Re: [PATCH 08/11] dma-buf: Restart reservation_object_wait_timeout_rcu() after writes

2016-09-23 Thread Daniel Vetter
On Mon, Aug 29, 2016 at 08:08:31AM +0100, Chris Wilson wrote: > In order to be completely generic, we have to double check the read > seqlock after acquiring a reference to the fence. If the driver is > allocating fences from a SLAB_DESTROY_BY_RCU, or similar freelist, then > within an RCU grace

[PATCH 08/11] dma-buf: Restart reservation_object_wait_timeout_rcu() after writes

2016-08-29 Thread Chris Wilson
In order to be completely generic, we have to double check the read seqlock after acquiring a reference to the fence. If the driver is allocating fences from a SLAB_DESTROY_BY_RCU, or similar freelist, then within an RCU grace period a fence may be freed and reallocated. The RCU read side critical