drm/i915: Added write-enable pte bit supportt

This adds support for a write-enable bit in the entry of GTT.
This is handled via a read-only flag in the GEM buffer object which
is then used to see how to set the bit when writing the GTT entries.
Currently by default the Batch buffer & Ring buffers are marked as read only.

v2: Moved the pte override code for read-only bit to 'byt_pte_encode'. (Chris)
    Fixed the issue of leaving 'gt_old_ro' as unused. (Chris)

v3: Removed the 'gt_old_ro' field, now setting RO bit only for Ring Buffers(Daniel).

v4: Added a new 'flags' parameter to all the pte(gen6) encode & insert_entries functions,
    in lieu of overloading the cache_level enum (Daniel).

v5: Removed the superfluous VLV check & changed the definition location of PTE_READ_ONLY flag (Imre)

Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 6f5d5a1..743ddfc 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -116,7 +116,7 @@
 
 static gen6_gtt_pte_t snb_pte_encode(dma_addr_t addr,
 				     enum i915_cache_level level,
-				     bool valid)
+				     bool valid, u32 unused)
 {
 	gen6_gtt_pte_t pte = valid ? GEN6_PTE_VALID : 0;
 	pte |= GEN6_PTE_ADDR_ENCODE(addr);
@@ -138,7 +138,7 @@
 
 static gen6_gtt_pte_t ivb_pte_encode(dma_addr_t addr,
 				     enum i915_cache_level level,
-				     bool valid)
+				     bool valid, u32 unused)
 {
 	gen6_gtt_pte_t pte = valid ? GEN6_PTE_VALID : 0;
 	pte |= GEN6_PTE_ADDR_ENCODE(addr);
@@ -162,7 +162,7 @@
 
 static gen6_gtt_pte_t byt_pte_encode(dma_addr_t addr,
 				     enum i915_cache_level level,
-				     bool valid)
+				     bool valid, u32 flags)
 {
 	gen6_gtt_pte_t pte = valid ? GEN6_PTE_VALID : 0;
 	pte |= GEN6_PTE_ADDR_ENCODE(addr);
@@ -170,7 +170,8 @@
 	/* Mark the page as writeable.  Other platforms don't have a
 	 * setting for read-only/writable, so this matches that behavior.
 	 */
-	pte |= BYT_PTE_WRITEABLE;
+	if (!(flags & PTE_READ_ONLY))
+		pte |= BYT_PTE_WRITEABLE;
 
 	if (level != I915_CACHE_NONE)
 		pte |= BYT_PTE_SNOOPED_BY_CPU_CACHES;
@@ -180,7 +181,7 @@
 
 static gen6_gtt_pte_t hsw_pte_encode(dma_addr_t addr,
 				     enum i915_cache_level level,
-				     bool valid)
+				     bool valid, u32 unused)
 {
 	gen6_gtt_pte_t pte = valid ? GEN6_PTE_VALID : 0;
 	pte |= HSW_PTE_ADDR_ENCODE(addr);
@@ -193,7 +194,7 @@
 
 static gen6_gtt_pte_t iris_pte_encode(dma_addr_t addr,
 				      enum i915_cache_level level,
-				      bool valid)
+				      bool valid, u32 unused)
 {
 	gen6_gtt_pte_t pte = valid ? GEN6_PTE_VALID : 0;
 	pte |= HSW_PTE_ADDR_ENCODE(addr);
@@ -307,7 +308,7 @@
 static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 				      struct sg_table *pages,
 				      uint64_t start,
-				      enum i915_cache_level cache_level)
+				      enum i915_cache_level cache_level, u32 unused)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
@@ -645,7 +646,7 @@
 	uint32_t pd_entry;
 	int pte, pde;
 
-	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC, true);
+	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC, true, 0);
 
 	pd_addr = (gen6_gtt_pte_t __iomem *)dev_priv->gtt.gsm +
 		ppgtt->pd_offset / sizeof(gen6_gtt_pte_t);
@@ -947,7 +948,7 @@
 	unsigned first_pte = first_entry % I915_PPGTT_PT_ENTRIES;
 	unsigned last_pte, i;
 
-	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC, true);
+	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC, true, 0);
 
 	while (num_entries) {
 		last_pte = first_pte + num_entries;
@@ -970,7 +971,7 @@
 static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
 				      struct sg_table *pages,
 				      uint64_t start,
-				      enum i915_cache_level cache_level)
+				      enum i915_cache_level cache_level, u32 flags)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
@@ -987,7 +988,8 @@
 
 		pt_vaddr[act_pte] =
 			vm->pte_encode(sg_page_iter_dma_address(&sg_iter),
-				       cache_level, true);
+				       cache_level, true, flags);
+
 		if (++act_pte == I915_PPGTT_PT_ENTRIES) {
 			kunmap_atomic(pt_vaddr);
 			pt_vaddr = NULL;
@@ -1224,8 +1226,12 @@
 	       enum i915_cache_level cache_level,
 	       u32 flags)
 {
+	/* Currently applicable only to VLV */
+	if (vma->obj->gt_ro)
+		flags |= PTE_READ_ONLY;
+
 	vma->vm->insert_entries(vma->vm, vma->obj->pages, vma->node.start,
-				cache_level);
+				cache_level, flags);
 }
 
 static void ppgtt_unbind_vma(struct i915_vma *vma)
@@ -1400,7 +1406,7 @@
 static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
 				     struct sg_table *st,
 				     uint64_t start,
-				     enum i915_cache_level level)
+				     enum i915_cache_level level, u32 unused)
 {
 	struct drm_i915_private *dev_priv = vm->dev->dev_private;
 	unsigned first_entry = start >> PAGE_SHIFT;
@@ -1446,7 +1452,7 @@
 static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 				     struct sg_table *st,
 				     uint64_t start,
-				     enum i915_cache_level level)
+				     enum i915_cache_level level, u32 flags)
 {
 	struct drm_i915_private *dev_priv = vm->dev->dev_private;
 	unsigned first_entry = start >> PAGE_SHIFT;
@@ -1458,7 +1464,7 @@
 
 	for_each_sg_page(st->sgl, &sg_iter, st->nents, 0) {
 		addr = sg_page_iter_dma_address(&sg_iter);
-		iowrite32(vm->pte_encode(addr, level, true), &gtt_entries[i]);
+		iowrite32(vm->pte_encode(addr, level, true, flags), &gtt_entries[i]);
 		i++;
 	}
 
@@ -1470,7 +1476,7 @@
 	 */
 	if (i != 0)
 		WARN_ON(readl(&gtt_entries[i-1]) !=
-			vm->pte_encode(addr, level, true));
+			vm->pte_encode(addr, level, true, flags));
 
 	/* This next bit makes the above posting read even more important. We
 	 * want to flush the TLBs only after we're certain all the PTE updates
@@ -1524,7 +1530,7 @@
 		 first_entry, num_entries, max_entries))
 		num_entries = max_entries;
 
-	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC, use_scratch);
+	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC, use_scratch, 0);
 
 	for (i = 0; i < num_entries; i++)
 		iowrite32(scratch_pte, &gtt_base[i]);
@@ -1573,6 +1579,10 @@
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj = vma->obj;
 
+	/* Currently applicable only to VLV */
+	if (obj->gt_ro)
+		flags |= PTE_READ_ONLY;
+
 	/* If there is no aliasing PPGTT, or the caller needs a global mapping,
 	 * or we have a global mapping already but the cacheability flags have
 	 * changed, set the global PTEs.
@@ -1589,7 +1599,7 @@
 		    (cache_level != obj->cache_level)) {
 			vma->vm->insert_entries(vma->vm, obj->pages,
 						vma->node.start,
-						cache_level);
+						cache_level, flags);
 			obj->has_global_gtt_mapping = 1;
 		}
 	}
@@ -1601,7 +1611,7 @@
 		appgtt->base.insert_entries(&appgtt->base,
 					    vma->obj->pages,
 					    vma->node.start,
-					    cache_level);
+					    cache_level, flags);
 		vma->obj->has_aliasing_ppgtt_mapping = 1;
 	}
 }