drm/nouveau/mmu/gk104,gk20a: implement new vmm backend

Adds support for:
- 64KiB big page size.
- System-memory PTs.
- LPTE "invalid" state.
- (Tegra) Use of video memory aperture.

Adds support for marking LPTEs invalid, resulting in the corresponding
SPTEs being ignored, which is supposed to speed up TLB invalidates.

On The Tegra side, this will switch to using the video memory aperture
for all mappings.  The HW will still target non-coherent system memory,
but this aperture needs to be selected in order to support compression.

Tegra's instmem backend somewhat cheated to get this effect previously.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

1 files changed, 18 insertions, 0 deletions

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk20a.c
index e000c3d1554a..8086994a0446 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk20a.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgk20a.c
@@ -21,10 +21,25 @@
  */
 #include "vmm.h"
 
+#include <core/memory.h>
+
+int
+gk20a_vmm_aper(enum nvkm_memory_target target)
+{
+	switch (target) {
+	case NVKM_MEM_TARGET_NCOH: return 0;
+	default:
+		return -EINVAL;
+	}
+}
+
 static const struct nvkm_vmm_func
 gk20a_vmm_17 = {
 	.join = gf100_vmm_join,
 	.part = gf100_vmm_part,
+	.aper = gf100_vmm_aper,
+	.valid = gf100_vmm_valid,
+	.flush = gf100_vmm_flush,
 	.page = {
 		{ 17, &gk104_vmm_desc_17_17[0], NVKM_VMM_PAGE_xxHC },
 		{ 12, &gk104_vmm_desc_17_12[0], NVKM_VMM_PAGE_xxHx },
@@ -36,6 +51,9 @@ static const struct nvkm_vmm_func
 gk20a_vmm_16 = {
 	.join = gf100_vmm_join,
 	.part = gf100_vmm_part,
+	.aper = gf100_vmm_aper,
+	.valid = gf100_vmm_valid,
+	.flush = gf100_vmm_flush,
 	.page = {
 		{ 16, &gk104_vmm_desc_16_16[0], NVKM_VMM_PAGE_xxHC },
 		{ 12, &gk104_vmm_desc_16_12[0], NVKM_VMM_PAGE_xxHx },