Matthew Dharm wrote: 
> Did the patch Alan posted make any difference?
> 
> I've been getting several reports recently of problems with usb-storage and
> EHCI controllers.  If you unload the ehci driver, your device will run at
> 1.1 speeds -- can you try that and see if it works?
> 
> Matt
> 
> On Sun, Jan 11, 2004 at 07:46:29PM +0100, Lukas Postupa wrote:
> > Alan Cox wrote:
> > > With the various fixes people had been posting USB storage
> > > writing was still hanging repeatedly when doing a 20Gb rsync
> > > to usb-storage disks with a low memory system. Doing things
> > > like while(true) sync() made it hang even more often.
> > > 
> > > After a bit of digging the following seems to fix it
> > > 
> > > Not sure if 2.6 needs this as well.
> > 
> > I have similiar problems with kernel 2.6.0 on Intel architecture with
> > 512 MB Ram.
> > 
> > My Abit IT7-MAX2 2.0 mainboard has two USB - EHCI controllers:
> > 
> > 00:1d.7 USB Controller: Intel Corp. 82801DB USB EHCI Controller (rev 02)
> > 
> > 02:06.2 USB Controller: VIA Technologies, Inc. USB 2.0 (rev 51)
> > 
> > I'm using rsync to transfer my data to usb storage.
> > Connecting usb storage to one of the 4 ports of VIA EHCI controller and
> > then transferring data to it works good.
> > 
> > But connecting usb storage to one of the 6 ports of INTEL EHCI
> > controller and then transferring data to it, will hang up usb storage:
> > 
> > Buffer I/O error on device sda1, logical block 121479
> > lost page write due to I/O error on sda1
> > Buffer I/O error on device sda1, logical block 121480
> > lost page write due to I/O error on sda1
> > ehci_hcd 0000:00:1d.7: GetStatus port 2 status 001002 POWER sig=se0  CSC
> > hub 6-0:1.0: port 2, status 100, change 1, 12 Mb/s
> > usb 6-2: USB disconnect, address 4
> > usb 6-2: usb_disable_device nuking all URBs
> > usb 6-2: unregistering interface 6-2:1.0
> > usb-storage: storage_disconnect() called
> > usb-storage: usb_stor_stop_transport called
> > usb-storage: -- dissociate_dev
> > usb-storage: -- sending exit command to thread
> > usb-storage: *** thread awakened.
> > usb-storage: -- exit command received
> > usb-storage: -- usb_stor_release_resources finished
> > usb 6-2: unregistering device
> > 
> > VIA EHCI controller uses interrupt 21.
> > INTEL EHCI controller uses interrupt 23.
> > 
> > cat /proc/interrupts:
> > 
> >            CPU0       
> >   0:   16675721    IO-APIC-edge  timer
> >   1:       6876    IO-APIC-edge  i8042
> >   2:          0          XT-PIC  cascade
> >   9:          5   IO-APIC-level  acpi
> >  12:     191706    IO-APIC-edge  i8042
> >  14:     251218    IO-APIC-edge  ide0
> >  15:          1    IO-APIC-edge  ide1
> >  16:    1446615   IO-APIC-level  uhci_hcd, nvidia
> >  18:          0   IO-APIC-level  uhci_hcd, uhci_hcd
> >  19:        278   IO-APIC-level  uhci_hcd, uhci_hcd, EMU10K1
> >  21:     257509   IO-APIC-level  ehci_hcd
> >  22:       7640   IO-APIC-level  eth0
> >  23:     322432   IO-APIC-level  ehci_hcd
> > NMI:          0 
> > LOC:   16675149 
> > ERR:          0
> > MIS:          0

The patch Alan posted is written for kernel 2.4, but i am using kernel
2.6.

Unloading module ehci_hcd after usb storage hung up does not work;
rmmod "hangs".

I have to switch off usb storage and reboot.

Lukas

Attachment: signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil

Reply via email to