[sage-devel] Re: R build error in 5.10 was Re: ATLAS build error on CentOS

[leif]

leif wrote:

leif wrote:

leif wrote:

Ursula Whitcher wrote:

...
gcc -std=gnu99 -shared -fopenmp   -o libR.so CConverters.o
CommandLineArgs.o Rdynload.o Renviron.o RNG.o agrep.o apply.o
arithmetic.o array.o attrib.o base.o bind.o builtin.o character.o
coerce.o colors.o complex.o connections.o context.o cov.o cum.o dcf.o
datetime.o debug.o deparse.o deriv.o devices.o dotcode.o dounzip.o
dstruct.o duplicate.o engine.o envir.o errors.o eval.o format.o
fourier.o gevents.o gram.o gram-ex.o gramLatex.o gramRd.o graphics.o
grep.o identical.o inlined.o inspect.o internet.o iosupport.o lapack.o
list.o localecharset.o logic.o main.o mapply.o match.o memory.o model.o
names.o objects.o optim.o optimize.o options.o par.o paste.o platform.o
plot.o plot3d.o plotmath.o print.o printarray.o printvector.o
printutils.o qsort.o random.o raw.o registration.o relop.o rlocale.o
saveload.o scan.o seq.o serialize.o size.o sort.o source.o split.o
sprintf.o startup.o subassign.o subscript.o subset.o summary.o
sysutils.o unique.o util.o version.o vfonts.o xxxpr.o
../unix/Rembedded.o ../unix/libunix.a ../appl/libappl.a
../nmath/libnmath.a ../extra/zlib/libz.a ../extra/bzip2/libbz2.a
../extra/pcre/libpcre.a ../extra/tre/libtre.a  ../extra/xz/liblzma.a
-L../../lib -lRblas -lgfortran -lm -lquadmath   -lreadline -lncurses
-lrt -ldl -lm
...
gcc -std=gnu99 -Wl,--export-dynamic -fopenmp   -o R.bin Rmain.o
-L../../lib -lR -lRblas
../../lib/libR.so: undefined reference to `rl_sort_completion_matches'
collect2: error: ld returned 1 exit status


This appears to be an upstream bug; despite that we configure with
'--with-readline="$SAGE_LOCAL"', the corresponding '-L...' is missing in
the linker command such that your system's libreadline gets picked up.


Or maybe not:

## Readline.

# Check whether --with-readline was given.
if test "${with_readline+set}" = set; then :
   withval=$with_readline; if test "${withval}" = no; then
   use_readline=no
else
   use_readline=yes
fi

else
   use_readline=yes
fi
...
[A couple of readline-related tests, probably disabling it.]
...


I.e., R's 'configure' only supports enabling/disabling use of readline,
but not specifying a non-standard location [that way].


This is actually a regression introduced by the "clean-up" in #13443
[1], which removed adding "-L${SAGE_LOCAL}/lib" to LDFLAGS (and similar
for CPPFLAGS, but the latter in contrast works);  if LIBRARY_PATH is
already set, ${SAGE_LOCAL}/lib gets *appended* instead of prepended (in
sage-env).


Ursula, can you confirm that `LIBRARY_PATH` is set in your shell?

E.g., what does the following give?

$ ./sage --sh -c 'echo $LIBRARY_PATH'


See http://trac.sagemath.org/sage_trac/ticket/14709

for why this may not work even if LIBRARY_PATH just contained 
${SAGE_LOCAL}/lib (i.e., was empty outside of the Sage environment) -- 
the linker might still pick up some system-wide readline library, which 
here was the cause of the error.



-leif

--
() The ASCII Ribbon Campaign
/\   Help Cure HTML E-Mail

--
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sage-devel] Re: R build error in 5.10 was Re: ATLAS build error on CentOS

[Ursula Whitcher]

On 6/5/2013 10:13 PM, leif wrote:


Ursula, can you confirm that `LIBRARY_PATH` is set in your shell?

E.g., what does the following give?

$ ./sage --sh -c 'echo $LIBRARY_PATH'



/home/whitchua/sage-5.10.rc0/local/lib

--Ursula.

--
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sage-devel] Re: R build error in 5.10 was Re: ATLAS build error on CentOS

[Ursula Whitcher]

On 6/5/2013 8:50 PM, leif wrote:


This appears to be an upstream bug; despite that we configure with
'--with-readline="$SAGE_LOCAL"', the corresponding '-L...' is missing in
the linker command such that your system's libreadline gets picked up.

You can try:

$ env LDFLAGS="-L/home/whitchua/sage-5.10.rc0/local/lib" ./sage -i
spkg/standard/r-2.15.2.p2.spkg

and if that worked, resume the normal build with

$ make


This worked!  Thank you so much for your help.

UAW

--
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-devel] Re: R build error in 5.10 was Re: ATLAS build error on CentOS

[leif]

leif wrote:

leif wrote:

Ursula Whitcher wrote:

...
gcc -std=gnu99 -shared -fopenmp   -o libR.so CConverters.o
CommandLineArgs.o Rdynload.o Renviron.o RNG.o agrep.o apply.o
arithmetic.o array.o attrib.o base.o bind.o builtin.o character.o
coerce.o colors.o complex.o connections.o context.o cov.o cum.o dcf.o
datetime.o debug.o deparse.o deriv.o devices.o dotcode.o dounzip.o
dstruct.o duplicate.o engine.o envir.o errors.o eval.o format.o
fourier.o gevents.o gram.o gram-ex.o gramLatex.o gramRd.o graphics.o
grep.o identical.o inlined.o inspect.o internet.o iosupport.o lapack.o
list.o localecharset.o logic.o main.o mapply.o match.o memory.o model.o
names.o objects.o optim.o optimize.o options.o par.o paste.o platform.o
plot.o plot3d.o plotmath.o print.o printarray.o printvector.o
printutils.o qsort.o random.o raw.o registration.o relop.o rlocale.o
saveload.o scan.o seq.o serialize.o size.o sort.o source.o split.o
sprintf.o startup.o subassign.o subscript.o subset.o summary.o
sysutils.o unique.o util.o version.o vfonts.o xxxpr.o
../unix/Rembedded.o ../unix/libunix.a ../appl/libappl.a
../nmath/libnmath.a ../extra/zlib/libz.a ../extra/bzip2/libbz2.a
../extra/pcre/libpcre.a ../extra/tre/libtre.a  ../extra/xz/liblzma.a
-L../../lib -lRblas -lgfortran -lm -lquadmath   -lreadline -lncurses
-lrt -ldl -lm
...
gcc -std=gnu99 -Wl,--export-dynamic -fopenmp   -o R.bin Rmain.o
-L../../lib -lR -lRblas
../../lib/libR.so: undefined reference to `rl_sort_completion_matches'
collect2: error: ld returned 1 exit status


This appears to be an upstream bug; despite that we configure with
'--with-readline="$SAGE_LOCAL"', the corresponding '-L...' is missing in
the linker command such that your system's libreadline gets picked up.


Or maybe not:

## Readline.

# Check whether --with-readline was given.
if test "${with_readline+set}" = set; then :
   withval=$with_readline; if test "${withval}" = no; then
   use_readline=no
else
   use_readline=yes
fi

else
   use_readline=yes
fi
...
[A couple of readline-related tests, probably disabling it.]
...


I.e., R's 'configure' only supports enabling/disabling use of readline,
but not specifying a non-standard location [that way].


This is actually a regression introduced by the "clean-up" in #13443 
[1], which removed adding "-L${SAGE_LOCAL}/lib" to LDFLAGS (and similar 
for CPPFLAGS, but the latter in contrast works);  if LIBRARY_PATH is 
already set, ${SAGE_LOCAL}/lib gets *appended* instead of prepended (in 
sage-env).



Ursula, can you confirm that `LIBRARY_PATH` is set in your shell?

E.g., what does the following give?

$ ./sage --sh -c 'echo $LIBRARY_PATH'


-leif

[1] http://trac.sagemath.org/sage_trac/ticket/13443

--
() The ASCII Ribbon Campaign
/\   Help Cure HTML E-Mail

--
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-devel] Re: R build error in 5.10 was Re: ATLAS build error on CentOS

[leif]

leif wrote:

Ursula Whitcher wrote:

...
gcc -std=gnu99 -shared -fopenmp   -o libR.so CConverters.o
CommandLineArgs.o Rdynload.o Renviron.o RNG.o agrep.o apply.o
arithmetic.o array.o attrib.o base.o bind.o builtin.o character.o
coerce.o colors.o complex.o connections.o context.o cov.o cum.o dcf.o
datetime.o debug.o deparse.o deriv.o devices.o dotcode.o dounzip.o
dstruct.o duplicate.o engine.o envir.o errors.o eval.o format.o
fourier.o gevents.o gram.o gram-ex.o gramLatex.o gramRd.o graphics.o
grep.o identical.o inlined.o inspect.o internet.o iosupport.o lapack.o
list.o localecharset.o logic.o main.o mapply.o match.o memory.o model.o
names.o objects.o optim.o optimize.o options.o par.o paste.o platform.o
plot.o plot3d.o plotmath.o print.o printarray.o printvector.o
printutils.o qsort.o random.o raw.o registration.o relop.o rlocale.o
saveload.o scan.o seq.o serialize.o size.o sort.o source.o split.o
sprintf.o startup.o subassign.o subscript.o subset.o summary.o
sysutils.o unique.o util.o version.o vfonts.o xxxpr.o
../unix/Rembedded.o ../unix/libunix.a ../appl/libappl.a
../nmath/libnmath.a ../extra/zlib/libz.a ../extra/bzip2/libbz2.a
../extra/pcre/libpcre.a ../extra/tre/libtre.a  ../extra/xz/liblzma.a
-L../../lib -lRblas -lgfortran -lm -lquadmath   -lreadline -lncurses
-lrt -ldl -lm
...
gcc -std=gnu99 -Wl,--export-dynamic -fopenmp   -o R.bin Rmain.o
-L../../lib -lR -lRblas
../../lib/libR.so: undefined reference to `rl_sort_completion_matches'
collect2: error: ld returned 1 exit status


This appears to be an upstream bug; despite that we configure with
'--with-readline="$SAGE_LOCAL"', the corresponding '-L...' is missing in
the linker command such that your system's libreadline gets picked up.


Or maybe not:

## Readline.

# Check whether --with-readline was given.
if test "${with_readline+set}" = set; then :
  withval=$with_readline; if test "${withval}" = no; then
  use_readline=no
else
  use_readline=yes
fi

else
  use_readline=yes
fi
...
[A couple of readline-related tests, probably disabling it.]
...


I.e., R's 'configure' only supports enabling/disabling use of readline, 
but not specifying a non-standard location [that way].



-leif

--
() The ASCII Ribbon Campaign
/\   Help Cure HTML E-Mail

--
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[sage-devel] Re: R build error in 5.10 was Re: ATLAS build error on CentOS

[leif]

Ursula Whitcher wrote:

On 6/5/2013 11:39 AM, William Stein wrote:


Given that she is having build issues, I strongly recommend just
building a clean sage-5.10.rc0 from scratch.  There may be (many?)
changes to the Sage library, etc., that are related to updated ATLAS
for the first time in years, and doing what you suggest above might
lead to a "broken" sage, in the sense that it doesn't pass its test
suite.


All right, this build failed too, but in a different way.  This time,
the offending package was R.



[SNIP]


gcc -std=gnu99 -shared -fopenmp   -o libR.so CConverters.o
CommandLineArgs.o Rdynload.o Renviron.o RNG.o agrep.o apply.o
arithmetic.o array.o attrib.o base.o bind.o builtin.o character.o
coerce.o colors.o complex.o connections.o context.o cov.o cum.o dcf.o
datetime.o debug.o deparse.o deriv.o devices.o dotcode.o dounzip.o
dstruct.o duplicate.o engine.o envir.o errors.o eval.o format.o
fourier.o gevents.o gram.o gram-ex.o gramLatex.o gramRd.o graphics.o
grep.o identical.o inlined.o inspect.o internet.o iosupport.o lapack.o
list.o localecharset.o logic.o main.o mapply.o match.o memory.o model.o
names.o objects.o optim.o optimize.o options.o par.o paste.o platform.o
plot.o plot3d.o plotmath.o print.o printarray.o printvector.o
printutils.o qsort.o random.o raw.o registration.o relop.o rlocale.o
saveload.o scan.o seq.o serialize.o size.o sort.o source.o split.o
sprintf.o startup.o subassign.o subscript.o subset.o summary.o
sysutils.o unique.o util.o version.o vfonts.o xxxpr.o
../unix/Rembedded.o ../unix/libunix.a ../appl/libappl.a
../nmath/libnmath.a ../extra/zlib/libz.a ../extra/bzip2/libbz2.a
../extra/pcre/libpcre.a ../extra/tre/libtre.a  ../extra/xz/liblzma.a
-L../../lib -lRblas -lgfortran -lm -lquadmath   -lreadline -lncurses
-lrt -ldl -lm
make[7]: Entering directory
`/home/whitchua/sage-5.10.rc0/spkg/build/r-2.15.2.p2/src/src/main'
mkdir -p --
/home/whitchua/sage-5.10.rc0/spkg/build/r-2.15.2.p2/src/bin/exec
make[7]: Leaving directory
`/home/whitchua/sage-5.10.rc0/spkg/build/r-2.15.2.p2/src/src/main'
make[6]: Leaving directory
`/home/whitchua/sage-5.10.rc0/spkg/build/r-2.15.2.p2/src/src/main'
make[6]: Entering directory
`/home/whitchua/sage-5.10.rc0/spkg/build/r-2.15.2.p2/src/src/main'
gcc -std=gnu99 -I../../src/extra/zlib -I../../src/extra/bzip2
-I../../src/extra/pcre -I../../src/extra  -I../../src/extra/xz/api -I.
-I../../src/include -I../../src/include  -DHAVE_CONFIG_H   -fopenmp
-fpic  -g -O2   -c Rmain.c -o Rmain.o
gcc -std=gnu99 -Wl,--export-dynamic -fopenmp   -o R.bin Rmain.o
-L../../lib -lR -lRblas
../../lib/libR.so: undefined reference to `rl_sort_completion_matches'
collect2: error: ld returned 1 exit status


This appears to be an upstream bug; despite that we configure with 
'--with-readline="$SAGE_LOCAL"', the corresponding '-L...' is missing in 
the linker command such that your system's libreadline gets picked up.


You can try:

$ env LDFLAGS="-L/home/whitchua/sage-5.10.rc0/local/lib" ./sage -i 
spkg/standard/r-2.15.2.p2.spkg


and if that worked, resume the normal build with

$ make


(Or before reissuing 'make' / 'sage -i ...', do e.g. 'export MAKE="make 
-j8"' to speed up things this time... :-) )



-leif

--
() The ASCII Ribbon Campaign
/\   Help Cure HTML E-Mail

--
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.