Do Existing Drivers Conflict with PREEMPT_RT? a 3-Year Survey
summary
A three‑year survey of PREEMPT_RT non‑core diffs examines sequence counters, tasklets, printk/console changes and fixes touching graphics, SCSI, networking, FireWire and SPI drivers. Sequence counters gain extra runtime sanity checks and a formalized latch sequence counter API; tasklet busy‑loops are replaced with condition variables and sleeping after a manual 400+ drivers surveyed audit that finds all except 10 call sites invoked tasklet delete from preemptable context. Coccinelle helps some refactors; goal is drivers that work unchanged under PREEMPT_RT.
AI-written summary · about
chapters (9)
- 0:00Talk introduction
- 1:31PREEMPT_RT benefits
- 4:01Sequence counters challenges
- 10:08Tasklets API issues
- 15:37Tasklets removal and compliance
- 18:52PREEMPT_RT with standard APIs
- 23:29Printk and console locking
- 25:16Q&A
- 29:14PREEMPT_RT vs static partitioning
related talks
34mDo Existing Drivers Conflict with PREEMPT_RT? a 3-Year Survey
47mWhat Every Driver Developer Should Know about RT
53mPREEMPT_RT - How Not to Break It.
24mPREEMPT_RT Q&A with tglx
47mReal-Time Linux: What Is Next?
50mInside the RT Patch
33mQ&A about Preempt RT
36mJourney of improving RT performance on Single CPU equipped PREEMPT_RT Linux