Bug 745197

Summary: New (hidden) binutils220 package for RHEL 5.8
Product: Red Hat Enterprise Linux 5 Reporter: Jeff Law <law>
Component: Package ReviewAssignee: Andreas Schwab <schwab>
Status: CLOSED ERRATA QA Contact:
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 5.8CC: jakub, notting, pm-rhel
Target Milestone: rc   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-10-27 16:26:34 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Bug Depends On:    
Bug Blocks: 188273, 743394, 749916    
Attachments:
Description Flags
binutils220.spec none

Description Jeff Law 2011-10-11 15:50:38 UTC
Description of problem:
gcc-4.4 in RHEL 5.7 will issue instructions that can not be handled by the
assembler available on RHEL 5.7 (see BZ 669190).

A few approaches were investigated to resolve this issue; the final/best
resolution is to include an additional version of binutils that will be used by
the gcc-4.4 compiler, but which will otherwise be hidden.

This is a formal request to include binutils220 as an approved package for RHEL
5.8 on x86/x86_64 based platforms.

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


How reproducible:


Steps to Reproduce:
1.
2.
3.
  
Actual results:


Expected results:


Additional info:

Comment 1 Jakub Jelinek 2011-10-21 17:05:45 UTC
Created attachment 529539 [details]
binutils220.spec

Comment 2 Jakub Jelinek 2011-10-21 17:31:32 UTC
The above is the latest binutils220.spec I found:

rpmlint:

binutils220.i386: W: spelling-error %description -l en_US objdump -> obj dump, obj-dump, objurgate
binutils220.ia64: W: spelling-error %description -l en_US objdump -> obj dump, obj-dump, objurgate
binutils220.ppc64: W: spelling-error %description -l en_US objdump -> obj dump, obj-dump, objurgate
binutils220.ppc: W: spelling-error %description -l en_US objdump -> obj dump, obj-dump, objurgate
binutils220.s390: W: spelling-error %description -l en_US objdump -> obj dump, obj-dump, objurgate
binutils220.s390x: W: spelling-error %description -l en_US objdump -> obj dump, obj-dump, objurgate
binutils220.src: W: spelling-error %description -l en_US objdump -> obj dump, obj-dump, objurgate
binutils220.x86_64: W: spelling-error %description -l en_US objdump -> obj dump, obj-dump, objurgate

Not a bug, objdump is the correct name of the program.

binutils220.ppc64: W: no-manual-page-for-binary embedspu220
binutils220.ppc: W: no-manual-page-for-binary embedspu220

This is just a small wrapper shell script, no need for man page (it doesn't have one upstream either).

binutils220.src: W: %ifarch-applied-patch Patch3: binutils-2.20.51.0.2-ia64-lib64.patch

Intentional.

binutils220.src: W: invalid-url Source0: ftp://ftp.kernel.org/pub/linux/devel/binutils/binutils-2.20.51.0.2.tar.bz2 <urlopen error ftp error: 550 Failed to change directory.>

kernel.org partially down.

binutils220-debuginfo.i386: E: incorrect-fsf-address /usr/src/debug/binutils-2.20.51.0.2/bfd/elf-vxworks.c
binutils220-debuginfo.ia64: E: incorrect-fsf-address /usr/src/debug/binutils-2.20.51.0.2/bfd/elf-vxworks.c
binutils220-debuginfo.ppc64: E: incorrect-fsf-address /usr/src/debug/binutils-2.20.51.0.2/bfd/elf-vxworks.c
binutils220-debuginfo.ppc: E: incorrect-fsf-address /usr/src/debug/binutils-2.20.51.0.2/bfd/elf-vxworks.c
binutils220-debuginfo.x86_64: E: incorrect-fsf-address /usr/src/debug/binutils-2.20.51.0.2/bfd/elf-vxworks.c
15 packages and 0 specfiles checked; 5 errors, 12 warnings.

This file isn't even compiled into the binaries in this package, so irrelevant.

OK      source files match upstream:

49a961cc925df5a5f5e1cbe52befd8bc2875cfa205e0b9ea6caaa63504fd1b33  binutils-2.20.51.0.2.tar.bz2  matches tarball in RHEL6 (kernel.org repo still not available after the outage).

OK      package meets naming and versioning guidelines
OK      specfile is properly named, is cleanly written and uses macros
consistently.
OK      dist tag is present.
OK      license field matches the actual license.
OK      license is open source-compatible.

Not OK  License text included in package.

Please include COPYING* in docdir.  While binutils perhaps has it in the info documentation, this one doesn't have info docs.

N/A     latest version is being packaged.

Not intended.

OK      BuildRequires are proper.
OK      compiler flags are appropriate.
OK      package builds in mock.
OK      debuginfo package looks complete.
OK      final provides and requires look sane.
OK      %check is present and all tests pass.

run in %build section instead of %check, but otherwise ok and passes on all arches.

N/A     shared libraries are added to the regular linker search paths with
proper scriptlets

libbfd and libopcodes is (intentionally) linked statically into both binaries.

OK      owns the directories it creates.
OK      doesn't own any directories it shouldn't.
OK      no duplicates in %files.
OK      file permissions are appropriate.
N/A     correct scriptlets present.
OK      code, not content.
OK      documentation is small, so no -docs subpackage is necessary.
OK      %docs are not necessary for the proper functioning of the package.
N/A     headers in -devel
N/A     pkgconfig files in -devel
OK      no libtool .la droppings.
OK      not a GUI app.
OK      obsoletes and provides of the obsoleted package are valid

Issues (other than the above COPYING*):
1) I thought we agreed on hiding the binaries more, so they should move away from /usr/bin to say /usr/libexec/binutils220/ - gcc44 will find them there
2) the spec file isn't uptodate with RHEL6 binutils.spec - this one is based on -5.16, RHEL 6 has -5.27.  Please bring in all the more recent patches.

Comment 3 Jeff Law 2011-10-26 04:47:32 UTC
For the record:

  David Cantrell confirms there should be no installer issues/changes necessary to support this package.

  The security team (Vincent Danen) indicated this shouldn't be a problem from a security standpoint, but they'll want to take a quick peek once it's checked into CVS.


So we're waiting on Andreas to fix the problems Jakub noted above.  It's my understanding nothing an move forward until those issues are fixed.  This does not bode well for getting everything done by Friday :(

Comment 4 Jeff Law 2011-10-27 16:26:34 UTC
Per step #6.  Moved to RHEL5.0-ACCEPT and closed.