newfs error in 5.0.2 with SanDisk SD card

[Rui-Xiang Guo <rxg%lavabit.com@localhost>]

Hi, all.
This is a 128Mb sd card with the usb adaptor. I try to format it with the
steps below:
tree# fdisk sd0
fdisk: Cannot determine the number of heads
Disk: /dev/rsd0d
NetBSD disklabel disk geometry:
cylinders: 120, heads: 64, sectors/track: 32 (2048 sectors/cylinder)
total sectors: 246016

BIOS disk geometry:
cylinders: 120, heads: 64, sectors/track: 32 (2048 sectors/cylinder)
total sectors: 246016

Partition table:
0: <UNUSED>
1: <UNUSED>
2: <UNUSED>
3: <UNUSED>
No active partition.
Drive serial number: 29265 (0x00007251)
tree# fdisk -u sd0
fdisk: Cannot determine the number of heads
Disk: /dev/rsd0d
NetBSD disklabel disk geometry:
cylinders: 120, heads: 64, sectors/track: 32 (2048 sectors/cylinder)
total sectors: 246016

BIOS disk geometry:
cylinders: 120, heads: 64, sectors/track: 32 (2048 sectors/cylinder)
total sectors: 246016

Do you want to change our idea of what BIOS thinks? [n]

Partition table:
0: <UNUSED>
1: <UNUSED>
2: <UNUSED>
3: <UNUSED>
No active partition.
Drive serial number: 29265 (0x00007251)
Which partition do you want to change?: [none] 3
The data for partition 3 is:
<UNUSED>
sysid: [0..255 default: 169] 
start: [0..120cyl default: 32, 0cyl, 0MB]
size: [0..120cyl default: 245984, 120cyl, 120MB] 
bootmenu: [] 
The bootselect code is not installed, do you want to install it now? [n]

Partition table:
0: <UNUSED>
1: <UNUSED>
2: <UNUSED>
3: NetBSD (sysid 169)
    start 32, size 245984 (120 MB, Cyls 0-120/7/32)
        PBR is not bootable: All bytes are identical (0x00)
No active partition.
Drive serial number: 29265 (0x00007251)
Which partition do you want to change?: [none]

Installed bootfile doesn't support required options.
Update the bootcode from /usr/mdec/mbr? [n]

We haven't written the MBR back to disk yet.  This is your last chance.
Partition table:
0: <UNUSED>
1: <UNUSED>
2: <UNUSED>
3: NetBSD (sysid 169)
    start 32, size 245984 (120 MB, Cyls 0-120/7/32)
        PBR is not bootable: All bytes are identical (0x00)
No active partition.
Drive serial number: 29265 (0x00007251)
Should we write new partition table? [n] y
tree# disklabel sd0
# /dev/rsd0d:
type: SCSI
disk: STORAGE DEVICE  
label: fictitious
flags: removable
bytes/sector: 512
sectors/track: 32
tracks/cylinder: 64
sectors/cylinder: 2048
cylinders: 120
total sectors: 246016
rpm: 3600
interleave: 1
trackskew: 0
cylinderskew: 0
headswitch: 0           # microseconds
track-to-track seek: 0  # microseconds
drivedata: 0 

5 partitions:
#        size    offset     fstype [fsize bsize cpg/sgs]
 a:    245984        32     4.2BSD      0     0     0  # (Cyl.      0*-    120*)
 c:    245984        32     unused      0     0        # (Cyl.      0*-    120*)
 d:    246016         0     unused      0     0        # (Cyl.      0 -    
120*)tree# newfs -O 2 /dev/rsd0a
/dev/rsd0a: 120.1MB (245984 sectors) block size 8192, fragment size 1024
        using 4 cylinder groups of 30.03MB, 3844 blks, 7232 inodes.
super-block backups (for fsck_ffs -b #) at:
wtfs: write error for sector 144: Input/output error

Then I use 4.0.1 boot disk to try it again and no error happen and the disk
can be mounted successfully.

After that, I take the disk under 5.0.2 and try again:
tree# fdisk sd0
Disk: /dev/rsd0d
NetBSD disklabel disk geometry:
cylinders: 120, heads: 64, sectors/track: 32 (2048 sectors/cylinder)
total sectors: 246016

BIOS disk geometry:
cylinders: 244, heads: 16, sectors/track: 63 (1008 sectors/cylinder)
total sectors: 246015

Partition table:
0: <UNUSED>
1: <UNUSED>
2: <UNUSED>
3: NetBSD (sysid 169)
    start 63, size 245953 (120 MB, Cyls 0-244/1/1)
        PBR is not bootable: All bytes are identical (0xf6)
No active partition.
Drive serial number: 29265 (0x00007251)
tree# disklabel sd0   
# /dev/rsd0d:
type: SCSI
disk: STORAGE DEVICE  
label: fictitious
flags: removable
bytes/sector: 512
sectors/track: 32
tracks/cylinder: 64
sectors/cylinder: 2048
cylinders: 120
total sectors: 246016
rpm: 3600
interleave: 1
trackskew: 0
cylinderskew: 0
headswitch: 0           # microseconds
track-to-track seek: 0  # microseconds
drivedata: 0 

5 partitions:
#        size    offset     fstype [fsize bsize cpg/sgs]
 a:    245889        63     4.2BSD      0     0     0  # (Cyl.      0*-    120*)
 c:    245953        63     unused      0     0        # (Cyl.      0*-    120*)
 d:    246016         0     unused      0     0        # (Cyl.      0 -    120*
tree# mount /dev/sd0a /mnt
mount_ffs: /dev/sd0a on /mnt: incorrect super block
tree# newfs -O 2 /dev/rsd0a
/dev/rsd0a: 120.1MB (245888 sectors) block size 8192, fragment size 1024
        using 4 cylinder groups of 30.02MB, 3842 blks, 7232 inodes.
super-block backups (for fsck_ffs -b #) at:
wtfs: write error for sector 144: Input/output error

I gor the same error. Does the kernels of 5.0.2 and 4.0.1 treat the disk
in much different way?

p.s. I cc this to port-i386 because I think it shall be happen on i386, too.
     Please also reply to my mailbox, thanks.

-rxg