All types of wireless in Linux are terrible and why the vendors should feel bad
summary
Linux Wi‑Fi and Bluetooth repeatedly break because vendor firmware updates arrive without reproducible testing: Intel Bluetooth firmware breaks Wi‑Fi across hundreds of machines, Broadcom/Cypress/Synaptics change hands and stop releasing usable blobs, and Realtek drivers are often left in staging or missing firmware. The concrete fixes proposed are a hardware-backed CI lab (Faraday cages, CKI/Beaker integration) and a policy to revert or block firmware updates that cause regressions so regressions get caught and vendors are held accountable.
AI-written summary · about
chapters (14)
- 0:00Introduction and background
- 1:09Why wireless is unpleasant
- 3:36Broadcom Wi-Fi firmware issues
- 6:43Vendor engagement and Synaptics
- 7:30Marvell and NXP firmware struggles
- 8:10Realtek and staging drivers
- 9:14How to fix wireless regressions
- 13:30Bluetooth stack problems
- 14:54Automated testing and CI solutions
- 15:58Qualcomm and MediaTek status
- 17:24Reverting firmware updates
- 18:08Public shaming and vendor accountability
- 20:17Kernel regression tracking
- 21:18Summary and call to action
related talks
55mBluetooth on Embedded Linux Systems Deep Dive
55mReliable Linux Wireless - Techniques for Debugging Wireless Module Integrations
45mBOF: Corporate Use of Embedded Linux
27mWIFI-Important Remote Attack Surface: Threat is Expanding
48mUnpacking the Linux WiFi Stack: Writing and Integrating Wireless Drivers
41mWearables and Wireless: Bluetooth Internals for the Brave
01h 04mBoF: State of Linux on ARM Laptops
06mQ&A