Fusion MPT regression, fixed or not?

Fusion MPT regression, fixed or not?

[Matt Sealey]

Hello World,

I re-stumbled on this installing Gentoo on VMware today;

http://lkml.org/lkml/2007/3/23/312

It doesn't seem to be fixed in 2.6.22-rc7, but I might be mistaken
somehow. Has anyone done any work on this or looked at the problem
in more detail? The trail seems to die at that thread.

It is affecting VMware for me mainly but a lot of heavy server
hardware seems to use the chip, in fact some I am looking at right
now to recommend purchasing on.. I'd really like to use the latest
kernel because virtualisation is the key (dynamic ticks and
tickless hz and the latest virtualisation options would help eke out
more value ;)

-- 
Matt Sealey <matt@genesi-usa.com>
Genesi, Manager, Developer Relations

Re: Fusion MPT regression, fixed or not?

[Petr Vandrovec]

Matt Sealey wrote:
> Hello World,
> 
> I re-stumbled on this installing Gentoo on VMware today;
> 
> http://lkml.org/lkml/2007/3/23/312
> 
> It doesn't seem to be fixed in 2.6.22-rc7, but I might be mistaken
> somehow. Has anyone done any work on this or looked at the problem
> in more detail? The trail seems to die at that thread.

Not that it is relevant to LKML, but you need either upgrade to VMware 
Workstation 6.0, or you need patch later in that thread: 
http://lkml.org/lkml/2007/3/23/345

> It is affecting VMware for me mainly but a lot of heavy server
> hardware seems to use the chip, in fact some I am looking at right
> now to recommend purchasing on.. I'd really like to use the latest
> kernel because virtualisation is the key (dynamic ticks and
> tickless hz and the latest virtualisation options would help eke out
> more value ;)

It affects only VMware's emulation, not real hardware (at least unless 
you hit one with broken firmware).
							Petr