MSI-X support

[cagatay-y]

cloud-hypervisor only supports MSI-X interrupts for PCI devices, so support for MSI-X is needed to support running on it. Additionally, MSI-X can allow us to set separate interrupt handlers for the configuration change interrupts and queue updates (even per-queue handlers once we have multiple queues) in the future.

[github-actions]

Benchmark Results

Details

Benchmark	Current: 938d062	Previous: ff18234	Performance Ratio
startup_benchmark Build Time	99.15 s	97.61 s	1.02 ❗
startup_benchmark File Size	0.86 MB	0.86 MB	1.00 ❗
Startup Time - 1 core	0.97 s (±0.04 s)	0.93 s (±0.03 s)	1.04
Startup Time - 2 cores	0.98 s (±0.02 s)	0.95 s (±0.03 s)	1.03
Startup Time - 4 cores	0.98 s (±0.04 s)	0.94 s (±0.03 s)	1.04
multithreaded_benchmark Build Time	103.38 s	99.88 s	1.04 ❗
multithreaded_benchmark File Size	0.96 MB	0.96 MB	1.00 ❗
Multithreaded Pi Efficiency - 2 Threads	87.55 % (±8.04 %)	88.00 % (±9.25 %)	0.99
Multithreaded Pi Efficiency - 4 Threads	43.53 % (±3.93 %)	43.12 % (±4.33 %)	1.01
Multithreaded Pi Efficiency - 8 Threads	25.02 % (±1.82 %)	24.87 % (±2.17 %)	1.01
micro_benchmarks Build Time	112.23 s	95.29 s	1.18 ❗
micro_benchmarks File Size	0.97 MB	0.97 MB	1.00 ❗
Scheduling time - 1 thread	71.60 ticks (±4.69 ticks)	66.64 ticks (±3.71 ticks)	1.07
Scheduling time - 2 threads	39.12 ticks (±5.06 ticks)	38.45 ticks (±5.98 ticks)	1.02
Micro - Time for syscall (getpid)	2.93 ticks (±0.29 ticks)	3.17 ticks (±0.47 ticks)	0.92
Memcpy speed - (built_in) block size 4096	62220.01 MByte/s (±44492.73 MByte/s)	66568.39 MByte/s (±47388.78 MByte/s)	0.93
Memcpy speed - (built_in) block size 1048576	29893.58 MByte/s (±24741.77 MByte/s)	29555.04 MByte/s (±24278.48 MByte/s)	1.01
Memcpy speed - (built_in) block size 16777216	25704.72 MByte/s (±21625.70 MByte/s)	28751.44 MByte/s (±23938.19 MByte/s)	0.89
Memset speed - (built_in) block size 4096	62363.88 MByte/s (±44585.36 MByte/s)	66780.03 MByte/s (±47520.66 MByte/s)	0.93
Memset speed - (built_in) block size 1048576	30697.52 MByte/s (±25205.32 MByte/s)	30326.51 MByte/s (±24715.31 MByte/s)	1.01
Memset speed - (built_in) block size 16777216	26486.35 MByte/s (±22117.96 MByte/s)	29524.28 MByte/s (±24375.39 MByte/s)	0.90
Memcpy speed - (rust) block size 4096	58148.66 MByte/s (±42771.28 MByte/s)	63679.07 MByte/s (±46857.40 MByte/s)	0.91
Memcpy speed - (rust) block size 1048576	29836.67 MByte/s (±24755.69 MByte/s)	29550.94 MByte/s (±24294.01 MByte/s)	1.01
Memcpy speed - (rust) block size 16777216	24288.25 MByte/s (±20508.25 MByte/s)	28708.33 MByte/s (±23916.49 MByte/s)	0.85
Memset speed - (rust) block size 4096	59449.21 MByte/s (±43690.98 MByte/s)	64665.18 MByte/s (±47525.69 MByte/s)	0.92
Memset speed - (rust) block size 1048576	30608.74 MByte/s (±25197.75 MByte/s)	30306.68 MByte/s (±24722.80 MByte/s)	1.01
Memset speed - (rust) block size 16777216	25002.99 MByte/s (±20957.96 MByte/s)	29488.78 MByte/s (±24360.76 MByte/s)	0.85
alloc_benchmarks Build Time	105.06 s	92.78 s	1.13 ❗
alloc_benchmarks File Size	0.94 MB	0.94 MB	1.00 ❗
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	9940.86 Ticks (±134.93 Ticks)	9546.98 Ticks (±130.84 Ticks)	1.04 ❗
Allocations - Average Allocation time (no fail)	9940.86 Ticks (±134.93 Ticks)	9546.98 Ticks (±130.84 Ticks)	1.04 ❗
Allocations - Average Deallocation time	925.79 Ticks (±445.38 Ticks)	1180.86 Ticks (±623.01 Ticks)	0.78
mutex_benchmark Build Time	104.92 s	98.39 s	1.07 ❗
mutex_benchmark File Size	0.97 MB	0.97 MB	1.00 ❗
Mutex Stress Test Average Time per Iteration - 1 Threads	12.86 ns (±0.60 ns)	12.90 ns (±0.54 ns)	1.00
Mutex Stress Test Average Time per Iteration - 2 Threads	16.24 ns (±1.07 ns)	13.42 ns (±0.72 ns)	1.21 ❗

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

[mkroening]

pci_types is updated as of #2220.