Messages in this thread |  | | Date | Wed, 6 Feb 2013 18:29:01 -0800 | Subject | Re: [PATCH 2/3] stop_machine: dequeue work before signal completion | From | Tejun Heo <> |
| |
Hello,
On Wed, Feb 6, 2013 at 6:21 PM, Namhyung Kim <namhyung@kernel.org> wrote: >> Why does this matter? It's inside spinlock. What's being made better >> by this change? > > IIUC the work should be deleted from the list, otherwise it'd trigger > BUG_ON when the cpu gets online again.
Ah, okay, the original code was missing list_del_init(), so we end up with trashy work list if CPU goes down while there are pending work items which will trigger BUG_ON() later when the CPU comes back on.
Hillf, can you please redo the patch description? I can't tell what the patch is about from the description at all. If it's a bug fix, describe the bug and maybe accompany with oops trace if possible, and then describe how it's fixed.
Thanks.
-- tejun
|  |