Re: linux-next: build failure after merge of the scsi tree

[Alexey Dobriyan <adobriyan@gmail.com>]

On Tue, Mar 27, 2012 at 03:22:16PM -0700, Greg KH wrote:
> On Wed, Mar 28, 2012 at 01:17:21AM +0300, Alexey Dobriyan wrote:
> > On Mon, Mar 26, 2012 at 07:07:21AM -0700, Greg KH wrote:
> > > On Mon, Mar 26, 2012 at 08:57:21AM +0100, James Bottomley wrote:
> > > > On Mon, 2012-03-26 at 12:17 +1100, Stephen Rothwell wrote:
> > > > > Hi James,
> > > > > 
> > > > > After merging the scsi tree, today's linux-next build (x86_64 allmodconfig)
> > > > > failed like this:
> > > > > 
> > > > > drivers/staging/keucr/scsiglue.c:349:2: error: unknown field 'proc_info' specified in initializer
> > > > > drivers/staging/rts_pstor/rtsx.c:258:2: error: unknown field 'proc_info' specified in initializer
> > > > > drivers/staging/rts5139/rts51x_scsi.c:2190:2: error: unknown field 'proc_info' specified in initializer
> > > > > 
> > > > > Caused by commit 104c4fe25dc9 ("[SCSI] remove scsi_host_template::proc_info").
> > > > > 
> > > > > Since this is a staging driver, I applied these following patches:
> > > > 
> > > > Yes, that looks about right, thanks.  We haven't seen anything about
> > > > these drivers on the SCSI list, so I've no idea where they are in
> > > > development.
> > > 
> > > Why is new patches going into your tree right now, during the 3.4 merge
> > > window?  API changes should have happened weeks ago, to let others fix
> > > up things like this.
> > > 
> > > As for the "where they are in development", they vary, but, a simple
> > > grep should have shown you that these in-kernel drivers should also be
> > > fixed up, or at the least, give me the heads up to let me do it for you.
> > > 
> > > Care to send me the patch that causes this problem so I can create a fix
> > > for this?
> > 
> > See commits 422f07001d6638fdde28f1909cc9162bc7f571d3..104c4fe25dc9bde823ba4591e910a77071b98ab5
> > Especially the first one.
> > 
> > Probably the best course of action is to remove proc_info code from
> > these drivers because the interface was deprecated for a long time.
> 
> If the interface is depreciated, yes, I can remove it now, just let me
> know and I will do so.
> 
> > The amount of code once removed from staging prevented me from doing
> > any work on them.
> > 
> > Looking at staging ->read_proc users this is going to be a problem for
> > its removal. :-(
> 
> Why?

Because if staging does count, I can't remove the interface
without breaking allmodconfig and it would take forever to convert
staging stuff. I don't have energy to do it anymore.
Mainline still have several _hard_ ->read_proc conversions.
I've tried several times and failed.

If staging doesn't count, I will break allmodconfig and all those nasty
emails will show up anyway implying that staging does count.