got it, looks like i did a accidental "dd" in vi :)
On Saturday 09 August 2008 08:35:14 pm Venky K Shankar wrote:
> hi kernel'ers :
>
> i have a kernel module in which i am trying to implement a simple device
> driver following the LDD3 book.
>
> currently i am only trying to open and close the device node. the module
> compiles fine and when insmod'ed it sets up the hooks using cdev_init() and
> cdev_add().
>
> then i create the device node in the /dev directory as per the major and
> minor number.
>
> but when i try to open() the device node i get the following Oops message :
> (/var/log/messages)
>
> ===========================================================================
>========================================= Aug 9 19:30:05 linux-ujq3 kernel:
> BUG: unable to handle kernel NULL pointer dereference at virtual address
> 00000005 Aug 9 19:30:05 linux-ujq3 kernel: printing eip:
> Aug 9 19:30:05 linux-ujq3 kernel: c017340a
> Aug 9 19:30:05 linux-ujq3 kernel: *pde = 00000000
> Aug 9 19:30:05 linux-ujq3 kernel: Oops: 0002 [#1]
> Aug 9 19:30:05 linux-ujq3 kernel: SMP
> Aug 9 19:30:05 linux-ujq3 kernel: last sysfs file:
> /devices/system/cpu/cpu1/online Aug 9 19:30:05 linux-ujq3 kernel: Modules
> linked in: skull snd_pcm_oss snd_mixer_oss snd_seq snd_seq_device
> iptable_filter ip_tables ip6table_filter ip6_tables x_tables ipv6 micr
> ocode firmware_class apparmor nls_iso8859_1 nls_cp437 vfat fat fuse loop
> dm_mod snd_hda_intel snd_pcm ohci1394 8139cp snd_timer ieee1394 8139too snd
> parport_pc ati_agp rtc_cmos mi i agpgart i2c_piix4 button rtc_core
> usb_storage soundcore sr_mod cdrom snd_page_alloc i2c_core ide_core rtc_lib
> parport sg sd_mod ohci_hcd ehci_hcd usbcore edd ext3 mbcache jbd fa n
> pata_atiixp sata_sil libata scsi_mod thermal processor
> Aug 9 19:30:05 linux-ujq3 kernel: CPU: 1
> Aug 9 19:30:05 linux-ujq3 kernel: EIP: 0060:[<c017340a>] Tainted: G
> N VLI Aug 9 19:30:05 linux-ujq3 kernel: EFLAGS: 00210246
> (2.6.22.17-0.1-default #1) Aug 9 19:30:05 linux-ujq3 kernel: EIP is at
> chrdev_open+0x79/0x133 Aug 9 19:30:05 linux-ujq3 kernel: eax: 00000001
> ebx: f3f6de44 ecx: f3f6de44 edx: f3d51f64 Aug 9 19:30:05 linux-ujq3
> kernel: esi: f3f6de44 edi: 00000000 ebp: f3d51e60 esp: f3f99ee0 Aug
> 9 19:30:05 linux-ujq3 kernel: ds: 007b es: 007b fs: 00d8 gs: 0033 ss:
> 0068 Aug 9 19:30:05 linux-ujq3 kernel: Process a.out (pid: 4364,
> ti=f3f98000 task=dff8b570 task.ti=f3f98000) Aug 9 19:30:05 linux-ujq3
> kernel: Stack: f421a3c0 00000000 f421a3c0 f3d51e60 f3f99f30 c0173391
> c016f8b7 dfbf13c0 Aug 9 19:30:05 linux-ujq3 kernel: f3e8dc6c
> f421a3c0 ffffff9c f3f99f30 00000003 c016f9e8 f421a3c0 00000000 Aug 9
> 19:30:05 linux-ujq3 kernel: 0804855b c016fa2e bfa9f9d8 f3f99f30
> f3e8dc6c dfbf13c0 13645c10 00000005 Aug 9 19:30:05 linux-ujq3 kernel: Call
> Trace:
> Aug 9 19:30:05 linux-ujq3 kernel: [<c0173391>] chrdev_open+0x0/0x133
> Aug 9 19:30:05 linux-ujq3 kernel: [<c016f8b7>] __dentry_open+0xc1/0x178
> Aug 9 19:30:05 linux-ujq3 kernel: [<c016f9e8>]
> nameidata_to_filp+0x24/0x33 Aug 9 19:30:05 linux-ujq3 kernel:
> [<c016fa2e>] do_filp_open+0x37/0x3e Aug 9 19:30:05 linux-ujq3 kernel:
> [<c016f79b>] get_unused_fd+0x57/0xb2 Aug 9 19:30:05 linux-ujq3 kernel:
> [<c016fa77>] do_sys_open+0x42/0xc8 Aug 9 19:30:05 linux-ujq3 kernel:
> [<c016fb36>] sys_open+0x1c/0x1e Aug 9 19:30:05 linux-ujq3 kernel:
> [<c0104e22>] sysenter_past_esp+0x6b/0xa9 Aug 9 19:30:05 linux-ujq3 kernel:
> [<c02c0000>] unix_find_other+0x75/0x161 Aug 9 19:30:05 linux-ujq3 kernel:
> =======================
> Aug 9 19:30:05 linux-ujq3 kernel: Code: c3 2d 15 00 8b b5 0c 01 00 00 85
> f6 75 36 89 9d 0c 01 00 00 8b 44 24 04 8d 95 04 01 00 00 89 de 31 ff 89 85
> 10 01 00 00 8b 43 48 <89> 50 0 4 89 85 04 01 00 00 8d 43 48 89 53 48 31 db
> 89 85 08 01
> Aug 9 19:30:05 linux-ujq3 kernel: EIP: [<c017340a>] chrdev_open+0x79/0x133
> SS:ESP 0068:f3f99ee0
> ===========================================================================
>============================================= the line : "Aug 9 19:30:05
> linux-ujq3 kernel: BUG: unable to handle kernel NULL pointer dereference at
> virtual address 00000005" comes from mm/fault.c (do_page_fault()), but i am
> unable to find what is causing the problem (may be a very very silly one
> :)).
>
> after this the system hangs and the only solution is a hard reboot.
>
> where am i going wrong or am i missing something.
> i am running : Linux linux-ujq3 2.6.22.17-0.1-default #1 SMP 2008/02/10
> 20:01:04 UTC i686 i686 i386 GNU/Linux
>
> source is attached : skull.c and skull.h
>
> thanks in advance.
> -Venky
