On Sun, May 30, Vojtech Pavlik wrote: > On Sun, May 30, 2004 at 02:52:27PM -0400, Alan Stern wrote: > > On Sun, 30 May 2004, Vojtech Pavlik wrote: > > > > > Okay, so can we all agree on this one? > > > > I don't know if it's really needed, but for consistency you might want to > > make a similar change around the call to usb_stor_Bulk_transport() near > > the start of isd200_invoke_transport() in isd200.c. > > > > Apart from that minor point, this looks okay to me. > > Ok, so here goes the final version. Greg, please apply. > > -- > Vojtech Pavlik > SuSE Labs, SuSE CR
> [EMAIL PROTECTED], 2004-05-30 22:43:21+02:00, [EMAIL PROTECTED] > usb: Based on a 2.4 patch from [EMAIL PROTECTED], this patch > serializes usb storage and usbfs operation, so that usbfs > cannot disturb storage by seemingly harmless control reads. > > Signed-off-by: Vojtech Pavlik <[EMAIL PROTECTED]> > > > transport.c | 11 +++++++++++ > isd200.c | 4 ++++ > 2 files changed, 15 insertions(+) This patch causes deadlocks if you unplug and replug an usb-storage device. Can someone please have a look at this? khubd D 00000000 0 19 1 32 3 (L-TLB) Call trace: [<c000a748>] __switch_to+0x48/0x70 [<c02bbb10>] schedule+0x19c/0x5bc [<c02bc1b0>] wait_for_completion+0x7c/0xec [<d226cda0>] scsi_wait_req+0x70/0xb8 [scsi_mod] [<d226d838>] scsi_probe_and_add_lun+0x138/0xa68 [scsi_mod] [<d226e354>] scsi_scan_target+0x74/0x468 [scsi_mod] [<d226e7c4>] scsi_scan_channel+0x7c/0x9c [scsi_mod] [<d226e8bc>] scsi_scan_host_selected+0xd8/0x17c [scsi_mod] [<d223e950>] storage_probe+0x5b4/0x940 [usb_storage] [<c0205f40>] usb_probe_interface+0x70/0x90 [<c0190a24>] bus_match+0x50/0x94 [<c0190ac0>] device_attach+0x58/0xd0 [<c0190c18>] bus_add_device+0x78/0xd8 [<c018f670>] device_add+0xb0/0x19c [<c020cae8>] usb_set_configuration+0x2a0/0x3b0 usb-storage D 00000000 0 11359 1 11360 10647 (L-TLB) Call trace: [<c000a748>] __switch_to+0x48/0x70 [<c02bbb10>] schedule+0x19c/0x5bc [<c02bb5f4>] __down+0xc0/0x120 [<d223d08c>] usb_stor_invoke_transport+0x324/0x338 [usb_storage] [<d223cae8>] usb_stor_transparent_scsi_command+0x1c/0xbc [usb_storage] [<d223eff8>] usb_stor_control_thread+0x1dc/0x258 [usb_storage] [<c0009c90>] kernel_thread+0x44/0x60 scsi_eh_1 D 00000000 0 11360 1 11359 (L-TLB) Call trace: [<c000a748>] __switch_to+0x48/0x70 [<c02bbb10>] schedule+0x19c/0x5bc [<c02bc1b0>] wait_for_completion+0x7c/0xec [<d223c66c>] command_abort+0x7c/0xf4 [usb_storage] [<d226ac78>] scsi_error_handler+0x600/0xe6c [scsi_mod] [<c0009c90>] kernel_thread+0x44/0x60 -- USB is for mice, FireWire is for men! sUse lINUX ag, nÃRNBERG
usb-dell-patch-4-ul.deadlock.txt.gz
Description: application/gunzip
