Feature: Delegate /tmp to uhyve

[jounathaen]

This feature delegates all /tmp file accesses to uhyve, which can handle them since hermit-os/uhyve#783
This might be desirable, if the tempfiles are larger than the assigned memory, and it is not desirable when high-performance is required.

This depends on #1976

CC @n0toose

[n0toose]

lgtm

[github-actions]

Benchmark Results

Benchmark	Current: 4a3a64b	Previous: 0a574f1	Performance Ratio
startup_benchmark Build Time	112.20 s	112.17 s	1.00
startup_benchmark File Size	0.90 MB	0.86 MB	1.05
Startup Time - 1 core	0.91 s (±0.02 s)	0.92 s (±0.03 s)	0.99
Startup Time - 2 cores	0.93 s (±0.03 s)	0.91 s (±0.03 s)	1.02
Startup Time - 4 cores	0.92 s (±0.03 s)	0.90 s (±0.03 s)	1.02
multithreaded_benchmark Build Time	110.08 s	110.29 s	1.00
multithreaded_benchmark File Size	0.96 MB	0.96 MB	1.00
Multithreaded Pi Efficiency - 2 Threads	87.03 % (±7.71 %)	90.21 % (±11.06 %)	0.96
Multithreaded Pi Efficiency - 4 Threads	42.16 % (±3.81 %)	43.77 % (±3.68 %)	0.96
Multithreaded Pi Efficiency - 8 Threads	24.36 % (±2.15 %)	25.74 % (±2.50 %)	0.95
micro_benchmarks Build Time	264.79 s	268.25 s	0.99
micro_benchmarks File Size	0.97 MB	0.97 MB	1.00
Scheduling time - 1 thread	158.17 ticks (±23.47 ticks)	148.09 ticks (±20.64 ticks)	1.07
Scheduling time - 2 threads	88.33 ticks (±14.63 ticks)	84.62 ticks (±21.43 ticks)	1.04
Micro - Time for syscall (getpid)	10.02 ticks (±4.93 ticks)	9.69 ticks (±3.51 ticks)	1.03
Memcpy speed - (built_in) block size 4096	62830.43 MByte/s (±44875.96 MByte/s)	60088.55 MByte/s (±43602.38 MByte/s)	1.05
Memcpy speed - (built_in) block size 1048576	15717.58 MByte/s (±12700.10 MByte/s)	15955.36 MByte/s (±12877.41 MByte/s)	0.99
Memcpy speed - (built_in) block size 16777216	11490.25 MByte/s (±9343.25 MByte/s)	11539.44 MByte/s (±9422.40 MByte/s)	1.00
Memset speed - (built_in) block size 4096	63481.55 MByte/s (±45307.35 MByte/s)	60692.81 MByte/s (±43930.14 MByte/s)	1.05
Memset speed - (built_in) block size 1048576	16095.58 MByte/s (±12925.74 MByte/s)	16329.57 MByte/s (±13087.08 MByte/s)	0.99
Memset speed - (built_in) block size 16777216	11729.25 MByte/s (±9463.81 MByte/s)	11768.44 MByte/s (±9537.48 MByte/s)	1.00
Memcpy speed - (rust) block size 4096	56885.31 MByte/s (±41408.66 MByte/s)	52603.46 MByte/s (±39907.35 MByte/s)	1.08
Memcpy speed - (rust) block size 1048576	14934.15 MByte/s (±11984.26 MByte/s)	15838.45 MByte/s (±12923.25 MByte/s)	0.94
Memcpy speed - (rust) block size 16777216	11644.17 MByte/s (±9432.96 MByte/s)	11496.86 MByte/s (±9346.63 MByte/s)	1.01
Memset speed - (rust) block size 4096	57754.10 MByte/s (±41946.64 MByte/s)	53627.63 MByte/s (±40204.50 MByte/s)	1.08
Memset speed - (rust) block size 1048576	15324.94 MByte/s (±12254.27 MByte/s)	16175.83 MByte/s (±13086.77 MByte/s)	0.95
Memset speed - (rust) block size 16777216	11868.40 MByte/s (±9546.29 MByte/s)	11803.55 MByte/s (±9534.43 MByte/s)	1.01
alloc_benchmarks Build Time	268.48 s	275.93 s	0.97
alloc_benchmarks File Size	0.96 MB	0.92 MB	1.05
Allocations - Allocation success	100.00 %	100.00 %	1
Allocations - Deallocation success	100.00 %	100.00 %	1
Allocations - Pre-fail Allocations	100.00 %	100.00 %	1
Allocations - Average Allocation time	17051.79 Ticks (±621.00 Ticks)	16993.86 Ticks (±868.21 Ticks)	1.00
Allocations - Average Allocation time (no fail)	17051.79 Ticks (±621.00 Ticks)	16993.86 Ticks (±868.21 Ticks)	1.00
Allocations - Average Deallocation time	2031.37 Ticks (±478.74 Ticks)	2094.62 Ticks (±499.20 Ticks)	0.97
mutex_benchmark Build Time	266.79 s	283.00 s	0.94
mutex_benchmark File Size	0.97 MB	0.97 MB	1.00
Mutex Stress Test Average Time per Iteration - 1 Threads	32.60 ns (±2.82 ns)	34.04 ns (±3.45 ns)	0.96
Mutex Stress Test Average Time per Iteration - 2 Threads	28.48 ns (±3.03 ns)	29.22 ns (±3.01 ns)	0.97

This comment was automatically generated by workflow using github-action-benchmark.

[mkroening]

Looks good to me! :)