Am 12.03.2018 um 08:53 schrieb Mathias Kresin:
10.03.2018 08:56, Sebastian Gottschall:

taken a look at the specific code, and from my point of view the code
that sets up the LED (including callback) is so trivial that it's simply
not worth dealing with adding the leds-gpio driver to the mix.
It adds extra complexity and an extra dependency for no reason at all.
There's no extra functionality to be gained by using it.
Stupid question: If the LED driver isn't using the GPIO subsystem
(when enabled), what happens if the user uses the GPIO subsystem to
fiddle with the pin the LED is connected to?

in our case here it can coexist and will not have a negative effect since the led will still run. (except if you reconfigure the gpio to input)
for sure it makes no sense.

Well, it will have negative effects as I noticed in OpenWrt. If the same GPIO is controlled by the internal LED function it can't be really used via the GPIO controller as there will be "bogus" changes of the GPIO pin state.
yes, it will have bogus changes, but it will not raise any other issues than the bogus changes your forcefully triggered. why disallowing something, if there is no reason for.
but i give you a example for a use.
if the wifi interface is down or not used, you can still take control about the led. so set the led to any state you want, no matter if the trigger is still installed

but however i can block the gpio for beeing reused if the led driver took it. this has been made in ath9k for instance.

Most likely I'm not aware of the upstream code you are talking about. But OpenWrt has a custom patch which registers the ath9k pins as GPIO controller.
thats the code i'm talking about, but as far as i know this code is upstream.

As soon as there are pins, manufactures will use them and the assumption about a default led connected to a specific pin will fail.
nope. manufactures are mainly using the qca propertiery driver and this qca driver uses fixed hardcoded led pins. 1 for qca988x and 17 for everything else. the led pins i assigned are following the qca specification for this chipsets and not to forget
i tested it on many devices

I've seen (home)routers routers using the "default" ath9k LED pin as button or for LEDs with a different purpose than "wireless". I never was able to find any kind of information which explains why exactly this specific pin is used for the ath9k-led. But that's another story.
i know. but thats not the case for ath10k so far and i'm following exact the same way qca is doing by themself. i just know 2 cases. vendors to use the led pin of the chipset or they dont and use standard system leds with any assignment
and in all cases with no exception the default led pin was used.

My solution for OpenWrt was a patch which prevents the creation of the ath9k-led if the "default" pin is used as GPIO.
i have seen that code, but decided no to follow it since any developer is responsible for configuring whatever he wants. if it doesnt lead to a crash i see no reason for preventing it.
but thats my philosophy.

If mach files are used, the ath9k led isn't created (ath9k led pin is set to -1) in case the platform data have a button or led using the same pin. If the device tree is used, the ath9k led isn't created (ath9k led pin is set to -1) if the devicetree node for the ath9k device has the gpio-controller property. I'm not really proud of the code and it can be most likely done better, but it fixes the issue outlined above.
sounds very inflexible or better "intransparent". from the black box view its hard to understand how its handled without reading the code. its has no clean logic. and your issue above is no issue. if a dev wants to make stupid things, let him do. he learns from his misstakes and he does not harm the hardware nor the stability. there is no reason for extra hyper security barriers to guide someone how todo it. but here are samples to reuse a gpio from 2 points as i explained at the beginning. i could imagine todo a manual error blinking or something if a interface is down. not sure, but other people might be more creative here.

I've done it this way since the use of the GPIO controller should always override any kind of default LEDs.

In the end ath[0|10]k-led is only required for hardware configurations where the wireless isn't fixed like with miniPCIe, USB ... wireless cards. If the configuration is fixed - like it is for most of the (home)routers - the GPIO controller can be registered via the devicetree and gpio-leds can be used. Something similar can be most likely done via mach files (I barely touch mach files).
ath79 uses mach files and no dts yet. qca ipq platforms are using dts.
and i have a mikrotik minipcie cards with wireless leds :-)

all 3 ways are valid.
and i have to note that i wrote that patch for embedded devices like openwrt does. this code is a requirement to get the leds to work on several devices supported by openwrt.

I'm aware of the issue since the first version of your patch. But since I'm very interested in getting the ath10k pins exposed as gpio controller, I planned to add a workaround similar to what we have for ath9k to OpenWrt.
feel free to make any addition patch. but consider that the driver already contains a gpio controler. but in that case just avoid using of the led trigger. but for universal use the gpio controler isnt enough unfortunatly. (ath79 for instance) i mean creating a additional platformdata for ath10k to initialize the pin in mach files is really not usefull if it can be done from userspace in sysfs


Mit freundlichen Grüssen / Regards

Sebastian Gottschall / CTO

NewMedia-NET GmbH - DD-WRT
Firmensitz:  Stubenwaldallee 21a, 64625 Bensheim
Registergericht: Amtsgericht Darmstadt, HRB 25473
Geschäftsführer: Peter Steinhäuser, Christian Scheele
Tel.: +496251-582650 / Fax: +496251-5826565

Reply via email to