/lib

add

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/openblas/lib

to your ~/.bashrc [for later use with nwchem and ecce, add /opt/openblas/lib to /etc/ld.so.conf and do sudo ldconfig -- you might want to make libopenblas.so and libopenblas.so.0 sym links to the main lib, libopenblas_bulldozer-r0.2.6.so]

---

single-precision gromacs 4.6 with both CPU and GPU

CUDA
If you have an nvidia card and want to enable GPU calcs, do

sudo apt-get install nvidia-cuda-toolkit gcc-4.6 g++-4.6

If /usr/lib/libcuda.so is nothing by a symmlink to /usr/lib/libcuda.so.1, and the file /usr/lib/libcuda.so.1 is missing (this was the case on my wheezy amd64), then do

sudo rm /usr/lib/libcuda.so
sudo ln -s /usr/lib/x86_64-linux-gnu/libcuda.so.1 /usr/lib/libcuda.so

You can also simply make sure that there/s no /usr/lib/libcuda.so.

Continue with the gromacs compilation:

cd ~/tmp
sudo apt-get install cmake
wget ftp://ftp.gromacs.org/pub/gromacs/gromacs-4.6.tar.gz
tar xvf gromacs-4.6.tar.gz
mkdir build_gromacs46
cd build_gromacs46
sudo mkdir /opt/gromacs
sudo chown ${USER} /opt/gromacs
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib/openmpi/lib:/opt/openblas/lib
export LDFLAGS="-L/opt/openblas/lib -lopenblas"
export CPPFLAGS="-I/opt/openblas/include"
export CC=/usr/bin/gcc-4.6 && export CXX=/usr/bin/g++-4.6 && cmake -DGMX_FFT_LIBRARY=fftw3 -DGMX_BUILD_OWN_FFTW=On -DGMX_DOUBLE=off -DCMAKE_INSTALL_PREFIX=/opt/gromacs/gromacs4.6_single -DGMX_EXTERNAL_BLAS=/opt/openblas/lib ../gromacs-4.6
make
make install

Note: for acml I used this instead:

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib/openmpi/lib:/opt/acml/acml5.2.0/gfortran64_fma4_int64/lib
export LDFLAGS="-L/opt/acml/acml5.2.0/gfortran64_fma4_int64/lib -lacml"
export CPPFLAGS="-I/opt/acml/acml5.2.0/gfortran64_fma4_int64/include"
export CC=/usr/bin/gcc-4.6 && export CXX=/usr/bin/g++-4.6 && cmake -DGMX_FFT_LIBRARY=fftw3 -DGMX_BUILD_OWN_FFTW=On -DGMX_DOUBLE=off -DCMAKE_INSTALL_PREFIX=/opt/gromacs/gromacs4.6_single -DGMX_EXTERNAL_BLAS=/opt/acml/acml5.2.0/gfortran64_fma4_int64/lib ../gromacs-4.6

---

Double-precision gromacs without GPU acceleration:

cd ~/tmp/build_gromacs46

rm * -rf
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib/openmpi/lib:/opt/openblas/lib
export LDFLAGS="-L/opt/openblas/lib -lopenblas"
export CPPFLAGS="-I/opt/openblas/include"
export CC=/usr/bin/gcc-4.6 && export CXX=/usr/bin/g++-4.6 && cmake -DGMX_FFT_LIBRARY=fftw3 -DGMX_BUILD_OWN_FFTW=On -DGMX_DOUBLE=on -DGMX_GPU=off -DCMAKE_INSTALL_PREFIX=/opt/gromacs/gromacs4.6_double -DGMX_EXTERNAL_BLAS=/opt/openblas/lib ../gromacs-4.6
make
make install

Note: for acml I used this instead:

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib/openmpi/lib:/opt/acml/acml5.2.0/gfortran64_fma4_int64/lib
export LDFLAGS="-L/opt/acml/acml5.2.0/gfortran64_fma4_int64/lib -lacml"
export CPPFLAGS="-I/opt/acml/acml5.2.0/gfortran64_fma4_int64/include"
export CC=/usr/bin/gcc-4.6 && export CXX=/usr/bin/g++-4.6 && cmake -DGMX_FFT_LIBRARY=fftw3 -DGMX_BUILD_OWN_FFTW=On -DGMX_DOUBLE=on -DGMX_GPU=off -DCMAKE_INSTALL_PREFIX=/opt/gromacs/gromacs4.6_double -DGMX_EXTERNAL_BLAS=/opt/acml/acml5.2.0/gfortran64_fma4_int64/lib ../gromacs-4.6

---

Add gromacs to path:

echo 'export PATH=$PATH:/opt/gromacs/gromacs4.6_single/bin:/opt/gromacs/gromacs4.6_double/bin' >> ~/.bashrc

---

Switching between GPU and CPU
You can use the same binary for both, but remember that only the single precision binaries have GPU support to begin with. To set gpu vs cpu, use the -nb option in mdrun:

-nb          enum   auto    Calculate non-bonded interactions on: auto, cpu,
                            gpu or gpu_cpu

---

Quick test:

cd ~/tmp
wget http://www.gromacs.org/@api/deki/files/128/=gromacs-gpubench-dhfr.tar.gz
tar xvf \=gromacs-gpubench-dhfr.tar.gz
cd dhfr/GPU/dhfr-solv-PME.bench
mdrun -nb cpu -s topol.tpr -testverlet

Hit ctrl+c to stop and get statistics. Then try

mdrun -nb gpu -s topol.tpr -testverlet

I got

XPU       ns/day
--------------
auto      7.4
GPU       7.7
CPU       4.1
gpu_cpu   7.5

where I have a 3 core 3.1 GHz AMD Athlon II X3 445 CPU and an NVIDIA GeForce GT 430 graphics card -- neither of which is anything special.

Note also that the ns/day values depended highly on how long I let the calc run, and as I didn't time it and make them run the same amount of time, I suspect that auto, GPU and gpu_cpu are all about the same.

395. Compiling the ck-kernel 3.8.x on debian (and patching i915)

What is the ck kernel? In short it is a patch for the mainline kernel which is supposed to render it more responsive for desktop use. See here: https://wiki.archlinux.org/index.php/Linux-ck

I'm not going to try to 'benchmark' this against the stock kernel, since I'm not clear what a good metric would be -- how do you measure 'responsiveness'? Come to think of it -- what's a good, objective definition?

Download the kernel version that you need from http://kernel.org and the ck path from Con's web page at http://users.on.net/~ckolivas/kernel/

I also ended up applying this patch to sort out recent i915 issues:
http://verahill.blogspot.com.au/2013/03/368-slow-mouse-and-keyboard-triggered.html
https://bbs.archlinux.org/viewtopic.php?pid=1248190
http://forums.gentoo.org/viewtopic-p-7278760.html
https://bbs.archlinux.org/viewtopic.php?pid=1254285#p1254285

I did it by creating a file called i915.patch and placing it in the kernel source root:

--- drivers/gpu/drm/drm_crtc_helper.c   2013-04-26 10:24:07.987942008 +1000
+++ ../../linux-3.8.8/drivers/gpu/drm/drm_crtc_helper.c 2013-04-17 15:11:28.000000000 +1000
@@ -1067,7 +1067,7 @@ void drm_helper_hpd_irq_event(struct drm
        enum drm_connector_status old_status;
        bool changed = false;
 
-       if (!dev->mode_config.poll_enabled || !drm_kms_helper_poll)
+       if (!dev->mode_config.poll_enabled)
                return;
 
        mutex_lock(&dev->mode_config.mutex);

Note that as far as I know it doesn't solve the issue, but it does allow you to disable polling by editing /etc/default/grub and adding

GRUB_CMDLINE_LINUX_DEFAULT="quiet drm_kms_helper.poll=N"

followed by running

sudo update-grub

So far it's working fine (after testing for ca one hour).

Anyway, here's kernel 3.8.5 (should work on all 3.8.x):

sudo apt-get install kernel-package fakeroot build-essential ncurses-dev
mkdir ~/tmp
cd ~/tmp
wget https://www.kernel.org/pub/linux/kernel/v3.x/linux-3.8.5.tar.xz
tar xvf linux-3.8.5.tar.xz
cd linux-3.8.5/
wget http://ck.kolivas.org/patches/3.0/3.8/3.8-ck1/patch-3.8-ck1.bz2
bunzip2 patch-3.8-ck1.bz2
patch -p1 < patch-3.8-ck1

patching file arch/powerpc/platforms/cell/spufs/sched.c
patching file Documentation/scheduler/sched-BFS.txt
patching file Documentation/sysctl/kernel.txt
patching file fs/proc/base.c
patching file include/linux/init_task.h
patching file include/linux/ioprio.h
patching file include/linux/sched.h
patching file init/Kconfig
patching file init/main.c
patching file kernel/delayacct.c
patching file kernel/exit.c
patching file kernel/posix-cpu-timers.c
patching file kernel/sysctl.c
patching file lib/Kconfig.debug
patching file include/linux/jiffies.h
patching file drivers/cpufreq/cpufreq.c
patching file drivers/cpufreq/cpufreq_ondemand.c
patching file drivers/cpufreq/cpufreq_conservative.c
patching file kernel/sched/bfs.c
patching file kernel/sched/Makefile
patching file include/uapi/linux/sched.h
patching file include/linux/swap.h
patching file mm/memory.c
patching file mm/swapfile.c
patching file mm/vmscan.c
patching file arch/x86/Kconfig
patching file kernel/Kconfig.hz
patching file kernel/Kconfig.preempt
patching file Makefile

patch -p0 -R < i915.patch

patching file drivers/gpu/drm/drm_crtc_helper.c
make-kpgk clean
cat /boot/config-`uname -r`>.config
make oldconfig
time fakeroot make-kpkg -j4 --initrd kernel_image kernel_headers
sudo dpkg -i ../linux*3.8.5-ck*.deb

where 4 is the number of cores on your machine (note: it only has to do with compiling -- you can use the compiled binaries on any number of cores).

That's it. Now all you have to do is reboot and you'll be using your new kernel. Simple, eh?

394. Eduroam using wicd and network-manager

Eduroam is a "secure international roaming service" which is used by a great number of universities in Europe, the US and Australia, as well as being used by a limited number of campuses in Asia and Africa.

It's a pretty neat idea since there's frequent exchange of visitors between universities (visiting researchers, seminar speakers, PhD students visiting to do measurements etc.). Getting an account to set up with IT for a temporary user is too much hassle -- and this takes care of that.

While Eduroam might be implemented slightly different on different campuses, this is what I've had luck with in Melbourne (https://wiki.aarnet.edu.au/display/eduroam/For+End+Users).

Here are a couple of screenshots and instructions for network-manager and wicd

---

network-manager

Security:             WPA & WPA2 Enterprise
Authentication:       Protected EAP (PEAP)
Anon. Identity:
CA certificate: 
Inner Authentication:  MSCHAPv2
Username:              uniusername
Password:              password

---

Wicd
I wiped my laptop a month or two ago and hadn't bothered reconnecting to eduroam after that, so when setting it up in wicd I first tried creating a script in /etc/wicd/encryption/templates, which I activated by including it in /etc/wicd/encryption/templates/active. I kept on getting 'bad password' errors though. Note that I've had no success whatsoever connecting to any network using wicd in virtualbox. What I show below works on physical hardware (i.e. my Thinkpad SL410) though.

Anyway, simple: click on Properties for the eduroam network you want to connect to, and set it up as shown below (version 1.7.2 as found in Wheezy)

For Encryption, pick PEAP with TKIP/MSCHAPv2, and input your university username and password.

You don't need to use the CA cert, so if you don't have access to the eduroam CA cert, don't worry, and don't include it (it improves security though).

---

Android
As a bonus, here are the settings in android.

---

Links to this post:
http://crunchbang.org/forums/viewtopic.php?pid=345087