Arch Linux

FeatherMonkey

Member

Registered: 2007-02-26

Posts: 313

Is this regression on flash devices? (exceeds device capacity)

This is actually a phone in flash mode now I would never say it has been smooth sailing but since trying recently I'm finding it is barely syncing.

In trying to investigate it I thought I caught something about bad device formatting is blocking but used be ignored.(Though since stumbling across it think lkml I'll be damned if I can find it again).

This is a SE phone when I look in dmesg after plugging in I get.

sdb: p1 size 4026336 exceeds device capacity, limited to end of disk
sd 4:0:0:0: [sdb] Test WP failed, assume Write Enabled
sd 4:0:0:0: [sdb] Assuming drive cache: write through
sd 4:0:0:0: [sdb] Attached SCSI removable disk

I've tried the dostools but these error

Seek to 2061483520:Invalid argument

gparted seems to be reporting(Also claims unallocated and something about no partitions)

Can't have a partition outside the disk!

fdisk -l

Disk /dev/sdb: 2061 MB, 2061484032 bytes
128 heads, 32 sectors/track, 982 cylinders
Units = cylinders of 4096 * 512 = 2097152 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb1   *           1         983     2013168    6  FAT16

I'm out of ideas I did try to see if it was regression but after blocking I/O during a reinstall of a kernel and having to chroot in to fix I'm dubious of my skill set to investigate further. I have tried enabling usbmon debug but to be honest it is gooble-de-gook, though I think I also caught this. I think it arises when unmounting/flushing but not sure it could be gparted related at the same time.

INFO: task umount:5463 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
umount        D 0000000000000000     0  5463   5462 0x00000000
 ffff8800674e1c98 0000000000000086 0000000000000000 000000000cefe8b4
 ffff88007c446ae0 ffff8800018952c0 ffff88007c446db0 ffff8800674e1c64
 ffff8800674e1c50 000000010000aade ffff8800674e1fd8 00000000000152c0
Call Trace:
 [<ffffffff81142e60>] ? bdi_sched_wait+0x0/0x10
 [<ffffffff81142e69>] bdi_sched_wait+0x9/0x10
 [<ffffffff81359497>] __wait_on_bit+0x57/0x80
 [<ffffffff81142e60>] ? bdi_sched_wait+0x0/0x10
 [<ffffffff81359533>] out_of_line_wait_on_bit+0x73/0x90
 [<ffffffff810733b0>] ? wake_bit_function+0x0/0x40
 [<ffffffff81142d9e>] ? bdi_queue_work+0x9e/0xd0
 [<ffffffff81142e53>] bdi_sync_writeback+0x83/0x90
 [<ffffffff81142e8d>] sync_inodes_sb+0x1d/0x100
 [<ffffffff811474ca>] __sync_filesystem+0x7a/0x80
 [<ffffffff81147693>] sync_filesystem+0x43/0x60
 [<ffffffff81122f62>] generic_shutdown_super+0x22/0x100
 [<ffffffff8112306c>] kill_block_super+0x2c/0x50
 [<ffffffff81123f55>] deactivate_super+0x85/0xa0
 [<ffffffff8113c6b3>] mntput_no_expire+0x63/0xb0
 [<ffffffff8113d1de>] sys_umount+0x5e/0x390
 [<ffffffff81009f02>] system_call_fastpath+0x16/0x1b

I can format it outside of the phone but the phone won't accept it and reformats it.

Really I'm looking for help in either working around/fixing it or investigating it further to actual produce a useful bug report.

Last edited by FeatherMonkey (2010-05-10 11:02:51)

FeatherMonkey

Member

Registered: 2007-02-26

Posts: 313

Re: Is this regression on flash devices? (exceeds device capacity)

Well not sure I've got anything else to add I'm seeing a few bits about this(Soe similar some about resume) some old some new, perhaps some one can tell me with the above output

<snip>INFO: task umount:5463 blocked for more than 120 seconds.</snip> which is repeatable should I attempt to report this upsteam or take it to the Arch bug tracker?