[Linux-Xtensa] USB host on linux-xtensa

Piet Delaney piet.delaney at gmail.com
Thu May 14 10:09:27 PDT 2009


Hi Sanu:

  How about just trying the 2.6.29-smp kernel, select 
CONFIG_CC_OPTIMIZE_FOR_DEBUGGING in the
kernel hacking section, placing a breakpoint at panic or the spinlock 
bad code and simply look the problem
more closely. Marc and I don't have a system  that we can exercise USB 
interrupts to reproduce this. It's
not very easy to understand a problem with a back trace that formal 
parameters being passed to functsions,
local parameters, and browsing the data structures involved in the 
problem. I like ddd for getting a quick and
clear view of whats going on. Presenting that in an hour ...

-piet
> Hi,
>
>   Has anyone tried enabling the USB support for xtensa-linux 2.6.24 ? I did the hook ups and when I run ,I see that the device descriptor request is reaching the device (the first eigth bytes after reset) and I have verified the data on the device side. I see that my target system (custom-xtensa-cpu) crashes and I get the following error
> -------------------------------------------------------------------------------------------
> BUG: spinlock bad magic on CPU#0, khubd/8
> Unable to handle kernel paging request at virtual address 000001d2
>  pc = d00844f4, ra = 900845b0
> Oops: sig: 11 [#1]
>
> a00: 900845b0 d3849a40 d3849e24 d012bd5c 00000002 00000000 000002b8 00000016
> a08: 900844e6 d38499f0 0000002d d012bd5c 00000004 d381d2b8 00000008 80008048
> pc: d00844f4, ps: 00060613, depc: 00000018, excvaddr: 000001d2
> lbeg: d00fca70, lend: d00fca7e lcount: 00000000, sar: 0000001e
>
> Stack: 00000002 00000000 000002b8 00000016 90007728 d3849a90 00060403 00000001
>        00000001 00000005 00060003 d014ed94 00000000 ffffffff d06798b4 d24494c0
>        900ab25c d3849ab0 d3849e20 00060003 d06798b4 00000000 d3848000 ffffffff
> Call Trace:
>  [<d00fe1f2>] _spin_lock_irqsave+0xe/0x10
>  [<d0007728>] complete+0x10/0x34
>  [<d00ab25c>] usb_api_blocking_completion+0x10/0x14
>  [<d00a9c1e>] usb_hcd_giveback_urb+0x3e/0x70
>  [<d00b3288>] ehci_urb_done+0x68/0x74
>  [<d00b3989>] qh_completions+0x289/0x300
>  [<d00b436d>] ehci_work+0xa9/0x4a0
>  [<d00b66e0>] ehci_irq+0x1f8/0x6f0
>  [<d00aa2c4>] usb_hcd_irq+0x48/0x50
>  [<d00229d0>] handle_IRQ_event+0x1c/0x40
>  [<d002396d>] handle_edge_irq+0x95/0x108
>  [<d0002668>] do_IRQ+0x38/0x44
>  [<d0003efc>] do_interrupt+0x9c/0xcc
>  [<d0001af1>] common_exception_return+0x0/0x3f
>
> Kernel panic - not syncing: Fatal exception in interrupt
> ------------------------------------------------------------------------------------------------
>
>    The issue may not be with Linux, my target itself is not stable. If anyone has seen similar issue and knows a fix, please pass your comments.
>
> Thanks and regards
> Sanu
>
>
> _______________________________________________
> linux-xtensa mailing list
> linux-xtensa at linux-xtensa.org
> http://lists.linux-xtensa.org/mailman/listinfo/linux-xtensa
>
>   



More information about the linux-xtensa mailing list