Re: [PATCH] sched/rt: RT_RUNTIME_GREED sched feature
From: Clark Williams <hidden>
Date: 2016-11-07 18:50:13
Also in:
lkml
Attachments
- (unnamed) [application/pgp-signature] 801 bytes
From: Clark Williams <hidden>
Date: 2016-11-07 18:50:13
Also in:
lkml
On Mon, 7 Nov 2016 13:30:46 -0500 Steven Rostedt [off-list ref] wrote:
On Mon, 7 Nov 2016 12:22:21 -0600 Clark Williams [off-list ref] wrote:quoted
I'm still reviewing the patch, but I have to wonder why bother with making it a scheduler feature? The SCHED_FIFO definition allows a fifo thread to starve others because a fifo task will run until it yields. Throttling was added as a safety valve to allow starved SCHED_OTHER tasks to get some cpu time. Adding this unconditionally gets us a safety valve for throttling a badly written fifo task, but allows the fifo task to continue to consume cpu cycles if it's not starving anyone. Or am I missing something that's blazingly obvious?Or I say make it the default. If people want the old behavior, they can modify SCHED_FEATURES to do so.
Ok, I can see wanting the previous behavior. Clark