Re: [PATCH v2 2/5] cpu: CPU notifier error injection

[Akinobu Mita]

2011/7/18 Américo Wang <xiyou.wangcong@xxxxxxxxx>:
> On Mon, Jul 18, 2011 at 9:16 AM, Akinobu Mita <akinobu.mita@xxxxxxxxx> wrote:
> ...
>> diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
>> index 72e2384..cf0b4b7 100644
>> --- a/Documentation/feature-removal-schedule.txt
>> +++ b/Documentation/feature-removal-schedule.txt
>> @@ -583,3 +583,11 @@ Why:       Superseded by the UVCIOC_CTRL_QUERY ioctl.
>>  Who:   Laurent Pinchart <laurent.pinchart@xxxxxxxxxxxxxxxx>
>>
>>  ----------------------------
>> +
>> +What:  Support for CPU_NOTIFIER_ERROR_INJECT (cpu-notifier-error-inject.ko)
>> +When:  3.3
>> +Why:   Replaced with more generic debugfs interface in
>> +       /sys/kernel/debug/cpu-notifier-error-inject/
>> +Who:   Akinobu Mita <akinobu.mita@xxxxxxxxx>
>
> Hi, Akinobu,
>
> I think it is okay to just remove it by this patchset, no need to defer it.

The only drawback is that new machanism has no way to specify notifier
priority and the priority is hard-coded to be -1.  So it can only inject
error at the tail of notifier chain.

My plan to fix that was the kernel parameters.  But it's not flexible
as it needs to reboot for changing the priority.

So I'm now considering keeping this kernel module style instead of
initializing in late_initcall().
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/