gpu: nvgpu: Use coherent aperture flag

When using a coherent DMA API wee must make sure to program
any aperture fields with the coherent aperture setting. To
do this the nvgpu_aperture_mask() function was modified to
take a third aperture mask argument, a coherent setting, so
that code can use this function to generate coherent aperture
settings.

The aperture choice is some what tricky: the default version
of this function uses the state of the DMA API to determine
what aperture to use for SYSMEM: either coherent or
non-coherent internally. Thus a kernel user need only specify
the normal nvgpu_mem struct and the correct mask should be
chosen. Due to many uses of nvgpu_mem structs not created
directly from the DMA API wrapper it's easier to translate
SYSMEM to SYSMEM_COH after creation.

However, the GMMU mapping code, will encounter buffers from
userspace with difference coerency attributes than the DMA
API. Thus the __nvgpu_aperture_mask() really respects the
aperture setting passed in regardless of the DMA API state.
This aperture setting is pulled from NVGPU_VM_MAP_IO_COHERENT
since this is either passed in from userspace or set by the
kernel when using coherent DMA. The aperture field in attrs
is upgraded to coh if this flag is set.

This change also adds a coherent sysmem mask everywhere that
it can. There's a couple places that do not have a coherent
register field defined yet. These need to eventually be
defined and added.

Lastly the aperture mask code has been mvoed from the Linux
vm.c code to the general vm.c code since this function has
no Linux dependencies.

Note: depends on https://git-master.nvidia.com/r/1664536 for
new register fields.

JIRA EVLR-2333

Change-Id: I4b347911ecb7c511738563fe6c34d0e6aa380d71
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1655220
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

2 files changed, 20 insertions, 14 deletions

diff --git a/drivers/gpu/nvgpu/gp10b/fifo_gp10b.c b/drivers/gpu/nvgpu/gp10b/fifo_gp10b.c
index c82fb1cc..1436a260 100644
--- a/drivers/gpu/nvgpu/gp10b/fifo_gp10b.c
+++ b/drivers/gpu/nvgpu/gp10b/fifo_gp10b.c
@@ -25,6 +25,7 @@
 #include <nvgpu/dma.h>
 #include <nvgpu/bug.h>
 #include <nvgpu/log2.h>
+#include <nvgpu/enabled.h>
 
 #include "fifo_gp10b.h"
 
@@ -78,8 +79,9 @@ int channel_gp10b_commit_userd(struct channel_gk20a *c)
         nvgpu_mem_wr32(g, &c->inst_block,
                        ram_in_ramfc_w() + ram_fc_userd_w(),
                        nvgpu_aperture_mask(g, &g->fifo.userd,
-                        pbdma_userd_target_sys_mem_ncoh_f(),
-                        pbdma_userd_target_vid_mem_f()) |
+                                           pbdma_userd_target_sys_mem_ncoh_f(),
+                                           pbdma_userd_target_sys_mem_coh_f(),
+                                           pbdma_userd_target_vid_mem_f()) |
                        pbdma_userd_addr_f(addr_lo));
 
         nvgpu_mem_wr32(g, &c->inst_block,
diff --git a/drivers/gpu/nvgpu/gp10b/mm_gp10b.c b/drivers/gpu/nvgpu/gp10b/mm_gp10b.c
index 0439dda9..7ff5f6a6 100644
--- a/drivers/gpu/nvgpu/gp10b/mm_gp10b.c
+++ b/drivers/gpu/nvgpu/gp10b/mm_gp10b.c
@@ -124,8 +124,9 @@ int gp10b_init_bar2_mm_hw_setup(struct gk20a *g)
 
         gk20a_writel(g, bus_bar2_block_r(),
                      nvgpu_aperture_mask(g, inst_block,
-                                bus_bar2_block_target_sys_mem_ncoh_f(),
-                                bus_bar2_block_target_vid_mem_f()) |
+                                         bus_bar2_block_target_sys_mem_ncoh_f(),
+                                         bus_bar2_block_target_sys_mem_coh_f(),
+                                         bus_bar2_block_target_vid_mem_f()) |
                      bus_bar2_block_mode_virtual_f() |
                      bus_bar2_block_ptr_f(inst_pa));
 
@@ -148,8 +149,9 @@ static void update_gmmu_pde3_locked(struct vm_gk20a *vm,
         phys_addr >>= gmmu_new_pde_address_shift_v();
 
         pde_v[0] |= nvgpu_aperture_mask(g, pd->mem,
-                        gmmu_new_pde_aperture_sys_mem_ncoh_f(),
-                        gmmu_new_pde_aperture_video_memory_f());
+                                        gmmu_new_pde_aperture_sys_mem_ncoh_f(),
+                                        gmmu_new_pde_aperture_sys_mem_coh_f(),
+                                        gmmu_new_pde_aperture_video_memory_f());
         pde_v[0] |= gmmu_new_pde_address_sys_f(u64_lo32(phys_addr));
         pde_v[0] |= gmmu_new_pde_vol_true_f();
         pde_v[1] |= phys_addr >> 24;
@@ -194,6 +196,7 @@ static void update_gmmu_pde0_locked(struct vm_gk20a *vm,
                         gmmu_new_dual_pde_address_small_sys_f(small_addr);
                 pde_v[2] |= nvgpu_aperture_mask(g, pd->mem,
                         gmmu_new_dual_pde_aperture_small_sys_mem_ncoh_f(),
+                        gmmu_new_dual_pde_aperture_small_sys_mem_coh_f(),
                         gmmu_new_dual_pde_aperture_small_video_memory_f());
                 pde_v[2] |= gmmu_new_dual_pde_vol_small_true_f();
                 pde_v[3] |= small_addr >> 24;
@@ -204,6 +207,7 @@ static void update_gmmu_pde0_locked(struct vm_gk20a *vm,
                 pde_v[0] |= gmmu_new_dual_pde_vol_big_true_f();
                 pde_v[0] |= nvgpu_aperture_mask(g, pd->mem,
                         gmmu_new_dual_pde_aperture_big_sys_mem_ncoh_f(),
+                        gmmu_new_dual_pde_aperture_big_sys_mem_coh_f(),
                         gmmu_new_dual_pde_aperture_big_video_memory_f());
                 pde_v[1] |= big_addr >> 28;
         }
@@ -240,11 +244,10 @@ static void __update_pte(struct vm_gk20a *vm,
                 gmmu_new_pte_address_sys_f(phys_shifted) :
                 gmmu_new_pte_address_vid_f(phys_shifted);
         u32 pte_tgt = __nvgpu_aperture_mask(g,
-                        attrs->aperture,
-                        attrs->coherent ?
-                                gmmu_new_pte_aperture_sys_mem_coh_f() :
-                                gmmu_new_pte_aperture_sys_mem_ncoh_f(),
-                        gmmu_new_pte_aperture_video_memory_f());
+                                        attrs->aperture,
+                                        gmmu_new_pte_aperture_sys_mem_ncoh_f(),
+                                        gmmu_new_pte_aperture_sys_mem_coh_f(),
+                                        gmmu_new_pte_aperture_video_memory_f());
 
         pte_w[0] = pte_valid | pte_addr | pte_tgt;
 
@@ -306,7 +309,7 @@ static void update_gmmu_pte_locked(struct vm_gk20a *vm,
                 page_size >> 10,
                 nvgpu_gmmu_perm_str(attrs->rw_flag),
                 attrs->kind_v,
-                nvgpu_aperture_str(attrs->aperture),
+                nvgpu_aperture_str(g, attrs->aperture),
                 attrs->cacheable ? 'C' : '-',
                 attrs->sparse    ? 'S' : '-',
                 attrs->priv      ? 'P' : '-',
@@ -428,8 +431,9 @@ void gp10b_mm_init_pdb(struct gk20a *g, struct nvgpu_mem *inst_block,
 
         nvgpu_mem_wr32(g, inst_block, ram_in_page_dir_base_lo_w(),
                 nvgpu_aperture_mask(g, vm->pdb.mem,
-                  ram_in_page_dir_base_target_sys_mem_ncoh_f(),
-                  ram_in_page_dir_base_target_vid_mem_f()) |
+                                ram_in_page_dir_base_target_sys_mem_ncoh_f(),
+                                ram_in_page_dir_base_target_sys_mem_coh_f(),
+                                ram_in_page_dir_base_target_vid_mem_f()) |
                 ram_in_page_dir_base_vol_true_f() |
                 ram_in_big_page_size_64kb_f() |
                 ram_in_page_dir_base_lo_f(pdb_addr_lo) |