Comment 5 for bug 389555

As these locking issues often are sensitive to timing, it could well be that other changes had some influence. The problem with the stable kernels is that, after the experience with 2.6.27.y which received a huge amount of updates and sometimes caused regressions, we now went back to cherry-pick patches to reported problems for which we can get help in testing. As those patches are upstream and also kept by Ted, I don't think it will help that much if Greg would maintain a 2.6.28.11. But I should make sure we better sync with Ted.