~/talk

Real-Time Linux on Embedded Multicore Processors

links:none on file
slides:none on file

Standard Linux plus PREEMPT-RT and real latency measurement with cyclictest move from single-core to NXP T2080/T4240 multicore hardware; results show clean speedup to about four or five physical cores then plateau or regress because SMT/virtual cores and a single vector unit per physical core create contention. Strategies that help are isolcpus and IRQ affinity and dedicated-core deployment, but long runs reveal rare outliers caused by shared L2 cache / fabric interference and DMA/I O contention — measure, prototype, and map threads to physical cores.

AI-written summary · about

  1. 0:00Start
  2. 4:16Linux Kernel
  3. 5:16Cyclic Test Simplified
  4. 6:42A Cyclic Test on the Old Single Core Processor System
  5. 15:13Interrupt
  6. 19:07Start a New Project
~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.