gpu-paging: Support asynchronous paging

- Fully enables *_ASYNC API
- Allows page mapping to be overlapped with I/O, resulting in an 11% speedup
  to synchronous reads

Benchmarks, 1,000 iters, before:
gpu_paging_speed, write: 185.5ms +/- 3.58
gpu_paging_speed, read: 180.5ms +/- 1.42
gpu_paging_overhead_speed, write start: 183.3ms +/- 3.89
gpu_paging_overhead_speed, write finish: 3.4ms +/- 2.61
gpu_paging_overhead_speed, read start: 181.6ms +/- 3.34
gpu_paging_overhead_speed, read finish: 41.1ms +/- 2.69

Benchmarks, 1,000 iters, after:
gpu_paging_speed, write: 185.8ms +/- 3.70
gpu_paging_speed, read: 161.3ms +/- 0.97
gpu_paging_overhead_speed, write start: 38.9ms +/- 5.47
gpu_paging_overhead_speed, write finish: 3.1ms +/- 2.42
gpu_paging_overhead_speed, read start: 79.4 +/- 6.42
gpu_paging_overhead_speed, read finish: 44.3 +/- 1.53

1 files changed, 1 insertions, 0 deletions

diff --git a/drivers/gpu/nvgpu/include/nvgpu/linux/vm.h b/drivers/gpu/nvgpu/include/nvgpu/linux/vm.h
index b86a428a..85abce6f 100644
--- a/drivers/gpu/nvgpu/include/nvgpu/linux/vm.h
+++ b/drivers/gpu/nvgpu/include/nvgpu/linux/vm.h
@@ -51,6 +51,7 @@ struct nvgpu_mapped_buf_priv {
         struct sg_table *sgt;
         // For fast reverse lookup (FD -> mapped_buf)
         struct list_head nvmap_priv_entry;
+        struct completion swap_io_done;
 };
 
 /* NVGPU_AS_MAP_BUFFER_FLAGS_DIRECT_KIND_CTRL must be set */