tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Processor sets, affinity, real-time extensions

On Jun 2,  9:47pm, "Mindaugas R." wrote:
} (John Nemeth) wrote:
} >      cpuctl(8) offline says:
} > 
} >                                 Bound LWPs will continue to be executed on
} > the CPU, and device interrupts routed to the CPU will continue to be han-
} >          dled.
} > 
} >      Is this still true, since it doesn't meet the needs mentioned above?
} Yes, however this is not accurate now. Bound LWPs, in this context, are LWPs
} with LW_BOUND flag set. This is used for kernel-LWPs only.

     Then this doesn't meet the requirements from my first post.  To
meet those requirements, we need something that guarantees absolutely
no code will be executed on the CPU in question and power be removed
from the CPU (if possible).

}-- End of excerpt from "Mindaugas R."

Home | Main Index | Thread Index | Old Index