Dear all,
Please note a small correction to the benchmark description in the assignment 0. The original statement referring to a floating-point divide operation was inaccurate. The code does not measure the duration of a divide operation. Instead, it is dominated by the square root computation inside the loop.
The previous statement was:
"Actually, this code can be used to measure the duration of a double-precision floating-point divide operation because the square root computation entirely dominates its runtime. Everything else can be 'hidden' behind the divide."
The corrected statement is:
"Actually, this code can be used to measure the duration of a double-precision floating-point square root operation because the square root computation entirely dominates its runtime. Everything else can be 'hidden' behind the square root."
This is the version you should use when interpreting your benchmark results.
All the best for your assignment submission!
Best,
Ayesha