------- Comment #7 from paolo dot carlini at oracle dot com 2008-09-01 11:09 ------- (In reply to comment #6) > The other thing is that __thread is emulated on targets that don't > support it.
That's interesting, I didn't notice we are actually doing that, right now (I remember some discussions...). Then, however, I'm missing more... Why we are running the check in libstdc++-v3 and libjava two times instead of just running it once and, if necessary, instantiating the emulation and, in case, passing down the info to the libraries?!? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37308
