Re: [PATCH 1/2] jbd2: add new tracepoint jbd2_sleep_on_shadow

["Theodore Y. Ts'o" <tytso@mit.edu>]

On Wed, Sep 11, 2019 at 02:52:51PM +0800, Xiaoguang Wang wrote:
> > I think maybe it might be better to use units of microseconds and then
> > change sleep to usleep so the units are clear?  This is a spinlock, so
> > it should be quick.
>
> Sorry, I may not quite understand you, do you mean that milliseconds is not precise, so
> should use microseconds? For these two patches, they do not use usleep or msleep to do
> real sleep work, they just record the duration which process takes to wait bh_shadow flag
> to be cleared or transaction to be unlocked.

Apologies, I should have been clear enough.  Yes, my concern that
milliseconds might not be fine-grained enough.  The sample results
which you showed had values of 2ms, 1ms, and 0ms.  And the single 0ms
result in particular raised the concern that we should use a
microseconds instead of milliseconds.

In fact, instead of "sleep", maybe "latency(us)" or "latency(ms)"
would be a better label?

Regards,

						- Ted