Skip to content

Conversation

@Gelbpunkt
Copy link
Member

@Gelbpunkt Gelbpunkt commented Sep 19, 2025

RISC-V has no PCI support at the moment, so we cannot test it there, but it will at least compile fine now.

@Gelbpunkt Gelbpunkt marked this pull request as draft September 19, 2025 14:40
Copy link
Contributor

@github-actions github-actions bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Benchmark Results

Benchmark Current: 0a19f43 Previous: 85b8266 Performance Ratio
startup_benchmark Build Time 111.66 s 111.00 s 1.01
startup_benchmark File Size 0.89 MB 0.89 MB 1.00
Startup Time - 1 core 0.90 s (±0.04 s) 0.89 s (±0.03 s) 1.02
Startup Time - 2 cores 0.92 s (±0.03 s) 0.91 s (±0.04 s) 1.02
Startup Time - 4 cores 0.92 s (±0.03 s) 0.93 s (±0.03 s) 0.99
multithreaded_benchmark Build Time 111.24 s 113.81 s 0.98
multithreaded_benchmark File Size 1.00 MB 1.00 MB 1.00
Multithreaded Pi Efficiency - 2 Threads 86.65 % (±10.17 %) 88.19 % (±8.58 %) 0.98
Multithreaded Pi Efficiency - 4 Threads 43.39 % (±3.89 %) 44.07 % (±3.30 %) 0.98
Multithreaded Pi Efficiency - 8 Threads 25.76 % (±2.47 %) 25.16 % (±2.37 %) 1.02
micro_benchmarks Build Time 246.18 s 246.86 s 1.00
micro_benchmarks File Size 1.01 MB 1.01 MB 1.00
Scheduling time - 1 thread 135.00 ticks (±31.77 ticks) 123.33 ticks (±32.30 ticks) 1.09
Scheduling time - 2 threads 74.28 ticks (±15.27 ticks) 84.29 ticks (±17.18 ticks) 0.88
Micro - Time for syscall (getpid) 7.59 ticks (±3.93 ticks) 7.77 ticks (±4.23 ticks) 0.98
Memcpy speed - (built_in) block size 4096 60643.59 MByte/s (±43489.60 MByte/s) 54859.20 MByte/s (±39610.07 MByte/s) 1.11
Memcpy speed - (built_in) block size 1048576 19221.77 MByte/s (±16237.01 MByte/s) 22004.52 MByte/s (±19296.28 MByte/s) 0.87
Memcpy speed - (built_in) block size 16777216 15032.53 MByte/s (±12496.46 MByte/s) 13425.53 MByte/s (±10918.47 MByte/s) 1.12
Memset speed - (built_in) block size 4096 60712.27 MByte/s (±43536.45 MByte/s) 54922.42 MByte/s (±39653.65 MByte/s) 1.11
Memset speed - (built_in) block size 1048576 19627.59 MByte/s (±16484.00 MByte/s) 22431.42 MByte/s (±19479.96 MByte/s) 0.88
Memset speed - (built_in) block size 16777216 15377.30 MByte/s (±12678.31 MByte/s) 13846.16 MByte/s (±11209.76 MByte/s) 1.11
Memcpy speed - (rust) block size 4096 56171.29 MByte/s (±40804.03 MByte/s) 53850.35 MByte/s (±39705.05 MByte/s) 1.04
Memcpy speed - (rust) block size 1048576 23239.81 MByte/s (±20376.92 MByte/s) 22919.02 MByte/s (±20666.28 MByte/s) 1.01
Memcpy speed - (rust) block size 16777216 14390.10 MByte/s (±11891.02 MByte/s) 14917.25 MByte/s (±12407.45 MByte/s) 0.96
Memset speed - (rust) block size 4096 56826.22 MByte/s (±41235.97 MByte/s) 54350.35 MByte/s (±40017.63 MByte/s) 1.05
Memset speed - (rust) block size 1048576 23496.48 MByte/s (±20465.28 MByte/s) 23276.03 MByte/s (±20803.11 MByte/s) 1.01
Memset speed - (rust) block size 16777216 14673.30 MByte/s (±12039.50 MByte/s) 15418.30 MByte/s (±12751.21 MByte/s) 0.95
alloc_benchmarks Build Time 242.06 s 248.01 s 0.98
alloc_benchmarks File Size 0.96 MB 0.96 MB 1.00
Allocations - Allocation success 100.00 % 100.00 % 1
Allocations - Deallocation success 100.00 % 70.02 % (±0.29 %) 1.43
Allocations - Pre-fail Allocations 100.00 % 100.00 % 1
Allocations - Average Allocation time 9596.98 Ticks (±371.58 Ticks) 10407.84 Ticks (±352.99 Ticks) 0.92
Allocations - Average Allocation time (no fail) 9596.98 Ticks (±371.58 Ticks) 10407.84 Ticks (±352.99 Ticks) 0.92
Allocations - Average Deallocation time 1624.91 Ticks (±332.92 Ticks) 1297.47 Ticks (±167.81 Ticks) 1.25
mutex_benchmark Build Time 236.78 s 247.17 s 0.96
mutex_benchmark File Size 1.01 MB 1.01 MB 1.00
Mutex Stress Test Average Time per Iteration - 1 Threads 26.84 ns (±6.10 ns) 27.44 ns (±6.40 ns) 0.98
Mutex Stress Test Average Time per Iteration - 2 Threads 25.04 ns (±3.24 ns) 25.80 ns (±3.69 ns) 0.97

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

This lets the driver be used on all architectures, not just x86_64.

While at it, drop the udelay requirement, it boils down to a spin loop
and we already have a spin loop here.

Signed-off-by: Jens Reidel <[email protected]>
Signed-off-by: Jens Reidel <[email protected]>
@Gelbpunkt Gelbpunkt marked this pull request as ready for review October 21, 2025 09:05
@mkroening mkroening self-assigned this Oct 23, 2025
@mkroening mkroening self-requested a review October 23, 2025 14:28
Copy link
Member

@mkroening mkroening left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks! :)

Copy link
Member

@mkroening mkroening left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks! :)

@mkroening mkroening added this pull request to the merge queue Oct 24, 2025
@github-merge-queue github-merge-queue bot removed this pull request from the merge queue due to no response for status checks Oct 24, 2025
@mkroening mkroening added this pull request to the merge queue Oct 25, 2025
Merged via the queue into hermit-os:main with commit cb0d623 Oct 25, 2025
41 of 49 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants