Error due to limited precision.txt

This is correct calc
prev_val + min_dist = new val
192009078601097601351680 +  1559054711628234752000 =  193568134720100719656960

new val - prev_val =  min_dist_c != min_dist
193568134720100719656960 - 192009078601097601351680 = 1559056119003118305280

Example of error due to precision of floating point numbers:
Even when double is used everywhere
0,2-->prev_val::219145517497640556691456
0,2-->old_prev_val_c::219145517497640556691456
0,2-->distances_c[lower+i]::469631602634976867123200
0,2-->local_dist_0::250486085137336310431744
0,2-->local_dist_1::24536580825498624
0,2-->w,i::(1,1)
0,2-->min_dist  ::24536580825498624
0,2-->min_dist_c::24536580825498624
0,2-->distances[lower+i]  ::219145542034221380403200
0,2-->distances_c[lower+i]::219145542034221380403200
0,2-->prev_val  ::219145542034221380403200
0,2-->prev_val_c::219145542034221380403200

219145517497640556691456
	   24536580825498624
------------------------
219145542034221382190080 -->Correct
219145542034221380403200 -->What we get

prev_val + min_dist != distances[lower+i] even when we actually add the two operands on LHS to get the RHS
This error is due to limited precision of floating point numbers , float has 7 digit and double had 15 (significant)digit precision.