Merge remote branch 'nouveau/drm-nouveau-next' of /ssd/git/drm-nouveau-next into drm-core-next

* 'nouveau/drm-nouveau-next' of /ssd/git/drm-nouveau-next:
  drm/nvc0: accelerate ttm buffer moves
  drm/nvc0: initial support for tiled buffer objects
  drm/nvc0: implement fbcon acceleration
  drm/nvc0: implement pgraph engine hooks
  drm/nvc0: implement pfifo engine hooks
  drm/nvc0: implement fencing
  drm/nvc0: fix channel dma init paths
  drm/nvc0: skip dma object creation for drm channel
  drm/nvc0: implement channel structure initialisation
  drm/nvc0: gpuobj_new need only check validity and init the relevant engine
  drm/nvc0: reject the notifier_alloc ioctl
  drm/nvc0: create shared channel vm
  drm/nvc0: initial vm implementation, use for bar1/bar3 management
  drm/nvc0: import initial vm backend
  drm/nouveau: modify vm to accomodate dual page tables for nvc0
  drm/nv50: add missing license header to nv50_fbcon.c
  drm/nv50: fix smatch warning in nv50_vram.c
  drm/nouveau: sizeof() vs ARRAY_SIZE()

30 files changed, 5044 insertions, 341 deletions

diff --git a/drivers/gpu/drm/nouveau/Makefile b/drivers/gpu/drm/nouveau/Makefile
index b1d8941e04d8..e12c97fd8db8 100644
--- a/drivers/gpu/drm/nouveau/Makefile
+++ b/drivers/gpu/drm/nouveau/Makefile
@@ -18,17 +18,19 @@ nouveau-y := nouveau_drv.o nouveau_state.o nouveau_channel.o nouveau_mem.o \
              nv04_fifo.o nv10_fifo.o nv40_fifo.o nv50_fifo.o nvc0_fifo.o \
              nv04_graph.o nv10_graph.o nv20_graph.o \
              nv40_graph.o nv50_graph.o nvc0_graph.o \
-             nv40_grctx.o nv50_grctx.o \
+             nv40_grctx.o nv50_grctx.o nvc0_grctx.o \
              nv84_crypt.o \
              nv04_instmem.o nv50_instmem.o nvc0_instmem.o \
              nv50_evo.o nv50_crtc.o nv50_dac.o nv50_sor.o \
-             nv50_cursor.o nv50_display.o nv50_fbcon.o \
+             nv50_cursor.o nv50_display.o \
              nv04_dac.o nv04_dfp.o nv04_tv.o nv17_tv.o nv17_tv_modes.o \
-             nv04_crtc.o nv04_display.o nv04_cursor.o nv04_fbcon.o \
+             nv04_crtc.o nv04_display.o nv04_cursor.o \
+             nv04_fbcon.o nv50_fbcon.o nvc0_fbcon.o \
              nv10_gpio.o nv50_gpio.o \
              nv50_calc.o \
              nv04_pm.o nv50_pm.o nva3_pm.o \
-             nv50_vram.o nv50_vm.o
+             nv50_vram.o nvc0_vram.o \
+             nv50_vm.o nvc0_vm.o
 
 nouveau-$(CONFIG_DRM_NOUVEAU_DEBUG) += nouveau_debugfs.o
 nouveau-$(CONFIG_COMPAT) += nouveau_ioc32.o
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 42d1ad62b381..a7fae26f4654 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -413,7 +413,7 @@ nouveau_bo_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
                 man->default_caching = TTM_PL_FLAG_CACHED;
                 break;
         case TTM_PL_VRAM:
-                if (dev_priv->card_type == NV_50) {
+                if (dev_priv->card_type >= NV_50) {
                         man->func = &nouveau_vram_manager;
                         man->io_reserve_fastpath = false;
                         man->use_io_reserve_lru = true;
@@ -515,6 +515,58 @@ nouveau_bo_mem_ctxdma(struct ttm_buffer_object *bo,
 }
 
 static int
+nvc0_bo_move_m2mf(struct nouveau_channel *chan, struct ttm_buffer_object *bo,
+                  struct ttm_mem_reg *old_mem, struct ttm_mem_reg *new_mem)
+{
+        struct drm_nouveau_private *dev_priv = nouveau_bdev(bo->bdev);
+        struct nouveau_bo *nvbo = nouveau_bo(bo);
+        u64 src_offset = old_mem->start << PAGE_SHIFT;
+        u64 dst_offset = new_mem->start << PAGE_SHIFT;
+        u32 page_count = new_mem->num_pages;
+        int ret;
+
+        if (!nvbo->no_vm) {
+                if (old_mem->mem_type == TTM_PL_VRAM)
+                        src_offset  = nvbo->vma.offset;
+                else
+                        src_offset += dev_priv->gart_info.aper_base;
+
+                if (new_mem->mem_type == TTM_PL_VRAM)
+                        dst_offset  = nvbo->vma.offset;
+                else
+                        dst_offset += dev_priv->gart_info.aper_base;
+        }
+
+        page_count = new_mem->num_pages;
+        while (page_count) {
+                int line_count = (page_count > 2047) ? 2047 : page_count;
+
+                ret = RING_SPACE(chan, 12);
+                if (ret)
+                        return ret;
+
+                BEGIN_NVC0(chan, 2, NvSubM2MF, 0x0238, 2);
+                OUT_RING  (chan, upper_32_bits(dst_offset));
+                OUT_RING  (chan, lower_32_bits(dst_offset));
+                BEGIN_NVC0(chan, 2, NvSubM2MF, 0x030c, 6);
+                OUT_RING  (chan, upper_32_bits(src_offset));
+                OUT_RING  (chan, lower_32_bits(src_offset));
+                OUT_RING  (chan, PAGE_SIZE); /* src_pitch */
+                OUT_RING  (chan, PAGE_SIZE); /* dst_pitch */
+                OUT_RING  (chan, PAGE_SIZE); /* line_length */
+                OUT_RING  (chan, line_count);
+                BEGIN_NVC0(chan, 2, NvSubM2MF, 0x0300, 1);
+                OUT_RING  (chan, 0x00100110);
+
+                page_count -= line_count;
+                src_offset += (PAGE_SIZE * line_count);
+                dst_offset += (PAGE_SIZE * line_count);
+        }
+
+        return 0;
+}
+
+static int
 nv50_bo_move_m2mf(struct nouveau_channel *chan, struct ttm_buffer_object *bo,
                   struct ttm_mem_reg *old_mem, struct ttm_mem_reg *new_mem)
 {
@@ -690,7 +742,10 @@ nouveau_bo_move_m2mf(struct ttm_buffer_object *bo, int evict, bool intr,
         if (dev_priv->card_type < NV_50)
                 ret = nv04_bo_move_m2mf(chan, bo, &bo->mem, new_mem);
         else
+        if (dev_priv->card_type < NV_C0)
                 ret = nv50_bo_move_m2mf(chan, bo, &bo->mem, new_mem);
+        else
+                ret = nvc0_bo_move_m2mf(chan, bo, &bo->mem, new_mem);
         if (ret == 0) {
                 ret = nouveau_bo_move_accel_cleanup(chan, nvbo, evict,
                                                     no_wait_reserve,
@@ -901,6 +956,7 @@ nouveau_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
         case TTM_PL_VRAM:
         {
                 struct nouveau_vram *vram = mem->mm_node;
+                u8 page_shift;
 
                 if (!dev_priv->bar1_vm) {
                         mem->bus.offset = mem->start << PAGE_SHIFT;
@@ -909,8 +965,14 @@ nouveau_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
                         break;