panic: uhci_abort_xfer: not in process context (was Re: uplcom / ucom problems on RELENG_5)

[Mike Tancsa]

At 09:16 AM 02/04/2005, Ian Dowse wrote:
>In message <6.2.1.2.0.20050331074641.04f72eb8 at 64.7.153.2>, Mike Tancsa writes:
> >panic: uhci_abort_xfer: not in process context
>...
> >#6  0xc04d2c1f in usbd_abort_pipe (pipe=0xc19c7e00)
> >#7  0xc04c3cf5 in ucomstopread (sc=0x0)
> >#8  0xc04c38ba in ucomstop (tp=0xc171c000, flag=1)
> >#9  0xc054b393 in ttyflush (tp=0xc171c000, rw=1)
> >#10 0xc054974d in ttyinput (c=26, tp=0xc171c000)
>
>In this case I wonder if the transfer needs to be aborted at all,
>since it is just restarted immediately afterwards. Mike, maybe you
>could try the following patch? Sorry, I haven't tested this, so I
>may be missing something obvious.


So far so good!  I have been running with this patch for almost 24hrs and 
still no panic and also the modem attached to the serial device has not 
hung!  I am however getting those
putc to a clist with no reserved cblocks
putc to a clist with no reserved cblocks
putc to a clist with no reserved cblocks
putc to a clist with no reserved cblocks
putc to a clist with no reserved cblocks

spews still.

However, on the


[hifn]% uname -a
FreeBSD hifn.sentex.net 6.0-CURRENT FreeBSD 6.0-CURRENT #0: Fri Apr  1 
02:20:32 EST 
2005     mdtancsa at hifn.sentex.net:/usr/obj/usr/src/sys/recycle  i386
[hifn]%
I am not getting any of those warning/error messages and it has not crashed 
either.  But, its different hardware, and its just stock CURRENT.

I am going to restart the test on RELENG_5 to try out some newer drivers 
and let it run for another 24hrs.

         ---Mike