Messages in this thread | | | Date | Tue, 29 Apr 2014 18:56:54 +0100 | From | Al Viro <> | Subject | Re: dcache shrink list corruption? |
| |
On Tue, Apr 29, 2014 at 06:01:39PM +0200, Miklos Szeredi wrote:
> Attached patch is just a starting point (untested). Not sure how to minimize > contention without adding too much complexity.
Contention isn't the worst problem here - I'd expect the cacheline ping-pong to hurt more... I agree with the analysis, but I'd really like to avoid that spinlock ;-/
Let me see if we can avoid that... Oh, well - at least that's a good excuse to take a break from fucking deadlock analysis around the damn acct(2), most of VFS and network filesystems ;-/
| |