[Intel-gfx] [PATCH 14/23] drm/i915: add support for 64K scratch page

[Matthew Auld]

Before we can fully enable 64K pages, we need to first support a 64K
scratch page if we intend to support the case where we have object sizes
< 2M, since any scratch PTE must also point to a 64K region.  Without
this our 64K usage is limited to objects which completely fill the
page-table, and therefore don't need any scratch.

v2: add reminder about why 48b PPGTT

Reported-by: Chris Wilson <chris at chris-wilson.co.uk>
Signed-off-by: Matthew Auld <matthew.auld at intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen at linux.intel.com>
Cc: Chris Wilson <chris at chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/i915_gem_gtt.c | 61 +++++++++++++++++++++++++++++++++++--
 drivers/gpu/drm/i915/i915_gem_gtt.h |  1 +
 2 files changed, 60 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 135d179c7b0b..1200afa38bb1 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -475,12 +475,69 @@ static void fill_page_dma_32(struct i915_address_space *vm,
 static int
 setup_scratch_page(struct i915_address_space *vm, gfp_t gfp)
 {
-	return __setup_page_dma(vm, &vm->scratch_page, gfp | __GFP_ZERO);
+	struct page *page = NULL;
+	dma_addr_t addr;
+	int order;
+
+	/* In order to utilize 64K pages for an object with a size < 2M, we will
+	 * need to support a 64K scratch page, given that every 16th entry for a
+	 * page-table operating in 64K mode must point to a properly aligned 64K
+	 * region, including any PTEs which happen to point to scratch.
+	 *
+	 * This is only relevant for the 48b PPGTT where we support
+	 * huge-gtt-pages, see also i915_vma_insert().
+	 */
+	if (i915_vm_is_48bit(vm) &&
+	    HAS_PAGE_SIZE(vm->i915, I915_GTT_PAGE_SIZE_64K)) {
+		order = get_order(I915_GTT_PAGE_SIZE_64K);
+		page = alloc_pages(gfp | __GFP_ZERO, order);
+		if (page) {
+			addr = dma_map_page(vm->dma, page, 0,
+					    I915_GTT_PAGE_SIZE_64K,
+					    PCI_DMA_BIDIRECTIONAL);
+			if (unlikely(dma_mapping_error(vm->dma, addr))) {
+				__free_pages(page, order);
+				page = NULL;
+			}
+
+			if (!IS_ALIGNED(addr, I915_GTT_PAGE_SIZE_64K)) {
+				dma_unmap_page(vm->dma, addr,
+					       I915_GTT_PAGE_SIZE_64K,
+					       PCI_DMA_BIDIRECTIONAL);
+				__free_pages(page, order);
+				page = NULL;
+			}
+		}
+	}
+
+	if (!page) {
+		order = 0;
+		page = alloc_page(gfp | __GFP_ZERO);
+		if (unlikely(!page))
+			return -ENOMEM;
+
+		addr = dma_map_page(vm->dma, page, 0, PAGE_SIZE,
+				    PCI_DMA_BIDIRECTIONAL);
+		if (unlikely(dma_mapping_error(vm->dma, addr))) {
+			__free_page(page);
+			return -ENOMEM;
+		}
+	}
+
+	vm->scratch_page.page = page;
+	vm->scratch_page.daddr = addr;
+	vm->scratch_page.order = order;
+
+	return 0;
 }
 
 static void cleanup_scratch_page(struct i915_address_space *vm)
 {
-	cleanup_page_dma(vm, &vm->scratch_page);
+	struct i915_page_dma *p = &vm->scratch_page;
+
+	dma_unmap_page(vm->dma, p->daddr, BIT(p->order) << PAGE_SHIFT,
+		       PCI_DMA_BIDIRECTIONAL);
+	__free_pages(p->page, p->order);
 }
 
 static struct i915_page_table *alloc_pt(struct i915_address_space *vm)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index aa4488637fc9..356fec26e8c9 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -212,6 +212,7 @@ struct i915_vma;
 
 struct i915_page_dma {
 	struct page *page;
+	int order;
 	union {
 		dma_addr_t daddr;
 
-- 
2.13.5