Messages in this thread | | | Date | Thu, 5 Jun 2014 17:24:30 +0300 | Subject | Re: [PATCH 5/9] perf/x86: implement cross-HT corruption bug workaround | From | Maria Dimakopoulou <> |
| |
On Thu, Jun 5, 2014 at 5:14 PM, Peter Zijlstra <peterz@infradead.org> wrote: > On Thu, Jun 05, 2014 at 04:11:50PM +0200, Peter Zijlstra wrote: >> > > x86_schedule_event() >> > > ->start_scheduling() >> > > spin_lock() >> > > ->get_event_constraints() >> > > intel_get_excl_constraints() >> > > kmalloc(.gfp=GFP_KERNEL) >> > > >> > > How can that ever work? >> >> > Are you saying it is illegal to call kmalloc() from >> > this context? >> >> Nobody will come and arrest you for it, so no. Broken though. GFP_KERNEL >> will attempt to sleep to wait for reclaim, and you're holding a >> spinlock.
Ok, got it.
> > Furthermore, even GFP_ATOMIC shouldn't be used because these are/should > be raw_spinlocks and the zone->lock is a regular spinlock. > > So please look into the preallocation thing.
We will preallocate.
| |