CBF Lib

[DuncanSneddon]

Hi all,

I've been given the task of using CBFLib to allow the reading and writing of CBF and imgCIF (Like there is a differance, pah!!) into the GDA at Diamond. I'm compiling the souse to make libraries and I'm getting an error when the build is being tested.

Here it is:

Code: Select all

diff -u test_fcb_read_testflatout.out test_fcb_read_testflatout_orig.out
--- test_fcb_read_testflatout.out       2009-01-30 11:55:38.000000000 +0000
+++ test_fcb_read_testflatout_orig.out  2009-01-30 11:55:21.000000000
+++ +0000
@@ -4,4 +4,4 @@
   1000 x 1000 I4 TEST, WITH -3 on diag and transpose
   1000 x 1000 I2 TEST, WITH -3 on diag and transpose
   50 x 60 x 70 3D_I4 TEST
-   FCB_READ_IMAGE_3D_I4 ERROR:           -4
+ TESTS COMPLETED
make: [extra] Error 1 (ignored)
echo testflatpackedout.cbf | time ./bin/test_fcb_read_image > test_fcb_read_testflatpackedout.out
1.25user 0.00system 0:01.26elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (0major+1834minor)pagefaults 0swaps
diff -u test_fcb_read_testflatpackedout.out test_fcb_read_testflatpackedout_orig.out
--- test_fcb_read_testflatpackedout.out 2009-01-30 11:55:39.000000000 +0000
+++ test_fcb_read_testflatpackedout_orig.out    2009-01-30 11:55:21.000000000 +0000
@@ -4,5 +4,4 @@
   1000 x 1000 I4 TEST, WITH -3 on diag and transpose
   1000 x 1000 I2 TEST, WITH -3 on diag and transpose
   50 x 60 x 70 3D_I4 TEST
- EARLY TERMINATION AT                146492
-   FCB_READ_IMAGE_3D_I4 ERROR:           -1
+ TESTS COMPLETED
make: [extra] Error 1 (ignored)

Have any of you guys used cbflib before? Should I just ignore the error, assume I'm never going to be using a 3D data set and move on? Any advice or moral boosting abuse of CBF will be very welcome.

Cheers

D

[pascalp]

We're going to need much more informations.

I compiled it and runned the tests without any problems.

I used gcc 4.3.3 on Archlinux, yes, the brand new one
Target: i686-pc-linux-gnu
Configuré avec: ../configure --prefix=/usr --enable-shared --enable-languages=c,c++,fortran,objc,obj-c++,treelang --enable-threads=posix --mandir=/usr/share/man --infodir=/usr/share/info --enable-__cxa_atexit --disable-multilib --libdir=/usr/lib --libexecdir=/usr/lib --enable-clocale=gnu --disable-libstdcxx-pch --with-tune=generic
Modèle de thread: posix
gcc version 4.3.3 (GCC)


What compiler are you using ? which version ? on what platform ? What modifications on the Makefile ?

[DuncanSneddon]

Sorry

We're on Red Hat 4EL at the moment and using gcc 4.1.2

We're using ciflib 0.8.0 and we haven't really modified the makefile but we did modify the compiler configuration to match the gfortran version.

D

[DuncanSneddon]

I think I fixed it, or more accurately my mate who know ahell of a lot more about Linux than me fixed it

Much thanks for help. The solution involved a hack of the Makefile and commenting out a few duplications?? in the cbf.h file