Re: [SCSI] fix media change events for polled devices

[James Bottomley <James.Bottomley@HansenPartnership.com>]

On Fri, 2008-03-21 at 16:42 -0400, Jeff Garzik wrote:
> Kay Sievers wrote:
> > On Fri, 2008-03-21 at 13:12 -0400, Jeff Garzik wrote:
> >> Linux Kernel Mailing List wrote:
> >>>     [SCSI] fix media change events for polled devices
> >>>     
> >>>     Commit:
> >>>       a341cd0f (SCSI: add asynchronous event notification API)
> >>>     breaks:
> >>>       285e9670 (sr,sd: send media state change modification events)
> >>>     by introducing an event filter, which is removed here, to make
> >>>     events, we are depending on, happen again.
> >>
> >> By simply reading the code history, it is trivial to verify that this 
> >> description is false:
> >>
> >> Commit 285e9670 depends on a341cd0f, so by definition it is 285e9670 -- 
> >> or rather the incomplete update of your original patch that resulted in 
> >> 285e9670 -- that is broken.
> > 
> > It worked fine with Kristen's patches, and that's where it is coming
> > from from.
> 
> Neither her patches nor yours went upstream verbatim at version one. 
> You need to look at what happens upstream, not what happened in your 
> private testing six months ago.
> 
> 
> > You mean the read-only sysfs attribute? :)
> 
> I mean the attribute with both 'show' (read) and 'store' (write) 
> functions.  The perms do need to change, thanks for noticing that bug.

That's not a bug.

For starters, we have transport classes that provide generic store
methods but can't pass the information on to drivers.  For these, we set
the attribute to read only even if there is a store method.  Even if
that weren't the case, how do you know which of UGO wants the write
setting?  Finally, if you look at the sysfs code, the place where mode
settings of the sysfs files are handled doesn't see the show/store
methods at all (these are handled at a different layering within the
generic driver model).

James