Hi, Eric

> If the CPU has been in softirq context for too long, then stop
> choosing softirq context and instead fall back to the traditional
> workqueue. This could help address objections about increased use of > softirq context.

Could you share me some example about this?

> When possible, make dm-verity do the work inline in the existing block
> softirq, instead of using the BH workqueue.  This would reduce the
> amount of overhead.  Note that dm-crypt does it this way.

> Let's just set good parameters by default instead.
> "try_verify_in_tasklet" is kind of useless as-is, and we should just
> fix it.  Especially since it only affects performance and not other
> behavior, I don't think there are any compatibility concerns with
> changing what it does exactly.

These ideas have been implemented in patch v2. Thanks for your suggestions.

- LongPing

Reply via email to