On Thu, Sep 02, 2021 at 10:22:23AM -0600, Jens Axboe wrote:
> Hi,
>
> Booting current -git yields the below splat. I'm assuming this is
> related to the new RT stuff, where spin_lock() can sleep. This obviously
> won't fly off IPI.
You want to turn off PROVE_RAW_LOCK_NESTING for the moment.
On 9/2/21 12:56 PM, Matthew Wilcox wrote:
> On Thu, Sep 02, 2021 at 10:22:23AM -0600, Jens Axboe wrote:
>> Hi,
>>
>> Booting current -git yields the below splat. I'm assuming this is
>> related to the new RT stuff, where spin_lock() can sleep. This obviously
>> won't fly off IPI.
>
> You want to turn off PROVE_RAW_LOCK_NESTING for the moment.
Because?
--
Jens Axboe
On Thu, Sep 02, 2021 at 12:59:33PM -0600, Jens Axboe wrote:
> On 9/2/21 12:56 PM, Matthew Wilcox wrote:
> > On Thu, Sep 02, 2021 at 10:22:23AM -0600, Jens Axboe wrote:
> >> Hi,
> >>
> >> Booting current -git yields the below splat. I'm assuming this is
> >> related to the new RT stuff, where spin_lock() can sleep. This obviously
> >> won't fly off IPI.
> >
> > You want to turn off PROVE_RAW_LOCK_NESTING for the moment.
>
> Because?
help
Enable the raw_spinlock vs. spinlock nesting checks which ensure
that the lock nesting rules for PREEMPT_RT enabled kernels are
not violated.
NOTE: There are known nesting problems. So if you enable this
option expect lockdep splats until these problems have been fully
addressed which is work in progress. This config switch allows to
identify and analyze these problems. It will be removed and the
check permanently enabled once the main issues have been fixed.
If unsure, select N.
On 9/2/21 1:02 PM, Matthew Wilcox wrote:
> On Thu, Sep 02, 2021 at 12:59:33PM -0600, Jens Axboe wrote:
>> On 9/2/21 12:56 PM, Matthew Wilcox wrote:
>>> On Thu, Sep 02, 2021 at 10:22:23AM -0600, Jens Axboe wrote:
>>>> Hi,
>>>>
>>>> Booting current -git yields the below splat. I'm assuming this is
>>>> related to the new RT stuff, where spin_lock() can sleep. This obviously
>>>> won't fly off IPI.
>>>
>>> You want to turn off PROVE_RAW_LOCK_NESTING for the moment.
>>
>> Because?
>
> help
> Enable the raw_spinlock vs. spinlock nesting checks which ensure
> that the lock nesting rules for PREEMPT_RT enabled kernels are
> not violated.
>
> NOTE: There are known nesting problems. So if you enable this
> option expect lockdep splats until these problems have been fully
> addressed which is work in progress. This config switch allows to
> identify and analyze these problems. It will be removed and the
> check permanently enabled once the main issues have been fixed.
>
> If unsure, select N.
This isn't a new text, it's been like that for quite a while. The warning
is new on -git with that option.
--
Jens Axboe
On Thu, Sep 02 2021 at 13:05, Jens Axboe wrote:
> On 9/2/21 1:02 PM, Matthew Wilcox wrote:
>> NOTE: There are known nesting problems. So if you enable this
>> option expect lockdep splats until these problems have been fully
>> addressed which is work in progress. This config switch allows to
>> identify and analyze these problems. It will be removed and the
>> check permanently enabled once the main issues have been fixed.
>>
>> If unsure, select N.
>
> This isn't a new text, it's been like that for quite a while. The warning
> is new on -git with that option.
I'm looking into it...