Messages in this thread | | | From | Catalin Marinas <> | Date | Wed, 7 Jan 2015 22:24:44 +0000 | Subject | Re: [PATCH] Revert 9fc2105aeaaf56b0cf75296a84702d0f9e64437b to fix pyaudio (and probably more) |
| |
On 7 January 2015 at 19:36, Linus Torvalds <torvalds@linux-foundation.org> wrote: > This whole thread has wasted more time than the whole original > argument for wasted time ever was. I still have no idea what the > original argument was, and why you guys want some made-up and > incorrect feel-good number rather than just document the fact that the > bogomips is simple dependent on the clocksource you use for delays.
One reason - consistency with kernels running on the *same* hardware prior to commit d0a533b18235d362 (ARM: 7452/1: delay: allow timer-based delay implementation to be selected). You may see BogoMIPS of 800 prior to that commit dropping to 6 after that commit on the *same* piece of hardware. If we don't break any user assumptions here, no further changes needed.
-- Catalin
| |