It was on the SuSE ftp site ftp://ftp.suse.com/pub/projects/gcc/8.2/ Zsolt Donn aka n5xwb Washburn wrote:
Lukacs Zsolt wrote:
Hello list
I have ask it on another list but there was no answer. So I will try it here.
I have upgraded to the new gcc3.3.1 from the SuSE ftp site my box (SuSE 8.2) and I tried to compile mplayer but it does not succeded. The compilation went just fine, but there was a problem with the linker. It seems that if a variable/function is declared static it does not see it resulting in an undefined reference.
I have upgraded the following packages: cpp-3.3.1-0.i586.rpm gcc-c++-3.3.1-0.i586.rpm libstdc++-3.3.1-0.i586.rpm gcc-3.3.1-0.i586.rpm libgcc-3.3.1-0.i586.rpm libstdc++-devel-3.3.1-0.i586.rpm
Do I have to install some other package?
Any help would be apreciated. Thanks
Zsolt
Hey Lukacs;
What ftp site did you find theses at? ftp://ftp.suse.com?
On Wednesday 10 September 2003 06:29, Lukacs Zsolt wrote:
I have upgraded to the new gcc3.3.1 from the SuSE ftp site my box (SuSE 8.2) and I tried to compile mplayer but it does not succeded. The compilation went just fine, but there was a problem with the linker. It seems that if a variable/function is declared static it does not see it resulting in an undefined reference.
I have upgraded the following packages: cpp-3.3.1-0.i586.rpm gcc-c++-3.3.1-0.i586.rpm libstdc++-3.3.1-0.i586.rpm gcc-3.3.1-0.i586.rpm libgcc-3.3.1-0.i586.rpm libstdc++-devel-3.3.1-0.i586.rpm
Do I have to install some other package?
Perhaps the problem is this: http://www.ussg.iu.edu/hypermail/linux/kernel/0309.0/1586.html (see also the begin of the thread, http://www.ussg.iu.edu/hypermail/linux/kernel/0309.0/1440.html ) We (I and, independently, a coworker) have found that SuSE's gcc-3.3.1-0 can't even compile the linux-2.4.20.SuSE kernel. Or Mantel's 21, or vanilla 22. Actually it compiles, the result just doesn't boot. But compiling these kernels with gcc3.3-43 (the 'final' 3.3.0, which recently was taken out from the /pub/projects/gcc) is ok. I also observed that compiling python2.3 with gcc3.3 beta from the 8.2 CDs is OK. It runs 'make test' smoothly. Now, compiling it with gcc-3.3.1-0, and running 'make test', gives you a segmentation fault at test_re. Bus this can be very well python issue, and not a gcc issue. I hope they fix this soon. Or, perhaphs, put 3.3-43 back at the ftp server. Regards, Ivan
Ivan Nazarenko wrote:
On Wednesday 10 September 2003 06:29, Lukacs Zsolt wrote:
Perhaps the problem is this: http://www.ussg.iu.edu/hypermail/linux/kernel/0309.0/1586.html
(see also the begin of the thread, http://www.ussg.iu.edu/hypermail/linux/kernel/0309.0/1440.html )
We (I and, independently, a coworker) have found that SuSE's gcc-3.3.1-0 can't even compile the linux-2.4.20.SuSE kernel. Or Mantel's 21, or vanilla 22. Actually it compiles, the result just doesn't boot.
But compiling these kernels with gcc3.3-43 (the 'final' 3.3.0, which recently was taken out from the /pub/projects/gcc) is ok.
I also observed that compiling python2.3 with gcc3.3 beta from the 8.2 CDs is OK. It runs 'make test' smoothly. Now, compiling it with gcc-3.3.1-0, and running 'make test', gives you a segmentation fault at test_re. Bus this can be very well python issue, and not a gcc issue.
I hope they fix this soon. Or, perhaphs, put 3.3-43 back at the ftp server.
Regards,
Ivan
I have a stock CD 8.2 GCC 3.# compiler and binutils system. I have found that most things seem to compile cleanly except for many "warnings" on this reload. I even got a stock "make cloneconfig" (which was edited from i586 to k6) and modules to compile. Modules did fail on rio.o. I removed it from .config and the kernel is running (after mkinitrd). Kernel Modules always seem to be the problem with a kernel compile. I have seen/heard that this is because different people write the modules and not all code is the same. I tried to compile LyX and it states there is a internal compiler problem. However, I tried "scribus" (www.scribus.com) and it compiled very clean. Do you suggest upgrading GCC? If so which version is best -- 73 de Donn Washburn __ " http://www.hal-pc.org/~n5xwb " Ham Callsign N5XWB / / __ __ __ __ __ __ __ 307 Savoy St. / /__ / / / \/ / / /_/ / \ \/ / Sugar Land, TX 77478 /_____/ /_/ /_/\__/ /_____/ /_/\_\ LL# 1.281.242.3256 a MSDOS Virus "Free Zone" OS Email: n5xwb@hal-pc.org Info: http://www.knoppix.net
On Wednesday 10 September 2003 11:10, Donn aka n5xwb Washburn wrote:
I have a stock CD 8.2 GCC 3.# compiler and binutils system. I have found that most things seem to compile cleanly except for many "warnings" on this reload. I even got a stock "make cloneconfig" (which was edited from i586 to k6) and modules to compile. Modules did fail on rio.o. I removed it from .config and the kernel is running (after mkinitrd). Kernel Modules always seem to be the problem with a kernel compile. I have seen/heard that this is because different people write the modules and not all code is the same. I tried to compile LyX and it states there is a internal compiler problem. However, I tried "scribus" (www.scribus.com) and it compiled very clean.
Do you suggest upgrading GCC? If so which version is best
The gcc3.3 beta from the CD is Ok. The problem is the upgrade that is now in ftp.suse.com/pub/projects/gcc/8.2, the gcc3.3.1-0, that is broken concerning linux-2.4.20.SuSE compiles. Regards, Ivan
participants (3)
-
Donn aka n5xwb Washburn -
Ivan Nazarenko -
Lukacs Zsolt