On 9/26/21 17:19, Damien Le Moal wrote:

Another thing: in patch 3, you are actually not handling the overflows. So
dd_queued() may return some very weird number (temporarily) when the inserted
count overflows before the completed count does. Since
dd_dispatch_aged_requests() does not care about the actual value of dd_queued(),
only if it is 0 or not, I am not 100% sure if it is useful to fix. Except maybe
for sysfs attributes ?

Hmm ... I'm not following. I think it can be proven mathematically that
dd_queued() returns a number in the range [0, max_queued_requests). Here is
an example:
* inserted = 1 (wrapped around).
* completed = UINT32_MAX - 1 (about to wrap but has not yet wrapped around).
* dd_queued() returns 2.

Bart.