can we drop the comedi serial2002 drivers from staging?

can we drop the comedi serial2002 drivers from staging?

[Christoph Hellwig]

As far as I can tell there has been no targeted work on the driver
since merging it at all, and since 2014 even the comedi-wide
cleanups stopped, laving just drive by tree wide changes since.

At the same time the driver badly abuses the tty layer and is one
of only two major abusers of the poll code for in-kernel waits.

Re: can we drop the comedi serial2002 drivers from staging?

[Greg Kroah-Hartman]

On Sat, Jun 16, 2018 at 11:06:45AM +0200, Christoph Hellwig wrote:
> As far as I can tell there has been no targeted work on the driver
> since merging it at all, and since 2014 even the comedi-wide
> cleanups stopped, laving just drive by tree wide changes since.
> 
> At the same time the driver badly abuses the tty layer and is one
> of only two major abusers of the poll code for in-kernel waits.

Yeah, it's an ugly driver.  It should be rewritten to use the new serdev
api instead of trying to open a tty device node from within the kernel.

I don't object to dropping it.  Ian and Hartley, do you know of any
users of this driver anymore?

thanks,

greg k-h

Re: can we drop the comedi serial2002 drivers from staging?

[Ian Abbott]

On 16/06/18 10:37, Greg Kroah-Hartman wrote:
> On Sat, Jun 16, 2018 at 11:06:45AM +0200, Christoph Hellwig wrote:
>> As far as I can tell there has been no targeted work on the driver
>> since merging it at all, and since 2014 even the comedi-wide
>> cleanups stopped, laving just drive by tree wide changes since.
>>
>> At the same time the driver badly abuses the tty layer and is one
>> of only two major abusers of the poll code for in-kernel waits.
> 
> Yeah, it's an ugly driver.  It should be rewritten to use the new serdev
> api instead of trying to open a tty device node from within the kernel.

I was thinking about that, but it looks tricky to use for a "normal" 
PC-based distro due to typical lack of device-tree support.

Another possibility is to support comedi drivers in userspace via CUSE, 
which would require a comedi-cuse interface library that user-space 
comedi drivers could link to similar to how kernel-mode comedi drivers 
link to the comedi core.

> I don't object to dropping it.  Ian and Hartley, do you know of any
> users of this driver anymore?

The only place I know it was in use was in Anders Blomdell's lab at Lund 
University.  I've added him to the Cc: list.

-- 
-=( Ian Abbott <abbotti@mev.co.uk> || Web: www.mev.co.uk )=-
-=( MEV Ltd. is a company registered in England & Wales. )=-
-=( Registered number: 02862268.  Registered address:    )=-
-=( 15 West Park Road, Bramhall, STOCKPORT, SK7 3JZ, UK. )=-

Re: can we drop the comedi serial2002 drivers from staging?

[Greg Kroah-Hartman]

On Sat, Jun 16, 2018 at 11:06:45AM +0200, Christoph Hellwig wrote:
> As far as I can tell there has been no targeted work on the driver
> since merging it at all, and since 2014 even the comedi-wide
> cleanups stopped, laving just drive by tree wide changes since.
> 
> At the same time the driver badly abuses the tty layer and is one
> of only two major abusers of the poll code for in-kernel waits.

I've dropped it now, if people really want it, we can revert it and
convert to use serdev.

thanks,

greg k-h