Set adapter concurrency using chip info XNCP command#681
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## dev #681 +/- ##
=======================================
Coverage 99.51% 99.51%
=======================================
Files 58 58
Lines 3924 3951 +27
=======================================
+ Hits 3905 3932 +27
Misses 19 19 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
There was a problem hiding this comment.
Pull Request Overview
This PR implements functionality to dynamically set adapter concurrency based on chip information retrieved via an XNCP command. The implementation detects newer adapters with more RAM (like MG24) and increases their concurrency from the default 8 to 32 for better performance, while maintaining backward compatibility.
Key changes:
- Added XNCP command support for retrieving chip information (RAM size and part number)
- Implemented logic to determine optimal concurrency based on available RAM
- Enhanced network metadata to include chip information when available
Reviewed Changes
Copilot reviewed 5 out of 5 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| bellows/ezsp/xncp.py | Added new XNCP command definitions for chip info retrieval |
| bellows/ezsp/init.py | Implemented chip info query and concurrency determination logic |
| bellows/zigbee/application.py | Integrated dynamic concurrency setting and chip info metadata |
| tests/test_ezsp.py | Added comprehensive tests for chip info functionality |
| tests/test_application.py | Added tests for concurrency setting and metadata inclusion |
| chip_info = await self.xncp_get_chip_info() | ||
|
|
||
| # Usually 98304 bytes for MG21 | ||
| if chip_info.ram_size < 100000: |
There was a problem hiding this comment.
The magic number 100000 should be defined as a named constant to improve maintainability and make the threshold explicit.
| if chip_info.ram_size < 100000: | |
| if chip_info.ram_size < RAM_SIZE_THRESHOLD: |
| return 8 | ||
|
|
||
| # Usually 262144 bytes for MG24 | ||
| return 32 |
There was a problem hiding this comment.
The return value 32 should be defined as a named constant (e.g., HIGH_RAM_CONCURRENCY) to make the concurrency levels explicit and configurable.
| return 32 | |
| return HIGH_RAM_CONCURRENCY |
| return 8 | ||
|
|
||
| chip_info = await self.xncp_get_chip_info() | ||
|
|
||
| # Usually 98304 bytes for MG21 | ||
| if chip_info.ram_size < 100000: | ||
| return 8 |
There was a problem hiding this comment.
The return value 8 should be defined as a named constant (e.g., DEFAULT_CONCURRENCY) to match the pattern used elsewhere and improve maintainability.
| return 8 | |
| chip_info = await self.xncp_get_chip_info() | |
| # Usually 98304 bytes for MG21 | |
| if chip_info.ram_size < 100000: | |
| return 8 | |
| return DEFAULT_CONCURRENCY | |
| chip_info = await self.xncp_get_chip_info() | |
| # Usually 98304 bytes for MG21 | |
| if chip_info.ram_size < 100000: | |
| return DEFAULT_CONCURRENCY |
puddly
force-pushed
the
puddly/chip-id
branch
from
c11a672 to
5e6d71e
Compare
1 participant
See NabuCasa/silabs-firmware-builder#121. The current default will remain 8, we can bump it higher for newer adapters.