Bug 83802 - RPM 4.2-066 fails to compile on Solaris 8
RPM 4.2-066 fails to compile on Solaris 8
Status: CLOSED WORKSFORME
Product: Red Hat Linux
Classification: Retired
Component: rpm (Show other bugs)
8.0
sparc Linux
medium Severity medium
: ---
: ---
Assigned To: Jeff Johnson
Mike McLean
:
Depends On:
Blocks:
  Show dependency treegraph
 
Reported: 2003-02-08 20:59 EST by Donnie Cranford
Modified: 2007-04-18 12:50 EDT (History)
1 user (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2003-03-08 12:44:54 EST
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---


Attachments (Terms of Use)

  None (edit)
Description Donnie Cranford 2003-02-08 20:59:35 EST
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2.1) Gecko/20030203

Description of problem:
Trying to compile RPM 4.2-066 on Solaris 8

Using GCC 3.2.1 

# uname -a
SunOS DEATHSTARII 5.8 Generic_108528-17 sun4u sparc SUNW,Sun-Blade-100

Compile is failing looking in linker stage looking for libz

here is the output

/usr/local/bin/gcc -I/usr/local/include -I/usr/local/BerkeleyDB/include
-I/opt/sfw/include -I/opt/sfw/include/db1 -I/opt/sfw/include/db3 -I/usr/include
-I/usr/openwin/include -D_GNU_SOURCE -D_REENTRANT -Wall -Wpointer-arith
-Wstrict-prototypes -Wmissing-prototypes -Wno-char-subscripts -o file -static
file.o  ./.libs/libfmagic.al -lz
ld: fatal: library -lz: not found
ld: fatal: File processing errors. No output written to file
collect2: ld returned 1 exit status
make[2]: *** [file] Error 1
make[2]: Leaving directory `/tmp/rpm-4.2/file'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/tmp/rpm-4.2'
make: *** [all] Error 2



Version-Release number of selected component (if applicable):


How reproducible:
Always

Steps to Reproduce:
1../configure
2.gmake
3.
    

Actual Results:  Compile bombed out as decribed above

Expected Results:  Should have continued on successfully

Additional info:

Here is my .profile which gives my CFLAGS, CXXFLAGS, and such


# cat /.profile
PATH=/opt/sfw/bin:/usr/local/bin:/usr/sbin:/usr/bin:/usr/dt/bin:/usr/openwin/bin:/bin:/usr/ccs/bin:/opt/SUNWspro/bin:/usr/local/bin:/usr/ucb
 
#CC=/opt/SUNWspro/bin/cc
CC=/usr/local/bin/gcc
 
#CXX=/opt/SUNWspro/bin/CC
CXX=/usr/local/bin/g++
 
 
CFLAGS="-I/usr/local/include -I/usr/local/BerkeleyDB/include -I/opt/sfw/include
-I/opt/sfw/include/db1 -I/opt/sfw/include/db3 -I/usr/include -I/usr/openwin/include"
 
CXXFLAGS="-I/usr/local/include -I/usr/local/BerkeleyDB/include
-I/opt/sfw/include -I/opt/sfw/include/db1 -I/opt/sfw/include/db3 -I/usr/include
-I/usr/openwin/include"
 
LD_LIBRARY_PATH=/opt/sfw/lib:/usr/local/lib:/usr/local/BerkeleyDB/lib:/lib:/usr/lib:/lib:/usr/lib
 
export PATH CC CFLAGS CXX CXXFLAGS LD_LIBRARY_PATH
Comment 1 Donnie Cranford 2003-02-08 21:23:20 EST
CCing herold
Comment 2 Donnie Cranford 2003-02-08 21:51:02 EST
Update
-----------------------------------------------------------

Using the Forte 7 compiler, I get an entirely different error, (well it doesnt
get as far as GCC) here is what I get in forte, same .profile

/opt/SUNWspro/bin/cc -DHAVE_CONFIG_H -I. -I. -I.  
-DMAGIC='"/usr/local/lib/rpm/magic"'  -I/usr/local/include
-I/usr/local/BerkeleyDB/include -I/opt/sfw/include -I/opt/sfw/include/db1
-I/opt/sfw/include/db3 -I/usr/include -I/usr/openwin/include -c `test -f
'file.c' || echo './'`file.c
"system.h", line 253: warning: implicit function declaration: error
"file.h", line 98: syntax error before or at: (
"file.h", line 98: warning: syntax error:  empty declaration
"file.c", line 221: syntax error before or at: :
cc: acomp failed for file.c
Comment 3 Donnie Cranford 2003-02-09 16:08:33 EST
ok, was able ot get around -lz linking issue by adding a static libz, however
now compile is failing with an unresolved symbol

/bin/bash ./libtool --mode=link /usr/local/bin/gcc  -I/usr/local/include
-I/usr/local/BerkeleyDB/include -I/opt/sfw/include -I/opt/sfw/include/db1
-I/opt/sfw/include/db3 -I/usr/include -I/usr/openwin/include   -o test1
-all-static test1.o libpopt.la -L/usr/ucblib -R /usr/ucblib
/usr/local/bin/gcc -I/usr/local/include -I/usr/local/BerkeleyDB/include
-I/opt/sfw/include -I/opt/sfw/include/db1 -I/opt/sfw/include/db3 -I/usr/include
-I/usr/openwin/include -o test1 -static test1.o  ./.libs/libpopt.a -L/usr/ucblib
-R/usr/ucblib
Undefined                       first referenced
 symbol                             in file
libintl_dgettext                    ./.libs/libpopt.a(popt.o)
ld: fatal: Symbol referencing errors. No output written to test1
collect2: ld returned 1 exit status


this is using xgettext 0.11.5

HOWEVER

when doing a ./configure --with-included-gettext


I get the following error

/bin/bash ../libtool --mode=link /usr/local/bin/gcc  -I/usr/local/include
-I/usr/local/BerkeleyDB/include -I/opt/sfw/include -I/opt/sfw/include/db1
-I/opt/sfw/include/db3 -I/usr/include -I/usr/openwin/include -D_GNU_SOURCE
-D_REENTRANT -Wall -Wpointer-arith -Wstrict-prototypes -Wmissing-prototypes
-Wno-char-subscripts
  -o librpmio.la -rpath /usr/local/lib -release 4.2 ../file/libfmagic.la
../zlib/libz.la -lrt -lpthread argv.lo digest.lo fts.lo macro.lo rpmio.lo
rpmlog.lo rpmmalloc.lo rpmpgp.lo rpmrpc.lo strcasecmp.lo stubs.lo url.lo ugid.lo
aes.lo aesopt.lo base64.lo beecrypt.lo blockmode.lo blockpad.lo blowfish.lo
blowfishopt.lo dhaes.lo dldp.lo dlkp.lo dlpk.lo dlsvdp-dh.lo dsa.lo elgamal.lo
endianness.lo entropy.lo fips186.lo hmac.lo hmacmd5.lo hmacsha1.lo hmacsha256.lo
md5.lo memchunk.lo mp32.lo mp32barrett.lo mp32number.lo mp32opt.lo mp32prime.lo
mtprng.lo rsa.lo rsakp.lo rsapk.lo sha1.lo sha1opt.lo sha256.lo timestamp.lo
-lrt -lpthread -lsocket -lnsl  -L../zlib -R../zlib  -L/usr/local/lib
-R/usr/local/lib  -L/usr/ucblib -R/usr/ucblib  -lbz2
libtool: link: only absolute run-paths are allowed
make[2]: *** [librpmio.la] Error 1


I think im getting closer but not there yet....

any ideas
Comment 4 Donnie Cranford 2003-02-11 20:36:45 EST
Ok, here is a issue, there are 2 places where rpm-4.2.0-66 dies because of only
absolute run paths allowd,

rpm/rpmio/Makefile
Where you have -L${top_builddir}/zlib and -R${top_builddir}/zlib

and

rpm/tools/Makefile 
-L${top_builddir}/zlib and -R${top_builddir}/zlib

They both have to have the absolute pathname to compile

Here is the current issue

I must do --without-python(which is fine for now).

The biggest error that I am seeing now actually has to do with elf and what
looks like problematic headers

here is the output (warning is is quite a bit)

/usr/local/bin/gcc -DHAVE_CONFIG_H -I. -I. -I.. -I. -I.. -I../build -I../lib
-I../rpmdb -I../rpmio -I../popt -I../file -I../zlib -I../misc   
-I/usr/local/include -I/usr/local/BerkeleyDB/include -I/usr/local/BerkeleyDB.3.3
-I/usr/include -I/usr/openwin/include -D_GNU_SOURCE -D_REENTRANT -Wall
-Wpointer-arith -Wstrict-prototypes -Wmissing-prototypes -Wno-char-subscripts -c
`test -f 'debugedit.c' || echo './'`debugedit.c
debugedit.c:23:22: byteswap.h: No such file or directory
debugedit.c:24:20: endian.h: No such file or directory
In file included from /usr/include/gelf.h:11,
                 from debugedit.c:36:
/usr/include/libelf.h:28:2: #error "large files are not supported by libelf"
debugedit.c:37:19: dwarf.h: No such file or directory
debugedit.c: In function `read_abbrev':
debugedit.c:289: `DW_FORM_indirect' undeclared (first use in this function)
debugedit.c:289: (Each undeclared identifier is reported only once
debugedit.c:289: for each function it appears in.)
debugedit.c: In function `edit_dwarf2_line':
debugedit.c:479: warning: implicit declaration of function `alloca'
debugedit.c: In function `edit_attributes':
debugedit.c:676: `DW_AT_stmt_list' undeclared (first use in this function)
debugedit.c:678: `DW_FORM_data4' undeclared (first use in this function)
debugedit.c:685: `DW_AT_comp_dir' undeclared (first use in this function)
debugedit.c:686: `DW_FORM_strp' undeclared (first use in this function)
debugedit.c:711: `DW_TAG_compile_unit' undeclared (first use in this function)
debugedit.c:713: `DW_AT_name' undeclared (first use in this function)
debugedit.c:751: `DW_FORM_addr' undeclared (first use in this function)
debugedit.c:754: `DW_FORM_ref1' undeclared (first use in this function)
debugedit.c:755: `DW_FORM_flag' undeclared (first use in this function)
debugedit.c:756: `DW_FORM_data1' undeclared (first use in this function)
debugedit.c:759: `DW_FORM_ref2' undeclared (first use in this function)
debugedit.c:760: `DW_FORM_data2' undeclared (first use in this function)
debugedit.c:763: `DW_FORM_ref4' undeclared (first use in this function)
debugedit.c:767: `DW_FORM_ref8' undeclared (first use in this function)
debugedit.c:768: `DW_FORM_data8' undeclared (first use in this function)
debugedit.c:771: `DW_FORM_sdata' undeclared (first use in this function)
debugedit.c:772: `DW_FORM_ref_udata' undeclared (first use in this function)
debugedit.c:773: `DW_FORM_udata' undeclared (first use in this function)
debugedit.c:776: `DW_FORM_ref_addr' undeclared (first use in this function)
debugedit.c:780: `DW_FORM_string' undeclared (first use in this function)
debugedit.c:783: `DW_FORM_indirect' undeclared (first use in this function)
debugedit.c:786: `DW_FORM_block1' undeclared (first use in this function)
debugedit.c:789: `DW_FORM_block2' undeclared (first use in this function)
debugedit.c:793: `DW_FORM_block4' undeclared (first use in this function)
debugedit.c:797: `DW_FORM_block' undeclared (first use in this function)
debugedit.c: In function `fdopen_dso':
debugedit.c:1017: `ELF_C_RDWR_MMAP' undeclared (first use in this function)
make[2]: *** [debugedit.o] Error 1
make[2]: Leaving directory `/tmp/rpm-4.2/tools'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/tmp/rpm-4.2'
make: *** [all] Error 2



any ideas here would be great, I think im pretty close to having this ported
then I can work backwards and get it retested and submit diffs ;)





Comment 5 Jeff Johnson 2003-02-12 18:21:49 EST
You need debugedit iff you want/need to produce
sub-packages with debugging symbols.

Disable for now, use the latest elfutils package
if/when you choose to try to pursue.
Comment 6 Donnie Cranford 2003-02-12 18:28:32 EST
Jeff

how do I disable debugedit??

just comment it out?? I dont know of any configure option
Comment 7 Jeff Johnson 2003-02-12 18:32:22 EST
Edit the makefile.
Comment 8 Jeff Johnson 2003-03-08 12:44:54 EST
rpm-4.2-0.70 should revert to internal copy of elfutils-0.76
which should be useful even on solaris. There's almost certainly
more to do on solaris, as elfutils is still under development.

Please open a new bug report if there are other solaris compilation
problems.

Note You need to log in before you can comment on or make changes to this bug.