pcmcia_ioctl.c compile error

pcmcia_ioctl.c compile error

[Adrian Bunk]

Commit 9c37066d888bf6e1b96ad12304971b3ddeabbad0 (proc: remove proc_bus) 
came through -mm that contains a patch implementing the overdue removal 
of the PCMCIA control ioctl that did not yet make it into Linus' tree:

<---  snip  -->

...
  CC      drivers/pcmcia/pcmcia_ioctl.o
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c: In function ‘pcmcia_setup_ioctl’:
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c:790: error: ‘proc_bus’ undeclared (first use in this function)
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c:790: error: (Each undeclared identifier is reported only once
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c:790: error: for each function it appears in.)
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c: In function ‘pcmcia_cleanup_ioctl’:
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c:801: error: ‘proc_bus’ undeclared (first use in this function)
make[3]: *** [drivers/pcmcia/pcmcia_ioctl.o] Error 1

<--  snip  -->

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: pcmcia_ioctl.c compile error

[Andrew Morton]

On Tue, 29 Apr 2008 21:27:10 +0300
Adrian Bunk <bunk@kernel.org> wrote:

> Commit 9c37066d888bf6e1b96ad12304971b3ddeabbad0 (proc: remove proc_bus) 
> came through -mm that contains a patch implementing the overdue removal 
> of the PCMCIA control ioctl that did not yet make it into Linus' tree:
> 
> <---  snip  -->
> 
> ...
>   CC      drivers/pcmcia/pcmcia_ioctl.o
> /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c: In function ___pcmcia_setup_ioctl___:
> /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c:790: error: ___proc_bus___ undeclared (first use in this function)
> /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c:790: error: (Each undeclared identifier is reported only once
> /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c:790: error: for each function it appears in.)
> /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c: In function ___pcmcia_cleanup_ioctl___:
> /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/pcmcia/pcmcia_ioctl.c:801: error: ___proc_bus___ undeclared (first use in this function)
> make[3]: *** [drivers/pcmcia/pcmcia_ioctl.o] Error 1
> 
> <--  snip  -->
> 

Yes, I fixed it but then fumbled the fix :(

I sent Linus the below an hour or so ago.

--- a/drivers/pcmcia/pcmcia_ioctl.c~drivers-pcmcia-pcmcia_ioctlc-fix-build
+++ a/drivers/pcmcia/pcmcia_ioctl.c
@@ -787,7 +787,7 @@ void __init pcmcia_setup_ioctl(void) {
 		major_dev = i;
 
 #ifdef CONFIG_PROC_FS
-	proc_pccard = proc_mkdir("pccard", proc_bus);
+	proc_pccard = proc_mkdir("bus/pccard", NULL);
 	if (proc_pccard)
 		create_proc_read_entry("drivers",0,proc_pccard,proc_read_drivers,NULL);
 #endif
@@ -798,7 +798,7 @@ void __exit pcmcia_cleanup_ioctl(void) {
 #ifdef CONFIG_PROC_FS
 	if (proc_pccard) {
 		remove_proc_entry("drivers", proc_pccard);
-		remove_proc_entry("pccard", proc_bus);
+		remove_proc_entry("bus/pccard", NULL);
 	}
 #endif
 	if (major_dev != -1)
_

Re: pcmcia_ioctl.c compile error

[Russell King]

On Tue, Apr 29, 2008 at 09:27:10PM +0300, Adrian Bunk wrote:
> Commit 9c37066d888bf6e1b96ad12304971b3ddeabbad0 (proc: remove proc_bus) 
> came through -mm that contains a patch implementing the overdue removal 
> of the PCMCIA control ioctl that did not yet make it into Linus' tree:

Okay, this means at least one of my platforms is now dead as far as
building new kernels for it, and nothing I can do about that.  Ho hum.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:

Re: pcmcia_ioctl.c compile error

[Randy Dunlap]

On Tue, 29 Apr 2008 21:17:04 +0100 Russell King wrote:

> On Tue, Apr 29, 2008 at 09:27:10PM +0300, Adrian Bunk wrote:
> > Commit 9c37066d888bf6e1b96ad12304971b3ddeabbad0 (proc: remove proc_bus) 
> > came through -mm that contains a patch implementing the overdue removal 
> > of the PCMCIA control ioctl that did not yet make it into Linus' tree:
> 
> Okay, this means at least one of my platforms is now dead as far as
> building new kernels for it, and nothing I can do about that.  Ho hum.

I don't get it.  Several of us had building problems for a few
days with some pasic driver.  We posted some bug reports and
some patches, but we put up with it for a few days...

---
~Randy

Re: pcmcia_ioctl.c compile error

[Harvey Harrison]

On Tue, 2008-04-29 at 13:59 -0700, Randy Dunlap wrote:
> On Tue, 29 Apr 2008 21:17:04 +0100 Russell King wrote:
> 
> > On Tue, Apr 29, 2008 at 09:27:10PM +0300, Adrian Bunk wrote:
> > > Commit 9c37066d888bf6e1b96ad12304971b3ddeabbad0 (proc: remove proc_bus) 
> > > came through -mm that contains a patch implementing the overdue removal 
> > > of the PCMCIA control ioctl that did not yet make it into Linus' tree:
> > 
> > Okay, this means at least one of my platforms is now dead as far as
> > building new kernels for it, and nothing I can do about that.  Ho hum.
> 
> I don't get it.  Several of us had building problems for a few
> days with some pasic driver.  We posted some bug reports and
> some patches, but we put up with it for a few days...
> 

I believe Russell is referring to the removal of the ioctl, not the
compile breakage.

Harvey

Re: pcmcia_ioctl.c compile error

[Randy Dunlap]

On Tue, 29 Apr 2008 14:03:07 -0700 Harvey Harrison wrote:

> On Tue, 2008-04-29 at 13:59 -0700, Randy Dunlap wrote:
> > On Tue, 29 Apr 2008 21:17:04 +0100 Russell King wrote:
> > 
> > > On Tue, Apr 29, 2008 at 09:27:10PM +0300, Adrian Bunk wrote:
> > > > Commit 9c37066d888bf6e1b96ad12304971b3ddeabbad0 (proc: remove proc_bus) 
> > > > came through -mm that contains a patch implementing the overdue removal 
> > > > of the PCMCIA control ioctl that did not yet make it into Linus' tree:
> > > 
> > > Okay, this means at least one of my platforms is now dead as far as
> > > building new kernels for it, and nothing I can do about that.  Ho hum.
> > 
> > I don't get it.  Several of us had building problems for a few
> > days with some pasic driver.  We posted some bug reports and
> > some patches, but we put up with it for a few days...
> > 
> 
> I believe Russell is referring to the removal of the ioctl, not the
> compile breakage.

Oh, thanks.  I couldn't tell from the text.

---
~Randy

Re: pcmcia_ioctl.c compile error

[Andrew Morton]

On Tue, 29 Apr 2008 14:03:07 -0700
Harvey Harrison <harvey.harrison@gmail.com> wrote:

> On Tue, 2008-04-29 at 13:59 -0700, Randy Dunlap wrote:
> > On Tue, 29 Apr 2008 21:17:04 +0100 Russell King wrote:
> > 
> > > On Tue, Apr 29, 2008 at 09:27:10PM +0300, Adrian Bunk wrote:
> > > > Commit 9c37066d888bf6e1b96ad12304971b3ddeabbad0 (proc: remove proc_bus) 
> > > > came through -mm that contains a patch implementing the overdue removal 
> > > > of the PCMCIA control ioctl that did not yet make it into Linus' tree:
> > > 
> > > Okay, this means at least one of my platforms is now dead as far as
> > > building new kernels for it, and nothing I can do about that.  Ho hum.
> > 
> > I don't get it.  Several of us had building problems for a few
> > days with some pasic driver.  We posted some bug reports and
> > some patches, but we put up with it for a few days...
> > 
> 
> I believe Russell is referring to the removal of the ioctl, not the
> compile breakage.
> 

That would be interesting information (although I have a vague feeling that
it has been discussed before).

I've been sitting on pcmcia-delete-obsolete-pcmcia_ioctl-feature.patch for
over a year, mainly waiting for Dominik to reappear and decide what to do
about it.

I could just drop it, shrug.

Re: pcmcia_ioctl.c compile error

[Russell King]

On Tue, Apr 29, 2008 at 01:59:02PM -0700, Randy Dunlap wrote:
> On Tue, 29 Apr 2008 21:17:04 +0100 Russell King wrote:
> 
> > On Tue, Apr 29, 2008 at 09:27:10PM +0300, Adrian Bunk wrote:
> > > Commit 9c37066d888bf6e1b96ad12304971b3ddeabbad0 (proc: remove proc_bus) 
> > > came through -mm that contains a patch implementing the overdue removal 
> > > of the PCMCIA control ioctl that did not yet make it into Linus' tree:
> > 
> > Okay, this means at least one of my platforms is now dead as far as
> > building new kernels for it, and nothing I can do about that.  Ho hum.
> 
> I don't get it.  Several of us had building problems for a few
> days with some pasic driver.  We posted some bug reports and
> some patches, but we put up with it for a few days...

What don't you get?

I've said each time the PCMCIA ioctl removal topic came up that it would
give me this precise problem.  I've given details of _why_ it would cause
me that problem.  It's gone ahead inspite of that.

Concerning the pasic driver.  For fuck sake, why are you bringing that
up.  I made it *perfectly* and *crystal* clear that I *WAS* *NOT* in a
position to fix that thing - BECAUSE I WASN'T GOING TO BE IN FRONT OF A
COMPUTER FOR VERY LONG ON THOSE DAYS.  Which I knew would be the case on
Monday morning.  I _tried_ to stop the thing being merged, but Linus
merged it anyway, and then told me he'd done that _AFTER_ I'd gone to
bed on Monday.  I tried to ensure that I was in a position to cover
_all_ eventualities (except the one which actually occurred.)

So, between Monday midnight and Thursday morning, when do you expect me
to sort the resulting mess out, given that I was not even able to be
in front of a computer until about 10pm on Tuesday and Wednesday?

Oh, maybe you expect me to give up sleeping just for you?

What do you want me to do?  Quit because of it?

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:

Re: pcmcia_ioctl.c compile error

[Randy Dunlap]

On Tue, 29 Apr 2008 14:09:26 -0700 Andrew Morton wrote:

> On Tue, 29 Apr 2008 14:03:07 -0700
> Harvey Harrison <harvey.harrison@gmail.com> wrote:
> 
> > On Tue, 2008-04-29 at 13:59 -0700, Randy Dunlap wrote:
> > > On Tue, 29 Apr 2008 21:17:04 +0100 Russell King wrote:
> > > 
> > > > On Tue, Apr 29, 2008 at 09:27:10PM +0300, Adrian Bunk wrote:
> > > > > Commit 9c37066d888bf6e1b96ad12304971b3ddeabbad0 (proc: remove proc_bus) 
> > > > > came through -mm that contains a patch implementing the overdue removal 
> > > > > of the PCMCIA control ioctl that did not yet make it into Linus' tree:
> > > > 
> > > > Okay, this means at least one of my platforms is now dead as far as
> > > > building new kernels for it, and nothing I can do about that.  Ho hum.
> > > 
> > > I don't get it.  Several of us had building problems for a few
> > > days with some pasic driver.  We posted some bug reports and
> > > some patches, but we put up with it for a few days...
> > > 
> > 
> > I believe Russell is referring to the removal of the ioctl, not the
> > compile breakage.
> > 
> 
> That would be interesting information (although I have a vague feeling that
> it has been discussed before).

Yes:  http://lkml.org/lkml/2008/2/27/291
(now that Harvey reminded me/us)


> I've been sitting on pcmcia-delete-obsolete-pcmcia_ioctl-feature.patch for
> over a year, mainly waiting for Dominik to reappear and decide what to do
> about it.
> 
> I could just drop it, shrug.


---
~Randy

Re: pcmcia_ioctl.c compile error

[Randy Dunlap]

On Tue, 29 Apr 2008 22:14:11 +0100 Russell King wrote:

> On Tue, Apr 29, 2008 at 01:59:02PM -0700, Randy Dunlap wrote:
> > On Tue, 29 Apr 2008 21:17:04 +0100 Russell King wrote:
> > 
> > > On Tue, Apr 29, 2008 at 09:27:10PM +0300, Adrian Bunk wrote:
> > > > Commit 9c37066d888bf6e1b96ad12304971b3ddeabbad0 (proc: remove proc_bus) 
> > > > came through -mm that contains a patch implementing the overdue removal 
> > > > of the PCMCIA control ioctl that did not yet make it into Linus' tree:
> > > 
> > > Okay, this means at least one of my platforms is now dead as far as
> > > building new kernels for it, and nothing I can do about that.  Ho hum.
> > 
> > I don't get it.  Several of us had building problems for a few
> > days with some pasic driver.  We posted some bug reports and
> > some patches, but we put up with it for a few days...
> 
> What don't you get?

Why you said "building new kernels for it".
So it confused me.  It's not strictly a build issue,
it's a run-time issue.


> I've said each time the PCMCIA ioctl removal topic came up that it would
> give me this precise problem.  I've given details of _why_ it would cause
> me that problem.  It's gone ahead inspite of that.


---
~Randy

Re: pcmcia_ioctl.c compile error

[Andrew Morton]

On Tue, 29 Apr 2008 14:15:20 -0700
Randy Dunlap <rdunlap@xenotime.net> wrote:

> > > I believe Russell is referring to the removal of the ioctl, not the
> > > compile breakage.
> > > 
> > 
> > That would be interesting information (although I have a vague feeling that
> > it has been discussed before).
> 
> Yes:  http://lkml.org/lkml/2008/2/27/291
> (now that Harvey reminded me/us)

oh, OK, whatever, that's easy.  I dropped the old patch and queued this
one:

--- a/Documentation/feature-removal-schedule.txt~x
+++ a/Documentation/feature-removal-schedule.txt
@@ -47,23 +47,6 @@ Who:	Mauro Carvalho Chehab <mchehab@infr
 
 ---------------------------
 
-What:	PCMCIA control ioctl (needed for pcmcia-cs [cardmgr, cardctl])
-When:	November 2005
-Files:	drivers/pcmcia/: pcmcia_ioctl.c
-Why:	With the 16-bit PCMCIA subsystem now behaving (almost) like a
-	normal hotpluggable bus, and with it using the default kernel
-	infrastructure (hotplug, driver core, sysfs) keeping the PCMCIA
-	control ioctl needed by cardmgr and cardctl from pcmcia-cs is
-	unnecessary, and makes further cleanups and integration of the
-	PCMCIA subsystem into the Linux kernel device driver model more
-	difficult. The features provided by cardmgr and cardctl are either
-	handled by the kernel itself now or are available in the new
-	pcmciautils package available at
-	http://kernel.org/pub/linux/utils/kernel/pcmcia/
-Who:	Dominik Brodowski <linux@brodo.de>
-
----------------------------
-
 What:	sys_sysctl
 When:	September 2010
 Option: CONFIG_SYSCTL_SYSCALL
_

Re: pcmcia_ioctl.c compile error

[Russell King]

On Tue, Apr 29, 2008 at 02:38:24PM -0700, Andrew Morton wrote:
> On Tue, 29 Apr 2008 14:15:20 -0700
> Randy Dunlap <rdunlap@xenotime.net> wrote:
> 
> > > > I believe Russell is referring to the removal of the ioctl, not the
> > > > compile breakage.
> > > > 
> > > 
> > > That would be interesting information (although I have a vague feeling that
> > > it has been discussed before).
> > 
> > Yes:  http://lkml.org/lkml/2008/2/27/291
> > (now that Harvey reminded me/us)
> 
> oh, OK, whatever, that's easy.  I dropped the old patch and queued this
> one:

I'll spend some time this coming weekend working out precisely what it
requires from the ioctl interface - maybe we can have a cut-down ioctl
interface that bolts straight on as an "add on" to the new controls
without being too invasive, while still allowing its PCMCIA bits to
work.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:

Re: pcmcia_ioctl.c compile error

[Alan Cox]

> I've said each time the PCMCIA ioctl removal topic came up that it would
> give me this precise problem.  I've given details of _why_ it would cause
> me that problem.  It's gone ahead inspite of that.

Which is reasonable.

> So, between Monday midnight and Thursday morning, when do you expect me
> to sort the resulting mess out, given that I was not even able to be
> in front of a computer until about 10pm on Tuesday and Wednesday?

How about delegate some of the control to other people ? There's something
wrong if it has to be you that sorts it out eveyr time - especially as you
effectively manage as many platforms as everyone else put together right
now.

Alan

Re: pcmcia_ioctl.c compile error

[Alan Cox]

> I could just drop it, shrug.

I'd favour pushing it. If it matters someone will then fix up the problem
platform. If it isn't pushed then it lets one obscure platform stall
progress for everyone else.

Alan (still trying to get all the termios ioctl changes into some ports)

Re: pcmcia_ioctl.c compile error

[Andrew Morton]

On Tue, 29 Apr 2008 22:56:20 +0100
Russell King <rmk+lkml@arm.linux.org.uk> wrote:

> On Tue, Apr 29, 2008 at 02:38:24PM -0700, Andrew Morton wrote:
> > On Tue, 29 Apr 2008 14:15:20 -0700
> > Randy Dunlap <rdunlap@xenotime.net> wrote:
> > 
> > > > > I believe Russell is referring to the removal of the ioctl, not the
> > > > > compile breakage.
> > > > > 
> > > > 
> > > > That would be interesting information (although I have a vague feeling that
> > > > it has been discussed before).
> > > 
> > > Yes:  http://lkml.org/lkml/2008/2/27/291
> > > (now that Harvey reminded me/us)
> > 
> > oh, OK, whatever, that's easy.  I dropped the old patch and queued this
> > one:
> 
> I'll spend some time this coming weekend working out precisely what it
> requires from the ioctl interface - maybe we can have a cut-down ioctl
> interface that bolts straight on as an "add on" to the new controls
> without being too invasive, while still allowing its PCMCIA bits to
> work.

umm, well,

a) as your machine still needed the ioctl code, we can assume that there
   are others out there.  So it's unclear that we _can_ delete it, or change
   its interfaces.

b) the rate of change in that code is very close to zero.  I'd say just let
   it be.  It's not a good use of one's time?

Re: pcmcia_ioctl.c compile error

[Russell King]

On Tue, Apr 29, 2008 at 10:58:45PM +0100, Alan Cox wrote:
> > So, between Monday midnight and Thursday morning, when do you expect me
> > to sort the resulting mess out, given that I was not even able to be
> > in front of a computer until about 10pm on Tuesday and Wednesday?
> 
> How about delegate some of the control to other people ? There's something
> wrong if it has to be you that sorts it out eveyr time - especially as you
> effectively manage as many platforms as everyone else put together right
> now.

I believe I pointed at the patch required to fix the problem via this
URL:

  http://www.arm.linux.org.uk/developer/patches/viewpatch.php?id=5010/1

A subset of that patch had also already been discussed between Philipp,
akpm and myself.  The URL also shows when it was applied (though not to
a branch of my tree I could've pushed into mainline at that stage.)

The fix was already known...  The fix had been pointed out by me during
the various mails I got about the breakage...  I'd effectively already
approved the fix by applying it...  And I had said I wouldn't be able to
sort it until Thursday.

Yes, maybe I should've said to Andrew "please grab this patch off this
URL and ensure it gets to Linus ASAP."  Hind sight is always a wonderful
thing.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of: