[PATCH][5/5] floppy.c: Fix device_create_file() warning

[PATCH][5/5] floppy.c: Fix device_create_file() warning

[Jesper Juhl]

This fixes the warning 
  warning: ignoring return value of `device_create_file', declared with attribute warn_unused_result
in function `floppy_init'. It does this by checking the return value and 
printing a warning message in case of no success.


Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
---

 floppy.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

--- linux-2.6.20/drivers/block/floppy.c.patch4	2007-02-05 23:22:26.000000000 +0100
+++ linux-2.6.20/drivers/block/floppy.c	2007-02-05 23:32:42.000000000 +0100
@@ -4302,7 +4302,12 @@
 		if (err)
 			goto out_flush_work;
 
-		device_create_file(&floppy_device[drive].dev,&dev_attr_cmos);
+		err = device_create_file(&floppy_device[drive].dev, &dev_attr_cmos);
+		if (err)
+			printk(KERN_WARNING "Unable to create sysfs attribute "
+				"file for floppy device: %s\n", 
+				floppy_device[drive].name);
+
 		/* to be cleaned up... */
 		disks[drive]->private_data = (void *)(long)drive;
 		disks[drive]->queue = floppy_queue;

Re: [PATCH][5/5] floppy.c: Fix device_create_file() warning

[Gene Heskett]

On Monday 05 February 2007 18:30, Jesper Juhl wrote:

And this is almost OT for this, but I'd like to interject here that in 
recent history, it has been very very difficult to properly access LSN0 
of a floppy.  Some formats, such as those laid down by a WD-1773 floppy 
controller, do indeed use a sector marking of '0' (zero) on all tracks.

This is akin to an old basic program that had a 'base' statement with an 
argument of zero or one.  We need, for legacy machines where a floppy 
disk sneakernet is in use, to maintain this "base 0" functionality in an 
easy to access format.  Ditto for sector sizes of 256 and even 128 bytes.

>This fixes the warning
>  warning: ignoring return value of `device_create_file', declared with
> attribute warn_unused_result in function `floppy_init'. It does this by
> checking the return value and printing a warning message in case of no
> success.
>
>
>Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
>---
>
> floppy.c |    7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
>--- linux-2.6.20/drivers/block/floppy.c.patch4	2007-02-05
> 23:22:26.000000000 +0100 +++
> linux-2.6.20/drivers/block/floppy.c	2007-02-05 23:32:42.000000000 +0100
> @@ -4302,7 +4302,12 @@
> 		if (err)
> 			goto out_flush_work;
>
>-		device_create_file(&floppy_device[drive].dev,&dev_attr_cmos);
>+		err = device_create_file(&floppy_device[drive].dev, &dev_attr_cmos);
>+		if (err)
>+			printk(KERN_WARNING "Unable to create sysfs attribute "
>+				"file for floppy device: %s\n",
>+				floppy_device[drive].name);
>+
> 		/* to be cleaned up... */
> 		disks[drive]->private_data = (void *)(long)drive;
> 		disks[drive]->queue = floppy_queue;
>
>
>
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel"
> in the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>Please read the FAQ at  http://www.tux.org/lkml/

-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Yahoo.com and AOL/TW attorneys please note, additions to the above
message by Gene Heskett are:
Copyright 2007 by Maurice Eugene Heskett, all rights reserved.

Re: [PATCH][5/5] floppy.c: Fix device_create_file() warning

[H. Peter Anvin]

Gene Heskett wrote:
> On Monday 05 February 2007 18:30, Jesper Juhl wrote:
> 
> And this is almost OT for this, but I'd like to interject here that in 
> recent history, it has been very very difficult to properly access LSN0 
> of a floppy.  Some formats, such as those laid down by a WD-1773 floppy 
> controller, do indeed use a sector marking of '0' (zero) on all tracks.
> 

I thought the sector numbers of floppy discs were software programmable 
(at formatting time.)  I believe different OSes used different offsets, 
with 1, 2, 3... being the way IBM did it in the PC, but all kinds of 
variants were used, including things like 0xa0, 0xa1, ... for no 
apparent reason.

	-hpa

Re: [PATCH][5/5] floppy.c: Fix device_create_file() warning

[Gene Heskett]

On Tuesday 06 February 2007 01:32, H. Peter Anvin wrote:
>Gene Heskett wrote:
>> On Monday 05 February 2007 18:30, Jesper Juhl wrote:
>>
>> And this is almost OT for this, but I'd like to interject here that in
>> recent history, it has been very very difficult to properly access
>> LSN0 of a floppy.  Some formats, such as those laid down by a WD-1773
>> floppy controller, do indeed use a sector marking of '0' (zero) on all
>> tracks.
>
>I thought the sector numbers of floppy discs were software programmable
>(at formatting time.)  I believe different OSes used different offsets,
>with 1, 2, 3... being the way IBM did it in the PC, but all kinds of
>variants were used, including things like 0xa0, 0xa1, ... for no
>apparent reason.
>
>	-hpa
>-
What I'm referring to hpa, is the suite of floppy tools that isn't even in 
the FC6 catalog for some reason, fdutils-x.x.  In the past when I was 
running fc2, I had to go find version 5.4, not 5.5 of that kit, build and 
install that, and then revert the floppy stuffs to something borrowed 
from a much older kernel before a command of 'setfdprm /dev/fd0 COCO720' 
could even get past the syntax error stage, and we traced it down to an 
inability to set a track format whose sector numbering scheme included a 
sector number of zero as the first sector of the track.

I have a copy of fdutils-5.4 here, which I have not done a make install of 
yet for FC6, so I should do a make clean;make;checkinstall --inspect and 
see if it will work.  I have not be able to do that since about 2.6.8 or 
so.

I just installed & tested it, and it appears to work correctly with the 
floppy driver in 2.6.20, but I'll see if I can dd a bootable floppy to 
it, the acid test for this user anyway... 10 minutes later, looking at 
the recovered image with khexedit, it looks good, everything is in the 
right place AFAICT, no little one byte offsets apparent.  So as long as 
I'm using the older fdutils version-5.4, it looks like it works.

If it does not work, then we have locked at least one vintage computer 
that I have several of, The TRS-80 Color Computer family, out of talking 
to linux by any means.  I have also spent the equ of several days trying 
to get its file transfer utility rzsz-3.36 based on Chuck Foresburgs 
code, to work through a pl2303 usb-serial convertor.  Something in the 
flow controls I think, I can type back and forth with terminal programs 
but nothing that sends large quantities of data works.

Maybe my fears are unfounded ATM, but I did want to voice them.

-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Yahoo.com and AOL/TW attorneys please note, additions to the above
message by Gene Heskett are:
Copyright 2007 by Maurice Eugene Heskett, all rights reserved.