On Thu, Sep 04, 2014 at 03:42:07PM +0800, Li Zefan wrote:
> On 2014/9/4 15:12, Li Zefan wrote:
> > Cc: Zhang Zhen <zhenzhang.zhang@xxxxxxxxxx>
> >
> > On 2014/9/4 7:29, Greg KH wrote:
> >> On Wed, Sep 03, 2014 at 04:25:30PM -0700, Greg KH wrote:
> >>> On Wed, Sep 03, 2014 at 04:01:22PM -0700, David Rientjes wrote:
> >>>> On Wed, 3 Sep 2014, gregkh@xxxxxxxxxxxxxxxxxxx wrote:
> >>>>
> >>>>>
> >>>>> The patch below does not apply to the 3.10-stable tree.
> >>>>
> >>>> It applies fine to linux-3.10.53.
> >>>
> >>> Well, yes, but it breaks the build :)
> >>>
> >>>>> If someone wants it applied there, or to any other stable or longterm
> >>>>> tree, then please email the backport, including the original git commit
> >>>>> id to <stable@xxxxxxxxxxxxxxx>.
> >>>>>
> >>>>
> >>>> I'm assuming that this is a report that the build fails when it's applied,
> >>>> although that's not clear. It requires commit
> >>>> 0c740d0afc3b ("introduce for_each_thread() to replace the buggy
> >>>> while_each_thread()") to build.
> >>>
> >>> Ok, I'll queue that up before this one and see if that works better...
> >>
> >> Nope, that one doesn't apply cleanly, and the backport wasn't "obvious"
> >> so I'll have to keep this one out of 3.10-stable for now. Unless
> >> someone wants to provide me a backport of 0c740d0afc3b that works?
> >>
> >
> > Coincidently we're backporting another bugfix that needs the same prerequisites
> > as this one does.
> >
> > You can cherry-pick those upstream commits:
> >
> > 80628ca06c5d kernel/fork.c:copy_process(): unify CLONE_THREAD-or-thread_group_leader code
> > 0c740d0afc3b introduce for_each_thread() to replace the buggy while_each_thread()
> > 1da4db0cd5c8 oom_kill: change oom_kill.c to use for_each_thread()
> > 4449a51a7c28 vm_is_stack: use for_each_thread() rather then buggy while_each_thread()
> >
> > The first two commits are prerequisites, and 1da4db0cd5c8 fixes a bug
> > we encountered.
> >
> > Credit goes to Zhang Zhen.
> >
>
> We just figured out another two small bugfixes we'd better also backport,
> they were sent with 0c740d0afc3b and 1da4db0cd5c8 as a patchset.
>
> 80628ca06c5d kernel/fork.c:copy_process(): unify CLONE_THREAD-or-thread_group_leader code
> 0c740d0afc3b introduce for_each_thread() to replace the buggy while_each_thread()
> 1da4db0cd5c8 oom_kill: change oom_kill.c to use for_each_thread()
> ad96244179fb oom_kill: has_intersects_mems_allowed() needs rcu_read_lock()
> 4d4048be8a93 oom_kill: add rcu_read_lock() into find_lock_task_mm()
> 4449a51a7c28 vm_is_stack: use for_each_thread() rather then buggy while_each_thread()
Thanks for the list, I'll look into these after this next round of
stable kernels are released.
greg k-h
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html