On 2023/7/13 12:27, Paul E. McKenney wrote:
On Thu, Jul 13, 2023 at 10:02:17AM +0800, Gao Xiang wrote:
... sorry forget some.
One additional question... What is your plan for kernels built with
CONFIG_PREEMPT_COUNT=n? After all, in such kernels, there is no way
that I know of for code to determine whether it is in an RCU read-side
critical section, holding a spinlock, or running with preemption disabled.
I'm not sure if Android (or all targetted) users use or care about
this configuration (CONFIG_PREEMPT_COUNT=n), personally I think
for such configuration we could just fall back to the workqueue
approach all the time.
Anyway, such optimization really comes from real workloads /
experience, users don't live well without such mitigation.
Thanks,
Gao Xiang
Thanx, Paul
