Re: [LKP] [lkp] [spi] 2baed30cb3: BUG: scheduling while atomic: systemd-udevd/134/0x00000002

[Sudip Mukherjee]

On Mon, Jan 25, 2016 at 10:15:27AM +0800, Huang, Ying wrote:
> Sudip Mukherjee <sudipm.mukherjee@xxxxxxxxx> writes:
> 
> > Hi Huang, Ying,
> > On Thu, Jan 21, 2016 at 11:36:52AM +0530, Sudip Mukherjee wrote:
> >> On Thu, Jan 21, 2016 at 01:47:10PM +0800, Huang, Ying wrote:
> >> > Sudip Mukherjee <sudipm.mukherjee@xxxxxxxxx> writes:
> >> > 
> >> > > On Wed, Jan 20, 2016 at 01:00:40PM +0800, Huang, Ying wrote:
> >> > >> Sudip Mukherjee <sudipm.mukherjee@xxxxxxxxx> writes:
> >> > >> 
> >> > >> > On Wed, Jan 20, 2016 at 08:44:37AM +0800, kernel test robot wrote:
> >> <snip>
> >> > >
> >> > > I am not able to reproduce this. Tested just with the kernel and
> >> > > yocto-minimal-i386.cgz filesystem and it booted properly.
> >> > >
> >> > > I guess I need atleast your job file to reproduce this.
> >> > 
> >> > This is a boot test so I did not attached the job file.  But the test
> >> > result may depends on specific root file system.  For example, the
> >> > process when BUG report is always systemd-udevd.  Maybe you need a
> >> > systemd based root file system.
> >> 
> >> So silly of me. Since you said 2baed30cb3, so i kept looking at that
> >> patch.
> >> Can you please test again after reverting:
> >> ebd43516d387 ("Staging: panel: usleep_range is preferred over udelay")
> >> 
> >> If it solves the problem then I will submit a formal patch.
> >
> > Did you get a chance to test it?
> 
> Sorry for late, as the new report just sent.  Reverting the specified
> commit can solve the issue.

Thanks for the confirmation. But I am still wondering how commit
2baed30cb3 can bring out some issues with ebd43516d387 ? AFAICS they are
totally unrelated.

regards
sudip