Subject: Re: Please Revert newlock2
To: None <firstname.lastname@example.org>
From: Bucky Katz <email@example.com>
Date: 02/20/2007 14:54:12
Jason Thorpe <firstname.lastname@example.org> writes:
> On Feb 18, 2007, at 11:28 AM, Thomas E. Spanjaard wrote:
>> Bucky Katz wrote:
>>> Thor Lancelot Simon <email@example.com> writes:
>>>> 2) Either cause them to perform acceptably for your application,
>>>> or else
>>>> endeavor to get M:N back in a supported state for uniprocessors.
>>> It has to be the later. The context switch overhead of 1:1 on ARM
>>> cause a severe performance degradation.
>> Doesn't OMAP support the v4/v5 hyped fast address space switching?
> That doesn't matter when you're switching within the same address space.
> Anyway, if there are performance problems with context switching on
> ARM, then those should be addressed independently.
The only ways you can address the performance problems with context
switching on ARM are to either a) reduce the number of context
switches you do on ARM or b) move to ARM11.
We're discussing hardware design deficiencies here that everyone who
puts a multitasking OS on ARM has to cope with.