Re: invalid drv data in show attribute

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, Sep 05, 2022 at 03:52:01PM +0000, Czerwacki, Eial wrote:
> >On Mon, Sep 05, 2022 at 12:56:29PM +0000, Czerwacki, Eial wrote:
> >> >> I'd prefer not adding complex code to the utils which need to traverse the sysfs tree to collect the right data
> >> >
> >> >for device in $(ls /sys/bus/vsmp/device/); do
> >> now I understand!
> >> that is what I was missing.
> >> so inside /sys/bus/vsmp/device/ I can find the following tree?
> >> 1
> >> 2
> >> 3
> >> ...
> >> n
> >> ?
> >
> >Yes, or how ever you want to name these "boards".
> /sys/bus/vsmp/boards/ or /sys/bus/vsmp/device/board#?

You can not create /sys/bus/vsmp/boards/ the driver model will not let
you do that.

You will have the device/ directory created automatically.  It's how you
want to name the board itself is up to you, it just has to be unique on
the bus.

> >> so I can create entries in /sys/hypervisor/vsmp for the global info
> >
> >Yes.
> >
> >> and link /sys/bus/vsmp/device/ to /sys/hypervisor/vsmp/boards?
> >
> >No symlink needed at all, they are independent things, right?  What
> >would require such a symlink?
> >
> >thanks,
> >
> >greg k-h
> I'd rather have all the info under one place than in multiple places if possible.
> is seams more logical to me

That's not how sysfs works, sorry.  :)

Hypervisors are not normal, the fact that you are using that location in
the first place is very very odd, only 1 other in-kernel user is using
that.  So maybe you don't even need that at all?

thanks,

greg k-h




[Index of Archives]     [Linux Driver Development]     [Linux Driver Backports]     [DMA Engine]     [Linux GPIO]     [Linux SPI]     [Video for Linux]     [Linux USB Devel]     [Linux Coverity]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [Yosemite Backpacking]
  Powered by Linux