----- long...@redhat.com wrote:
> On 01/11/2019 12:06 AM, Zhenzhong Duan wrote: > > > > > > On 2019/1/10 22:43, Waiman Long wrote: > >> On 01/10/2019 03:02 AM, Zhenzhong Duan wrote: > >>> Hi Maintainer, > >>> > >>> > >>> There is a question confused me for days. Appreciate an answer. > >>> > >>> In below code, the comment says we never have more than 4 nested > >>> contexts. > >>> > >>> What happen if debug and mce exceptions nest with the four, or we > >>> ensure it never happen? > >>> > >>> > >>> /* > >>> * Per-CPU queue node structures; we can never have more than 4 > nested > >>> * contexts: task, softirq, hardirq, nmi. > >>> * > >>> * Exactly fits one 64-byte cacheline on a 64-bit architecture. > >>> * > >>> * PV doubles the storage and uses the second cacheline for PV > state. > >>> */ > >>> static DEFINE_PER_CPU_ALIGNED(struct qnode, qnodes[MAX_NODES]); > >>> > >> Yes, both debug and mce exceptions are some kind of NMIs. So > >> theoretically, it is possible to have more than four. Are you aware > of > >> any debug and MCE exception handlers that need to take a spinlock > for > >> synchronization? > > Not for debug exception, for MCE exception handler I found below > two: > > > > do_machine_check->mce_report_event->schedule_work > > do_machine_check->force_sig->force_sig_info > > > > schedule_work() and force_sig_info() take spinlocks. > > -- > > Thanks > > Zhenzhong > > The comment for do_machine_scheck() has already state that: > > * This is executed in NMI context not subject to normal locking > rules. This > * implies that most kernel services cannot be safely used. Don't > even > * think about putting a printk in there! > > So even if it doesn't exceed the MAX_NODES limit, it could hit > deadlock > and other kind of locking hazards. So supporting MCE is not a reason > strong enough to extend MAX_NODES. Agree. > > In hindsight, we should have added a "BUG_ON(idx >= MAX_NODES);" > check > to avoid silent corruption because of that issue. Looks a good idea if it's hard to avoid using spinlock in MCE handler. Thanks Zhenzhong
