Bug 29202
| Summary: | 2.6.38-rc: fuse BUG, maybe related to detaching external usb drive without umounting | ||
|---|---|---|---|
| Product: | File System | Reporter: | Florian Mickler (florian) |
| Component: | Other | Assignee: | fs_other |
| Status: | CLOSED CODE_FIX | ||
| Severity: | normal | CC: | acpi-bugzilla, acpi_power-battery, caravena, florian, kovariadam, maciej.rutecki, miklos, phill, rjw, rockorequin, tj |
| Priority: | P1 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Kernel Version: | 2.6.38-rc3 | Subsystem: | |
| Regression: | Yes | Bisected commit-id: | |
| Bug Depends on: | |||
| Bug Blocks: | 27352 | ||
| Attachments: | Tested fix | ||
|
Description
Florian Mickler
2011-02-15 22:05:47 UTC
*** Bug 29352 has been marked as a duplicate of this bug. *** New stack trace from -rc5: ------------[ cut here ]------------ kernel BUG at fs/inode.c:1421! invalid opcode: 0000 [#2] PREEMPT SMP last sysfs file: /sys/devices/pci0000:00/0000:00:19.0/net/eth0/carrier CPU 3 Modules linked in: aesni_intel cryptd aes_x86_64 aes_generic ses enclosure usb_storage uas uvcvideo videodev v4l2_compat_ioctl32 snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_mixer_oss arc4 ecb snd_hda_codec_hdmi iwlagn i915 iwlcore drm_kms_helper snd_hda_codec_idt drm mac80211 ehci_hcd snd_hda_intel ppdev snd_hda_codec e1000e serio_raw firewire_ohci pcspkr psmouse i2c_algo_bit cfg80211 parport_pc usbcore ipv6 dell_wmi dell_laptop firewire_core sg lp iTCO_wdt sdhci_pci iTCO_vendor_support sdhci video evdev dcdbas sparse_keymap intel_agp i2c_i801 parport intel_gtt i2c_core mmc_core ac processor button container intel_ips battery wmi snd_hwdep snd_pcm rfkill snd_timer snd soundcore snd_page_alloc crc_itu_t fuse ext4 mbcache jbd2 crc16 sr_mod cdrom sd_mod ahci libahci libata scsi_mod Pid: 4313, comm: umount Tainted: G D W 2.6.38-rc5-mainline #1 Dell Inc. Latitude E6410/0667CC RIP: 0010:[<ffffffff8114c5f0>] [<ffffffff8114c5f0>] iput+0x240/0x2a0 RSP: 0018:ffff880070bcbda8 EFLAGS: 00010202 RAX: 0000000000000000 RBX: ffff880117a70b80 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffffffff81562580 RDI: ffff880117a70b80 RBP: ffff880070bcbdb8 R08: 2222222222222222 R09: 2222222222222222 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 R13: 0000000000000083 R14: ffff880117a70aa0 R15: ffff880117828a80 FS: 00007f35b184a740(0000) GS:ffff8800db4c0000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000001c0f0a8 CR3: 0000000116d43000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process umount (pid: 4313, threadinfo ffff880070bca000, task ffff88008ee6b780) Stack: ffff880117a70a80 0000000000000000 ffff880070bcbe08 ffffffff811668a7 ffff880070bcbdf8 ffffffffa0026e80 ffff880070bcbf28 ffff880117a70a80 0000000000000083 ffff880117a70aa0 ffff880070bcbf01 ffff88011577d000 Call Trace: [<ffffffff811668a7>] __blkdev_put+0x137/0x1c0 [<ffffffff81166a01>] blkdev_put+0xd1/0x170 [<ffffffff811354c9>] kill_block_super+0x49/0x80 [<ffffffffa012198d>] fuse_kill_sb_blk+0x4d/0x60 [fuse] [<ffffffff81135825>] deactivate_locked_super+0x45/0x70 [<ffffffff81136615>] deactivate_super+0x45/0x60 [<ffffffff81150b8a>] mntput_no_expire+0x9a/0xe0 [<ffffffff81151687>] sys_umount+0x67/0x390 [<ffffffff8100bdd2>] system_call_fastpath+0x16/0x1b Code: 41 00 83 05 c6 97 5f 00 01 48 89 05 6b 8a 41 00 48 89 42 08 48 89 93 88 00 00 00 48 c7 83 90 00 00 00 40 50 56 81 e9 4f fe ff ff <0f> 0b be 6e 05 00 00 48 c7 c7 9a 09 48 81 e8 cd b3 f0 ff 48 8b RIP [<ffffffff8114c5f0>] iput+0x240/0x2a0 RSP <ffff880070bcbda8> dell-wmi: Received unknown WMI event (0x11) [drm:ironlake_crtc_disable] *ERROR* failed to disable transcoder ---[ end trace d096156fa0643175 ]--- Hmm. That may be fixed in the current Linus' tree, by patches that went in after -rc5. This does not happen with vfat drive. It is not fixed in -rc5. Well, clearly FUSE is involved then. Reassigning. It still would be good to verify if -rc6 still has the problem, when it's out, there have been a few fixes in the VFS area merged since -rc5. See: bug #29492 , It is not fixed in -rc5 It is not fixed in -rc6. Created attachment 48872 [details]
Tested fix
Handled-By : Miklos Szeredi <miklos@szeredi.hu> Patch : https://bugzilla.kernel.org/attachment.cgi?id=48872 *** Bug 29792 has been marked as a duplicate of this bug. *** merged in .38-rc7: commit 3c522cedb572bb8d2e4867f358bdaa7d0c53d88c Author: Miklos Szeredi <mszeredi@suse.cz> Date: Thu Feb 24 15:45:41 2011 +0100 block: fix refcounting in BLKBSZSET *** Bug 29492 has been marked as a duplicate of this bug. *** |