fix: Proper network wakers #2086
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Gelbpunkt
force-pushed
the
wakers
branch
2 times, most recently
from
9c652d1 to
c03dc8e
Compare
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Benchmark Results
Details
| Benchmark | Current: 7a4cd75 | Previous: e645fec | Performance Ratio |
|---|---|---|---|
| startup_benchmark Build Time | 100.16 s |
99.81 s |
1.00 ❗ |
| startup_benchmark File Size | 0.87 MB |
0.86 MB |
1.01 ❗ |
| Startup Time - 1 core | 0.93 s (±0.04 s) |
0.93 s (±0.04 s) |
0.99 |
| Startup Time - 2 cores | 0.94 s (±0.05 s) |
0.96 s (±0.03 s) |
0.98 |
| Startup Time - 4 cores | 0.93 s (±0.03 s) |
0.94 s (±0.03 s) |
0.99 |
| multithreaded_benchmark Build Time | 101.65 s |
102.13 s |
1.00 ❗ |
| multithreaded_benchmark File Size | 0.97 MB |
0.97 MB |
1.00 ❗ |
| Multithreaded Pi Efficiency - 2 Threads | 89.13 % (±9.40 %) |
88.34 % (±8.04 %) |
1.01 |
| Multithreaded Pi Efficiency - 4 Threads | 43.34 % (±3.79 %) |
44.56 % (±3.57 %) |
0.97 |
| Multithreaded Pi Efficiency - 8 Threads | 25.08 % (±1.89 %) |
25.91 % (±1.86 %) |
0.97 |
| micro_benchmarks Build Time | 104.32 s |
100.32 s |
1.04 ❗ |
| micro_benchmarks File Size | 0.97 MB |
0.97 MB |
1.00 ❗ |
| Scheduling time - 1 thread | 68.99 ticks (±4.37 ticks) |
68.60 ticks (±3.48 ticks) |
1.01 |
| Scheduling time - 2 threads | 39.57 ticks (±5.51 ticks) |
40.01 ticks (±5.48 ticks) |
0.99 |
| Micro - Time for syscall (getpid) | 4.02 ticks (±0.54 ticks) |
3.30 ticks (±0.47 ticks) |
1.22 ❗ |
| Memcpy speed - (built_in) block size 4096 | 64052.43 MByte/s (±45393.26 MByte/s) |
66423.24 MByte/s (±47211.72 MByte/s) |
0.96 |
| Memcpy speed - (built_in) block size 1048576 | 29611.83 MByte/s (±24277.61 MByte/s) |
29366.89 MByte/s (±24195.28 MByte/s) |
1.01 |
| Memcpy speed - (built_in) block size 16777216 | 25591.21 MByte/s (±21445.88 MByte/s) |
24865.20 MByte/s (±20794.46 MByte/s) |
1.03 |
| Memset speed - (built_in) block size 4096 | 64881.68 MByte/s (±45983.33 MByte/s) |
66997.01 MByte/s (±47611.29 MByte/s) |
0.97 |
| Memset speed - (built_in) block size 1048576 | 30365.20 MByte/s (±24710.05 MByte/s) |
30137.28 MByte/s (±24628.24 MByte/s) |
1.01 |
| Memset speed - (built_in) block size 16777216 | 26345.61 MByte/s (±21915.59 MByte/s) |
25636.38 MByte/s (±21296.15 MByte/s) |
1.03 |
| Memcpy speed - (rust) block size 4096 | 61323.25 MByte/s (±45247.39 MByte/s) |
60802.24 MByte/s (±44540.11 MByte/s) |
1.01 |
| Memcpy speed - (rust) block size 1048576 | 29509.17 MByte/s (±24249.73 MByte/s) |
29548.98 MByte/s (±24182.47 MByte/s) |
1.00 |
| Memcpy speed - (rust) block size 16777216 | 24764.69 MByte/s (±20860.48 MByte/s) |
25616.27 MByte/s (±21548.07 MByte/s) |
0.97 |
| Memset speed - (rust) block size 4096 | 62007.22 MByte/s (±45683.42 MByte/s) |
61752.87 MByte/s (±45214.15 MByte/s) |
1.00 |
| Memset speed - (rust) block size 1048576 | 30214.18 MByte/s (±24655.61 MByte/s) |
30310.81 MByte/s (±24622.20 MByte/s) |
1.00 |
| Memset speed - (rust) block size 16777216 | 25523.71 MByte/s (±21348.85 MByte/s) |
26370.24 MByte/s (±22014.77 MByte/s) |
0.97 |
| alloc_benchmarks Build Time | 94.93 s |
94.69 s |
1.00 ❗ |
| alloc_benchmarks File Size | 0.94 MB |
0.93 MB |
1.01 ❗ |
| 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 | 8696.55 Ticks (±228.59 Ticks) |
9799.65 Ticks (±144.68 Ticks) |
0.89 ❗ |
| Allocations - Average Allocation time (no fail) | 8696.55 Ticks (±228.59 Ticks) |
9799.65 Ticks (±144.68 Ticks) |
0.89 ❗ |
| Allocations - Average Deallocation time | 1249.65 Ticks (±727.81 Ticks) |
919.79 Ticks (±324.70 Ticks) |
1.36 |
| mutex_benchmark Build Time | 93.15 s |
93.42 s |
1.00 ❗ |
| mutex_benchmark File Size | 0.97 MB |
0.97 MB |
1.00 ❗ |
| Mutex Stress Test Average Time per Iteration - 1 Threads | 12.62 ns (±0.56 ns) |
12.86 ns (±0.77 ns) |
0.98 |
| Mutex Stress Test Average Time per Iteration - 2 Threads | 92.90 ns (±4.65 ns) |
13.22 ns (±0.90 ns) |
7.03 ❗ |
This comment was automatically generated by workflow using github-action-benchmark.
Gelbpunkt
force-pushed
the
wakers
branch
8 times, most recently
from
390c379 to
466882e
Compare
Gelbpunkt
force-pushed
the
wakers
branch
3 times, most recently
from
f48f443 to
839554d
Compare
mkroening
requested changes
Jan 20, 2026
Member
mkroening
left a comment
mkroening
left a comment
There was a problem hiding this comment.
Thanks a lot; this is looking great! :)
Comment on lines
+586
to
+584
| create_timer_abs(Source::Scheduler, wt); | ||
| return; | ||
| } | ||
|
|
||
| cursor.move_next(); | ||
| } | ||
|
|
||
| set_oneshot_timer(); | ||
| create_timer_abs(Source::Scheduler, wt); |
Member
There was a problem hiding this comment.
The docs say that wakeup_time is relative. This was also done wrongly before, but maybe we should follow up to fix this.
Gelbpunkt
force-pushed
the
wakers
branch
2 times, most recently
from
e1f9653 to
d5f9bef
Compare
This refactors our timer interrupt handling into a separate module that all timer interrupts are configured through and handled by. It allows us to track why a timer interrupt was fired. We make use of this to now only conditionally wake the network task's waker when necessary. This is either because we sent some network packets, a network device driver received an interrupt, or because a timer interrupt that we configured so we can poll smoltcp in the future was handled.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This refactors our timer interrupt handling into a separate module that all timer interrupts are configured through and handled by. It allows us to track why a timer interrupt was fired.
We make use of this to now only conditionally wake the network task's waker when necessary. This is either because we sent some network packets, a network device driver received an interrupt, or because a timer interrupt that we configured so we can poll smoltcp in the future was handled.
Closes #2126.