Re: Regression with a0747a859ef6 ("bdi: add error handle for bdi_debug_register")

[weiping zhang]

2017-12-15 9:44 GMT+08:00 Bruno Wolff III <bruno@xxxxxxxx>:
> On Fri, Dec 15, 2017 at 09:22:21 +0800,
>  weiping zhang <zwp10758@xxxxxxxxx> wrote:
>>
>>
>> Thanks your testing, but I cann't find WARN_ON in device_add_disk from
>> this boot1.log, could you help reproduce that issue? And does this issue
>> can be
>> triggered at every bootup ?
>
>
> I don't know what you need for the first question. When I am physically at
> the machine I can do test reboots. If you have something specific you want
> me to try I should be able to.
>
> Every time I boot with the problem commit, the boot never completes. However
> it does seem to get pretty far. I get multiple register dumps every time.
> After a while (a few minutes) I reboot to a wrking kernel.
>
> The output I included is from: journalctl -k -b -1
> If you think it would be better to see more than dmesg output let me know.
I just want to know WARN_ON WHAT in device_add_disk,
if bdi_register_owner return error code, it may fail at any step of following:

bdi_debug_root is NULL
bdi->debug_dir is NULL
bdi->debug_stats is NULL

so I want see the WARN_ON as you paste before, also my DEBUG log will help
to find which step fail.