Padraig Brady wrote:
> Owen Taylor wrote:
> 
>> Padraig Brady <padraig@antefacto.com> writes:
>>
>>> I'm sure it will :-)
>>>
>>> However this it just masking the "problem"
>>
>>
>> Well, the question is, "what is the problem"?
>> Your problem is that a debug message is being output by the kernel and
>> filling your logs. If the debug message doesn't do anybody any good
>> (and it doesn't) then removing the debug message is a fine way of
>> solving the problem.
> 
> 
> True. But I thought there might be (future) side affects of
> cdrom_media_changed() always returning true. Why is it there at all?

I had a quick look at this and basically cdrom_media_changed()
is only valid when there is a disc in the drive. If the drive
is open or empty then it always returns true. There are a couple
of comments saying it's like this to catch buggy drives, but
hence the interface is broken IMHO and should be removed/fixed?

The attached test program illustrates the problem.

Padraig.