NvDrv/nvhost-as-gpu: Ensure that the object passed to MapBufferEx has already been allocated.

Also added a consistency check and a comment for the case when the object id is different than its handle. The real nvservices doesn't make a distinction between ids and handles, each object gets an unique handle which doubles as its id.
This commit is contained in:
Subv 2018-04-23 11:10:00 -05:00
parent e4bd0bddea
commit 0d6eafe11a
1 changed files with 10 additions and 0 deletions

View File

@ -108,6 +108,16 @@ u32 nvhost_as_gpu::MapBufferEx(const std::vector<u8>& input, std::vector<u8>& ou
auto object = nvmap_dev->GetObject(params.nvmap_handle);
ASSERT(object);
// We can only map objects that have already been assigned a CPU address.
ASSERT(object->status == nvmap::Object::Status::Allocated);
ASSERT(params.buffer_offset == 0);
// The real nvservices doesn't make a distinction between handles and ids, and
// object can only have one handle and it will be the same as its id. Assert that this is the
// case to prevent unexpected behavior.
ASSERT(object->id == params.nvmap_handle);
auto& gpu = Core::System::GetInstance().GPU();
if (params.flags & 1) {