Red Hat Bugzilla – Bug 17393
rpm --rebuild glibc-2.1.3-19.src.rpm --> no _start
Last modified: 2016-11-24 10:06:11 EST
what went wrong?
[/usr/lib]$ nm crt1.o
nm: crt1.o: no symbols
I have gcc-2.95.2, binutils-2.10, redhat 5.0+many updates.
I have compiled glibc many times, but this is the first time crt*.o broke.
now I installed glibc-2.1.3-21 rpms from Redhat and they seem to work.
however, if I compile cdrdao-1.1.4 and run it, it segfaults. gdb shows:
#0 __pthread_mutex_lock (mutex=0x0) at mutex.c:82
#1 0x2aad5c2b in __flockfile (stream=0x80b8cc0) at lockfile.c:32
#2 0x2ab5d760 in buffered_vfprintf (s=0x80b8cc0, format=0x8097e83 "",
#3 0x2ab58966 in _IO_vfprintf (s=0x80b8cc0, format=0x8097e83 "",
ap=0x7fffe758) at vfprintf.c:1029
#4 0x804b9e8 in message (level=0,
fmt=0x804b801 "t", '\220' <repeats 14 times>,
"U\211e\203l\b\203=h<\n\b") at main.cc:258
#5 0x2ab2b9cb in __libc_start_main (main=0x804ebb0 <main>, argc=1,
init=0x804ad68 <_init>, fini=0x8097c6c <_fini>, rtld_fini=0x2aab5e60
stack_end=0x7fffe8ac) at ../sysdeps/generic/libc-start.c:92
I don't know what's broken... is this related to the mutex fix/change in
cdrdao binary from cdrdao homepage works ok.
I forgot to tell one more prob I had with self-compiled 2.1.3-19.
cdrecord failed with
cdrecord: Input/output error. send opc: scsi sendcmd: retryable error
status: 0x0 (GOOD STATUS)
CDB: 54 01 00 00 00 00 00 00 00 00
cdrecord: Resource temporarily unavailable. OPC failed.
after upgrading to glibc-2.1.3-21, cdrecord works ok.
what could cause this kind of behaviour?
This is for an very old glibc version. If you have the same problem with RHL9
we'd like to hear about it. Otherwise I'll close the bug soon.
No reply in almost six months. Closing as not a bug.