Merge tag 'drm-intel-fixes-2022-07-13' of...

Merge tag 'drm-intel-fixes-2022-07-13' of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes

- Selftest fix (Andrzej)
- TTM fix sg_table construction (Matt Auld)
- Error return fixes (Dan)
- Fix a performance regression related to waitboost (Chris)
- Fix GT resets (Chris)
Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/Ys87yMujcG2sJC1R@intel.com

drivers/gpu/drm/i915/gem/i915_gem_ttm.c

@@ -620,10 +620,15 @@ i915_ttm_resource_get_st(struct drm_i915_gem_object *obj,
 			 struct ttm_resource *res)
 {
 	struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
+	u64 page_alignment;
 
 	if (!i915_ttm_gtt_binds_lmem(res))
 		return i915_ttm_tt_get_st(bo->ttm);
 
+	page_alignment = bo->page_alignment << PAGE_SHIFT;
+	if (!page_alignment)
+		page_alignment = obj->mm.region->min_page_size;
+
 	/*
 	 * If CPU mapping differs, we need to add the ttm_tt pages to
 	 * the resulting st. Might make sense for GGTT.
@@ -634,7 +639,8 @@ i915_ttm_resource_get_st(struct drm_i915_gem_object *obj,
 			struct i915_refct_sgt *rsgt;
 
 			rsgt = intel_region_ttm_resource_to_rsgt(obj->mm.region,
-								 res);
+								 res,
+								 page_alignment);
 			if (IS_ERR(rsgt))
 				return rsgt;
 
@@ -643,7 +649,8 @@ i915_ttm_resource_get_st(struct drm_i915_gem_object *obj,
 		return i915_refct_sgt_get(obj->ttm.cached_io_rsgt);
 	}
 
-	return intel_region_ttm_resource_to_rsgt(obj->mm.region, res);
+	return intel_region_ttm_resource_to_rsgt(obj->mm.region, res,
+						 page_alignment);
 }
 
 static int i915_ttm_truncate(struct drm_i915_gem_object *obj)

drivers/gpu/drm/i915/gem/i915_gem_wait.c

@@ -9,6 +9,7 @@
 #include <linux/jiffies.h>
 
 #include "gt/intel_engine.h"
+#include "gt/intel_rps.h"
 
 #include "i915_gem_ioctls.h"
 #include "i915_gem_object.h"
@@ -31,6 +32,37 @@ i915_gem_object_wait_fence(struct dma_fence *fence,
 				      timeout);
 }
 
+static void
+i915_gem_object_boost(struct dma_resv *resv, unsigned int flags)
+{
+	struct dma_resv_iter cursor;
+	struct dma_fence *fence;
+
+	/*
+	 * Prescan all fences for potential boosting before we begin waiting.
+	 *
+	 * When we wait, we wait on outstanding fences serially. If the
+	 * dma-resv contains a sequence such as 1:1, 1:2 instead of a reduced
+	 * form 1:2, then as we look at each wait in turn we see that each
+	 * request is currently executing and not worthy of boosting. But if
+	 * we only happen to look at the final fence in the sequence (because
+	 * of request coalescing or splitting between read/write arrays by
+	 * the iterator), then we would boost. As such our decision to boost
+	 * or not is delicately balanced on the order we wait on fences.
+	 *
+	 * So instead of looking for boosts sequentially, look for all boosts
+	 * upfront and then wait on the outstanding fences.
+	 */
+
+	dma_resv_iter_begin(&cursor, resv,
+			    dma_resv_usage_rw(flags & I915_WAIT_ALL));
+	dma_resv_for_each_fence_unlocked(&cursor, fence)
+		if (dma_fence_is_i915(fence) &&
+		    !i915_request_started(to_request(fence)))
+			intel_rps_boost(to_request(fence));
+	dma_resv_iter_end(&cursor);
+}
+
 static long
 i915_gem_object_wait_reservation(struct dma_resv *resv,
 				 unsigned int flags,
@@ -40,6 +72,8 @@ i915_gem_object_wait_reservation(struct dma_resv *resv,
 	struct dma_fence *fence;
 	long ret = timeout ?: 1;
 
+	i915_gem_object_boost(resv, flags);
+
 	dma_resv_iter_begin(&cursor, resv,
 			    dma_resv_usage_rw(flags & I915_WAIT_ALL));
 	dma_resv_for_each_fence_unlocked(&cursor, fence) {

drivers/gpu/drm/i915/gt/intel_gt.c

@@ -1209,6 +1209,20 @@ void intel_gt_invalidate_tlbs(struct intel_gt *gt)
 	mutex_lock(&gt->tlb_invalidate_lock);
 	intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL);
 
+	spin_lock_irq(&uncore->lock); /* serialise invalidate with GT reset */
+
+	for_each_engine(engine, gt, id) {
+		struct reg_and_bit rb;
+
+		rb = get_reg_and_bit(engine, regs == gen8_regs, regs, num);
+		if (!i915_mmio_reg_offset(rb.reg))
+			continue;
+
+		intel_uncore_write_fw(uncore, rb.reg, rb.bit);
+	}
+
+	spin_unlock_irq(&uncore->lock);
+
 	for_each_engine(engine, gt, id) {
 		/*
 		 * HW architecture suggest typical invalidation time at 40us,
@@ -1223,7 +1237,6 @@ void intel_gt_invalidate_tlbs(struct intel_gt *gt)
 		if (!i915_mmio_reg_offset(rb.reg))
 			continue;
 
-		intel_uncore_write_fw(uncore, rb.reg, rb.bit);
 		if (__intel_wait_for_register_fw(uncore,
 						 rb.reg, rb.bit, 0,
 						 timeout_us, timeout_ms,

drivers/gpu/drm/i915/gt/intel_reset.c

@@ -300,7 +300,7 @@ static int gen6_hw_domain_reset(struct intel_gt *gt, u32 hw_domain_mask)
 	return err;
 }
 
-static int gen6_reset_engines(struct intel_gt *gt,
+static int __gen6_reset_engines(struct intel_gt *gt,
 				intel_engine_mask_t engine_mask,
 				unsigned int retry)
 {
@@ -321,6 +321,20 @@ static int gen6_reset_engines(struct intel_gt *gt,
 	return gen6_hw_domain_reset(gt, hw_mask);
 }
 
+static int gen6_reset_engines(struct intel_gt *gt,
+			      intel_engine_mask_t engine_mask,
+			      unsigned int retry)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&gt->uncore->lock, flags);
+	ret = __gen6_reset_engines(gt, engine_mask, retry);
+	spin_unlock_irqrestore(&gt->uncore->lock, flags);
+
+	return ret;
+}
+
 static struct intel_engine_cs *find_sfc_paired_vecs_engine(struct intel_engine_cs *engine)
 {
 	int vecs_id;
@@ -487,7 +501,7 @@ static void gen11_unlock_sfc(struct intel_engine_cs *engine)
 	rmw_clear_fw(uncore, sfc_lock.lock_reg, sfc_lock.lock_bit);
 }
 
-static int gen11_reset_engines(struct intel_gt *gt,
+static int __gen11_reset_engines(struct intel_gt *gt,
 				 intel_engine_mask_t engine_mask,
 				 unsigned int retry)
 {
@@ -583,8 +597,11 @@ static int gen8_reset_engines(struct intel_gt *gt,
 	struct intel_engine_cs *engine;
 	const bool reset_non_ready = retry >= 1;
 	intel_engine_mask_t tmp;
+	unsigned long flags;
 	int ret;
 
+	spin_lock_irqsave(&gt->uncore->lock, flags);
+
 	for_each_engine_masked(engine, gt, engine_mask, tmp) {
 		ret = gen8_engine_reset_prepare(engine);
 		if (ret && !reset_non_ready)
@@ -612,17 +629,19 @@ static int gen8_reset_engines(struct intel_gt *gt,
 	 * This is best effort, so ignore any error from the initial reset.
 	 */
 	if (IS_DG2(gt->i915) && engine_mask == ALL_ENGINES)
-		gen11_reset_engines(gt, gt->info.engine_mask, 0);
+		__gen11_reset_engines(gt, gt->info.engine_mask, 0);
 
 	if (GRAPHICS_VER(gt->i915) >= 11)
-		ret = gen11_reset_engines(gt, engine_mask, retry);
+		ret = __gen11_reset_engines(gt, engine_mask, retry);
 	else
-		ret = gen6_reset_engines(gt, engine_mask, retry);
+		ret = __gen6_reset_engines(gt, engine_mask, retry);
 
 skip_reset:
 	for_each_engine_masked(engine, gt, engine_mask, tmp)
 		gen8_engine_reset_cancel(engine);
 
+	spin_unlock_irqrestore(&gt->uncore->lock, flags);
+
 	return ret;
 }
 

drivers/gpu/drm/i915/gt/selftest_lrc.c

@@ -176,8 +176,8 @@ static int live_lrc_layout(void *arg)
 			continue;
 
 		hw = shmem_pin_map(engine->default_state);
-		if (IS_ERR(hw)) {
-			err = PTR_ERR(hw);
+		if (!hw) {
+			err = -ENOMEM;
 			break;
 		}
 		hw += LRC_STATE_OFFSET / sizeof(*hw);
@@ -365,8 +365,8 @@ static int live_lrc_fixed(void *arg)
 			continue;
 
 		hw = shmem_pin_map(engine->default_state);
-		if (IS_ERR(hw)) {
-			err = PTR_ERR(hw);
+		if (!hw) {
+			err = -ENOMEM;
 			break;
 		}
 		hw += LRC_STATE_OFFSET / sizeof(*hw);

drivers/gpu/drm/i915/gvt/cmd_parser.c

@@ -3117,9 +3117,9 @@ void intel_gvt_update_reg_whitelist(struct intel_vgpu *vgpu)
 			continue;
 
 		vaddr = shmem_pin_map(engine->default_state);
-		if (IS_ERR(vaddr)) {
-			gvt_err("failed to map %s->default state, err:%zd\n",
-				engine->name, PTR_ERR(vaddr));
+		if (!vaddr) {
+			gvt_err("failed to map %s->default state\n",
+				engine->name);
 			return;
 		}
 

drivers/gpu/drm/i915/i915_scatterlist.c

@@ -68,6 +68,7 @@ void i915_refct_sgt_init(struct i915_refct_sgt *rsgt, size_t size)
  * drm_mm_node
  * @node: The drm_mm_node.
  * @region_start: An offset to add to the dma addresses of the sg list.
+ * @page_alignment: Required page alignment for each sg entry. Power of two.
  *
  * Create a struct sg_table, initializing it from a struct drm_mm_node,
  * taking a maximum segment length into account, splitting into segments
@@ -77,15 +78,18 @@ void i915_refct_sgt_init(struct i915_refct_sgt *rsgt, size_t size)
  * error code cast to an error pointer on failure.
  */
 struct i915_refct_sgt *i915_rsgt_from_mm_node(const struct drm_mm_node *node,
-					      u64 region_start)
+					      u64 region_start,
+					      u64 page_alignment)
 {
-	const u64 max_segment = SZ_1G; /* Do we have a limit on this? */
+	const u64 max_segment = round_down(UINT_MAX, page_alignment);
 	u64 segment_pages = max_segment >> PAGE_SHIFT;
 	u64 block_size, offset, prev_end;
 	struct i915_refct_sgt *rsgt;
 	struct sg_table *st;
 	struct scatterlist *sg;
 
+	GEM_BUG_ON(!max_segment);
+
 	rsgt = kmalloc(sizeof(*rsgt), GFP_KERNEL);
 	if (!rsgt)
 		return ERR_PTR(-ENOMEM);
@@ -112,6 +116,8 @@ struct i915_refct_sgt *i915_rsgt_from_mm_node(const struct drm_mm_node *node,
 				sg = __sg_next(sg);
 
 			sg_dma_address(sg) = region_start + offset;
+			GEM_BUG_ON(!IS_ALIGNED(sg_dma_address(sg),
+					       page_alignment));
 			sg_dma_len(sg) = 0;
 			sg->length = 0;
 			st->nents++;
@@ -138,6 +144,7 @@ struct i915_refct_sgt *i915_rsgt_from_mm_node(const struct drm_mm_node *node,
  * i915_buddy_block list
  * @res: The struct i915_ttm_buddy_resource.
  * @region_start: An offset to add to the dma addresses of the sg list.
+ * @page_alignment: Required page alignment for each sg entry. Power of two.
  *
  * Create a struct sg_table, initializing it from struct i915_buddy_block list,
  * taking a maximum segment length into account, splitting into segments
@@ -147,11 +154,12 @@ struct i915_refct_sgt *i915_rsgt_from_mm_node(const struct drm_mm_node *node,
  * error code cast to an error pointer on failure.
  */
 struct i915_refct_sgt *i915_rsgt_from_buddy_resource(struct ttm_resource *res,
-						     u64 region_start)
+						     u64 region_start,
+						     u64 page_alignment)
 {
 	struct i915_ttm_buddy_resource *bman_res = to_ttm_buddy_resource(res);
 	const u64 size = res->num_pages << PAGE_SHIFT;
-	const u64 max_segment = rounddown(UINT_MAX, PAGE_SIZE);
+	const u64 max_segment = round_down(UINT_MAX, page_alignment);
 	struct drm_buddy *mm = bman_res->mm;
 	struct list_head *blocks = &bman_res->blocks;
 	struct drm_buddy_block *block;
@@ -161,6 +169,7 @@ struct i915_refct_sgt *i915_rsgt_from_buddy_resource(struct ttm_resource *res,
 	resource_size_t prev_end;
 
 	GEM_BUG_ON(list_empty(blocks));
+	GEM_BUG_ON(!max_segment);
 
 	rsgt = kmalloc(sizeof(*rsgt), GFP_KERNEL);
 	if (!rsgt)
@@ -191,6 +200,8 @@ struct i915_refct_sgt *i915_rsgt_from_buddy_resource(struct ttm_resource *res,
 					sg = __sg_next(sg);
 
 				sg_dma_address(sg) = region_start + offset;
+				GEM_BUG_ON(!IS_ALIGNED(sg_dma_address(sg),
+						       page_alignment));
 				sg_dma_len(sg) = 0;
 				sg->length = 0;
 				st->nents++;

drivers/gpu/drm/i915/i915_scatterlist.h

@@ -213,9 +213,11 @@ static inline void __i915_refct_sgt_init(struct i915_refct_sgt *rsgt,
 void i915_refct_sgt_init(struct i915_refct_sgt *rsgt, size_t size);
 
 struct i915_refct_sgt *i915_rsgt_from_mm_node(const struct drm_mm_node *node,
-					      u64 region_start);
+					      u64 region_start,
+					      u64 page_alignment);
 
 struct i915_refct_sgt *i915_rsgt_from_buddy_resource(struct ttm_resource *res,
-						     u64 region_start);
+						     u64 region_start,
+						     u64 page_alignment);
 
 #endif

drivers/gpu/drm/i915/intel_region_ttm.c

@@ -152,6 +152,7 @@ int intel_region_ttm_fini(struct intel_memory_region *mem)
  * Convert an opaque TTM resource manager resource to a refcounted sg_table.
  * @mem: The memory region.
  * @res: The resource manager resource obtained from the TTM resource manager.
+ * @page_alignment: Required page alignment for each sg entry. Power of two.
  *
  * The gem backends typically use sg-tables for operations on the underlying
  * io_memory. So provide a way for the backends to translate the
@@ -161,16 +162,19 @@ int intel_region_ttm_fini(struct intel_memory_region *mem)
  */
 struct i915_refct_sgt *
 intel_region_ttm_resource_to_rsgt(struct intel_memory_region *mem,
-				  struct ttm_resource *res)
+				  struct ttm_resource *res,
+				  u64 page_alignment)
 {
 	if (mem->is_range_manager) {
 		struct ttm_range_mgr_node *range_node =
 			to_ttm_range_mgr_node(res);
 
 		return i915_rsgt_from_mm_node(&range_node->mm_nodes[0],
-					      mem->region.start);
+					      mem->region.start,
+					      page_alignment);
 	} else {
-		return i915_rsgt_from_buddy_resource(res, mem->region.start);
+		return i915_rsgt_from_buddy_resource(res, mem->region.start,
+						     page_alignment);
 	}
 }
 

drivers/gpu/drm/i915/intel_region_ttm.h

@@ -24,7 +24,8 @@ int intel_region_ttm_fini(struct intel_memory_region *mem);
 
 struct i915_refct_sgt *
 intel_region_ttm_resource_to_rsgt(struct intel_memory_region *mem,
-				  struct ttm_resource *res);
+				  struct ttm_resource *res,
+				  u64 page_alignment);
 
 void intel_region_ttm_resource_free(struct intel_memory_region *mem,
 				    struct ttm_resource *res);

drivers/gpu/drm/i915/selftests/i915_gem_gtt.c

@@ -742,7 +742,7 @@ static int pot_hole(struct i915_address_space *vm,
 		u64 addr;
 
 		for (addr = round_up(hole_start + min_alignment, step) - min_alignment;
-		     addr <= round_down(hole_end - (2 * min_alignment), step) - min_alignment;
+		     hole_end > addr && hole_end - addr >= 2 * min_alignment;
 		     addr += step) {
 			err = i915_vma_pin(vma, 0, 0, addr | flags);
 			if (err) {

drivers/gpu/drm/i915/selftests/intel_memory_region.c

@@ -451,7 +451,6 @@ static int igt_mock_splintered_region(void *arg)
 
 static int igt_mock_max_segment(void *arg)
 {
-	const unsigned int max_segment = rounddown(UINT_MAX, PAGE_SIZE);
 	struct intel_memory_region *mem = arg;
 	struct drm_i915_private *i915 = mem->i915;
 	struct i915_ttm_buddy_resource *res;
@@ -460,7 +459,10 @@ static int igt_mock_max_segment(void *arg)
 	struct drm_buddy *mm;
 	struct list_head *blocks;
 	struct scatterlist *sg;
+	I915_RND_STATE(prng);
 	LIST_HEAD(objects);
+	unsigned int max_segment;
+	unsigned int ps;
 	u64 size;
 	int err = 0;
 
@@ -472,7 +474,13 @@ static int igt_mock_max_segment(void *arg)
 	 */
 
 	size = SZ_8G;
-	mem = mock_region_create(i915, 0, size, PAGE_SIZE, 0, 0);
+	ps = PAGE_SIZE;
+	if (i915_prandom_u64_state(&prng) & 1)
+		ps = SZ_64K; /* For something like DG2 */
+
+	max_segment = round_down(UINT_MAX, ps);
+
+	mem = mock_region_create(i915, 0, size, ps, 0, 0);
 	if (IS_ERR(mem))
 		return PTR_ERR(mem);
 
@@ -498,12 +506,21 @@ static int igt_mock_max_segment(void *arg)
 	}
 
 	for (sg = obj->mm.pages->sgl; sg; sg = sg_next(sg)) {
+		dma_addr_t daddr = sg_dma_address(sg);
+
 		if (sg->length > max_segment) {
 			pr_err("%s: Created an oversized scatterlist entry, %u > %u\n",
 			       __func__, sg->length, max_segment);
 			err = -EINVAL;
 			goto out_close;
 		}
+
+		if (!IS_ALIGNED(daddr, ps)) {
+			pr_err("%s: Created an unaligned scatterlist entry, addr=%pa, ps=%u\n",
+			       __func__,  &daddr, ps);
+			err = -EINVAL;
+			goto out_close;
+		}
 	}
 
 out_close:

drivers/gpu/drm/i915/selftests/mock_region.c

@@ -33,7 +33,8 @@ static int mock_region_get_pages(struct drm_i915_gem_object *obj)
 		return PTR_ERR(obj->mm.res);
 
 	obj->mm.rsgt = intel_region_ttm_resource_to_rsgt(obj->mm.region,
-							 obj->mm.res);
+							 obj->mm.res,
+							 obj->mm.region->min_page_size);
 	if (IS_ERR(obj->mm.rsgt)) {
 		err = PTR_ERR(obj->mm.rsgt);
 		goto err_free_resource;