drm/i915/blt: support copying objects

We can already clear an object with the blt, so try to do the same to
support copying from one object backing store to another. Really this is
just object -> object, which is not that useful yet, what we really want
is two backing stores, but that will require some vma rework first,
otherwise we are stuck with "tmp" objects.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20190810174338.19810-1-chris@chris-wilson.co.uk

drivers/gpu/drm/i915/gem/i915_gem_object_blt.c

@@ -33,8 +33,10 @@ struct i915_vma *intel_emit_vma_fill_blt(struct intel_context *ce,
 	size = (1 + 8 * count) * sizeof(u32);
 	size = round_up(size, PAGE_SIZE);
 	pool = intel_engine_pool_get(&ce->engine->pool, size);
-	if (IS_ERR(pool))
+	if (IS_ERR(pool)) {
+		err = PTR_ERR(pool);
 		goto out_pm;
+	}
 
 	cmd = i915_gem_object_pin_map(pool->obj, I915_MAP_WC);
 	if (IS_ERR(cmd)) {
@@ -189,6 +191,207 @@ int i915_gem_object_fill_blt(struct drm_i915_gem_object *obj,
 	return err;
 }
 
+struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce,
+					 struct i915_vma *src,
+					 struct i915_vma *dst)
+{
+	struct drm_i915_private *i915 = ce->vm->i915;
+	const u32 block_size = S16_MAX * PAGE_SIZE;
+	struct intel_engine_pool_node *pool;
+	struct i915_vma *batch;
+	u64 src_offset, dst_offset;
+	u64 count, rem;
+	u32 size, *cmd;
+	int err;
+
+	GEM_BUG_ON(src->size != dst->size);
+
+	GEM_BUG_ON(intel_engine_is_virtual(ce->engine));
+	intel_engine_pm_get(ce->engine);
+
+	count = div_u64(dst->size, block_size);
+	size = (1 + 11 * count) * sizeof(u32);
+	size = round_up(size, PAGE_SIZE);
+	pool = intel_engine_pool_get(&ce->engine->pool, size);
+	if (IS_ERR(pool)) {
+		err = PTR_ERR(pool);
+		goto out_pm;
+	}
+
+	cmd = i915_gem_object_pin_map(pool->obj, I915_MAP_WC);
+	if (IS_ERR(cmd)) {
+		err = PTR_ERR(cmd);
+		goto out_put;
+	}
+
+	rem = src->size;
+	src_offset = src->node.start;
+	dst_offset = dst->node.start;
+
+	do {
+		size = min_t(u64, rem, block_size);
+		GEM_BUG_ON(size >> PAGE_SHIFT > S16_MAX);
+
+		if (INTEL_GEN(i915) >= 9) {
+			*cmd++ = GEN9_XY_FAST_COPY_BLT_CMD | (10 - 2);
+			*cmd++ = BLT_DEPTH_32 | PAGE_SIZE;
+			*cmd++ = 0;
+			*cmd++ = size >> PAGE_SHIFT << 16 | PAGE_SIZE / 4;
+			*cmd++ = lower_32_bits(dst_offset);
+			*cmd++ = upper_32_bits(dst_offset);
+			*cmd++ = 0;
+			*cmd++ = PAGE_SIZE;
+			*cmd++ = lower_32_bits(src_offset);
+			*cmd++ = upper_32_bits(src_offset);
+		} else if (INTEL_GEN(i915) >= 8) {
+			*cmd++ = XY_SRC_COPY_BLT_CMD | BLT_WRITE_RGBA | (10 - 2);
+			*cmd++ = BLT_DEPTH_32 | BLT_ROP_SRC_COPY | PAGE_SIZE;
+			*cmd++ = 0;
+			*cmd++ = size >> PAGE_SHIFT << 16 | PAGE_SIZE / 4;
+			*cmd++ = lower_32_bits(dst_offset);
+			*cmd++ = upper_32_bits(dst_offset);
+			*cmd++ = 0;
+			*cmd++ = PAGE_SIZE;
+			*cmd++ = lower_32_bits(src_offset);
+			*cmd++ = upper_32_bits(src_offset);
+		} else {
+			*cmd++ = SRC_COPY_BLT_CMD | BLT_WRITE_RGBA | (6 - 2);
+			*cmd++ = BLT_DEPTH_32 | BLT_ROP_SRC_COPY | PAGE_SIZE;
+			*cmd++ = size >> PAGE_SHIFT << 16 | PAGE_SIZE;
+			*cmd++ = dst_offset;
+			*cmd++ = PAGE_SIZE;
+			*cmd++ = src_offset;
+		}
+
+		/* Allow ourselves to be preempted in between blocks. */
+		*cmd++ = MI_ARB_CHECK;
+
+		src_offset += size;
+		dst_offset += size;
+		rem -= size;
+	} while (rem);
+
+	*cmd = MI_BATCH_BUFFER_END;
+	intel_gt_chipset_flush(ce->vm->gt);
+
+	i915_gem_object_unpin_map(pool->obj);
+
+	batch = i915_vma_instance(pool->obj, ce->vm, NULL);
+	if (IS_ERR(batch)) {
+		err = PTR_ERR(batch);
+		goto out_put;
+	}
+
+	err = i915_vma_pin(batch, 0, 0, PIN_USER);
+	if (unlikely(err))
+		goto out_put;
+
+	batch->private = pool;
+	return batch;
+
+out_put:
+	intel_engine_pool_put(pool);
+out_pm:
+	intel_engine_pm_put(ce->engine);
+	return ERR_PTR(err);
+}
+
+static int move_to_gpu(struct i915_vma *vma, struct i915_request *rq, bool write)
+{
+	struct drm_i915_gem_object *obj = vma->obj;
+
+	if (obj->cache_dirty & ~obj->cache_coherent)
+		i915_gem_clflush_object(obj, 0);
+
+	return i915_request_await_object(rq, obj, write);
+}
+
+int i915_gem_object_copy_blt(struct drm_i915_gem_object *src,
+			     struct drm_i915_gem_object *dst,
+			     struct intel_context *ce)
+{
+	struct drm_gem_object *objs[] = { &src->base, &dst->base };
+	struct i915_address_space *vm = ce->vm;
+	struct i915_vma *vma[2], *batch;
+	struct ww_acquire_ctx acquire;
+	struct i915_request *rq;
+	int err, i;
+
+	vma[0] = i915_vma_instance(src, vm, NULL);
+	if (IS_ERR(vma[0]))
+		return PTR_ERR(vma[0]);
+
+	err = i915_vma_pin(vma[0], 0, 0, PIN_USER);
+	if (unlikely(err))
+		return err;
+
+	vma[1] = i915_vma_instance(dst, vm, NULL);
+	if (IS_ERR(vma[1]))
+		goto out_unpin_src;
+
+	err = i915_vma_pin(vma[1], 0, 0, PIN_USER);
+	if (unlikely(err))
+		goto out_unpin_src;
+
+	batch = intel_emit_vma_copy_blt(ce, vma[0], vma[1]);
+	if (IS_ERR(batch)) {
+		err = PTR_ERR(batch);
+		goto out_unpin_dst;
+	}
+
+	rq = intel_context_create_request(ce);
+	if (IS_ERR(rq)) {
+		err = PTR_ERR(rq);
+		goto out_batch;
+	}
+
+	err = intel_emit_vma_mark_active(batch, rq);
+	if (unlikely(err))
+		goto out_request;
+
+	err = drm_gem_lock_reservations(objs, ARRAY_SIZE(objs), &acquire);
+	if (unlikely(err))
+		goto out_request;
+
+	for (i = 0; i < ARRAY_SIZE(vma); i++) {
+		err = move_to_gpu(vma[i], rq, i);
+		if (unlikely(err))
+			goto out_unlock;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(vma); i++) {
+		unsigned int flags = i ? EXEC_OBJECT_WRITE : 0;
+
+		err = i915_vma_move_to_active(vma[i], rq, flags);
+		if (unlikely(err))
+			goto out_unlock;
+	}
+
+	if (rq->engine->emit_init_breadcrumb) {
+		err = rq->engine->emit_init_breadcrumb(rq);
+		if (unlikely(err))
+			goto out_unlock;
+	}
+
+	err = rq->engine->emit_bb_start(rq,
+					batch->node.start, batch->node.size,
+					0);
+out_unlock:
+	drm_gem_unlock_reservations(objs, ARRAY_SIZE(objs), &acquire);
+out_request:
+	if (unlikely(err))
+		i915_request_skip(rq, err);
+
+	i915_request_add(rq);
+out_batch:
+	intel_emit_vma_release(ce, batch);
+out_unpin_dst:
+	i915_vma_unpin(vma[1]);
+out_unpin_src:
+	i915_vma_unpin(vma[0]);
+	return err;
+}
+
 #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
 #include "selftests/i915_gem_object_blt.c"
 #endif

drivers/gpu/drm/i915/gem/i915_gem_object_blt.h

@@ -19,6 +19,10 @@ struct i915_vma *intel_emit_vma_fill_blt(struct intel_context *ce,
 					 struct i915_vma *vma,
 					 u32 value);
 
+struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce,
+					 struct i915_vma *src,
+					 struct i915_vma *dst);
+
 int intel_emit_vma_mark_active(struct i915_vma *vma, struct i915_request *rq);
 void intel_emit_vma_release(struct intel_context *ce, struct i915_vma *vma);
 
@@ -26,4 +30,8 @@ int i915_gem_object_fill_blt(struct drm_i915_gem_object *obj,
 			     struct intel_context *ce,
 			     u32 value);
 
+int i915_gem_object_copy_blt(struct drm_i915_gem_object *src,
+			     struct drm_i915_gem_object *dst,
+			     struct intel_context *ce);
+
 #endif

drivers/gpu/drm/i915/gem/selftests/i915_gem_object_blt.c

@@ -103,10 +103,116 @@ static int igt_fill_blt(void *arg)
 	return err;
 }
 
+static int igt_copy_blt(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct intel_context *ce = i915->engine[BCS0]->kernel_context;
+	struct drm_i915_gem_object *src, *dst;
+	struct rnd_state prng;
+	IGT_TIMEOUT(end);
+	u32 *vaddr;
+	int err = 0;
+
+	prandom_seed_state(&prng, i915_selftest.random_seed);
+
+	do {
+		const u32 max_block_size = S16_MAX * PAGE_SIZE;
+		u32 sz = min_t(u64, ce->vm->total >> 4, prandom_u32_state(&prng));
+		u32 phys_sz = sz % (max_block_size + 1);
+		u32 val = prandom_u32_state(&prng);
+		u32 i;
+
+		sz = round_up(sz, PAGE_SIZE);
+		phys_sz = round_up(phys_sz, PAGE_SIZE);
+
+		pr_debug("%s with phys_sz= %x, sz=%x, val=%x\n", __func__,
+			 phys_sz, sz, val);
+
+		src = huge_gem_object(i915, phys_sz, sz);
+		if (IS_ERR(src)) {
+			err = PTR_ERR(src);
+			goto err_flush;
+		}
+
+		vaddr = i915_gem_object_pin_map(src, I915_MAP_WB);
+		if (IS_ERR(vaddr)) {
+			err = PTR_ERR(vaddr);
+			goto err_put_src;
+		}
+
+		memset32(vaddr, val,
+			 huge_gem_object_phys_size(src) / sizeof(u32));
+
+		i915_gem_object_unpin_map(src);
+
+		if (!(src->cache_coherent & I915_BO_CACHE_COHERENT_FOR_READ))
+			src->cache_dirty = true;
+
+		dst = huge_gem_object(i915, phys_sz, sz);
+		if (IS_ERR(dst)) {
+			err = PTR_ERR(dst);
+			goto err_put_src;
+		}
+
+		vaddr = i915_gem_object_pin_map(dst, I915_MAP_WB);
+		if (IS_ERR(vaddr)) {
+			err = PTR_ERR(vaddr);
+			goto err_put_dst;
+		}
+
+		memset32(vaddr, val ^ 0xdeadbeaf,
+			 huge_gem_object_phys_size(dst) / sizeof(u32));
+
+		if (!(dst->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE))
+			dst->cache_dirty = true;
+
+		mutex_lock(&i915->drm.struct_mutex);
+		err = i915_gem_object_copy_blt(src, dst, ce);
+		mutex_unlock(&i915->drm.struct_mutex);
+		if (err)
+			goto err_unpin;
+
+		i915_gem_object_lock(dst);
+		err = i915_gem_object_set_to_cpu_domain(dst, false);
+		i915_gem_object_unlock(dst);
+		if (err)
+			goto err_unpin;
+
+		for (i = 0; i < huge_gem_object_phys_size(dst) / sizeof(u32); ++i) {
+			if (vaddr[i] != val) {
+				pr_err("vaddr[%u]=%x, expected=%x\n", i,
+				       vaddr[i], val);
+				err = -EINVAL;
+				goto err_unpin;
+			}
+		}
+
+		i915_gem_object_unpin_map(dst);
+
+		i915_gem_object_put(src);
+		i915_gem_object_put(dst);
+	} while (!time_after(jiffies, end));
+
+	goto err_flush;
+
+err_unpin:
+	i915_gem_object_unpin_map(dst);
+err_put_dst:
+	i915_gem_object_put(dst);
+err_put_src:
+	i915_gem_object_put(src);
+err_flush:
+	if (err == -ENOMEM)
+		err = 0;
+
+	return err;
+}
+
 int i915_gem_object_blt_live_selftests(struct drm_i915_private *i915)
 {
 	static const struct i915_subtest tests[] = {
 		SUBTEST(igt_fill_blt),
+		SUBTEST(igt_copy_blt),
 	};
 
 	if (intel_gt_is_wedged(&i915->gt))

drivers/gpu/drm/i915/gt/intel_gpu_commands.h

@@ -186,11 +186,12 @@
 #define GFX_OP_DRAWRECT_INFO     ((0x3<<29)|(0x1d<<24)|(0x80<<16)|(0x3))
 #define GFX_OP_DRAWRECT_INFO_I965  ((0x7900<<16)|0x2)
 
-#define COLOR_BLT_CMD			(2<<29 | 0x40<<22 | (5-2))
+#define COLOR_BLT_CMD			(2 << 29 | 0x40 << 22 | (5 - 2))
 #define XY_COLOR_BLT_CMD		(2 << 29 | 0x50 << 22)
-#define SRC_COPY_BLT_CMD		((2<<29)|(0x43<<22)|4)
-#define XY_SRC_COPY_BLT_CMD		((2<<29)|(0x53<<22)|6)
-#define XY_MONO_SRC_COPY_IMM_BLT	((2<<29)|(0x71<<22)|5)
+#define SRC_COPY_BLT_CMD		(2 << 29 | 0x43 << 22)
+#define GEN9_XY_FAST_COPY_BLT_CMD	(2 << 29 | 0x42 << 22)
+#define XY_SRC_COPY_BLT_CMD		(2 << 29 | 0x53 << 22)
+#define XY_MONO_SRC_COPY_IMM_BLT	(2 << 29 | 0x71 << 22 | 5)
 #define   BLT_WRITE_A			(2<<20)
 #define   BLT_WRITE_RGB			(1<<20)
 #define   BLT_WRITE_RGBA		(BLT_WRITE_RGB | BLT_WRITE_A)

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

@@ -1136,7 +1136,7 @@ i830_emit_bb_start(struct i915_request *rq,
 		 * stable batch scratch bo area (so that the CS never
 		 * stumbles over its tlb invalidation bug) ...
 		 */
-		*cs++ = SRC_COPY_BLT_CMD | BLT_WRITE_RGBA;
+		*cs++ = SRC_COPY_BLT_CMD | BLT_WRITE_RGBA | (6 - 2);
 		*cs++ = BLT_DEPTH_32 | BLT_ROP_SRC_COPY | 4096;
 		*cs++ = DIV_ROUND_UP(len, 4096) << 16 | 4096;
 		*cs++ = cs_offset;