drm/nouveau/mmu: shift wrapping bug in gf100_vm_map()

Commit Message

Dan Carpenter Feb. 9, 2015, 1:58 p.m. UTC

Since "1" is an int, then it means we can't use the high bits of this
u64.

Signed-off-by: Dan Carpenter <[e-mail]>

Patch

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gf100.c
index 294cda3..b10c90a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gf100.c
@@ -106,7 +106,7 @@  static void
 gf100_vm_map(struct nvkm_vma *vma, struct nvkm_gpuobj *pgt,
 	     struct nvkm_mem *mem, u32 pte, u32 cnt, u64 phys, u64 delta)
 {
-	u64 next = 1 << (vma->node->type - 8);
+	u64 next = 1ULL << (vma->node->type - 8);
 
 	phys  = gf100_vm_addr(vma, phys, mem->memtype, 0);
 	pte <<= 3;