Has Skype 1.4.0.64 been shown to work with glibc 2.5?

Andypoo stated in glibc 2.4 dependence that glibc 2.5 should be fine, but there was a post in How to get Skype 1.4 Alhpa to work in Slackware 11? stating that this was untrue/unlikely. I'm running glibc 2.5 and have had no luck with the new skype and there seem to be no hints other than glibc as to what's the problem

I am using skype(1.4.0.64) with slackware-current and it is working ok.

And which version is your glibc?

2.5

Skype 1.4 releases have all worked with glibc 2.5, and this is still the case.

The requirement was glibc 2.4. This has been relaxed to 2.3 in the next release. This binary still works on glibc 2.4 and glibc 2.5 systems.

The comments in that thread were related to CXXABI in libstdc++, which I'm trying to confirm if there are any issues outstanding.

Andrew.

Thanks for letting us know.


Interesting. Then you possibly noticed that I posted under that topic (I'm assuming you're referring to the Slackware thread, although you didn't specify) stating that I was experiencing the same problem.

Do you plan on suggesting anything for non-slackware users? Do I have to post this under a new (non-slackware specific) topic or file a bug-report to get information about this issue?



Many thanks. It's good to know that such a fundamental part of the system isn't the problem.

In light of Andypoo's last message, I was hoping I could also bother you for your libstdc++ version if you're still watching this topic.

How can I check what version it is?

"grep libstd *" on /var/log/packages gives me:

audacious-plugins-1.3.4-i486-2:usr/lib/audacious/Container/libstdio.so
cxxlibs-6.0.8-i486-3:cxxlibs: linked C++ binaries linked with older versions of libstdc++.
cxxlibs-6.0.8-i486-3:usr/lib/libstdc++.so.6.0.8
cxxlibs-6.0.8-i486-3:usr/i486-slackware-linux/lib/libstdc++.so.4.0.0
cxxlibs-6.0.8-i486-3:usr/i486-slackware-linux/lib/libstdc++.so.5.0.7
cxxlibs-6.0.8-i486-3:usr/i486-slackware-linux/lib/libstdc++-3-libc6.2-2-2.10.0.so
gcc-4.1.2-i486-1:usr/share/locale/de/LC_MESSAGES/libstdc++.mo
gcc-4.1.2-i486-1:usr/share/locale/fr/LC_MESSAGES/libstdc++.mo
gcc-g++-4.1.2-i486-1:usr/doc/gcc-4.1.2/libstdc++-v3/
gcc-g++-4.1.2-i486-1:usr/doc/gcc-4.1.2/libstdc++-v3/README
gcc-g++-4.1.2-i486-1:usr/doc/gcc-4.1.2/libstdc++-v3/faq.html
gcc-g++-4.1.2-i486-1:usr/doc/gcc-4.1.2/libstdc++-v3/ChangeLog
gcc-g++-4.1.2-i486-1:usr/lib/libstdc++.la
gcc-g++-4.1.2-i486-1:usr/lib/libstdc++.a
gcc-g++-4.1.2-i486-1:usr/lib/libstdc++.so.6.0.8
kdevelop-3.4.1-i486-1:usr/share/apps/kdevdocumentation/tocs/libstdc++.toc

Could it be 6.0.8?

slackware-current is 6.0.8 (cxxlibs). I'm running the same as Toni.

Let us know your version and we'll go from there.

Andypoo.

Usually, I just use my package manager's tools to query installed or available packages' versions.


Frankly, I've never quite understood the meaning of the numeral suffix on libraries. I seem to be running version 3.3

I also found the libstdc++.6.0.8 library, but not in '/usr/lib/'. With a sudden burst of hope, I symlinked to the library, only to get the same problem as before.

Thanks for all the info, though. It's much appreciated.


Thanks for your reply Andypoo.

From a quick Google and specifically checking the Gentoo and Debian package search machines it seems that cxxlibs is a Slackware-specific package.

I do, however, (as I mentioned in response to Toni's post) have libstdc++.so.6.0.8.


Again, the output of the error is:

It seems Skype may simply be using the wrong libstdc++.so.6 symlink (pointing to 6.0.3).

The real question is which version:
/usr/lib/libstdc++.so.6
links to

Because this is probably what Skype is finding first and using from your LD_LIBRARY_PATH.

Having said all that, this conversation is getting a little too indepth now as we've already changed the core library requirements when we relaxed the dependency on GLIBC 2.4. So just wait a little bit for the next release and we can continue this conversation then

BTW, this v3.3 package actually contains libstdc++.so.5, which is provided for pre-gcc-3.4 compatibility, and is not relevant here.

Thanks,
Andypoo.

Good suggestion. But alas:
... :-(

Any other suggestions?


I might agree with you if I were running glibc 2.3. As far as any developer has told me, this issue has nothing to do with any previous 2.4 dependence and so isn't an issue that might have been solved.