gpu: nvgpu: Initial MAP_BUFFER_BATCH implementation

Add batch support for mapping and unmapping. Batching essentially
helps transform some per-map/unmap overhead to per-batch overhead,
namely gk20a_busy()/gk20a_idle() calls, GPU L2 flushes, and GPU TLB
invalidates. Batching with size 64 has been measured to yield >20x
speed-up in low-level fixed-address mapping microbenchmarks.

Bug 1614735
Bug 1623949

Change-Id: Ie22b9caea5a7c3fc68a968d1b7f8488dfce72085
Signed-off-by: Sami Kiminki <skiminki@nvidia.com>
Reviewed-on: http://git-master/r/733231
(cherry picked from commit de4a7cfb93e8228a4a0c6a2815755a8df4531c91)
Reviewed-on: http://git-master/r/763812
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>

1 files changed, 5 insertions, 3 deletions

diff --git a/drivers/gpu/nvgpu/vgpu/mm_vgpu.c b/drivers/gpu/nvgpu/vgpu/mm_vgpu.c
index 855aac0d..be1fa47d 100644
--- a/drivers/gpu/nvgpu/vgpu/mm_vgpu.c
+++ b/drivers/gpu/nvgpu/vgpu/mm_vgpu.c
@@ -66,7 +66,8 @@ static u64 vgpu_locked_gmmu_map(struct vm_gk20a *vm,
                                 u32 flags,
                                 int rw_flag,
                                 bool clear_ctags,
-                                bool sparse)
+                                bool sparse,
+                                struct vm_gk20a_mapping_batch *batch)
 {
         int err = 0;
         struct device *d = dev_from_vm(vm);
@@ -130,7 +131,8 @@ static void vgpu_locked_gmmu_unmap(struct vm_gk20a *vm,
                                 int pgsz_idx,
                                 bool va_allocated,
                                 int rw_flag,
-                                bool sparse)
+                                bool sparse,
+                                struct vm_gk20a_mapping_batch *batch)
 {
         struct gk20a *g = gk20a_from_vm(vm);
         struct gk20a_platform *platform = gk20a_get_platform(g->dev);
@@ -182,7 +184,7 @@ static void vgpu_vm_remove_support(struct vm_gk20a *vm)
         while (node) {
                 mapped_buffer =
                         container_of(node, struct mapped_buffer_node, node);
-                gk20a_vm_unmap_locked(mapped_buffer);
+                gk20a_vm_unmap_locked(mapped_buffer, NULL);
                 node = rb_first(&vm->mapped_buffers);
         }