revive: Rework the instruction benchmark

[athei]

Fixes #6157

This fixes the last remaining benchmark that was not correct since it was too low level to be written in Rust. Instead, we opted.

This PR changes the benchmark that determines the scaling from ref_time to PolkaVM Gas by benchmarking the absolute worst case of an instruction: One that causes two cache misses by touching two cache lines.

The Contract itself is designed to be as simple as possible. It does random unaligned reads in a loop until the r (repetition) number is reached. The randomness is fully generated by the host and written to the guests memory before the benchmark is run. This allows the benchmark to determine the influence of one loop iteration via linear regression.

[athei]

/cmd bench --runtime dev --pallet pallet_revive --clean

[github-actions]

Command "bench --runtime dev --pallet pallet_revive --clean" has started 🚀 See logs here

[github-actions]

Command "bench --runtime dev --pallet pallet_revive --clean" has finished ✅ See logs here

Subweight results:

File	Extrinsic	Old	New	Change [%]
substrate/frame/revive/src/weights.rs	instr	742.93us	1.31ms	+76.30
substrate/frame/revive/src/weights.rs	seal_origin	272.00ns	308.00ns	+13.24
substrate/frame/revive/src/weights.rs	seal_value_transferred	276.00ns	308.00ns	+11.59
substrate/frame/revive/src/weights.rs	seal_call_data_size	261.00ns	287.00ns	+9.96
substrate/frame/revive/src/weights.rs	seal_caller	273.00ns	300.00ns	+9.89
substrate/frame/revive/src/weights.rs	seal_balance	4.45us	4.83us	+8.73
substrate/frame/revive/src/weights.rs	seal_return_data_size	260.00ns	280.00ns	+7.69
substrate/frame/revive/src/weights.rs	seal_caller_is_origin	318.00ns	342.00ns	+7.55
substrate/frame/revive/src/weights.rs	set_transient_storage_full	1.73us	1.84us	+6.71
substrate/frame/revive/src/weights.rs	seal_gas_limit	397.00ns	423.00ns	+6.55
substrate/frame/revive/src/weights.rs	seal_ecdsa_recover	47.34us	49.80us	+5.20
substrate/frame/revive/src/weights.rs	seal_minimum_balance	294.00ns	277.00ns	-5.78
substrate/frame/revive/src/weights.rs	seal_caller_is_root	290.00ns	272.00ns	-6.21
substrate/frame/revive/src/weights.rs	seal_address	287.00ns	264.00ns	-8.01
substrate/frame/revive/src/weights.rs	instr_empty_loop		7.32ms	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	set_service_head		131.21us	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	service_queue_base		127.86us	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	service_page_item		505.09us	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	service_page_base_no_completion		130.00us	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	service_page_base_completion		129.87us	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	reap_page		374.70us	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	ready_ring_unknit		387.70us	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	ready_ring_knit		289.26us	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	execute_overweight_page_updated		467.36us	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	execute_overweight_page_removed		403.02us	Added
cumulus/parachains/runtimes/contracts/contracts-rococo/src/weights/pallet_message_queue.rs	bump_service_head		157.66us	Added

Command output:

✅ Successful benchmarks of runtimes/pallets:
-- dev: ['pallet_revive']

[xermicus]

Suggested change

	// we skip the first address since it is out entry point
	// we skip the first address since it is our entry point