Bug 217361
Summary: | OOo finds wrong libjvm.so on multi-arch gcj 4.1.1 installations | ||||||
---|---|---|---|---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | David Woodhouse <dwmw2> | ||||
Component: | openoffice.org | Assignee: | Caolan McNamara <caolanm> | ||||
Status: | CLOSED ERRATA | QA Contact: | |||||
Severity: | medium | Docs Contact: | |||||
Priority: | medium | ||||||
Version: | 6 | ||||||
Target Milestone: | --- | ||||||
Target Release: | --- | ||||||
Hardware: | ppc64 | ||||||
OS: | Linux | ||||||
Whiteboard: | |||||||
Fixed In Version: | 2.0.4-5.5.10 | Doc Type: | Bug Fix | ||||
Doc Text: | Story Points: | --- | |||||
Clone Of: | Environment: | ||||||
Last Closed: | 2007-01-04 08:44:46 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: | |||||||
Attachments: |
|
Description
David Woodhouse
2006-11-27 14:43:45 UTC
Created attachment 142176 [details] java bytecode can I get the output of this attachment saved as findhome.class and > java findhome I expect to see... /usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre and to see.. that /usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre/lib/ppc/client/libjvm.so is a link to /usr/lib/gcj-4.1.1/libjvm.so I think the problem is in the little piece of logic to bridge during this transition period from the old style libgcj.so providing the jvm to the new style libjvm.so. The problem being that vanilla 4.1.1 provides libgcj.so while we use libjvm.so, and I "keeping vanilla working" breaks on a nonvanilla 4.1.1 in a multiarch libjvm installed scenario pending seeing expected results, fix for this will appear in next scheduled FC-6 update and next rawhide push pmac /home/dwmw2 $ rpm -q --qf %{NAME}-%{VERSION}-%{RELEASE}.%{ARCH}\\n libgcj libgcj-4.1.1-30.ppc pmac /home/dwmw2 $ java findhome /usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre file:///usr/lib pmac /home/dwmw2 $ ls -l /usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre/lib/ppc/client/libjvm.so lrwxrwxrwx 1 root root 28 Nov 27 16:13 /usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre/lib/ppc/client/libjvm.so -> /usr/lib/gcj-4.1.1/libjvm.so pmac /home/dwmw2 $ rpm -q --qf %{NAME}-%{VERSION}-%{RELEASE}.%{ARCH}\\n libgcj libgcj-4.1.1-30.ppc libgcj-4.1.1-30.ppc64 pmac /home/dwmw2 $ java findhome /usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre file:///usr/lib64 pmac /home/dwmw2 $ ls -l /usr/lib{,64}/jvm/java-1.4.2-gcj-1.4.2.0/jre/lib/ppc/client/libjvm.so ls: /usr/lib64/jvm/java-1.4.2-gcj-1.4.2.0/jre/lib/ppc/client/libjvm.so: No such file or directory lrwxrwxrwx 1 root root 28 Nov 27 16:15 /usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre/lib/ppc/client/libjvm.so -> /usr/lib/gcj-4.1.1/libjvm.so all in order, fix should work as expected If I run the 32-bit version of /usr/bin/gij, which is normally overwritten by the installation of libgcj.ppc64, then I see the same results as when only libgcj.ppc is installed. ION}-%{RELEASE}.%{ARCH}\\n libgcjlibgcj-4.1.1-30.ppc libgcj-4.1.1-30.ppc64 pmac /home/dwmw2 $ file -L /usr/bin/gij /usr/bin/gij: ELF 32-bit MSB executable, PowerPC or cisco 4500, version 1 (SYSV), for GNU/Linux 2.6.9, dynamically linked (uses shared libs), for GNU/Linux 2.6.9, stripped pmac /home/dwmw2 $ java findhome /usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre file:///usr/lib OpenOffice works correctly like this too. |