Interpreted libriscv vs. Luau (both actual sandboxes)
June 8, 2025 ยท View on GitHub
$ ./bench
-
All benchmark results are measured in 200x1200 samples benchmark overhead median 0ns lowest: 0ns highest: 0ns Self-test running ELF entry at 0x10C20 Calling into event loop...! event_loop: Checking for work event_loop: Going to sleep! add_work: Adding work Resuming event loop...! event_loop: Checking for work work: Doing some work! event_loop: Going to sleep! event_loop: Checking for work event_loop: Going to sleep! add_work: Adding work event_loop: Checking for work work: Doing some work! event_loop: Going to sleep! RISC-V self-test OK
-
Loading program of size 202739 from 0x75827123c100 to virtual 0x10000 -> 0x417f3
-
Program segment readable: 1 writable: 0 executable: 1
-
Loading program of size 2096 from 0x75827126d8f4 to virtual 0x427f4 -> 0x43024
-
Program segment readable: 1 writable: 1 executable: 0
-
Using program-provided exit function at 0x119c8
-
Entry is at 0x10c20 libriscv: mips instr 1600020000, time 1.549093, 1032.875527 mip/s libriscv: call overhead median 2ns lowest: 2ns highest: 3ns libriscv: call args=1 median 2ns lowest: 2ns highest: 3ns libriscv: call args=2 median 2ns lowest: 2ns highest: 4ns libriscv: call args=3 median 2ns lowest: 2ns highest: 8ns libriscv: call args=4 median 2ns lowest: 2ns highest: 2ns libriscv: call args=5 median 2ns lowest: 2ns highest: 5ns libriscv: call args=6 median 2ns lowest: 2ns highest: 3ns libriscv: call args=7 median 2ns lowest: 2ns highest: 6ns libriscv: call args=8 median 3ns lowest: 3ns highest: 3ns libriscv: preempt overhead median 10ns lowest: 10ns highest: 15ns libriscv: resume execution median 3ns lowest: 3ns highest: 5ns luau: call overhead median 87ns lowest: 86ns highest: 116ns luau: call args=1 median 96ns lowest: 95ns highest: 125ns luau: call args=2 median 100ns lowest: 98ns highest: 132ns luau: call args=3 median 102ns lowest: 99ns highest: 133ns luau: call args=4 median 103ns lowest: 101ns highest: 136ns luau: call args=5 median 104ns lowest: 102ns highest: 111ns luau: call args=6 median 106ns lowest: 105ns highest: 199ns luau: call args=7 median 111ns lowest: 109ns highest: 216ns luau: call args=8 median 110ns lowest: 109ns highest: 145ns
libriscv: syscall args=0 median 3ns lowest: 3ns highest: 7ns libriscv: syscall args=1 median 3ns lowest: 3ns highest: 94ns libriscv: syscall args=2 median 3ns lowest: 3ns highest: 6ns libriscv: syscall args=3 median 3ns lowest: 3ns highest: 6ns libriscv: syscall args=4 median 3ns lowest: 3ns highest: 5ns libriscv: syscall args=5 median 3ns lowest: 3ns highest: 4ns libriscv: syscall args=6 median 4ns lowest: 4ns highest: 93ns libriscv: syscall args=7 median 4ns lowest: 3ns highest: 5ns libriscv: syscall string median 6ns lowest: 6ns highest: 10ns luau: syscall args=0 median 105ns lowest: 103ns highest: 111ns luau: syscall args=1 median 113ns lowest: 111ns highest: 117ns luau: syscall args=2 median 125ns lowest: 123ns highest: 216ns luau: syscall args=3 median 130ns lowest: 128ns highest: 219ns luau: syscall args=4 median 139ns lowest: 136ns highest: 201ns luau: syscall args=5 median 149ns lowest: 147ns highest: 174ns luau: syscall args=6 median 157ns lowest: 155ns highest: 190ns luau: syscall args=7 median 167ns lowest: 164ns highest: 191ns luau: syscall string median 114ns lowest: 112ns highest: 145ns
After this point, call overhead is subtracted from every benchmark result for both libriscv and Lua
libriscv: array append median 3ns lowest: 3ns highest: 8ns
libriscv: vector append median 7ns lowest: 7ns highest: 10ns
luau: table append median 26ns lowest: 25ns highest: 57ns
libriscv: 8x arguments median 48ns lowest: 40ns highest: 58ns
libriscv: 8x stored args median 35ns lowest: 31ns highest: 41ns
libriscv: 8x prepared args median 42ns lowest: 39ns highest: 52ns
luau: 8x arguments median 224ns lowest: 220ns highest: 315ns
libriscv: integer math median 3ns lowest: 3ns highest: 7ns
libriscv: fp math median 13ns lowest: 10ns highest: 17ns
libriscv: exp math median 13ns lowest: 12ns highest: 22ns
libriscv: fib(40) median 220ns lowest: 204ns highest: 328ns
libriscv: taylor(1K) median 6746ns lowest: 5811ns highest: 7409ns
luau: integer math median 38ns lowest: 34ns highest: 72ns
luau: fp math median 46ns lowest: 43ns highest: 77ns
luau: exp math median 50ns lowest: 47ns highest: 76ns
luau: fib(40) median 489ns lowest: 477ns highest: 535ns
luau: taylor(1K) median 10122ns lowest: 8297ns highest: 16641ns
libriscv: complex syscall median 35ns lowest: 34ns highest: 42ns
luau: complex syscall median 674ns lowest: 659ns highest: 708ns
libriscv: micro threads median 91ns lowest: 89ns highest: 105ns
luau: coroutines median 178ns lowest: 173ns highest: 205ns
libriscv: micro thread args median 137ns lowest: 133ns highest: 149ns
libriscv: full thread args median 146ns lowest: 141ns highest: 240ns
luau: coroutine args median 226ns lowest: 220ns highest: 338ns
luau: coroutine args median 223ns lowest: 217ns highest: 316ns
libriscv: naive memcpy median 888ns lowest: 819ns highest: 1010ns
libriscv: syscall memcpy median 18ns lowest: 18ns highest: 130ns
luau: memcpy median 2170ns lowest: 2140ns highest: 2293ns
libriscv: syscall memset median 11ns lowest: 11ns highest: 15ns
libriscv: naive memset median 176ns lowest: 170ns highest: 217ns
luau: memset median 1673ns lowest: 1660ns highest: 1758ns
libriscv: sieve(10M) median 138ms lowest: 138ms highest: 138ms
luau: sieve(10M) median 574ms lowest: 574ms highest: 574ms