Re: [PATCH v2 00/16] Common Dell SMBIOS API

[Darren Hart <dvhart@infradead.org>]

On Mon, Feb 08, 2016 at 11:29:20AM -0800, Greg Kroah-Hartman wrote:
> On Mon, Feb 08, 2016 at 11:20:14AM -0800, Darren Hart wrote:
> > On Fri, Jan 22, 2016 at 03:48:51PM +0100, Pali Rohár wrote:
> > > On Friday 22 January 2016 15:27:12 Michał Kępień wrote:
> > > > Note:
> > > > 
> > > >     In this series (both v1 and v2) I tried to stick to the overall
> > > >     concept used in dell-laptop, but in the v1 thread me and Pali also
> > > >     briefly discussed his alternative ideas [1][2] as to what this API
> > > >     could look like, so feel free to suggest a different approach.
> > > > 
> > > > [1] http://www.spinics.net/lists/platform-driver-x86/msg08260.html
> > > > [2] http://www.spinics.net/lists/platform-driver-x86/msg08268.html
> > > 
> > > I would like to hear opinion about dell-smbios API also from other
> > > people. Darren, can you look and comment it?
> > 
> > This is an excellently prepared series, nice work Michał.
> > 
> > Most of my concerns were addressed by later patches in the series. I have pushed
> > a version of this 1/7 fixed per lkp (linux/io.h) and 7/16 with a corrected body
> > as I sent in reply to that patch. This is on my tree as the dell-smbios branch.
> > 
> > My only major concern is module load order dependencies. Inter-module
> > dependencies are frowned upon with good reason, the kernel load ordering is
> > non-deterministic and it's possible, for example, for dell-laptop to fail to
> > find the symbols exported by dell-smbios under certain conditions.
> > 
> > I have worked around this in the past with things like the following:
> > 
> > #ifdef MODULE
> > #ifdef CONFIG_FOO_MODULE
> > 	if (request_module("foo"))
> > 		return -ENODEV;
> > #endif
> > #endif
> > 
> > Something like the above may be necessary for dell-smbios in dell-laptop,
> > dell-wmi, and dell-leds now that they depend on the dell-smbios exported
> > functions.
> > 
> > Cc Greg in case there is a better way to handle this that I'm not aware of.
> 
> No need to request_module anything, the symbol resolution should pull
> the dependant module in automagically when you do a 'modprobe',
> otherwise the symbols would never be found.

OK, Thank you Greg. Based on this, I have no concerns that haven't been addressed in the
pdx86/dell-smbios branch.

Michał, did you have any changes you wanted to make?

Pali, are you happy enough with this to add your reviewed-by?

I'm satisfied with the series and can push to next tomorrow if we're all in
agreement.

-- 
Darren Hart
Intel Open Source Technology Center