rick at svlug.org
Sat Jan 17 17:54:55 PST 2015
Alison Chaiken wrote:
> Steve and Rick, if you look in your dmesg log, you'll find that the order
> of devices coming up at boot under the control of the kernel is quite
> asynchronous. Kernel developers have a limited ability to sequence
> operations. If one device driver doesn't call the methods of another, or
> take a reference on the other, they may come up in any old order. Linux
> has been becoming increasing asynchronous, preemptible and unpredictable
> with each passing kernel.
Yes, this is basically the point I've been dangling in front of Steve. ;->
Near as I can tell, the change in this area, within the kernel and
associated driver code, spiked upwards dramatically starting around the
beginning of this decade, which is why it's taken me a bit by surprise.
Thanks for the details and data points about power, CPU speed, CPd idle,
and the like. Brown's slides also hint at another area that is probably
significant: more-effective kernel support for ACPI functions. Even though
I've long said that ACPI is about the clearest case of Second System Effect
yet seen in the IA32 world, sadly it cannot be ignored, and doubtless
introduces yet more variability in device timing and presence.
De-lurk any time!
More information about the svlug