On 18/03/2019 15:59, Roger Riggs wrote:
Hi Alan,

I checked the bytecode counts of the two versions and did not see
a material difference so it is due to the lack of optimizations in fastdebug.
What I meant is whether profiling has hinted at anything obvious, maybe an expense assert or some non-product code. In any case, I think what you have is fine and it would be good to get it off the exclude list (it should not have been put there in the first place of course).

-Alan

Reply via email to