On Sun, Oct 10, 2010 at 08:54:51AM +0200, Andi Kleen wrote: > That would be over 6 months just to make even a little progress. I think that's unfair. There's been absolutely no work from Nick to get things mergeable since 2.6.35-rc days where we gave him that feedback. We now have had Dave pick it up and sort out various issues with the third or so of the patchset he needed most to sort the lock contention problems in the workloads he saw, and we'll get large improvements for those for .37. The dcache_lock splitup alone is another massive task that needs a lot more work, too. I've started reviewing it and already fixed tons issues in in and the surrounding code. > Sorry, I am not convinced yet that any progress in this area has to be > that glacial. Linus indicated last time he wanted to move faster on the > VFS improvements. And the locking as it stands today is certainly a > major problem. > > Maybe it's possible to come up with a way to integrate this faster? Certainly not for .37, where even the inode_lock splitup is pretty damn later. Nick disappearing for a few weeks and others having to pick up the work to sort it out certainly doesn't help. And the dcache_lock splitup is a much larget task than that anyway. Getting that into .38 is the enabler for doing more fancy things. And as Dave mentioned at least in the writeback area it's much better to sort out the algorithmic problems now than to blindly split some locks up more. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>