Messages in this thread | | | Subject | Re: [PATCH 0/6] support "dataplane" mode for nohz_full | From | Mike Galbraith <> | Date | Wed, 27 May 2015 05:28:01 +0200 |
| |
On Tue, 2015-05-26 at 15:51 -0400, Chris Metcalf wrote:
> On balance I suspect it's still better to make command line arguments > handle the common cases most succinctly.
I prefer user specifies precisely, but yeah, that entails more typing.
Idle curiosity: can SGI monster from hell boot a NO_HZ_FULL_ALL kernel, w/wo it implying isolcpus? Readers having same and a reactor to power it in their basement, please test.
-Mike
| |