[PATCH] 2.5.32 floppy init and misc fixes

[PATCH] 2.5.32 floppy init and misc fixes

[Mikolaj J. Habryn]

The floppy driver doesn't clean up it's register_sys_device if it fails
to init correctly. The below patch adds the appropriate
unregister_sys_device in the various failure paths, and also includes
Ewan MacMahon's one-liner for working VCs on devfs-only machines, and a
couple of missing #includes. Minimum required to get 2.5.32 working on
my Portege.

m.

--- linux-2.5.32/drivers/char/console.c.orig	2002-08-29 14:07:59.000000000 +1000
+++ linux-2.5.32/drivers/char/console.c	2002-08-29 13:42:07.000000000 +1000
@@ -2526,6 +2526,7 @@
 
 	kbd_init();
 	console_map_init();
+	con_init_devfs();
 	vcs_init();
 	return 0;
 }
--- linux-2.5.32/drivers/block/floppy.c.orig	2002-08-29 16:17:30.000000000 +1000
+++ linux-2.5.32/drivers/block/floppy.c	2002-08-29 16:19:20.000000000 +1000
@@ -4235,6 +4235,7 @@
 	devfs_handle = devfs_mk_dir (NULL, "floppy", NULL);
 	if (register_blkdev(MAJOR_NR,"fd",&floppy_fops)) {
 		printk("Unable to get major %d for floppy\n",MAJOR_NR);
+		unregister_sys_device(&device_floppy);
 		return -EBUSY;
 	}
 
@@ -4269,6 +4270,7 @@
 		unregister_blkdev(MAJOR_NR,"fd");
 		del_timer(&fd_timeout);
 		blk_cleanup_queue(BLK_DEFAULT_QUEUE(MAJOR_NR));
+		unregister_sys_device(&device_floppy);
 		return -ENODEV;
 	}
 #if N_FDC > 1
@@ -4280,6 +4282,7 @@
 		del_timer(&fd_timeout);
 		blk_cleanup_queue(BLK_DEFAULT_QUEUE(MAJOR_NR));
 		unregister_blkdev(MAJOR_NR,"fd");
+		unregister_sys_device(&device_floppy);
 		return -EBUSY;
 	}
 
@@ -4343,6 +4346,7 @@
 			floppy_release_irq_and_dma();
 		blk_cleanup_queue(BLK_DEFAULT_QUEUE(MAJOR_NR));
 		unregister_blkdev(MAJOR_NR,"fd");
+		unregister_sys_device(&device_floppy);
 	}
 	
 	for (drive = 0; drive < N_DRIVE; drive++) {
--- linux-2.5.32/drivers/net/wireless/hermes.c.orig	2002-08-29 14:55:28.000000000 +1000
+++ linux-2.5.32/drivers/net/wireless/hermes.c	2002-08-29 14:56:03.000000000 +1000
@@ -45,6 +45,7 @@
 #include <linux/threads.h>
 #include <linux/smp.h>
 #include <asm/io.h>
+#include <linux/sched.h>
 #include <linux/ptrace.h>
 #include <linux/delay.h>
 #include <linux/init.h>
--- linux-2.5.32/drivers/char/toshiba.c.orig	2002-08-29 14:09:44.000000000 +1000
+++ linux-2.5.32/drivers/char/toshiba.c	2002-08-29 14:10:09.000000000 +1000
@@ -69,6 +69,7 @@
 #include <linux/init.h>
 #include <linux/stat.h>
 #include <linux/proc_fs.h>
+#include <linux/interrupt.h>
 
 #include <linux/toshiba.h>
 

Re: [PATCH] 2.5.32 floppy init and misc fixes

[Mikael Pettersson]

Mikolaj J. Habryn writes:
 > The floppy driver doesn't clean up it's register_sys_device if it fails
 > to init correctly. The below patch adds the appropriate
 > unregister_sys_device in the various failure paths, and also includes
 > Ewan MacMahon's one-liner for working VCs on devfs-only machines, and a
 > couple of missing #includes. Minimum required to get 2.5.32 working on
 > my Portege.
...
 > --- linux-2.5.32/drivers/block/floppy.c.orig	2002-08-29 16:17:30.000000000 +1000
 > +++ linux-2.5.32/drivers/block/floppy.c	2002-08-29 16:19:20.000000000 +1000
 > @@ -4235,6 +4235,7 @@
 >  	devfs_handle = devfs_mk_dir (NULL, "floppy", NULL);
 >  	if (register_blkdev(MAJOR_NR,"fd",&floppy_fops)) {
 >  		printk("Unable to get major %d for floppy\n",MAJOR_NR);
 > +		unregister_sys_device(&device_floppy);
 >  		return -EBUSY;
 >  	}
 >  
etc.

Floppy has many more problems.
Repeadedly loading and unloading the floppy.o module corrupts
sys_device data structures.
Writing to floppy can give ENOSPC errors even though space exists.
Writing to floppy can OOPS the kernel due to a NULL pointer error.
VFS-over-floppy corrupts data since 2.5.13.
Putting lilo on ext2 on floppy can cause a kernel hang due to an
infinite loop of "buffer layer error".

I have a patch which fixes the {,un}register_sys_device() bugs, NULL
queue bug, zero i_size bug, and fixes read/write enough that raw media
access (e.g. tar or dd to/from /dev/fd0) works. It's in the 2.5-dj tree,
and separately in <http://www.csd.uu.se/~mikpe/linux/patches/2.5/>.

I havent' pushed this to Linus since it's meaningless as long as
the VFS data corruption exists. It was broken by blkdev/VFS changes,
but those responsible haven't yet bothered to repair it.

/Mikael

Re: [PATCH] 2.5.32 floppy init and misc fixes

[Randy.Dunlap]

On Thu, 29 Aug 2002, Mikael Pettersson wrote:

| Floppy has many more problems.
| Repeadedly loading and unloading the floppy.o module corrupts
| sys_device data structures.
| Writing to floppy can give ENOSPC errors even though space exists.
| Writing to floppy can OOPS the kernel due to a NULL pointer error.
| VFS-over-floppy corrupts data since 2.5.13.
| Putting lilo on ext2 on floppy can cause a kernel hang due to an
| infinite loop of "buffer layer error".
|
| I have a patch which fixes the {,un}register_sys_device() bugs, NULL
| queue bug, zero i_size bug, and fixes read/write enough that raw media
| access (e.g. tar or dd to/from /dev/fd0) works. It's in the 2.5-dj tree,
| and separately in <http://www.csd.uu.se/~mikpe/linux/patches/2.5/>.
|
| I havent' pushed this to Linus since it's meaningless as long as
| the VFS data corruption exists. It was broken by blkdev/VFS changes,
| but those responsible haven't yet bothered to repair it.

I would add one more: select delay timings are same as in 2.4:
#define SEL_DLY (2*HZ/100)
but HZ is not the same as in 2.4...

-- 
~Randy

Re: [PATCH] 2.5.32 floppy init and misc fixes

[Mikael Pettersson]

On Thu, 29 Aug 2002 10:30:14 -0700 (PDT), Randy.Dunlap wrote:
>On Thu, 29 Aug 2002, Mikael Pettersson wrote:
>
>| Floppy has many more problems.
...
>
>I would add one more: select delay timings are same as in 2.4:
>#define SEL_DLY (2*HZ/100)
>but HZ is not the same as in 2.4...

I took a quick glance through floppy.c, and all HZ-dependent
delays seem to be in jiffies, so I think they are Ok.

/Mikael

Re: [PATCH] 2.5.32 floppy init and misc fixes

[Randy.Dunlap]

On Sat, 31 Aug 2002, Mikael Pettersson wrote:

| On Thu, 29 Aug 2002 10:30:14 -0700 (PDT), Randy.Dunlap wrote:
| >On Thu, 29 Aug 2002, Mikael Pettersson wrote:
| >
| >| Floppy has many more problems.
| ...
| >
| >I would add one more: select delay timings are same as in 2.4:
| >#define SEL_DLY (2*HZ/100)
| >but HZ is not the same as in 2.4...
|
| I took a quick glance through floppy.c, and all HZ-dependent
| delays seem to be in jiffies, so I think they are Ok.

I agree except for SEL_DLY as listed above.
It divides by 100 (assuming HZ is 100 I believe).
Same #define in 2.4.current and 2.5.32, so the value
is different in them (with HZ = 1000 in 2.5.32).

-- 
~Randy