https://bugs.kde.org/show_bug.cgi?id=371491
--- Comment #3 from Michael Daniels <mdani...@blackberry.com> --- The testing that I had done, which I realize I was not made clear in the description, was specific to our OS (QNX Neutrino). Our kernel is putting the per-cpu data above the 4GB boundary and sets the fs base address to that value. Since the fs base address is always over 4GB, I can reproduce the behaviour quite easily locally. I thought of different ways this might be testable, but to my knowledge fs is not writable in user-mode / valgrind and at least for Linux it looks like it's 0. So I am not aware of a reliable / portable way to force fs + 32bit addr over 4GB to validate the truncation. I am open to suggestions if anyone else has ideas. -- You are receiving this mail because: You are watching all bug changes.
