Is changing serial from /dev/ttySx to /dev/ttyOx a good idea?

Hi All

With the change to the new omap-serial stuff, the device names have changed from /dev/ttySn to /dev/ttyOn.

Using the "fit/form/function" argument, this would seem to be a bad idea.

The kernel does not exist in a vacuum. It is surrounded by u-boot scripts, command line options, init scripts etc etc. Flipping back and forth between 2.6.36 and 2.6.37 kernels for testing purposes needs these to be changed. Annoying and anti-productive.

Isn't it just better to just replace the ttyOn name with ttyS. That would make a far smoother transition.

Thoughts?

Hi,

2010/12/20 Charles Manning <[email protected]>:

Quoted 12 lines

Hi All

With the change to the new omap-serial stuff, the device names have changed from /dev/ttySn to /dev/ttyOn.

Using the "fit/form/function" argument, this would seem to be a bad idea.

The kernel does not exist in a vacuum. It is surrounded by u-boot scripts, command line options, init scripts etc etc. Flipping back and forth between 2.6.36 and 2.6.37 kernels for testing purposes needs these to be changed. Annoying and anti-productive.

Quoted 4 lines

Isn't it just better to just replace the ttyOn name with ttyS. That would make a far smoother transition.

Thoughts?

In fact, the transition is not friendly indeed for a user. I don't know why the transition is introduced, for what purpose? Who can give a explanation?

Hi,

Quoted 3 lines · Ming Lei

In fact, the transition is not friendly indeed for a user. I don't know why the transition is introduced, for what purpose? Who can give a explanation?

Patch itself has one explanation, kernel commit is b612633b5928077441b979471869753bfa93d41a

http://www.spinics.net/lists/linux-serial/msg02788.html

Hi,

2010/12/20 Abraham Arce <[email protected]>:

Quoted 7 lines

Hi,

Quoted 3 lines · Ming Lei

In fact, the transition is not friendly indeed for a user. I don't know why the transition is introduced, for what purpose? Who can give a explanation?

Quoted 4 lines

Patch itself has one explanation, kernel commit is b612633b5928077441b979471869753bfa93d41a

http://www.spinics.net/lists/linux-serial/msg02788.html

The commit does not explain the cause of renaming ttyS to ttyO, does it?

Hi,

2010/12/20 Charles Manning<[email protected]>:

Quoted 12 lines

Hi All

With the change to the new omap-serial stuff, the device names have changed from /dev/ttySn to /dev/ttyOn.

Using the "fit/form/function" argument, this would seem to be a bad idea.

The kernel does not exist in a vacuum. It is surrounded by u-boot scripts, command line options, init scripts etc etc. Flipping back and forth between 2.6.36 and 2.6.37 kernels for testing purposes needs these to be changed. Annoying and anti-productive.

Quoted 4 lines

Isn't it just better to just replace the ttyOn name with ttyS. That would make a far smoother transition.

Thoughts?

In fact, the transition is not friendly indeed for a user. I don't know why the transition is introduced, for what purpose? Who can give a explanation?

thanks,