Hi Hiro, On Wed, 10 Oct 2001 17:21:23 -0500 (CDT) Hironori Ito <hito@students.phsx.ukans.edu> wrote concerning "Re: new PC and gcc": > HELP!!! I'll try :-) > Ok. With all suggestions, I tried the following. > > 1. export CXX='g++ -V egcs-2.91.66' > 2. export CC='gcc -V egcs-2.91.66' > > Then, do the usual things to compile brat. It failed because it > complains about header files. Then, I tried > > 1. export CXX='g++ -V egcs-2.91.66 -I/usr/i386-glibc21-linux/include' > 2. export CC='g++ -V egcs-2.91.66 -I/usr/i386-glib21-linux/include' > > Then, it failed again with the same message. The following is the output > of the compilation. > > > make[2]: Entering directory `/home/hito/temp/brat/data/abc' > /bin/sh ../../libtool --mode=compile g++ -V egcs-2.91.66 > -I/usr/i386-glibc21-linux/include > -DPACKAGE=\"brat\" -DVERSION=\"2.1.9\" -DONL_unix=1 > -I. -I. -I. -D_REENTRANT > -I/afs/rhic/opt/brahms/root/root_v3.01.06/include -I../../data/abc Argh! Do not use ROOT 3.01/06 - it has a serious bug that makes it unusable for us. Find the appropiate mail on brahms-dev-l. Also, do not use the ROOT installed in /afs/rhic/opt/brahms/root/root_v3.01.06 since it's not supported. Use the one in /afs/rhic/opt/brahms/[pro|new] Just insert /afs/rhic/opt/brahms/[pro|new]/bin in your PATH environment variable. > -I../../data/params -I../../util > -I../../managers -I../../modules/abc > -I../../modules/util -DMAJOR=2 -DMINOR=1 -DREVISION=9 -g -O2 > -c BrClonesArray.cxx > g++ -V egcs-2.91.66 -I/usr/i386-glibc21-linux/include > -DPACKAGE=\"brat\" -DVERSION=\"2.1.9\" -DONL_unix=1 > -I. -I. -I. -D_REENTRANT -I/afs/rhic/opt/brahms/root/root_v3.01.06/include > -I../../data/abc -I../../data/params -I../../util -I../../managers > -I../../modules/abc -I../../modules/util -DMAJOR=2 -DMINOR=1 -DREVISION=9 > -g -O2 -Wp,-MD,.deps/BrClonesArray.pp -c BrClonesArray.cxx -fPIC -DPIC -o > BrClonesArray.lo > /usr/i386-glibc21-linux/include/bits/string2.h: In function `char * > __strcpy_small(char *, __uint16_t, __uint16_t, __uint32_t, __uint32_t, > size_t)': > In file included from /usr/i386-glibc21-linux/include/string.h:346, > from > /afs/rhic/opt/brahms/root/root_v3.01.06/include/Bytes.h:40, > from > /afs/rhic/opt/brahms/root/root_v3.01.06/include/TBuffer.h:28, > from > /afs/rhic/opt/brahms/root/root_v3.01.06/include/TObject.h:192, > from > /afs/rhic/opt/brahms/root/root_v3.01.06/include/TCollection.h:26, > from > /afs/rhic/opt/brahms/root/root_v3.01.06/include/TSeqCollection.h:26, > from > /afs/rhic/opt/brahms/root/root_v3.01.06/include/TObjArray.h:26, > from > /afs/rhic/opt/brahms/root/root_v3.01.06/include/TClonesArray.h:26, > from BrClonesArray.h:45, > from BrClonesArray.cxx:13: > /usr/i386-glibc21-linux/include/bits/string2.h:408: warning: ANSI C++ > forbids implicit conversion from `void *' in initialization > /usr/i386-glibc21-linux/include/bits/string2.h:419: warning: ANSI C++ > forbids using pointer of type `void *' in arithmetic > /usr/i386-glibc21-linux/include/bits/string2.h:419: warning: ANSI C++ > forbids implicit conversion from `void *' in assignment > /usr/i386-glibc21-linux/include/bits/string2.h:427: warning: ANSI C++ > forbids using pointer of type `void *' in arithmetic > /usr/i386-glibc21-linux/include/bits/string2.h:427: warning: ANSI C++ > forbids implicit conversion from `void *' in assignment > /usr/i386-glibc21-linux/include/bits/string2.h:432: warning: ANSI C++ > forbids using pointer of type `void *' in arithmetic This error I've not seen while compiling BRAT with ROOT 3.<xx>/<xx> using GCC 2.91.<xx>, GCC 2.95.2, nor 'GCC 2.96'. Try tracking down the actual location where __strcpy_small is used. Try taking BrClonesArray out of the compilation (Makefile.am) - this class should not be used anymore. If it is, it should be replaced by either a TClonesArray or a proper nested class, a la BrMultRdo, BrTofRdo, and BrChkvRdo. > Ok. Then, I found the script called i386-glibc21-linux-env.sh in > /usr/i386-glibc21-linux/bin directory. Therefore, I run this things, and > tried again. It still failed with the same message. (I tried some other > things, but the results are the same.) Are you using a SH like shell? Did you run it as prompt% . /usr/i386-glibc21-linux/bin/i386-glibc21-linux-env.sh or as prompt% /usr/i386-glibc21-linux/bin/i386-glibc21-linux-env.sh The former is the right way, since the later starts a sub-shell and does not modify the environment in the mother shell. If you're running a CSH like shell, do in the prompt: setenv LDEMULATION elf_i386_glibc21 setenv GCC_EXEC_PREFIX /usr/i386-glibc21-linux/lib/gcc-lib/ You can also do directly in a SH like shell export LDEMULATION=elf_i386_glibc21 export GCC_EXEC_PREFIX=/usr/i386-glibc21-linux/lib/gcc-lib/ I'm not sure what you mean in the above. Did you see the same messages - i.e., the ones above - when you compiled with CXX='g++ -V egcs-2.91.66' CC='gcc -V egcs-2.91.66' > Do someone have any idea to make it work??? I am tring at a new PC > called "concerto.hirg.bnl.gov". I can't get access to that machine. Is it accesible from the outside of BNL? Ok, I just did prompt% rpm -qa | grep egcs compat-egcs-6.2-1.1.2.14 compat-egcs-g77-6.2-1.1.2.14 compat-egcs-c++-6.2-1.1.2.14 compat-egcs-objc-6.2-1.1.2.14 prompt% rpm -ql compat-egcs-6.2-1.1.2.14 | less /usr/bin/egcs /usr/bin/i386-glibc21-linux-gcc /usr/bin/kgcc /usr/i386-glibc21-linux /usr/i386-glibc21-linux/bin /usr/i386-glibc21-linux/bin/i386-glibc21-linux-env.sh ... prompt% rpm -ql compat-egcs-c++-6.2-1.1.2.14 | less /usr/bin/egcs++ /usr/bin/i386-glibc21-linux-c++ /usr/bin/i386-glibc21-linux-g++ on a Red Hat 7.1 machine. Notice the programs egcs++ egcs As a small test I did prompt% echo "int main() { return 0; }" > foo.c prompt% egcs -v foo.c Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/specs gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) ... #include "..." search starts here: #include <...> search starts here: /usr/local/include /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/include /usr/include End of search list. ... So it seems to me that you may get away with doing CC=egcs CXX=egcs++ ./configure [Do remember to remove config.cache] I hope this helps you. Yours, Christian Holm Christensen ------------------------------------------- Address: Sankt Hansgade 23, 1. th. Phone: (+45) 35 35 96 91 DK-2200 Copenhagen N Cell: (+45) 28 82 16 23 Denmark Office: (+45) 353 25 305 Email: cholm@nbi.dk Web: www.nbi.dk/~cholm
This archive was generated by hypermail 2b30 : Thu Oct 11 2001 - 06:53:53 EDT