This may be a bug in gcc, but I was unable to reproduce this with a small example.  I'm not bothering with a Gentoo bug right now because they apparently don't care about gcc 4.7 yet. :(  Essentially, it appears that we're generating a warning when memset is called and it's "size" (not "length" as the error message says) parameter is compile-time constant that evaulates to zero.  However, rather than simply generating a warning, it actually causes ld to return non-zero.

/usr/lib64/ccache/bin/x86_64-pc-linux-gnu-g++  -fPIC -march=native -O2 -ggdb  -Wnon-virtual-dtor -Wno-long-long -Wundef -Wcast-align -Wchar-subscripts -Wall -W -Wpointer-arith -Wform
at-security -fno-exceptions -DQT_NO_EXCEPTIONS -fno-check-new -fno-common -Woverloaded-virtual -fno-threadsafe-statics -fvisibility=hidden -fvisibility-inlines-hidden -Wl,--enable-ne
w-dtags -Wl,--fatal-warnings -Wl,--no-undefined -lc  -Wl,-O1 -Wl,--as-needed -shared -Wl,-soname,kopete_qq.so -o ../../../lib/kopete_qq.so CMakeFiles/kopete_qq.dir/kopete_qq_automoc.
o CMakeFiles/kopete_qq.dir/ui/qqwebcamdialog.o CMakeFiles/kopete_qq.dir/ui/qqeditaccountwidget.o CMakeFiles/kopete_qq.dir/ui/dlgqqvcard.o CMakeFiles/kopete_qq.dir/qqprotocol.o CMakeF
iles/kopete_qq.dir/qqcontact.o CMakeFiles/kopete_qq.dir/qqaccount.o CMakeFiles/kopete_qq.dir/qqaddcontactpage.o CMakeFiles/kopete_qq.dir/qqsocket.o CMakeFiles/kopete_qq.dir/qqnotifys
ocket.o CMakeFiles/kopete_qq.dir/qqchatsession.o CMakeFiles/kopete_qq.dir/libeva.o CMakeFiles/kopete_qq.dir/evautil.o CMakeFiles/kopete_qq.dir/md5.o CMakeFiles/kopete_qq.dir/crypt.o 
-L/usr/lib64/qt4 -L/tmp/portage/kde-base/kopete-4.9.3/work/kopete-4.9.3_build/lib /usr/lib64/libkio.so.5.9.3 /usr/lib64/qt4/libQtNetwork.so /usr/lib64/qt4/libQt3Support.so ../../../l
ib/libkopete.so.4.9.3 ../../../lib/libkopete_videodevice.so.4.9.3 /usr/lib64/qt4/libQt3Support.so /usr/lib64/libkio.so.5.9.3 /usr/lib64/qt4/libQtNetwork.so /usr/lib64/qt4/libQtXml.so
 /usr/lib64/libkdeui.so.5.9.3 /usr/lib64/qt4/libQtGui.so /usr/lib64/qt4/libQtSvg.so /usr/lib64/libkdecore.so.5.9.3 /usr/lib64/qt4/libQtDBus.so /usr/lib64/qt4/libQtCore.so -lpthread -
Wl,-rpath,/usr/lib64/qt4:/tmp/portage/kde-base/kopete-4.9.3/work/kopete-4.9.3_build/lib:
CMakeFiles/kopete_qq.dir/evautil.o: In function `memset':
/usr/include/bits/string3.h:82: warning: memset used with constant zero length parameter; this could be due to transposed parameters
collect2: error: ld returned 1 exit status
make[2]: *** [lib/kopete_qq.so] Error 1
make[2]: Leaving directory `/tmp/portage/kde-base/kopete-4.9.3/work/kopete-4.9.3_build'
make[1]: *** [kopete/protocols/qq/CMakeFiles/kopete_qq.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....

The offending error message originates from here: http://gcc.gnu.org/git/?p=gcc.git;a=blob;f=gcc/testsuite/gcc.dg/memset-1.c;h=677ae91bd0e7f5f04b4ce3c1a93dfcbcee188d44;hb=HEAD

The work around is cheesy, it's to check the value of pos prior to calling memset.  In practice, this will likely get compiled-out.  While I haven't examined either Eva::Packet or ByteArray closely, I presume that the Eva::Packet::encrypt() is called somewhere (or how) in a manner that gcc is able to determine the outcome of the expression "text.size()" at compile time, thus, when the memset inline is expanded, it determines that it's both a compile-time constant and zero and gives is the warning, that ends up acting like an error.

Here's the work-around:

--- kopete/protocols/qq/evautil.cpp.orig        2012-12-03 11:47:00.466594261 -0600
+++ kopete/protocols/qq/evautil.cpp     2012-12-03 11:47:26.719613014 -0600
@@ -78,7 +78,9 @@
                plain[0] = ( rand() & 0xf8 ) | pos;
                memset( plain_pre, 0, 8 );
                memset( crypted_pre, 0, 8 );
-               memset( plain+1, rand()& 0xff, pos++ );
+               if( pos )
+                       memset( plain+1, rand()& 0xff, pos );
+               ++pos;
 
                // pad 2 bytes
                for( i = 0; i< 2; i++ )


Reproducible: Always

Steps to Reproduce:
see Details
Actual Results:  
see Details

Expected Results:  
see Details

see Details