I've been trying to get USB up to test a audio device and just managed
to get it all working to some extent.  When using xmms to play audio
(usb audio module - oss soundcore) I see the following kernel messages
repeatedly, maybe once a second or so:


Oct 26 09:15:56 sea kernel: Debug: sleeping function called from illegal
context at mm/slab.c:1374
Oct 26 09:15:56 sea kernel: Call Trace:
Oct 26 09:15:56 sea kernel:  [<c013fd0b>] __kmem_cache_alloc+0x17b/0x180
Oct 26 09:15:56 sea kernel:  [<e0a28bb5>] ohci_urb_enqueue+0xb5/0x2d0
[ohci-hcd]
Oct 26 09:15:56 sea kernel:  [<e09b9017>] hcd_submit_urb+0x117/0x180
[usbcore]
Oct 26 09:15:56 sea kernel:  [<e09c9d20>] usb_hcd_operations+0x0/0x40
[usbcore]
Oct 26 09:15:56 sea kernel:  [<e09b993b>] usb_submit_urb+0x1db/0x250
[usbcore]
Oct 26 09:15:56 sea kernel:  [<e0a30b75>] usbout_completed+0xe5/0x110
[audio]
Oct 26 09:15:56 sea kernel:  [<e09b9409>] usb_hcd_giveback_urb+0x19/0x30
[usbcore]
Oct 26 09:15:56 sea kernel:  [<e0a28add>] dl_done_list+0x13d/0x160
[ohci-hcd]
Oct 26 09:15:56 sea kernel:  [<e0a2945b>] ohci_irq+0xfb/0x160 [ohci-hcd]
Oct 26 09:15:56 sea kernel:  [<e09b9442>] usb_hcd_irq+0x22/0x60
[usbcore]
Oct 26 09:15:56 sea kernel:  [<c010b8c5>] handle_IRQ_event+0x45/0x70
Oct 26 09:15:56 sea kernel:  [<c010bb6a>] do_IRQ+0xba/0x160
Oct 26 09:15:56 sea kernel:  [<c0107100>] default_idle+0x0/0x50
Oct 26 09:15:56 sea kernel:  [<c010a1ba>] common_interrupt+0x42/0x58
Oct 26 09:15:56 sea kernel: 
Oct 26 09:15:57 sea kernel: Debug: sleeping function called from illegal
context at mm/slab.c:1374
Oct 26 09:15:57 sea kernel: Call Trace:
Oct 26 09:15:57 sea kernel:  [<c013fd0b>] __kmem_cache_alloc+0x17b/0x180
Oct 26 09:15:57 sea kernel:  [<e0a28bb5>] ohci_urb_enqueue+0xb5/0x2d0
[ohci-hcd]
Oct 26 09:15:57 sea kernel:  [<e09b9017>] hcd_submit_urb+0x117/0x180
[usbcore]
Oct 26 09:15:57 sea kernel:  [<e09c9d20>] usb_hcd_operations+0x0/0x40
[usbcore]
Oct 26 09:15:57 sea kernel:  [<e09b993b>] usb_submit_urb+0x1db/0x250
[usbcore]
Oct 26 09:15:57 sea kernel:  [<e0a30b75>] usbout_completed+0xe5/0x110
[audio]
Oct 26 09:15:57 sea kernel:  [<e09b9409>] usb_hcd_giveback_urb+0x19/0x30
[usbcore]
Oct 26 09:15:57 sea kernel:  [<e0a28add>] dl_done_list+0x13d/0x160
[ohci-hcd]
Oct 26 09:15:57 sea kernel:  [<e0a2945b>] ohci_irq+0xfb/0x160 [ohci-hcd]
Oct 26 09:15:57 sea kernel:  [<e09b9442>] usb_hcd_irq+0x22/0x60
[usbcore]
Oct 26 09:15:57 sea kernel:  [<c010b8c5>] handle_IRQ_event+0x45/0x70
Oct 26 09:15:57 sea kernel:  [<c010bb6a>] do_IRQ+0xba/0x160
Oct 26 09:15:57 sea kernel:  [<c0107100>] default_idle+0x0/0x50
Oct 26 09:15:57 sea kernel:  [<c010a1ba>] common_interrupt+0x42/0x58

Hardware:
Athlon MP - AMD 760MP
USB audio - Sound Blaster Extigy

I also notice that usb audio dies when closing the Netscape browser. 
Has anyone seen this?  I haven't run across other trouble reports about
this, and it also occurs on 2.4.1[89] kernels.  rmmod'ing the USB layer
and reloading it fixes the problem until it gets killed again on 2.4. 
Looks like it might be the hub, but not sure yet.

-- 
Kevin