site_archiver@lists.apple.com Delivered-To: darwin-dev@lists.apple.com User-agent: Microsoft-Entourage/11.0.0.040405 On Wed, 29 Sep 2004 16:27:38 +0200, Markus Hitter <mah@jump-ing.de> wrote:
Am 29.09.2004 um 15:54 schrieb John Siracusa:
One theory presented to explain the results was that the kernel is taking CPU temperature into account when doling out CPU cycles.
Well, I'm not really a kernel expert, but the scheduler's task is to decide which process/thread gets the next CPU quantum, where a CPU quantum is about 10 ms. If there's no process in need of processing time, it sends the CPU into doze mode.
The observed result of the test program is that it runs at X iterations/sec for some time, then drops to X - N iterations/sec for the remainder. The drop seems to coincide with a CPU temp that passes some threshold. This drop is not seen on any Macs other than the dual 2.5GHz Power Mac. The concern is that people buying a 2.5GHz Mac (at great expense) cannot use the full capacity of the CPUs for more than a short period of time.
I thought it might be the CPU hog "punisher" in the kernel
Surely, there are more mechanisms going on in the kernel, independent from the scheduler. If the CPU supports running at different speeds, you should be able to switch between them without telling the scheduler.
This is how I understand things but again, I'm no expert :-)
The goal is not to throttle the CPU, but to make it run a particular task using all available CPU resources. Obviously there is some overhead for the OS, etc., but the one-time drop-off in speed on the 2.5GHz Power Mac (and nowhere else when running the same code, same OS, etc.) has led some people to believe that the issue is CPU heat, and that the OS (or maybe even the hardware itself) is compensating for it by slowing down slightly. -John _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.appl... This email sent to site_archiver@lists.apple.com
participants (1)
-
John Siracusa