usb_remove_hcd calls free_irq on kernel 3.3
Brought to you by:
m_singer
Menu
▾
▴
#6 usb_remove_hcd calls free_irq on kernel 3.3
why does it do that? we don't use any irq.
[ 114.278232] usb_vhci_hcd usb_vhci_hcd.0: remove, state 1
[ 114.279356] usb usb2: USB disconnect, device number 1
[ 114.279390] usb 2-1: USB disconnect, device number 2
[ 114.309283] usb_vhci_hcd usb_vhci_hcd.0: stopped
[ 114.309292] ------------[ cut here ]------------
[ 114.309298] WARNING: at kernel/irq/manage.c:1163 __free_irq+0xa3/0x240()
[ 114.309304] Hardware name: VirtualBox
[ 114.309306] Trying to free already-free IRQ 0
[ 114.309307] Modules linked in: usb_vhci_iocifc(O) usb_vhci_hcd(O) be2iscsi iscsi_boot_sysfs bnx2i cnic uio cxgb4i cxgb4 lockd cxgb3i cxgb3 mdio sunrpc libcxgbi ib_iser rdma_cm ib_addr iw_cm ib_cm ib_sa ib_mad ib_core iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack ip6table_filter ip6_tables ppdev microcode snd_intel8x0 snd_ac97_codec ac97_bus snd_pcm snd_page_alloc i2c_piix4 i2c_core parport_pc virtio_net snd_timer snd soundcore parport [last unloaded: scsi_wait_scan]
[ 114.309353] Pid: 1130, comm: lt-virtual_devi Tainted: G O 3.3.0-0.rc5.git3.1.fc17.x86_64 #1
[ 114.309356] Call Trace:
[ 114.309369] [<ffffffff81060d6f>] warn_slowpath_common+0x7f/0xc0
[ 114.309373] [<ffffffff81060e66>] warn_slowpath_fmt+0x46/0x50
[ 114.309376] [<ffffffff8110bda5>] ? __free_irq+0x55/0x240
[ 114.309379] [<ffffffff8110bdf3>] __free_irq+0xa3/0x240
[ 114.309381] [<ffffffff8110bfe4>] free_irq+0x54/0xc0
[ 114.309385] [<ffffffff81486020>] usb_remove_hcd+0x130/0x170
[ 114.309390] [<ffffffffa02b6017>] vhci_hcd_remove+0x1a7/0x200 [usb_vhci_hcd]
[ 114.309394] [<ffffffff81413b27>] platform_drv_remove+0x17/0x20
[ 114.309397] [<ffffffff814120ac>] __device_release_driver+0x7c/0xe0
[ 114.309400] [<ffffffff8141213e>] device_release_driver+0x2e/0x40
[ 114.309403] [<ffffffff814117f2>] bus_remove_device+0x102/0x180
[ 114.309406] [<ffffffff8140ea87>] device_del+0x137/0x1c0
[ 114.309409] [<ffffffff8141405d>] platform_device_del+0x1d/0x90
[ 114.309412] [<ffffffff81414436>] platform_device_unregister+0x16/0x30
[ 114.309416] [<ffffffffa02b5c34>] usb_vhci_hcd_unregister+0x34/0x80 [usb_vhci_hcd]
[ 114.309419] [<ffffffffa02bd8f7>] device_release+0x37/0x90 [usb_vhci_iocifc]
[ 114.309426] [<ffffffff811bdda1>] fput+0x131/0x280
[ 114.309430] [<ffffffff811b8a49>] filp_close+0x69/0xa0
[ 114.309433] [<ffffffff81065255>] put_files_struct+0x115/0x3e0
[ 114.309436] [<ffffffff81065195>] ? put_files_struct+0x55/0x3e0
[ 114.309439] [<ffffffff810655e2>] exit_files+0x52/0x60
[ 114.309441] [<ffffffff81065b45>] do_exit+0x1a5/0xaa0
[ 114.309445] [<ffffffff810a2c0f>] ? local_clock+0x6f/0x80
[ 114.309449] [<ffffffff8169dbd0>] ? _raw_spin_unlock_irq+0x30/0x50
[ 114.309452] [<ffffffff8106678c>] do_group_exit+0x4c/0xc0
[ 114.309455] [<ffffffff8107a406>] get_signal_to_deliver+0x2b6/0x820
[ 114.309460] [<ffffffff81019288>] do_signal+0x68/0x7c0
[ 114.309463] [<ffffffffa02bf05e>] ? device_ioctl+0xe/0x220 [usb_vhci_iocifc]
[ 114.309467] [<ffffffff811d0339>] ? do_vfs_ioctl+0x99/0x5a0
[ 114.309470] [<ffffffff816a6971>] ? sysret_signal+0x5/0x4e
[ 114.309472] [<ffffffff81019a90>] do_notify_resume+0x90/0xc0
[ 114.309475] [<ffffffff816a6c22>] int_signal+0x12/0x17
[ 114.309477] ---[ end trace a3d73769d312b0af ]---
[ 114.309565] usb_vhci_hcd usb_vhci_hcd.0: USB bus 2 deregistered
Just before the call to usb_remove_hcd in vhci_hcd.c
i did a
hcd->irq=-1;
and it fixed this problem. (at least superficially).
I learned about it here
http://comments.gmane.org/gmane.linux.usb.general/59317
and
with references here
https://access.redhat.com/site/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Performance_Tuning_Guide/s-cpu-irq.html
Thank you very much for your help.
I'll use this to fix this bug in the next release.
Thank you,
Michael
This was a bug in the kernel. It is fixed already.