I can't get plugins to work, when mozilla is compiled with gcc 3.0. When i try a flash-site mozilla crashes: ./mozilla-bin: error while loading shared libraries: /windows/hda12/CS/mozilla/plugins/libflashplayer.so: undefined symbol: __builtin_vec_new when I try the java-plugin from blackdown-jre 1.3.1, it even doesn't show up in about:plugins another question i'm asking myself: is my english really so bad?

What compiler is the plugin compiled with? gcc 2.95 or lower and gcc 3.0 are not binary-compatible. That means that a program compiled with gcc 3.0 cannot be linked against a library (which is what a plugin is) that was compiled with gcc 2.95 or lower. Call it a gcc bug if you wish.....

I don't know the compiler which it was compiled with.. it's closed source. So what to do? I still want to use gcc 3.0... is there not another possibility to get this to work then using gcc 2.95.x? and please could you answer my other question :)

As far as I can tell, your options are: 1) Compile mozilla with 2.95 2) Make the plugin vendor compile with 3.0 There is also 3) The plugin is not using the C++ ABI and this is a Mozilla bug. I don't know whether 3) holds, so setting status to NEW and ccing blizzard and leaf because they may know something about the compiler situation And your English seems to be pretty much fine...

related or dup of 58491

In this case it's a plugin that's written in C++, so no.

All the compilers produce code with different ABIs. You're screwed.

*** Bug 93672 has been marked as a duplicate of this bug. ***

*** Bug 108379 has been marked as a duplicate of this bug. ***

Ok, the C++ plugins compiled with different compiler version can't work. And this couldn't be resolved. But Mozilla should recover fine from this bad plugin condition. It definitely shouldn't crash. Or at least it should somehow detect the wrong plugin and don't even try to use it.

>It definitely shouldn't crash. how is seeing mozilla's crash?

please disregard my previous question, I was using 2.95 compiled binary:(

*** Bug 118601 has been marked as a duplicate of this bug. ***

--- Mass reassigning Unix bugs to serge ---

Changing summary to something to which could be found solution. The original issue isn't possible to resolve. Also nominating for 1.0, but as resolving this problem probably involves some low level hacking I don't believe it gets fixed. But maybe I'm wrong.

for gcc 3.x there is reasonable workaround see bug 124006.

*** Bug 136780 has been marked as a duplicate of this bug. ***

*** This bug has been marked as a duplicate of 154206 ***

mass duplicate verifications . For filtering purposes, pls use keywd "massdupverification"