Skip to contentSkip to conversation

Page fault in umount

DISCUSSION Pierre Ossman 2 participants 3 messages Jan 22, 2005 – Jan 25, 2005

lore
Share & tools
lore
#0 lore raw

When I yank out my MP3 player, the programs trying to umount the disk cause the following page fault:

usb 1-5: USB disconnect, address 2
scsi0 (0:0): rejecting I/O to dead device
FAT bread failed in fat_clusters_flush

Unable to handle kernel paging request at virtual address 6b6b6b6b
 printing eip:
e0a0ecaf
*pde = 00000000
Oops: 0000 [#1]
PREEMPT
Modules linked in: radeon pcspkr iptable_filter tun parport_pc lp parport irport irnet ppp_generic slhc ircomm_tty ircomm irda crc_ccitt sd_mod autofs4 hidp rfcomm l2cap bluetooth pcmcia sunrpc ipt_MASQUERADE iptable_nat usb_storage scsi_mod ip_conntrack ip_tables microcode binfmt_misc nls_iso8859_1 nls_cp437 vfat fat ext3 jbd video button battery ac ohci1394 ieee1394 yenta_socket pcmcia_core uhci_hcd ehci_hcd i2c_i801 i2c_core snd_intel8x0m snd_intel8x0 snd_ac97_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd soundcore snd_page_alloc 8139cp mii
CPU:    0
EIP:    0060:[<e0a0ecaf>]    Not tainted VLI
EFLAGS: 00010296   (2.6.10)

EIP is at scsi_device_put+0xf/0x70 [scsi_mod]
eax: 6b6b6b6b ebx: d8994150 ecx: 00000000 edx: 00000000
esi: 6b6b6b6b edi: dfcae46c ebp: 00000000 esp: ce6b3ee4
ds: 007b es: 007b ss: 0068
Process umount (pid: 7190, threadinfo=ce6b2000 task=cbe5c000)

Stack: d883b9f4 d8994150 d883b9f4 e0b38103 6b6b6b6b e0b3a190 d8994150 e0b386f7
       d8994150 00000000 dc6a1764 dc6a17f8 c0195e56 dc6a17f8 00000000 
dc6a19b8
       c0472680 00000000 00000000 c0195ed0 dc6a1764 c0193acd d89645f8 
e09f6ec0
Call Trace:
 [<e0b38103>] scsi_disk_put+0x33/0x50 [sd_mod]
 [<e0b3a190>] scsi_disk_release+0x0/0x1b0 [sd_mod]
 [<e0b386f7>] sd_release+0x47/0x90 [sd_mod]
 [<c0195e56>] blkdev_put+0xc6/0x170
 [<c0195ed0>] blkdev_put+0x140/0x170
 [<c0193acd>] kill_block_super+0x3d/0x60
 [<c019144c>] deactivate_super+0xac/0x120
 [<c01b4f95>] __mntput+0x25/0x40
 [<c01b5c9f>] sys_umount+0x3f/0xa0
 [<c01746a4>] sys_munmap+0x44/0x70
 [<c0103c59>] sysenter_past_esp+0x52/0x75
Code: 34 24 e8 05 ba 90 df ba fa ff ff ff eb e0 8d b4 26 00 00 00 00 8d bc 27 00 00 00 00 83 ec 0c 89 74 24 08 8b 74 24 10 89 5c 24 04 <8b> 06 8b 80 b4 00 00 00 8b 00 85 c0 74 1f bb 00 e0 ff ff 21 e3

The device is mounted in sync so shouldn't this be safe as long as the device isn't busy? This works on other USB devices I have. A page fault seems severe either way.

Rgds
Pierre

#1 lore raw
Quoted 6 lines · Pierre Ossman

When I yank out my MP3 player, the programs trying to umount the disk cause the following page fault:

...
EIP is at scsi_device_put+0xf/0x70 [scsi_mod]

This should be fixed in 2.6.11-rc2-mm1, via bk-scsi-rc-fixes.patch. Could you retest with either

ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.11-rc2/2.6.11-rc2-mm1/2.6.11-rc2-mm1.bz2

or

ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.11-rc2/2.6.11-rc2-mm1/broken-out/bk-scsi-rc-fixes.patch

applied?

#2 lore raw
Quoted 13 lines · Andrew Morton
Quoted 6 lines · Pierre Ossman

When I yank out my MP3 player, the programs trying to umount the disk cause the following page fault:

...
EIP is at scsi_device_put+0xf/0x70 [scsi_mod]

This should be fixed in 2.6.11-rc2-mm1, via bk-scsi-rc-fixes.patch. Could you retest with either

Quoted 9 lines

ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.11-rc2/2.6.11-rc2-mm1/2.6.11-rc2-mm1.bz2

or

ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.11-rc2/2.6.11-rc2-mm1/broken-out/bk-scsi-rc-fixes.patch

applied?

The patch seems to solve the problem. Thanks.

Rgds
Pierre