bug in umass?

[Hans Petter Selasky]

Hi Alexander,

Does dumping core on a USB stick from KDB require any threads?

 From the USB point of view we are doing polling in "umass_cam_poll()".

--HPS

On 05/16/16 07:02, K. Macy wrote:
> I'm not able to complete a coredump in i915 to a USB key. The
> backtrace in the log looks like a bug in umass.
>
>
> May 15 21:57:10 beastie kernel: cmap[0]=0 cmap[1]=7f0000 cmap[2]=7f00
> cmap[3]=c4a000
> May 15 21:57:10 beastie kernel: end FB_INFO
> May 15 21:57:10 beastie kernel: drmn0: fb0: inteldrmfb frame buffer device
> May 15 21:57:10 beastie kernel: ..3%
> May 15 21:58:16 beastie syslogd: kernel boot file is /boot/kernel/kernel
> May 15 21:58:16 beastie kernel: trap_fatal() at trap_fatal+0x2d/frame
> 0xfffffe01e2fd6350
> May 15 21:58:16 beastie kernel: trap() at trap+0xc48/frame 0xfffffe01e2fd6690
> May 15 21:58:16 beastie kernel: trap_check() at trap_check+0x4a/frame
> 0xfffffe01e2fd66b0
> May 15 21:58:16 beastie kernel: calltrap() at calltrap+0x8/frame
> 0xfffffe01e2fd66b0
> May 15 21:58:16 beastie kernel: --- trap 0x9, rip =
> 0xffffffff80f5a950, rsp = 0xfffffe01e2fd6780, rbp = 0xfffffe01e2fd6810
> ---
> May 15 21:58:16 beastie kernel: __mtx_lock_flags() at
> __mtx_lock_flags+0xd0/frame 0xfffffe01e2fd6810
> May 15 21:58:16 beastie kernel: xpt_done_process() at
> xpt_done_process+0x495/frame 0xfffffe01e2fd68c0
> May 15 21:58:16 beastie kernel: xpt_done_td() at
> xpt_done_td+0x1c0/frame 0xfffffe01e2fd6930
> May 15 21:58:16 beastie kernel: fork_exit() at fork_exit+0x13b/frame
> 0xfffffe01e2fd69b0
> May 15 21:58:16 beastie kernel: fork_trampoline() at
> fork_trampoline+0xe/frame 0xfffffe01e2fd69b0
> May 15 21:58:16 beastie kernel: --- trap 0, rip = 0, rsp = 0, rbp = 0 ---
>