Description of Problem: The compilation of a bzImage for any 2.4.xx kernel fails due to a segmentation fault in program as from binutils(2.11.93.0.2-7 and the previous version). I get a segmentation fault using gcc(3.1.0-22 and the previous version) and with kgcc 1.1.2-40. gcc -D__ASEMBLY__ -D__KERNEL__ -I/usr/src/linux/include -traditional -c entry.S -o entry.o gcc: Internal error: Segmentation fault (program as) Please submit a full bug report. See <URL:http://bugzilla.redhat.com/bugzilla/> for instructions. make[1]: *** [entry.o] Error 1 make[1]: Leaving directory '/usr/src/linux/arch/i386/kernel' make: *** [_dir_arch/i386/kernel[ Error 2 kgcc -D__ASSEMBLY__ -D__KERNEL__ -I/usr/src/linux/include -traditional -c trampoline.S -o trampoline.o kgcc: Internal compiler error: program as got fatal signal 11 make[1] *** [trampoline.o] Error 1 make[1] Leaving directory '/usr/src/linux/arch/i386/kernel' make: *** [_dir_arch/i386/kernel] Error 2 make CFLAGS="-D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -Wno-trigraphs -02 -fomit-frame-pointer -fno-strict-aliasing -fno-common -pipe -march=i686 " -C arch/i386/kernel make[1]: Entering directory '/usr/src/linux/arch/i386/kernel' rm -f kernel.o ld -m elf_i386 -r -o kernel.o process.o semaphore.o signal.o entry.o traps.o irq.o vm86.o ptrace.o i8259.o ioport.o ldt.o setup.o time.o sys_i386.o pci-dma.o i386_ksyms.o i387.o bluesmoke.o dmi_scan.o pci-i386.o pci-pc.o pci-irq.o smp.o smpboot.o trampoline.o mpparse.o apic.o nmi.o io_spic.o acpitable.o entry.o: file not recognized: File truncated make[1]: *** [kernel.o] Error 1 make[1]: Leaving directory '/usr/src/linux/arch/i386/kernel' make: *** [_dir_arch/i386/kernel] Error 2 The compilation usually halts with some message reporting an error with entry.o. If using kgcc I get an error at trampoline.o, if I do make bzImage again it stops with the above error (entry.o truncated), all further attempts halt at this same location. I have tried to determine if the problem is due to bad RAM using "dd if=/dev/hda4 of=/dev/null bs-1024k count=768" and then "make bzImage" again; and the program halts at the exact same point again. Thus I do not think that the RAM is bad (at least I hope). I do not have any other problems using gcc-3.1.0 and most programs compile beautifully. Version-Release number of selected component (if applicable): gcc-3.1.0-22 and previous Rawhide release kgcc-1.1.2-40 binutils-2.11.93.0.2-7 and previous Rawhide release. RedHat Kernels: 2.4.7-10, 2.4.9-31, 2.4.17-0.18 and regular kernel 2.4.17 How Reproducible: Very reproducible. It does not matter how soon or long after the error I try again, the compilation halts at the same place and gives the same error. Steps to Reproduce: 1. any form of make config (make xconfig, make config. make oldconfig) 2. make dep with gcc or kgcc 3. make bzImage (or zImage) (with kgcc it firsts reports and error at the compilation of trampoline.o due to a segmentation fault in program as, all further attempts report that stop at kernel.o with the error entry.o: file is not recognized: File truncated.) Actual Results: Expected Results: Additional Information: I have a Dell Dimension 8100 with a 1.4 GHz Pentium 4, 768 Mb RDRAM, 500 Mb Swap space, Rawhide 2.0 (with a few non gcc or kernel related RedHat 7.2 and other distribution components), XFree86-4.2.0-32, on a ~40Gb IDE partition.
I also tried the two sets of RDRAM chips alone (2x128Mb and 2x256Mb) both resulted in the same error. In addition I tried a UPS system and also got the same error.
It is as that is segfaulting, not gcc. Can you please run gcc -save-temps -D__ASEMBLY__ -D__KERNEL__ -I/usr/src/linux/include -traditional -c entry.S -o entry.o and include here its output and entry.s it creates? Thanks.
Created attachment 48040 [details] output of requested command
Created attachment 48041 [details] Ignore above file. The requested file, entry.o, could not be sent because it is empty.
Cannot reproduce this. What glibc are you using? If glibc-2.2.90-{4,5,6,7}, please upgrade to 2.2.90-8.
The upgrade to glibc-2.2.90-8 worked perfectly (I am surprised that I missed the upgrade release). I guess the problem does not matter any more. Thanks for your time anyway.