Bug 678818
| Summary: | i686 libSDL marked having executable stack | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Mario Blättermann <mario.blaettermann> | ||||
| Component: | SDL | Assignee: | Petr Pisar <ppisar> | ||||
| Status: | CLOSED ERRATA | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||
| Severity: | unspecified | Docs Contact: | |||||
| Priority: | unspecified | ||||||
| Version: | 14 | CC: | adellam, hicham.haouari, kevin, lonelywoolf, otte, ppisar, twoerner | ||||
| Target Milestone: | --- | ||||||
| Target Release: | --- | ||||||
| Hardware: | i686 | ||||||
| OS: | Unspecified | ||||||
| Whiteboard: | |||||||
| Fixed In Version: | SDL-1.2.14-10.fc14 | Doc Type: | Bug Fix | ||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2011-03-03 03:26:32 UTC | Type: | --- | ||||
| Regression: | --- | Mount Type: | --- | ||||
| Documentation: | --- | CRM: | |||||
| Verified Versions: | Category: | --- | |||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||
| Embargoed: | |||||||
| Bug Depends On: | |||||||
| Bug Blocks: | 669844 | ||||||
| Attachments: |
|
||||||
|
Description
Mario Blättermann
2011-02-19 22:24:41 UTC
I'm not aware about any changes in SDL that could introduce this change. Both previous and current version have the stack marked as unexecutable. More ever it works for me. Can you read section <http://www.crypt.gen.nz/selinux/faq.html#CP.19> and check tour library of permission for stack (e.g. by execstack -q). Are you sure that downgrading SDL to previous version dismiss the problem? Bellow are unpacked packages downloaded from koji and no executable bit on stack segment is set:
$ scanelf --header SDL-1.2.14-{8,9}.fc14.x86_64/usr/lib64/libSDL-1.2.so.0.11.3
TYPE STK/REL/PTL FILE
ET_DYN RW- --- RW- SDL-1.2.14-8.fc14.x86_64/usr/lib64/libSDL-1.2.so.0.11.3
ET_DYN RW- --- RW- SDL-1.2.14-9.fc14.x86_64/usr/lib64/libSDL-1.2.so.0.11.3
However it's visible on i686 build:
$ scanelf --header SDL-1.2.14-{8,9}.fc14.i686/usr/lib/libSDL-1.2.so.0
TYPE STK/REL/PTL FILE
ET_DYN RW- --- RW- SDL-1.2.14-8.fc14.i686/usr/lib/libSDL-1.2.so.0
ET_DYN RWX --- RW- SDL-1.2.14-9.fc14.i686/usr/lib/libSDL-1.2.so.0
AFAIK i686 architecture has no executable protection in CPU implemented so the marker is just informative.
I guess somebody changed tool chain in F14. I will check it.
TYPE STK/REL/PTL FILE ET_DYN RW- --- RW- SDL-1.2.14-9.fc13.i686/usr/lib/libSDL-1.2.so.0 ET_DYN RW- --- RW- SDL-1.2.14-8.fc14.i686/usr/lib/libSDL-1.2.so.0 ET_DYN RWX --- RW- SDL-1.2.14-8.fc14.i686-rebuild/usr/lib/libSDL-1.2.so.0 ET_DYN RWX --- RW- SDL-1.2.14-9.fc14.i686/usr/lib/libSDL-1.2.so.0 ET_DYN RWX --- RW- SDL-1.2.14-11.fc16.i686/usr/lib/libSDL-1.2.so.0 The SDL-1.2.14-8.fc14.i686-rebuild (http://koji.fedoraproject.org/koji/taskinfo?taskID=2857019) built from the same sources as SDL-1.2.14-8.fc14.i686 differ in the bit. So it looks like a tool chain issue. Same issue here, gnash can't be run at all on i686 due to this I think we need this regression addressed before pushing any SDL update. I suspect a regression or an incompatible change in NASM. The good build was done with nasm-2.08.01-2.fc14, the bad one with nasm-2.09.03-2.fc14. Documented changes: http://www.nasm.us/doc/nasmdocc.html While we really need to get NASM fixed since this can also affect other packages (in fact, at least one package in a third-party repository has a workaround for what's probably the same issue), could we try using YASM instead, which appears to be SDL upstream's preferred choice? *** Bug 652297 has been marked as a duplicate of this bug. *** nasm-2.08.02-1.fc14.i686 passes. nasm-2.09-1.fc14.i686 produces affected objects. I found it. Newer nasm will not define `__OUTPUT_FORMAT__' macro as `elf', but as `elf32' if invoked as `nasm -f elf'. This is due to `elf' became alias for `elf32' in nasm-2.09:
Short aliases \c{win}, \c{elf} and \c{macho} for output formats are
introduced. Each stands for \c{win32}, \c{elf32} and \c{macho32}
accordingly.
I guess SDL sources should be patched to write:
%ifidn __OUTPUT_FORMAT__,elf32
section .note.GNU-stack noalloc noexec nowrite progbits
%endif
instead of
%ifidn __OUTPUT_FORMAT__,elf
section .note.GNU-stack noalloc noexec nowrite progbits
%endif
SDL-1.2.14-11.fc15 has been submitted as an update for Fedora 15. https://admin.fedoraproject.org/updates/SDL-1.2.14-11.fc15 SDL-1.2.14-10.fc14 has been submitted as an update for Fedora 14. https://admin.fedoraproject.org/updates/SDL-1.2.14-10.fc14 SDL-1.2.14-10.fc13 has been submitted as an update for Fedora 13. https://admin.fedoraproject.org/updates/SDL-1.2.14-10.fc13 SDL-1.2.14-11.fc15 has been pushed to the Fedora 15 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with su -c 'yum --enablerepo=updates-testing update SDL'. You can provide feedback for this update here: https://admin.fedoraproject.org/updates/SDL-1.2.14-11.fc15 SDL-1.2.14-11.fc15 has been pushed to the Fedora 15 stable repository. If problems still persist, please make note of it in this bug report. SDL-1.2.14-10.fc13 has been pushed to the Fedora 13 stable repository. If problems still persist, please make note of it in this bug report. SDL-1.2.14-10.fc14 has been pushed to the Fedora 14 stable repository. If problems still persist, please make note of it in this bug report. Created attachment 578707 [details]
Ooops same problem with Firefox
(In reply to comment #17) > Created attachment 578707 [details] > Ooops same problem with Firefox I'm not aware that Firefox would use SDL library. Even SDL is is supposed to be fixed now. You should report your issue against Firefox component and provide all necessary details like Firefox package version. |