On Wed, 2006-03-29 at 16:41 -0800, Andrew Morton wrote: > hm, you've extended lock_kernel() coverage (why? Does this help fix the > race??) but we still have a cond_resched() inside the now-newly-locked > region. If that cond_resched() drops the bkl, is the race reopened? No. The race fix is not dependent on the extension in BKL coverage. I removed the unlock_kernel()/lock_kernel() pair simply because they are redundant: the locks_alloc_lock() and cond_resched() will automatically release the BKL for you. Might as well save a couple of bytes. Cheers, Trond - To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
