[DTrace-devel] [PATCH v2 4/5] Implement the support code for generating aggregation data

[Eugene Loh]

On 12/01/2020 10:37 AM, Kris Van Hees wrote:

> Let's agree to disagree?

Guess so.  I did want to clarify that whole business of "one seq# per 
copy", which is neither how we do things nor, so far as we know, how we 
will do things.  So, that seems settled.

So:
         Reviewed-by: Eugene Loh <eugene.loh at oracle.com>
subject to those few typos and that vestigial FIXME.

Also, please consider my suggestion to remove the lquantize_impl() 
stuff, which is just a lot of code that's going to disappear in the next 
(or thereabouts) patch anyhow.  That soon-to-be-short-lived code simply 
does not reflect where the quantize functions (or for that matter stddev 
with its 128-bit multiplies) are going.  Just having count_impl() is a 
sufficient -- in fact, clearer -- exemplar.