On Fri, Sep 07, 2012 at 10:55:00AM -0600, Stephen Warren wrote:

Yes, that does solve the problem (well, with s/late_init/late_initcall/).

As you imply, that change wouldn't help if cpu-tegra.c was built as a
module.

I doubt that.  Have you confirmed it with testing?  When you install
module cpu-tegra, the pcie initialization has been done, right?

So, is this something you can work around in your patch?

Before we head to a solution, we need to identify the cause.

One thing I note is there is no loops_per_jiffy updating in cpu-tegra
driver.  But if I understand correctly, the driver is running on
MP system.  It should update loops_per_jiffy when cpu frequency
changes, right?  I'm worrying about that the timeout in tegra pcie
driver just happens to work with a wrong loops_per_jiffy, but becomes
broken when loops_per_jiffy gets correct.

-- 
Regards,
Shawn