Subject: Re: Please Revert newlock2
To: None <email@example.com>
From: Bucky Katz <firstname.lastname@example.org>
Date: 02/18/2007 19:02:19
Matt Thomas <email@example.com> writes:
> Bucky Katz wrote:
>> "Thomas E. Spanjaard" <firstname.lastname@example.org> writes:
>>> 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 will
>>>> cause a severe performance degradation.
>>> Doesn't OMAP support the v4/v5 hyped fast address space switching?
>> I dunno. I don't recognize it by that name and can't find any term
>> like that in the processor TRM.
>> If you mean (and I suspect you do) The "Fast Context Switch
>> FCSE, then yes the OMAP processors we are currently using support it.
> Except that FCSE is useless for NetBSD. It reduces each process to
> a 64MB (25 bit address space).
Which would be OK in our world, since we don't swap and only have 64mb
of memory, but would be far too intrusive to justify for just one