bug: cpu work in ops benchmark of core is optimized out

[NKID00]

Describe the bug

The following code in bench_read_parallel is optimized out before benchmark:

opendal/core/benches/ops/read.rs

Lines 98 to 103 in e869a70

	let mut d = 0;
	// mock same little cpu work
	for c in offset..offset + 100u64 {
	d += c & (0x1f1f1f1f + c % 256);
	}
	let _ = d;

Steps to Reproduce

The entire loop is optimized out in emitted llvm ir displayed using cargo-show-asm:

cargo asm -p opendal --bench ops --features tests --release --llvm ops::read::bench_read_parallel 12

Expected Behavior

This piece of code should not be optimized out.

Additional Context

No response

Are you willing to submit a PR to fix this bug?

• Yes, I would like to submit a PR.