So far we had the approach of using a directed acyclic graph with the dma_resv obj.
This turned out to have many downsides, especially it means that every single driver and user of this interface needs to be aware of this restriction when adding fences. If the rules for the DAG are not followed then we end up with potential hard to debug memory corruption, information leaks or even elephant big security holes because we allow userspace to access freed up memory.
Since we already took a step back from that by always looking at all fences we now go a step further and stop dropping the shared fences when a new exclusive one is added.
Signed-off-by: Christian König christian.koenig@amd.com --- drivers/dma-buf/dma-resv.c | 16 +--------------- include/linux/dma-resv.h | 22 +++++----------------- 2 files changed, 6 insertions(+), 32 deletions(-)
diff --git a/drivers/dma-buf/dma-resv.c b/drivers/dma-buf/dma-resv.c index 16ad98fd3130..e04afc30bc0a 100644 --- a/drivers/dma-buf/dma-resv.c +++ b/drivers/dma-buf/dma-resv.c @@ -356,35 +356,21 @@ EXPORT_SYMBOL(dma_resv_replace_fences); * @fence: the exclusive fence to add * * Add a fence to the exclusive slot. @obj must be locked with dma_resv_lock(). - * Note that this function replaces all fences attached to @obj, see also - * &dma_resv.fence_excl for a discussion of the semantics. + * See also &dma_resv.fence_excl for a discussion of the semantics. */ void dma_resv_add_excl_fence(struct dma_resv *obj, struct dma_fence *fence) { struct dma_fence *old_fence = dma_resv_excl_fence(obj); - struct dma_resv_list *old; - u32 i = 0;
dma_resv_assert_held(obj);
- old = dma_resv_shared_list(obj); - if (old) - i = old->shared_count; - dma_fence_get(fence);
write_seqcount_begin(&obj->seq); /* write_seqcount_begin provides the necessary memory barrier */ RCU_INIT_POINTER(obj->fence_excl, fence); - if (old) - old->shared_count = 0; write_seqcount_end(&obj->seq);
- /* inplace update, no shared fences */ - while (i--) - dma_fence_put(rcu_dereference_protected(old->shared[i], - dma_resv_held(obj))); - dma_fence_put(old_fence); } EXPORT_SYMBOL(dma_resv_add_excl_fence); diff --git a/include/linux/dma-resv.h b/include/linux/dma-resv.h index 20e13f36710a..ecb697d4d861 100644 --- a/include/linux/dma-resv.h +++ b/include/linux/dma-resv.h @@ -93,23 +93,11 @@ struct dma_resv { * * The exclusive fence, if there is one currently. * - * There are two ways to update this fence: - * - * - First by calling dma_resv_add_excl_fence(), which replaces all - * fences attached to the reservation object. To guarantee that no - * fences are lost, this new fence must signal only after all previous - * fences, both shared and exclusive, have signalled. In some cases it - * is convenient to achieve that by attaching a struct dma_fence_array - * with all the new and old fences. - * - * - Alternatively the fence can be set directly, which leaves the - * shared fences unchanged. To guarantee that no fences are lost, this - * new fence must signal only after the previous exclusive fence has - * signalled. Since the shared fences are staying intact, it is not - * necessary to maintain any ordering against those. If semantically - * only a new access is added without actually treating the previous - * one as a dependency the exclusive fences can be strung together - * using struct dma_fence_chain. + * To guarantee that no fences are lost, this new fence must signal + * only after the previous exclusive fence has signalled. If + * semantically only a new access is added without actually treating the + * previous one as a dependency the exclusive fences can be strung + * together using struct dma_fence_chain. * * Note that actual semantics of what an exclusive or shared fence mean * is defined by the user, for reservation objects shared across drivers