Optimize buffer_line by calculating before wait-for-free-block

[thinkyhead]

This optimization of buffer_line moves the wait-for-buffer test farther down so that calculations that don't need to write to the block can occur ahead of time. This saves lots of time because the calculations will already have been done when the block finally becomes available. Significantly more time may be saved by doing other expensive calculations earlier also.

[Roxy-3D]

I agree! It basically gives us back the CPU cycles to process one command. However... The only time it gives us these CPU cycles is when the Planner's buffer is already full...

[thinkyhead]

The only time it gives us these CPU cycles is when the Planner's buffer is already full...

@Roxy-3D Which will be most of the time in an active print job, so I think it will have noticeable benefits.

[thinkyhead]

Also moved calculation of dm and esteps ahead of the wait-for-block loop.

[Roxy-3D]

Which will be most of the time in an active print job, so I think it will have noticeable benefits.

I certainly believe this is worth doing! But I'm not sure it is going to be noticeable. Another way of looking at this is it increases our buffer depth from 16 to 17. When we have all 16 locations filled, it allows us to get through the crunching of one more.

But increasing the depth of the buffer by even one can stop some number of stutters!