Bug 769664 - Kernel corrupts ext4 file system operation on resume from hibernate
Summary: Kernel corrupts ext4 file system operation on resume from hibernate
Keywords:
Status: CLOSED DUPLICATE of bug 744275
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 16
Hardware: i686
OS: Linux
unspecified
high
Target Milestone: ---
Assignee: Kernel Maintainer List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2011-12-21 16:29 UTC by Andrew Duggan
Modified: 2011-12-21 19:38 UTC (History)
5 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2011-12-21 19:38:13 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)

Description Andrew Duggan 2011-12-21 16:29:41 UTC
Description of problem:
After resume from hibernate new files cannot be create on the file system.  Applications crash that get failed file create failures. 

Version-Release number of selected component (if applicable):
Fedora 16 kernel 3.1.5-10.f16.i686.PAE

How reproducible:
Always

Steps to Reproduce:
1. Hibernate /resume
2. echo a /tmp a.txt
3. 
  
Actual results:
   -bash: /tmp/a.txt: invalid argument

and 

[  717.070748] ------------[ cut here ]------------
[  717.070767] WARNING: at fs/inode.c:884 unlock_new_inode+0x42/0x50()
[  717.070772] Hardware name: MM061                           
[  717.070776] Modules linked in: ppdev parport_pc lp parport fuse ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 ip6table_filter ip6_tables nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack snd_hda_codec_idt arc4 snd_hda_intel b43 snd_hda_codec r852 mac80211 sm_common snd_hwdep nand snd_seq snd_seq_device b44 iTCO_wdt nand_ids snd_pcm cfg80211 iTCO_vendor_support mii nand_ecc r592 bcma memstick mtd snd_timer dell_wmi dell_laptop i2c_i801 rfkill snd ssb microcode joydev uinput soundcore sparse_keymap dcdbas snd_page_alloc sdhci_pci sdhci mmc_core firewire_ohci firewire_core crc_itu_t wmi i915 drm_kms_helper drm i2c_algo_bit i2c_core video [last unloaded: scsi_wait_scan]
[  717.070885] Pid: 1899, comm: bash Tainted: G        W   3.1.5-10.fc16.i686.PAE #1
[  717.070890] Call Trace:
[  717.070900]  [<c092086a>] ? printk+0x2d/0x2f
[  717.070909]  [<c045b452>] warn_slowpath_common+0x72/0xa0
[  717.070916]  [<c054e982>] ? unlock_new_inode+0x42/0x50
[  717.070923]  [<c054e982>] ? unlock_new_inode+0x42/0x50
[  717.070930]  [<c045b4a2>] warn_slowpath_null+0x22/0x30
[  717.070937]  [<c054e982>] unlock_new_inode+0x42/0x50
[  717.070945]  [<c059f3ba>] ext4_new_inode+0xa3a/0xf30
[  717.070953]  [<c05d9053>] ? jbd2__journal_start+0xb3/0xf0
[  717.070961]  [<c05aa601>] ext4_create+0xb1/0x120
[  717.070967]  [<c054495b>] vfs_create+0x9b/0x100
[  717.070973]  [<c054674f>] do_last+0x5bf/0x820
[  717.070979]  [<c0546a94>] path_openat+0xa4/0x350
[  717.070985]  [<c0546e51>] do_filp_open+0x31/0x80
[  717.070993]  [<c050e7ce>] ? handle_mm_fault+0x15e/0x2c0
[  717.071001]  [<c0655938>] ? strncpy_from_user+0x38/0x70
[  717.071043]  [<c05512ac>] ? alloc_fd+0x3c/0xe0
[  717.071050]  [<c0538aa6>] do_sys_open+0xe6/0x1b0
[  717.071057]  [<c0538b9e>] sys_open+0x2e/0x40
[  717.071064]  [<c09310df>] sysenter_do_call+0x12/0x28
[  717.071069] ---[ end trace bc24e69362c068fd ]---

Expected results:
Everything should just work

Additional info:

Those taints are totally bogus, I don't have anything in the kernel to taint it, no foreign modules.  

Happens on every 3.1 kernel I've tried in Fedora 16.  Same problem occurs on Fedora 15 with 2.6.40.x if and only if a windows partition is mounted when the hibernate and suspend occurs,  On 3.1.x it happens with only / and /boot mounted, plus all of the default cgroup and debugfs "mounts".  

Already disabled all the NFS in systemd just in case.

Comment 1 Josh Boyer 2011-12-21 19:38:13 UTC

*** This bug has been marked as a duplicate of bug 744275 ***


Note You need to log in before you can comment on or make changes to this bug.