[RFC] Adding MTD to device tree

[Sergei Shtylyov]

Hello.

Milton Miller wrote:

>>>>> +   h) MTD nodes
>>>>> +
>>>>> +   Memory Technology Devices are flash, ROM, and similar chips, 
>>>>> often used
>>>>> +   for solid state file systems on embedded devices.
>>>>> +
>>>>> +   Required properties:
>>>>> +
>>>>> +    - device_type : has to be "mtd"
>>>>> +    - compatible : Should be the name of the MTD driver. 
>>>>> Currently, this is
>>>>> +      most likely to be "physmap".
>>>>> +    - reg : Offset and length of the register set for the device.

>>>> I would prefer to call them something different in the device tree.
>>>> The name 'mtd' is very specific to Linux, but the device tree
>>>> is a more generic concept.

>>    "Memory type devices" are specific to Linux? Doubt it. :-)
>>    In fact, device type "flash" sounds too restrictive.

> First of all, I was only quoting that.

    I'm not ascribing them to you. :-)

> Second, no, memory type devices are not specific to Linux.  But the term 
> MTD
> probably is[1].

    I doubt Google based conslusions qualify here.

>  If you don't want to call it flash, then call it rom.  The
> fact that its programable rom is a detail.

    Important detail, I'd say.

>>>> I understand that the booting-without-of.txt file is by definition
>>>> Linux specific as well, but we should be prepared for making parts
>>>> of it a OS independent binding at the point where we put the same
>>>> device nodes into actual OF implementations that able to boot
>>>> different operating systems.

>>>> I would prefer a naming that has

>>>>   Required properties:
>>>>    - device_type : one of "nand-flash", "nor-flash", or "rom".
>>>>    - model : an identifier for the actual controller chip used.
>>>>    - compatible : Should be the name of the MTD driver. For
>>>>      type "rom", this is most likely "physmap".

>>> I'm with your suggestion for device_type and model, but not compatable.
>>>  "physmap"?  What kind of device is that?  A

>>    Directly mapped NOR flash or ROM I think.

> I don't know how you got there, without your linux implementation 
> knowledge.

    Indeed, this was compeletely Linux specific name.

>>> command set name, maybe with a width, would be

>>    That'd be pretty useless if you don't let Linux know which MTD 
>> *map* driver
>> to use. And I have specified the "bank-width" prop.

> The fact that its directly mapped should be evident from its location in
> the device tree and its properties such as reg.

    Nohow. The flash could be be indirectly accessible and still have the reg 
prop in the device node. It also may need nasty byte swapping tricks (in which 
case it's not physmap compatible).

> I think you are getting caught up the layering of the linux mtd 
> subsystem, and not
> thinking about describing the hardware.

    If we end up with the node spec that is correct from the point of 
description the hardware but absolutely useless when it comes to selectiong 
the proper driver, what use it would be?

>>> appropriate.   Physmap is the name of another linux driver.

>>    And the role of the "compatible" prop is exactly to help OS in 
>> selecting the driver.

> Yes, help select.  But not say "here is the name of the driver to use." 
>  Its to specify the programing interface.

    Well, you may be wrong here. See below. :-)

> Putting "cfi-xxx", "cfi", "rom" would show that it should be probed by cfi
> instead of by jedec for example.

    Yes. But that's just no use with the current platform_device base 
implementation of physmap. So we'd either need to extend this implementation 
or make physmap parse all that from the device tree by itself.

> Are we still talking about doing a driver that calls physmap with the 
> appropriate
> address gathered from of_device?

    Well, this would not exactly be a driver then.

> It would be the one to select the nodes compatable
> nodes, finds the relevant parms, and calls the mtd layer.

    In fact, I indended to probe the MTD/PPC people which way they like best: 
sticking with platform_device still, or making physmap of_device aware by 
itself (although we'd still need to register the flash device somewhere in 
arch/powerpc/, since there's not code anywhere that register *all* devices 
found in the device tree)...

>>>  Even rom would be better than physmap.

>>    Doubt it since the ROM is the only one thing (and even the least 
>> probable) that we're
>> going to support.

> But it would make sense to be a later compatible entry ... we can treat 
> the flash as a
> true read-only memory and it will work, we just will not have the 
> feature of writing
> to it.

    Indeed, if "compatible" incorporated the chip interfaces (probe types for 
MTD/physmap drivers), that'd have made sense. Alas, with platform_device based 
approach, we have no way of passing this info to MTD subsys currently.

> As segher pointed out, compatible is a (specific to generic) list of 
> hints to the
> programming interface.

    If you look into the OF specs, "compatible" and "name" are treated as the 
*driver* names here. There are whole passages there warning about not 
selecting too generic driver names like "network".

> I'll admit I'm not that familiar with the mtd subsystem.  But I did go 
> and browse the
> redboot, ebony, and ocetea map drivers, along with physmap.  With the 
> addresses filled
> out, they would appear to collapse into one driver, with multiple device 
> nodes for each
> one.  Except for the partitioning information.

    Then they should have indeed collapsed into it. The thing is physmap 
driver might be actually newer than those ones, so they're kept for historical 
reasons (and for the partition info which actually should move into the 
platform setup code).

> Ok that brings up partitions: if you need partitions, and don't have a 
> table system
> like redboot, then we need some way to convey that in the device tree.  My
> first instinct is that would be via device nodes under the map device in 
> the tree.

    Erm, what map device?

> The nodes would get a reg with relative address and size, and the name 
> would be
> the name of the partition.

    MTD partitions are *not* devices, so I guess the first instinct is wrong 
(I had to struggle with it too :-).

>   We could also do it in properties on the 
> mtd node, but
> then would not not have a good place to put the name.   (Well, we could 
> do a list of
> strings and a parallel list of partition sizes.   I would prefer we not 
> encode both
> in one property).

    'dtc' wouldn't able to chew it anyway although the OF spec. promised 
multi-type properties.
    Sorry, have you bothered looking into my node spec? :-)

>  Yes, we can rely on cmdline partitioning ... but that 
> means a long
> argument in your device tree you don't want edited.

    I finally decided that if we have the ability to store it in the device 
tree, this is better than have user to type and store the bootargs (or worse 
yet, rebuild U-Boot changing the default value of this env. var.).

> milton

> [1] Google seems to confirm this: the first several hits are for a power 
> equipment
> manufacturer, and then linux-mtd site by itself, without other os's 
> "mtd" projects
> in the first 4 pages of hits.

    Hm, MTD is a short for "memory type device", what Linux has to do with it 
at all? However, let the MTD experts correct me...

WBR, Sergei