Messages in this thread | | | Subject | Re: [RFC][PATCH 0/5] Introduce checks for preemptable code for this_cpu_read/write() | From | Steven Rostedt <> | Date | Mon, 19 Sep 2011 23:17:46 -0400 |
| |
On Mon, 2011-09-19 at 23:12 -0400, Steven Rostedt wrote:
> Actually, that's part of the issue. RT has made spin_locks not migrate. > But this has also increased the overhead of those same spinlocks. I'm > hoping to do away with the big hammer approach (although Thomas is less > interested in this). I would like to have areas that require per-cpu > variables to be annotated, and not have every spinlock disable > preemption. >
The point is, if I do make this change. It is even more essential that I detect the preempt enable use of this_cpu() and friends. As the spin_locks() will no longer be giving that guarantee.
-- Steve
| |