Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 8 May 2010 16:00:44 -0700
From:      "K. Macy" <kmacy@freebsd.org>
To:        bf1783@gmail.com
Cc:        alc@freebsd.org, freebsd-current@freebsd.org
Subject:   Re: Recent sys/vm/ changes and nvidia-driver
Message-ID:  <m2j82c4140e1005081600u80823e01k756981ffd2c27931@mail.gmail.com>
In-Reply-To: <AANLkTil0Gs3shexWm28gceSFpvIeJKfMAYEk9_t4jmzH@mail.gmail.com>
References:  <AANLkTil0Gs3shexWm28gceSFpvIeJKfMAYEk9_t4jmzH@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, May 8, 2010 at 2:20 PM, b. f. <bf1783@googlemail.com> wrote:
> On 05/08/10 13:36, Alan Cox wrote:
>> Doug Barton wrote:
>>> On 05/05/10 11:56, Alan Cox wrote:
>>>
>>>> I'm afraid that I would advise waiting a few days. =A0This round of
>>>> changes
>>>> are not yet complete.
>
> What performance differences, if any, can we expect on uniprocessors
> from the vm page lock-related changes? Kip's original post on -arch
> mentioned some performance improvements and no significant
> regressions, but on a dual 4-core machine.

I wouldn't actually worry about UP since the overhead can largely be
disabled by building without SMP. I think we need to be looking at how
a dual-core system performs, trading off any regressions there against
current processor trends of ever higher core and thread count.


Cheers,
Kip



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?m2j82c4140e1005081600u80823e01k756981ffd2c27931>