[Intel-gfx] [PATCH 60/68] drm/i915/bdw: Add 4 level switching infrastructure

Ben Widawsky benjamin.widawsky at intel.com
Fri Aug 22 05:12:23 CEST 2014


Map is easy, it's the same register as the PDP descriptor 0, but it only
has one entry. Also, the mapping code is now trivial thanks to all of
the prep patches.

This is the mostly just cleaning up lose ends of 4lvl before we can
cleanup all the code to rip out the legacy page table update functions
insert and clear entries.

Signed-off-by: Ben Widawsky <ben at bwidawsk.net>
---
 drivers/gpu/drm/i915/i915_gem_gtt.c | 60 ++++++++++++++++++++++++++++++++-----
 drivers/gpu/drm/i915/i915_gem_gtt.h |  4 ++-
 drivers/gpu/drm/i915/i915_reg.h     |  1 +
 3 files changed, 57 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 9b3358f..12c42ea 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -103,6 +103,9 @@ static inline gen8_ppgtt_pde_t gen8_pde_encode(struct drm_device *dev,
 	return pde;
 }
 
+#define gen8_pdpe_encode gen8_pde_encode
+#define gen8_pml4e_encode gen8_pde_encode
+
 static gen6_gtt_pte_t snb_pte_encode(dma_addr_t addr,
 				     enum i915_cache_level level,
 				     bool valid, u32 unused)
@@ -522,9 +525,9 @@ static int gen8_write_pdp(struct intel_engine_cs *ring,
 	return 0;
 }
 
-static int gen8_mm_switch(struct i915_hw_ppgtt *ppgtt,
-			  struct intel_engine_cs *ring,
-			  bool synchronous)
+static int gen8_legacy_mm_switch(struct i915_hw_ppgtt *ppgtt,
+				 struct intel_engine_cs *ring,
+				 bool synchronous)
 {
 	int i, ret;
 
@@ -541,6 +544,13 @@ static int gen8_mm_switch(struct i915_hw_ppgtt *ppgtt,
 	return 0;
 }
 
+static int gen8_48b_mm_switch(struct i915_hw_ppgtt *ppgtt,
+			      struct intel_engine_cs *ring,
+			      bool synchronous)
+{
+	return gen8_write_pdp(ring, 0, ppgtt->pml4.daddr, synchronous);
+}
+
 static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 				   uint64_t start,
 				   uint64_t length,
@@ -668,6 +678,37 @@ static void gen8_map_pagetable_range(struct i915_address_space *vm,
 	kunmap_atomic(pagedir);
 }
 
+static void gen8_map_page_directory(struct i915_pagedirpo *pdp,
+				    struct i915_pagedir *pd,
+				    int index,
+				    struct drm_device *dev)
+{
+	gen8_ppgtt_pdpe_t *pagedirpo;
+	gen8_ppgtt_pdpe_t pdpe;
+
+	/* We do not need to clflush because no platform requiring flush
+	 * supports 64b pagetables. */
+	if (!HAS_48B_PPGTT(dev))
+		return;
+
+	pagedirpo = kmap_atomic(pdp->page);
+	pdpe = gen8_pdpe_encode(dev, pd->daddr, I915_CACHE_LLC);
+	pagedirpo[index] = pdpe;
+	kunmap_atomic(pagedirpo);
+}
+
+static void gen8_map_page_directory_pointer(struct i915_pml4 *pml4,
+					    struct i915_pagedirpo *pdp,
+					    int index,
+					    struct drm_device *dev)
+{
+	gen8_ppgtt_pml4e_t *pagemap = kmap_atomic(pml4->page);
+	gen8_ppgtt_pml4e_t pml4e = gen8_pml4e_encode(dev, pdp->daddr, I915_CACHE_LLC);
+	BUG_ON(!HAS_48B_PPGTT(dev));
+	pagemap[index] = pml4e;
+	kunmap_atomic(pagemap);
+}
+
 static bool gen8_teardown_va_range_3lvl(struct i915_address_space *vm,
 					struct i915_pagedirpo *pdp,
 					uint64_t start, uint64_t length,
@@ -1118,6 +1159,7 @@ static int gen8_alloc_va_range_3lvl(struct i915_address_space *vm,
 
 		set_bit(pdpe, pdp->used_pdpes);
 		gen8_map_pagetable_range(vm, pd, start, length);
+		gen8_map_page_directory(pdp, pd, pdpe, dev);
 		pd->zombie = 0;
 	}
 
@@ -1191,6 +1233,8 @@ static int gen8_alloc_va_range_4lvl(struct i915_address_space *vm,
 		ret = gen8_alloc_va_range_3lvl(vm, pdp, start, length);
 		if (ret)
 			goto err_out;
+
+		gen8_map_page_directory_pointer(pml4, pdp, pml4e, vm->dev);
 	}
 
 	bitmap_or(pml4->used_pml4es, new_pdps, pml4->used_pml4es,
@@ -1252,9 +1296,7 @@ static int gen8_ppgtt_init_common(struct i915_hw_ppgtt *ppgtt, uint64_t size)
 	ppgtt->base.total = size;
 	ppgtt->base.cleanup = gen8_ppgtt_cleanup;
 	ppgtt->base.insert_entries = gen8_ppgtt_insert_entries;
-
 	ppgtt->enable = gen8_ppgtt_enable;
-	ppgtt->switch_mm = gen8_mm_switch;
 
 	if (HAS_48B_PPGTT(ppgtt->base.dev)) {
 		int ret = pml4_init(ppgtt);
@@ -1262,6 +1304,8 @@ static int gen8_ppgtt_init_common(struct i915_hw_ppgtt *ppgtt, uint64_t size)
 			free_pt_scratch(ppgtt->scratch_pd, ppgtt->base.dev);
 			return ret;
 		}
+
+		ppgtt->switch_mm = gen8_48b_mm_switch;
 	} else {
 		int ret = __pdp_init(&ppgtt->pdp, false);
 		if (ret) {
@@ -1269,7 +1313,7 @@ static int gen8_ppgtt_init_common(struct i915_hw_ppgtt *ppgtt, uint64_t size)
 			return ret;
 		}
 
-		ppgtt->switch_mm = gen8_mm_switch;
+		ppgtt->switch_mm = gen8_legacy_mm_switch;
 		trace_i915_pagedirpo_alloc(&ppgtt->base, 0, 0, GEN8_PML4E_SHIFT);
 	}
 
@@ -1299,6 +1343,7 @@ static int gen8_aliasing_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 		return ret;
 	}
 
+	/* FIXME: PML4 */
 	gen8_for_each_pdpe(pd, pdp, start, size, temp, pdpe)
 		gen8_map_pagetable_range(&ppgtt->base, pd,start, size);
 
@@ -1536,8 +1581,9 @@ static int gen8_ppgtt_enable(struct i915_hw_ppgtt *ppgtt)
 	int j, ret;
 
 	for_each_ring(ring, dev_priv, j) {
+		u32 four_level = HAS_48B_PPGTT(dev) ? GEN8_GFX_PPGTT_64B : 0;
 		I915_WRITE(RING_MODE_GEN7(ring),
-			   _MASKED_BIT_ENABLE(GFX_PPGTT_ENABLE));
+			   _MASKED_BIT_ENABLE(GFX_PPGTT_ENABLE | four_level));
 
 		/* We promise to do a switch later with FULL PPGTT. If this is
 		 * aliasing, this is the one and only switch we'll do */
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index ba103bd..f4c611e 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -36,7 +36,9 @@
 
 typedef uint32_t gen6_gtt_pte_t;
 typedef uint64_t gen8_gtt_pte_t;
-typedef gen8_gtt_pte_t gen8_ppgtt_pde_t;
+typedef gen8_gtt_pte_t		gen8_ppgtt_pde_t;
+typedef gen8_ppgtt_pde_t	gen8_ppgtt_pdpe_t;
+typedef gen8_ppgtt_pdpe_t	gen8_ppgtt_pml4e_t;
 
 /* GEN Agnostic defines */
 #define I915_PAGE_SIZE			4096
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 7a6cc69..afb2515 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1227,6 +1227,7 @@ enum punit_power_well {
 #define   GFX_REPLAY_MODE		(1<<11)
 #define   GFX_PSMI_GRANULARITY		(1<<10)
 #define   GFX_PPGTT_ENABLE		(1<<9)
+#define   GEN8_GFX_PPGTT_64B		(1<<7)
 
 #define VLV_DISPLAY_BASE 0x180000
 #define VLV_MIPI_BASE VLV_DISPLAY_BASE
-- 
2.0.4




More information about the Intel-gfx mailing list