ld: 0711-593 SEVERE ERROR: Symbol C_BSTAT - anyone got a better solution?

[David Kirkby]

There appears to be a problem in some updates to AIX (at least 5.3 and
6.1) which causes problems when building code with gcc. For example:

gcc -c -DHAVE_CONFIG_H -Ded_PROGRAM=\"/usr/bin/ed\" -I. -I. -g -O2
util.c
gcc -c -DHAVE_CONFIG_H -Ded_PROGRAM=\"/usr/bin/ed\" -I. -I. -g -O2
version.c
gcc -c -DHAVE_CONFIG_H -Ded_PROGRAM=\"/usr/bin/ed\" -I. -I. -g -O2
xmalloc.c
gcc -o patch -g -O2 error.o malloc.o realloc.o addext.o argmatch.o
backupfile.o basename.o dirname.o getopt.o getopt1.o inp.o maketime.o
partime.o patch.o pch.o quote.o quotearg.o quotesys.o util.o version.o
xmalloc.o
ld: 0711-593 SEVERE ERROR: Symbol C_BSTAT (entry 408) in object
error.o:
The symbol refers to a csect with symbol number 0, which was
not
found. The new symbol cannot be associated with a csect and
is being ignored.
ld: 0711-593 SEVERE ERROR: Symbol C_BSTAT (entry 411) in object
error.o:
The symbol refers to a csect with symbol number 0, which was
not
found. The new symbol cannot be associated with a csect and
is being ignored.


This problem is noted on an IBM forum

http://www.ibm.com/developerworks/forums/thread.jspa?threadID=348558

its been reported as a gcc bug

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46072

but from comments there, it appears this might not be a gcc issue at
all, as it appears one of IBM's AIX patches may have triggered this.

Does anyone know of a proper fix to this? (Compiling without debugging
information, or using -g0 is not always very practical, as some open-
source projects add -g by default. )

Has this been reported officially to IBM as a bug? If so, is there a
link available to it?

I'd like to try to port the Sage maths software to AIX

http://www.sagemath.org/

but this bug is a real show stopper, as it stops one building many
bits of Sage. The GNU 'patch' program is one that's affected. That can
be fixed by setting CFLAGS to -g0, but it would be a huge task to
change all of the software to disable debug information, apart from
the fact it would make debugging harder!!

I'm using AIX 5.2, Technology level 12, service pack 2. i.e.
5300-12-02-1036. I'm aware this is not the latest, but having looked
through a list of the fixes in the latest service pack, there does no
appear to be anything related to this.

Dave

[Thomas Braunbeck]

Post by David Kirkby
There appears to be a problem in some updates to AIX (at least 5.3 and

Check for IZ91228 and IZ94359 - the fix should be in 5.3 TL12 SP3
But then there is still IZ98134 and an ifix for 5.3 TL12 SP2 seems
to be available end of this week. I would wait for this ifix. It is
easier to 'test' such an ifix than updating AIX to the next SP (and
still not having an fix for IZ98134)

[No Body]

Post by Thomas Braunbeck

Post by David Kirkby
There appears to be a problem in some updates to AIX (at least 5.3 and

Check for IZ91228 and IZ94359 - the fix should be in 5.3 TL12 SP3
But then there is still IZ98134 and an ifix for 5.3 TL12 SP2 seems
to be available end of this week. I would wait for this ifix. It is
easier to 'test' such an ifix than updating AIX to the next SP (and
still not having an fix for IZ98134)

definitely chase the ifix if you have a support contract; it does fix
the issue. i found it in the apar db, requested it for my level
(61tl6sp4) and had it in a week (that part surprised me; kinda expected
it to be a little faster, especially when i opened my apar
electronically and pretty much said 'i'm hitting issue <x> mentioned in
apar <y> and need an ifix for os level <z>'.. but i also still have two
pmrs with runtimes of a year+.. :)

-r