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]
Date:	Tue, 17 Feb 2015 16:52:45 -0800
From:	Bryan Wu <cooloney@...il.com>
To:	Ricardo Ribalda Delgado <ricardo.ribalda@...il.com>,
	Rob Herring <robh+dt@...nel.org>
Cc:	Richard Purdie <rpurdie@...ys.net>,
	Linux LED Subsystem <linux-leds@...r.kernel.org>,
	lkml <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] led/led-class: Handle LEDs with the same name

On Tue, Feb 17, 2015 at 4:32 PM, Ricardo Ribalda Delgado
<ricardo.ribalda@...il.com> wrote:
> Hello Bryan
>
> On Wed, Feb 18, 2015 at 12:59 AM, Bryan Wu <cooloney@...il.com> wrote:
>>
>> DT just describe the hardware, so if it's a different hardware, they
>> should have different name.
>> red0 for GPIO 0, red1 for GPIO 1 or choose other good name instead of 0 and 1.
>
> I think I have not managed to explain myself properly.
>
> We have a host computer. with 2 pcie slots. The host is described with
> a DT that looks like:
>
> &axi1 {
>
> pci0{
>   reg = < 0x20000000 0x10000000 >
> }
>
> pci1{
>   reg = < 0x30000000 0x10000000 >
> }
> }
>
> The user can connect anything to the pci slots. (pci0 and pci1)
>
>
> Lets say that we have a type of add-on card. Described by this DT
> overlay (card.dtb):
>

I think who write this card.dtb should understand this issue. And
choose the right name.

> &pci {
>
> gpio_0: gpio_0 {
What happen if you just use name 'gpio: gpio {' here.? Any conflicts
or kernel oops?

> #gpio-cells = <2>;
> compatible = "xlnx,xps-gpio-1.00.a";
> reg = < 0x30040000 010000 >;
> };
>
>
> /*Leds*/
> leds {
> reg = < 0x30040000 010000 >;
> compatible = "gpio-leds";
> red {
> gpios = <&gpio_0 0 0>;
> linux,default-trigger = "drop-qt5023_video0";
> };
> }
>
> }
>
> The user connects two of those cards to the system (at locations pci0 and pci1).
>
> Then we have TWO gpios chip. Each of them have a led named red. When
> the second gpio-led is probed we have an error. Everything else
> (address offset, phandle, device renaming) is handled properly already
> by the kernel.
>
> On this system I cannot control card.dtb, or which type of cards will
> the user connect to the system. The DT is generated in run-time based
> on the hardware connected to the pci slots.
>
So you're supposed to get 2 card.dtb files for 2 PCI cards, right?
They should be different and you need to choose different name for the
hardware.

> I humbly believe that the issue here is that the subsystem does not
> protect ourselves against name collisions, because a month ago a
> device tree was considered immutable and in full control of the system
> designer, unfortunately this is not the case anymore.
>

>From device tree point of view, I believe different device should got
different name although they can match to same compatible string. Let
me invite DT folks for help.

Thanks,
-Bryan
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists