[Intel-gfx] [PATCH v2 3/4] drm/i915: create compact dma scatter lists for gem objects

[Imre Deak]

So far we created a sparse dma scatter list for gem objects, where each
scatter list entry represented only a single page. In the future we'll
have to handle compact scatter lists too where each entry can consist of
multiple pages, for example for objects imported through PRIME.

The previous patches have already fixed up all other places where the
i915 driver _walked_ these lists. Here we have the corresponding fix to
_create_ compact lists. It's not a performance or memory footprint
improvement, but it helps to better exercise the new logic.

Reference: http://www.spinics.net/lists/dri-devel/msg33917.html
Signed-off-by: Imre Deak <imre.deak at intel.com>
---
 drivers/gpu/drm/i915/i915_gem.c |   31 ++++++++++++++++++++++---------
 1 file changed, 22 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 03037cf..ec2f11f 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1625,9 +1625,8 @@ i915_gem_object_is_purgeable(struct drm_i915_gem_object *obj)
 static void
 i915_gem_object_put_pages_gtt(struct drm_i915_gem_object *obj)
 {
-	int page_count = obj->base.size / PAGE_SIZE;
-	struct scatterlist *sg;
-	int ret, i;
+	struct sg_page_iter sg_iter;
+	int ret;
 
 	BUG_ON(obj->madv == __I915_MADV_PURGED);
 
@@ -1647,8 +1646,8 @@ i915_gem_object_put_pages_gtt(struct drm_i915_gem_object *obj)
 	if (obj->madv == I915_MADV_DONTNEED)
 		obj->dirty = 0;
 
-	for_each_sg(obj->pages->sgl, sg, page_count, i) {
-		struct page *page = sg_page(sg);
+	for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0) {
+		struct page *page = sg_iter.page;
 
 		if (obj->dirty)
 			set_page_dirty(page);
@@ -1749,7 +1748,9 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 	struct address_space *mapping;
 	struct sg_table *st;
 	struct scatterlist *sg;
+	struct sg_page_iter sg_iter;
 	struct page *page;
+	unsigned long last_pfn = 0;	/* suppress gcc warning */
 	gfp_t gfp;
 
 	/* Assert that the object is not currently in any GPU domain. As it
@@ -1779,7 +1780,9 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 	gfp = mapping_gfp_mask(mapping);
 	gfp |= __GFP_NORETRY | __GFP_NOWARN | __GFP_NO_KSWAPD;
 	gfp &= ~(__GFP_IO | __GFP_WAIT);
-	for_each_sg(st->sgl, sg, page_count, i) {
+	sg = st->sgl;
+	st->nents = 0;
+	for (i = 0; i < page_count; i++) {
 		page = shmem_read_mapping_page_gfp(mapping, i, gfp);
 		if (IS_ERR(page)) {
 			i915_gem_purge(dev_priv, page_count);
@@ -1802,9 +1805,18 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 			gfp &= ~(__GFP_IO | __GFP_WAIT);
 		}
 
-		sg_set_page(sg, page, PAGE_SIZE, 0);
+		if (!i || page_to_pfn(page) != last_pfn + 1) {
+			if (i)
+				sg = sg_next(sg);
+			st->nents++;
+			sg_set_page(sg, page, PAGE_SIZE, 0);
+		} else {
+			sg->length += PAGE_SIZE;
+		}
+		last_pfn = page_to_pfn(page);
 	}
 
+	sg_mark_end(sg);
 	obj->pages = st;
 
 	if (i915_gem_object_needs_bit17_swizzle(obj))
@@ -1813,8 +1825,9 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 	return 0;
 
 err_pages:
-	for_each_sg(st->sgl, sg, i, page_count)
-		page_cache_release(sg_page(sg));
+	sg_mark_end(sg);
+	for_each_sg_page(st->sgl, &sg_iter, st->nents, 0)
+		page_cache_release(sg_iter.page);
 	sg_free_table(st);
 	kfree(st);
 	return PTR_ERR(page);
-- 
1.7.10.4