linux-kernel - Re: [RFC PATCH 0/6] Support ROHM BD96801 scalable PMIC

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <0674ca23-2cf2-48a6-84d3-e0936d50dd8c@gmail.com>
Date: Thu, 9 May 2024 10:03:43 +0300
From: Matti Vaittinen <mazziesaccount@...il.com>
To: Mark Brown <broonie@...nel.org>
Cc: Matti Vaittinen <matti.vaittinen@...rohmeurope.com>,
 Lee Jones <lee@...nel.org>, Rob Herring <robh@...nel.org>,
 Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
 Conor Dooley <conor+dt@...nel.org>, Liam Girdwood <lgirdwood@...il.com>,
 Wim Van Sebroeck <wim@...ux-watchdog.org>, Guenter Roeck
 <linux@...ck-us.net>, devicetree@...r.kernel.org,
 linux-kernel@...r.kernel.org, linux-watchdog@...r.kernel.org,
 Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [RFC PATCH 0/6] Support ROHM BD96801 scalable PMIC

On 5/9/24 08:08, Mark Brown wrote:
> On Mon, Apr 22, 2024 at 01:52:27PM +0300, Matti Vaittinen wrote:
>> On 4/5/24 12:19, Matti Vaittinen wrote:
>>> On 4/4/24 16:15, Matti Vaittinen wrote:
> 
>>>>> I would expect each parent interrupt to show up as a separate remap_irq.
> 
>>>>> So if we arrange to supply a name when we register multiple domains
>>>>> things should work fine?
> 
>>> After my latest findings, yes, I think so. How to do this correctly is
>>> beyond me though. The __irq_domain_create() seems to me that the name is
>>> meant to be the dt-node name when the controller is backed by a real
>>> dt-node. Naming of the irq_domain_alloc_named_fwnode() sounds to me like
> 
> ...
> 
>> If we wanted to support multiple HWIRQs / regmap-IRQ controller, it would
>> require us to duplicate almost everything in the struct regmap_irq_chip for
>> every new parent IRQ. The status/mask register information, IRQ type, etc.
>> Naturally, it would require also duplicating lot of the data contained in
>> the struct regmap_irq_chip_data. I am not sure if this could be done so the
>> change is not reflected in the existing IRQ data initialization macros etc.
>> Furthermore, some API changes would be required like changes to
>> regmap_irq_get_domain().
> 
> I don't understand what the difficulty is here - we're creating multiple
> interrupt controllers so I'd expect to have to have full definitions of
> each, and since everything is referenced by name from the root
> regmap_irq_chip which gets registered it's just a case of supplying
> different names and all the helpers should be fine?
> 
>> Thus, forcing the regmap-IRQ to support multiple parents instead of having
>> own regmap-IRQ instance / parent IRQ feels like fitting square item to a
>> round hole. I am sure fixing all the bugs I caused would give donate a lot
>> of EXP-points though :rolleyes:
> 
> Right, my suggestion is to register multiple regmap_irq instrances - one
> per parent - and supply a name that allows all the display/debugfs stuff
> that currently uses the dev_name() to deduplicate.  You'd end up
> sticking -primary, -secondary or whatever name was supplied onto the
> names we currently use.
> 
>> Another option I see, is trying to think if irq-domain name could be
>> changed. (This is what the RFC v3 does, [ab]using the
>> irq_domain_update_bus_token()). I was a bit put off by the idea of
>> 'instantiating' multiple domains (or regmap-IRQ controllers) from a single
>> node, but more I think of this, more I lean towards it. Besides, this is not
> 
> Yes, register mutliple controllers with different names.

Thanks for the guidance Mark. The controller name is not a problem. 
Problem is that I don't see a (proper) way to supply a name for the IRQ 
domain which gets registered by regmap-IRQ. IRQ domain code picks the 
name for the domain by the device-tree node. Both of our IRQ controllers 
would be instantiated from same node => the IRQ domain will get same 
name => debugfs will conflict.

My "solution" was simply dropping the ERRB IRQ from the driver (for now 
at least). I did send that as a series without 'RFC' - but made a 
mistake and restarted the versioning from v1. I am currently working 
with 2 other PMICs, one of them does also provide similar setup of two 
IRQ lines. Thus, I think being able to provide a name (suffix?) for IRQ 
domain when registering it instead of just using the name of the DT node 
is something I should look into. It's just nice to know someone else 
thinks it is valid approach.

Thanks for the input!

Yours,
	-- Matti



-- 
Matti Vaittinen
Linux kernel developer at ROHM Semiconductors
Oulu Finland

~~ When things go utterly wrong vim users can always type :help! ~~