Crap. Hit send before I was done. > -----Original Message----- > From: Matthew Wilcox [mailto:matthew@xxxxxx] > Sent: Friday, October 12, 2007 2:28 PM > To: Little, Chris > Cc: kernel-janitors@xxxxxxxxxxxxxxx > Subject: Re: removing lock_kernel() from fs/fat > > > Why the spin lock if it already has the bkl? > > Because you have to protect against other CPUs reading this > inode, some of which may have the BKL and others may have the > inode_hash_lock. That makes sense. > Another mistake you've made is using an rwlock. This is a > classic beginners mistake, as the problem with it is highly > non-obvious. The trouble with rwlocks is that they're unfair > to writers. A storm of readers can prevent a writer from > ever getting the lock. It's normally a good thing to use a > plain spinlock and only convert to an rwlock if you're sure > it's a good idea. > > A further mistake is picking a filesystem as a good candidate > for BKL removal. Take a look at > Documentation/filesystems/Locking. You'll see there's a lot > of places where the VFS takes the BKL on behalf of the > filesystem. So taking the BKL inside a filesystem implicitly > locks out a lot of places that might call into the filesystem. Question: Are there appropriate times to use rwlock, then? Doesn't it imply that there are writers? Many thanks for your comments. Things to file away. Pardon me if I keep throwing stuff up. One day I might just get good at this. :) - To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
