Merge tag 'drm-intel-next-2017-11-17-1' of git://anongit.freedesktop.org/drm/drm-intel into drm-next

More change sets for 4.16:

- Many improvements for selftests and other igt tests (Chris)
- Forcewake with PUNIT->PMIC bus fixes and robustness (Hans)
- Define an engine class for uABI (Tvrtko)
- Context switch fixes and improvements (Chris)
- GT powersavings and power gating simplification and fixes (Chris)
- Other general driver clean-ups (Chris, Lucas, Ville)
- Removing old, useless and/or bad workarounds (Chris, Oscar, Radhakrishna)
- IPS, pipe config, etc in preparation for another Fast Boot attempt (Maarten)
- OA perf fixes and support to Coffee Lake and Cannonlake (Lionel)
- Fixes around GPU fault registers (Michel)
- GEM Proxy (Tina)
- Refactor of Geminilake and Cannonlake plane color handling (James)
- Generalize transcoder loop (Mika Kahola)
- New HW Workaround for Cannonlake and Geminilake (Rodrigo)
- Resume GuC before using GEM (Chris)
- Stolen Memory handling improvements (Ville)
- Initialize entry in PPAT for older compilers (Chris)
- Other fixes and robustness improvements on execbuf (Chris)
- Improve logs of GEM_BUG_ON (Mika Kuoppala)
- Rework with massive rename of GuC functions and files (Sagar)
- Don't sanitize frame start delay if pipe is off (Ville)
- Cannonlake clock fixes (Rodrigo)
- Cannonlake HDMI 2.0 support (Rodrigo)
- Add a GuC doorbells selftest (Michel)
- Add might_sleep() check to our wait_for() (Chris)

Many GVT changes for 4.16:

- CSB HWSP update support (Weinan)
- GVT debug helpers, dyndbg and debugfs (Chuanxiao, Shuo)
- full virtualized opregion (Xiaolin)
- VM health check for sane fallback (Fred)
- workload submission code refactor for future enabling (Zhi)
- Updated repo URL in MAINTAINERS (Zhenyu)
- other many misc fixes

* tag 'drm-intel-next-2017-11-17-1' of git://anongit.freedesktop.org/drm/drm-intel: (260 commits)
  drm/i915: Update DRIVER_DATE to 20171117
  drm/i915: Add a policy note for removing workarounds
  drm/i915/selftests: Report ENOMEM clearly for an allocation failure
  Revert "drm/i915: Display WA #1133 WaFbcSkipSegments:cnl, glk"
  drm/i915: Calculate g4x intermediate watermarks correctly
  drm/i915: Calculate vlv/chv intermediate watermarks correctly, v3.
  drm/i915: Pass crtc_state to ips toggle functions, v2
  drm/i915: Pass idle crtc_state to intel_dp_sink_crc
  drm/i915: Enable FIFO underrun reporting after initial fastset, v4.
  drm/i915: Mark the userptr invalidate workqueue as WQ_MEM_RECLAIM
  drm/i915: Add might_sleep() check to wait_for()
  drm/i915/selftests: Add a GuC doorbells selftest
  drm/i915/cnl: Extend HDMI 2.0 support to CNL.
  drm/i915/cnl: Simplify dco_fraction calculation.
  drm/i915/cnl: Don't blindly replace qdiv.
  drm/i915/cnl: Fix wrpll math for higher freqs.
  drm/i915/cnl: Fix, simplify and unify wrpll variable sizes.
  drm/i915/cnl: Remove useless conversion.
  drm/i915/cnl: Remove spurious central_freq.
  drm/i915/selftests: exercise_ggtt may have nothing to do
  ...

14 files changed, 461 insertions, 49 deletions

diff --git a/drivers/gpu/drm/i915/selftests/huge_pages.c b/drivers/gpu/drm/i915/selftests/huge_pages.c
index 5cc8101..01af540 100644
--- a/drivers/gpu/drm/i915/selftests/huge_pages.c
+++ b/drivers/gpu/drm/i915/selftests/huge_pages.c
@@ -1159,6 +1159,9 @@ static int igt_ppgtt_exhaust_huge(void *arg)
 	int n, i;
 	int err = -ENODEV;
 
+	if (supported == I915_GTT_PAGE_SIZE_4K)
+		return 0;
+
 	/*
 	 * Sanity check creating objects with a varying mix of page sizes --
 	 * ensuring that our writes lands in the right place.
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_coherency.c b/drivers/gpu/drm/i915/selftests/i915_gem_coherency.c
index 35d778d..7a0d1e1 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_coherency.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_coherency.c
@@ -33,7 +33,7 @@ static int cpu_set(struct drm_i915_gem_object *obj,
 {
 	unsigned int needs_clflush;
 	struct page *page;
-	typeof(v) *map;
+	u32 *map;
 	int err;
 
 	err = i915_gem_obj_prepare_shmem_write(obj, &needs_clflush);
@@ -59,7 +59,7 @@ static int cpu_get(struct drm_i915_gem_object *obj,
 {
 	unsigned int needs_clflush;
 	struct page *page;
-	typeof(v) map;
+	u32 *map;
 	int err;
 
 	err = i915_gem_obj_prepare_shmem_read(obj, &needs_clflush);
@@ -82,7 +82,7 @@ static int gtt_set(struct drm_i915_gem_object *obj,
 		   u32 v)
 {
 	struct i915_vma *vma;
-	typeof(v) *map;
+	u32 __iomem *map;
 	int err;
 
 	err = i915_gem_object_set_to_gtt_domain(obj, true);
@@ -98,7 +98,7 @@ static int gtt_set(struct drm_i915_gem_object *obj,
 	if (IS_ERR(map))
 		return PTR_ERR(map);
 
-	map[offset / sizeof(*map)] = v;
+	iowrite32(v, &map[offset / sizeof(*map)]);
 	i915_vma_unpin_iomap(vma);
 
 	return 0;
@@ -109,7 +109,7 @@ static int gtt_get(struct drm_i915_gem_object *obj,
 		   u32 *v)
 {
 	struct i915_vma *vma;
-	typeof(v) map;
+	u32 __iomem *map;
 	int err;
 
 	err = i915_gem_object_set_to_gtt_domain(obj, false);
@@ -125,7 +125,7 @@ static int gtt_get(struct drm_i915_gem_object *obj,
 	if (IS_ERR(map))
 		return PTR_ERR(map);
 
-	*v = map[offset / sizeof(*map)];
+	*v = ioread32(&map[offset / sizeof(*map)]);
 	i915_vma_unpin_iomap(vma);
 
 	return 0;
@@ -135,7 +135,7 @@ static int wc_set(struct drm_i915_gem_object *obj,
 		  unsigned long offset,
 		  u32 v)
 {
-	typeof(v) *map;
+	u32 *map;
 	int err;
 
 	err = i915_gem_object_set_to_wc_domain(obj, true);
@@ -156,7 +156,7 @@ static int wc_get(struct drm_i915_gem_object *obj,
 		  unsigned long offset,
 		  u32 *v)
 {
-	typeof(v) map;
+	u32 *map;
 	int err;
 
 	err = i915_gem_object_set_to_wc_domain(obj, false);
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/selftests/i915_gem_context.c
index def5052..c82780a 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_context.c
@@ -325,7 +325,7 @@ static int igt_ctx_exec(void *arg)
 	LIST_HEAD(objects);
 	unsigned long ncontexts, ndwords, dw;
 	bool first_shared_gtt = true;
-	int err;
+	int err = -ENODEV;
 
 	/* Create a few different contexts (with different mm) and write
 	 * through each ctx/mm using the GPU making sure those writes end
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
index 9da0c9f..6491cf0 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
@@ -216,13 +216,21 @@ static int lowlevel_hole(struct drm_i915_private *i915,
 		hole_size = (hole_end - hole_start) >> size;
 		if (hole_size > KMALLOC_MAX_SIZE / sizeof(u32))
 			hole_size = KMALLOC_MAX_SIZE / sizeof(u32);
-		count = hole_size;
+		count = hole_size >> 1;
+		if (!count) {
+			pr_debug("%s: hole is too small [%llx - %llx] >> %d: %lld\n",
+				 __func__, hole_start, hole_end, size, hole_size);
+			break;
+		}
+
 		do {
-			count >>= 1;
 			order = i915_random_order(count, &prng);
-		} while (!order && count);
-		if (!order)
-			break;
+			if (order)
+				break;
+		} while (count >>= 1);
+		if (!count)
+			return -ENOMEM;
+		GEM_BUG_ON(!order);
 
 		GEM_BUG_ON(count * BIT_ULL(size) > vm->total);
 		GEM_BUG_ON(hole_start + count * BIT_ULL(size) > hole_end);
@@ -267,7 +275,9 @@ static int lowlevel_hole(struct drm_i915_private *i915,
 			mock_vma.node.size = BIT_ULL(size);
 			mock_vma.node.start = addr;
 
+			intel_runtime_pm_get(i915);
 			vm->insert_entries(vm, &mock_vma, I915_CACHE_NONE, 0);
+			intel_runtime_pm_put(i915);
 		}
 		count = n;
 
@@ -697,18 +707,26 @@ static int drunk_hole(struct drm_i915_private *i915,
 		unsigned int *order, count, n;
 		struct i915_vma *vma;
 		u64 hole_size;
-		int err;
+		int err = -ENODEV;
 
 		hole_size = (hole_end - hole_start) >> size;
 		if (hole_size > KMALLOC_MAX_SIZE / sizeof(u32))
 			hole_size = KMALLOC_MAX_SIZE / sizeof(u32);
-		count = hole_size;
+		count = hole_size >> 1;
+		if (!count) {
+			pr_debug("%s: hole is too small [%llx - %llx] >> %d: %lld\n",
+				 __func__, hole_start, hole_end, size, hole_size);
+			break;
+		}
+
 		do {
-			count >>= 1;
 			order = i915_random_order(count, &prng);
-		} while (!order && count);
-		if (!order)
-			break;
+			if (order)
+				break;
+		} while (count >>= 1);
+		if (!count)
+			return -ENOMEM;
+		GEM_BUG_ON(!order);
 
 		/* Ignore allocation failures (i.e. don't report them as
 		 * a test failure) as we are purposefully allocating very
@@ -956,7 +974,7 @@ static int exercise_ggtt(struct drm_i915_private *i915,
 	u64 hole_start, hole_end, last = 0;
 	struct drm_mm_node *node;
 	IGT_TIMEOUT(end_time);
-	int err;
+	int err = 0;
 
 	mutex_lock(&i915->drm.struct_mutex);
 restart:
@@ -1047,6 +1065,7 @@ static int igt_ggtt_page(void *arg)
 		goto out_remove;
 	}
 
+	intel_runtime_pm_get(i915);
 	for (n = 0; n < count; n++) {
 		u64 offset = tmp.start + order[n] * PAGE_SIZE;
 		u32 __iomem *vaddr;
@@ -1086,6 +1105,7 @@ static int igt_ggtt_page(void *arg)
 			break;
 		}
 	}
+	intel_runtime_pm_put(i915);
 
 	kfree(order);
 out_remove:
@@ -1160,7 +1180,7 @@ static int igt_gtt_reserve(void *arg)
 	struct drm_i915_gem_object *obj, *on;
 	LIST_HEAD(objects);
 	u64 total;
-	int err;
+	int err = -ENODEV;
 
 	/* i915_gem_gtt_reserve() tries to reserve the precise range
 	 * for the node, and evicts if it has to. So our test checks that
@@ -1351,7 +1371,7 @@ static int igt_gtt_insert(void *arg)
 	}, *ii;
 	LIST_HEAD(objects);
 	u64 total;
-	int err;
+	int err = -ENODEV;
 
 	/* i915_gem_gtt_insert() tries to allocate some free space in the GTT
 	 * to the node, evicting if required.
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_object.c b/drivers/gpu/drm/i915/selftests/i915_gem_object.c
index 1b8774a..f32aa6b 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_object.c
@@ -317,6 +317,7 @@ static int igt_partial_tiling(void *arg)
 	}
 
 	mutex_lock(&i915->drm.struct_mutex);
+	intel_runtime_pm_get(i915);
 
 	if (1) {
 		IGT_TIMEOUT(end);
@@ -418,6 +419,7 @@ next_tiling: ;
 	}
 
 out_unlock:
+	intel_runtime_pm_put(i915);
 	mutex_unlock(&i915->drm.struct_mutex);
 	i915_gem_object_unpin_pages(obj);
 out:
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_request.c b/drivers/gpu/drm/i915/selftests/i915_gem_request.c
index a999161..6bce990 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_request.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_request.c
@@ -332,7 +332,7 @@ static int live_nop_request(void *arg)
 	struct intel_engine_cs *engine;
 	struct live_test t;
 	unsigned int id;
-	int err;
+	int err = -ENODEV;
 
 	/* Submit various sized batches of empty requests, to each engine
 	 * (individually), and wait for the batch to complete. We can check
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_timeline.c b/drivers/gpu/drm/i915/selftests/i915_gem_timeline.c
index 4795877..3000e6a 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_timeline.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_timeline.c
@@ -79,7 +79,7 @@ static int igt_sync(void *arg)
 	}, *p;
 	struct intel_timeline *tl;
 	int order, offset;
-	int ret;
+	int ret = -ENODEV;
 
 	tl = mock_timeline(0);
 	if (!tl)
diff --git a/drivers/gpu/drm/i915/selftests/i915_live_selftests.h b/drivers/gpu/drm/i915/selftests/i915_live_selftests.h
index d7dd98a..088f45b 100644
--- a/drivers/gpu/drm/i915/selftests/i915_live_selftests.h
+++ b/drivers/gpu/drm/i915/selftests/i915_live_selftests.h
@@ -20,3 +20,4 @@ selftest(evict, i915_gem_evict_live_selftests)
 selftest(hugepages, i915_gem_huge_page_live_selftests)
 selftest(contexts, i915_gem_context_live_selftests)
 selftest(hangcheck, intel_hangcheck_live_selftests)
+selftest(guc, intel_guc_live_selftest)
diff --git a/drivers/gpu/drm/i915/selftests/i915_syncmap.c b/drivers/gpu/drm/i915/selftests/i915_syncmap.c
index bcab3d0..47f4ae1 100644
--- a/drivers/gpu/drm/i915/selftests/i915_syncmap.c
+++ b/drivers/gpu/drm/i915/selftests/i915_syncmap.c
@@ -333,7 +333,7 @@ static int igt_syncmap_join_below(void *arg)
 {
 	struct i915_syncmap *sync;
 	unsigned int step, order, idx;
-	int err;
+	int err = -ENODEV;
 
 	i915_syncmap_init(&sync);
 
@@ -402,7 +402,7 @@ static int igt_syncmap_neighbours(void *arg)
 	I915_RND_STATE(prng);
 	IGT_TIMEOUT(end_time);
 	struct i915_syncmap *sync;
-	int err;
+	int err = -ENODEV;
 
 	/*
 	 * Each leaf holds KSYNCMAP seqno. Check that when we create KSYNCMAP
@@ -447,7 +447,7 @@ static int igt_syncmap_compact(void *arg)
 {
 	struct i915_syncmap *sync;
 	unsigned int idx, order;
-	int err;
+	int err = -ENODEV;
 
 	i915_syncmap_init(&sync);
 
diff --git a/drivers/gpu/drm/i915/selftests/i915_vma.c b/drivers/gpu/drm/i915/selftests/i915_vma.c
index 2e86ec1..eb89e30 100644
--- a/drivers/gpu/drm/i915/selftests/i915_vma.c
+++ b/drivers/gpu/drm/i915/selftests/i915_vma.c
@@ -150,7 +150,7 @@ static int igt_vma_create(void *arg)
 	IGT_TIMEOUT(end_time);
 	LIST_HEAD(contexts);
 	LIST_HEAD(objects);
-	int err;
+	int err = -ENOMEM;
 
 	/* Exercise creating many vma amonst many objections, checking the
 	 * vma creation and lookup routines.
diff --git a/drivers/gpu/drm/i915/selftests/intel_guc.c b/drivers/gpu/drm/i915/selftests/intel_guc.c
new file mode 100644
index 0000000..f10029e
--- /dev/null
+++ b/drivers/gpu/drm/i915/selftests/intel_guc.c
@@ -0,0 +1,367 @@
+/*
+ * Copyright © 2017 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+
+#include "../i915_selftest.h"
+
+/* max doorbell number + negative test for each client type */
+#define ATTEMPTS (GUC_NUM_DOORBELLS + GUC_CLIENT_PRIORITY_NUM)
+
+struct intel_guc_client *clients[ATTEMPTS];
+
+static bool available_dbs(struct intel_guc *guc, u32 priority)
+{
+	unsigned long offset;
+	unsigned long end;
+	u16 id;
+
+	/* first half is used for normal priority, second half for high */
+	offset = 0;
+	end = GUC_NUM_DOORBELLS / 2;
+	if (priority <= GUC_CLIENT_PRIORITY_HIGH) {
+		offset = end;
+		end += offset;
+	}
+
+	id = find_next_zero_bit(guc->doorbell_bitmap, end, offset);
+	if (id < end)
+		return true;
+
+	return false;
+}
+
+static int check_all_doorbells(struct intel_guc *guc)
+{
+	u16 db_id;
+
+	pr_info_once("Max number of doorbells: %d", GUC_NUM_DOORBELLS);
+	for (db_id = 0; db_id < GUC_NUM_DOORBELLS; ++db_id) {
+		if (!doorbell_ok(guc, db_id)) {
+			pr_err("doorbell %d, not ok\n", db_id);
+			return -EIO;
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * Basic client sanity check, handy to validate create_clients.
+ */
+static int validate_client(struct intel_guc_client *client,
+			   int client_priority,
+			   bool is_preempt_client)
+{
+	struct drm_i915_private *dev_priv = guc_to_i915(client->guc);
+	struct i915_gem_context *ctx_owner = is_preempt_client ?
+			dev_priv->preempt_context : dev_priv->kernel_context;
+
+	if (client->owner != ctx_owner ||
+	    client->engines != INTEL_INFO(dev_priv)->ring_mask ||
+	    client->priority != client_priority ||
+	    client->doorbell_id == GUC_DOORBELL_INVALID)
+		return -EINVAL;
+	else
+		return 0;
+}
+
+/*
+ * Check that guc_init_doorbell_hw is doing what it should.
+ *
+ * During GuC submission enable, we create GuC clients and their doorbells,
+ * but after resetting the microcontroller (resume & gpu reset), these
+ * GuC clients are still around, but the status of their doorbells may be
+ * incorrect. This is the reason behind validating that the doorbells status
+ * expected by the driver matches what the GuC/HW have.
+ */
+static int igt_guc_init_doorbell_hw(void *args)
+{
+	struct drm_i915_private *dev_priv = args;
+	struct intel_guc *guc;
+	DECLARE_BITMAP(db_bitmap_bk, GUC_NUM_DOORBELLS);
+	int i, err = 0;
+
+	GEM_BUG_ON(!HAS_GUC(dev_priv));
+	mutex_lock(&dev_priv->drm.struct_mutex);
+
+	guc = &dev_priv->guc;
+	if (!guc) {
+		pr_err("No guc object!\n");
+		err = -EINVAL;
+		goto unlock;
+	}
+
+	err = check_all_doorbells(guc);
+	if (err)
+		goto unlock;
+
+	/*
+	 * Get rid of clients created during driver load because the test will
+	 * recreate them.
+	 */
+	guc_clients_destroy(guc);
+	if (guc->execbuf_client || guc->preempt_client) {
+		pr_err("guc_clients_destroy lied!\n");
+		err = -EINVAL;
+		goto unlock;
+	}
+
+	err = guc_clients_create(guc);
+	if (err) {
+		pr_err("Failed to create clients\n");
+		goto unlock;
+	}
+
+	err = validate_client(guc->execbuf_client,
+			      GUC_CLIENT_PRIORITY_KMD_NORMAL, false);
+	if (err) {
+		pr_err("execbug client validation failed\n");
+		goto out;
+	}
+
+	err = validate_client(guc->preempt_client,
+			      GUC_CLIENT_PRIORITY_KMD_HIGH, true);
+	if (err) {
+		pr_err("preempt client validation failed\n");
+		goto out;
+	}
+
+	/* each client should have received a doorbell during alloc */
+	if (!has_doorbell(guc->execbuf_client) ||
+	    !has_doorbell(guc->preempt_client)) {
+		pr_err("guc_clients_create didn't create doorbells\n");
+		err = -EINVAL;
+		goto out;
+	}
+
+	/*
+	 * Basic test - an attempt to reallocate a valid doorbell to the
+	 * client it is currently assigned should not cause a failure.
+	 */
+	err = guc_init_doorbell_hw(guc);
+	if (err)
+		goto out;
+
+	/*
+	 * Negative test - a client with no doorbell (invalid db id).
+	 * Each client gets a doorbell when it is created, after destroying
+	 * the doorbell, the db id is changed to GUC_DOORBELL_INVALID and the
+	 * firmware will reject any attempt to allocate a doorbell with an
+	 * invalid id (db has to be reserved before allocation).
+	 */
+	destroy_doorbell(guc->execbuf_client);
+	if (has_doorbell(guc->execbuf_client)) {
+		pr_err("destroy db did not work\n");
+		err = -EINVAL;
+		goto out;
+	}
+
+	err = guc_init_doorbell_hw(guc);
+	if (err != -EIO) {
+		pr_err("unexpected (err = %d)", err);
+		goto out;
+	}
+
+	if (!available_dbs(guc, guc->execbuf_client->priority)) {
+		pr_err("doorbell not available when it should\n");
+		err = -EIO;
+		goto out;
+	}
+
+	/* clean after test */
+	err = create_doorbell(guc->execbuf_client);
+	if (err) {
+		pr_err("recreate doorbell failed\n");
+		goto out;
+	}
+
+	/*
+	 * Negative test - doorbell_bitmap out of sync, will trigger a few of
+	 * WARN_ON(!doorbell_ok(guc, db_id)) but that's ok as long as the
+	 * doorbells from our clients don't fail.
+	 */
+	bitmap_copy(db_bitmap_bk, guc->doorbell_bitmap, GUC_NUM_DOORBELLS);
+	for (i = 0; i < GUC_NUM_DOORBELLS; i++)
+		if (i % 2)
+			test_and_change_bit(i, guc->doorbell_bitmap);
+
+	err = guc_init_doorbell_hw(guc);
+	if (err) {
+		pr_err("out of sync doorbell caused an error\n");
+		goto out;
+	}
+
+	/* restore 'correct' db bitmap */
+	bitmap_copy(guc->doorbell_bitmap, db_bitmap_bk, GUC_NUM_DOORBELLS);
+	err = guc_init_doorbell_hw(guc);
+	if (err) {
+		pr_err("restored doorbell caused an error\n");
+		goto out;
+	}
+
+out:
+	/*
+	 * Leave clean state for other test, plus the driver always destroy the
+	 * clients during unload.
+	 */
+	guc_clients_destroy(guc);
+	guc_clients_create(guc);
+unlock:
+	mutex_unlock(&dev_priv->drm.struct_mutex);
+	return err;
+}
+
+/*
+ * Create as many clients as number of doorbells. Note that there's already
+ * client(s)/doorbell(s) created during driver load, but this test creates
+ * its own and do not interact with the existing ones.
+ */
+static int igt_guc_doorbells(void *arg)
+{
+	struct drm_i915_private *dev_priv = arg;
+	struct intel_guc *guc;
+	int i, err = 0;
+	u16 db_id;
+
+	GEM_BUG_ON(!HAS_GUC(dev_priv));
+	mutex_lock(&dev_priv->drm.struct_mutex);
+
+	guc = &dev_priv->guc;
+	if (!guc) {
+		pr_err("No guc object!\n");
+		err = -EINVAL;
+		goto unlock;
+	}
+
+	err = check_all_doorbells(guc);
+	if (err)
+		goto unlock;
+
+	for (i = 0; i < ATTEMPTS; i++) {
+		clients[i] = guc_client_alloc(dev_priv,
+					      INTEL_INFO(dev_priv)->ring_mask,
+					      i % GUC_CLIENT_PRIORITY_NUM,
+					      dev_priv->kernel_context);
+
+		if (!clients[i]) {
+			pr_err("[%d] No guc client\n", i);
+			err = -EINVAL;
+			goto out;
+		}
+
+		if (IS_ERR(clients[i])) {
+			if (PTR_ERR(clients[i]) != -ENOSPC) {
+				pr_err("[%d] unexpected error\n", i);
+				err = PTR_ERR(clients[i]);
+				goto out;
+			}
+
+			if (available_dbs(guc, i % GUC_CLIENT_PRIORITY_NUM)) {
+				pr_err("[%d] non-db related alloc fail\n", i);
+				err = -EINVAL;
+				goto out;
+			}
+
+			/* expected, ran out of dbs for this client type */
+			continue;
+		}
+
+		/*
+		 * The check below is only valid because we keep a doorbell
+		 * assigned during the whole life of the client.
+		 */
+		if (clients[i]->stage_id >= GUC_NUM_DOORBELLS) {
+			pr_err("[%d] more clients than doorbells (%d >= %d)\n",
+			       i, clients[i]->stage_id, GUC_NUM_DOORBELLS);
+			err = -EINVAL;
+			goto out;
+		}
+
+		err = validate_client(clients[i],
+				      i % GUC_CLIENT_PRIORITY_NUM, false);
+		if (err) {
+			pr_err("[%d] client_alloc sanity check failed!\n", i);
+			err = -EINVAL;
+			goto out;
+		}
+
+		db_id = clients[i]->doorbell_id;
+
+		/*
+		 * Client alloc gives us a doorbell, but we want to exercise
+		 * this ourselves (this resembles guc_init_doorbell_hw)
+		 */
+		destroy_doorbell(clients[i]);
+		if (clients[i]->doorbell_id != GUC_DOORBELL_INVALID) {
+			pr_err("[%d] destroy db did not work!\n", i);
+			err = -EINVAL;
+			goto out;
+		}
+
+		err = __reserve_doorbell(clients[i]);
+		if (err) {
+			pr_err("[%d] Failed to reserve a doorbell\n", i);
+			goto out;
+		}
+
+		__update_doorbell_desc(clients[i], clients[i]->doorbell_id);
+		err = __create_doorbell(clients[i]);
+		if (err) {
+			pr_err("[%d] Failed to create a doorbell\n", i);
+			goto out;
+		}
+
+		/* doorbell id shouldn't change, we are holding the mutex */
+		if (db_id != clients[i]->doorbell_id) {
+			pr_err("[%d] doorbell id changed (%d != %d)\n",
+			       i, db_id, clients[i]->doorbell_id);
+			err = -EINVAL;
+			goto out;
+		}
+
+		err = check_all_doorbells(guc);
+		if (err)
+			goto out;
+	}
+
+out:
+	for (i = 0; i < ATTEMPTS; i++)
+		if (!IS_ERR_OR_NULL(clients[i]))
+			guc_client_free(clients[i]);
+unlock:
+	mutex_unlock(&dev_priv->drm.struct_mutex);
+	return err;
+}
+
+int intel_guc_live_selftest(struct drm_i915_private *dev_priv)
+{
+	static const struct i915_subtest tests[] = {
+		SUBTEST(igt_guc_init_doorbell_hw),
+		SUBTEST(igt_guc_doorbells),
+	};
+
+	if (!i915_modparams.enable_guc_submission)
+		return 0;
+
+	return i915_subtests(tests, dev_priv);
+}
diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c
index 3cac22e..2f63676 100644
--- a/drivers/gpu/drm/i915/selftests/intel_uncore.c
+++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c
@@ -120,10 +120,10 @@ static int intel_uncore_check_forcewake_domains(struct drm_i915_private *dev_pri
 	    !IS_CHERRYVIEW(dev_priv))
 		return 0;
 
-	if (IS_VALLEYVIEW(dev_priv)) /* XXX system lockup! */
-		return 0;
-
-	if (IS_BROADWELL(dev_priv)) /* XXX random GPU hang afterwards! */
+	/*
+	 * This test may lockup the machine or cause GPU hangs afterwards.
+	 */
+	if (!IS_ENABLED(CONFIG_DRM_I915_SELFTEST_BROKEN))
 		return 0;
 
 	valid = kzalloc(BITS_TO_LONGS(FW_RANGE) * sizeof(*valid),
@@ -148,7 +148,10 @@ static int intel_uncore_check_forcewake_domains(struct drm_i915_private *dev_pri
 	for_each_set_bit(offset, valid, FW_RANGE) {
 		i915_reg_t reg = { offset };
 
+		iosf_mbi_punit_acquire();
 		intel_uncore_forcewake_reset(dev_priv, false);
+		iosf_mbi_punit_release();
+
 		check_for_unclaimed_mmio(dev_priv);
 
 		(void)I915_READ(reg);
diff --git a/drivers/gpu/drm/i915/selftests/mock_engine.c b/drivers/gpu/drm/i915/selftests/mock_engine.c
index 331c2b0..55c0e2c 100644
--- a/drivers/gpu/drm/i915/selftests/mock_engine.c
+++ b/drivers/gpu/drm/i915/selftests/mock_engine.c
@@ -32,6 +32,13 @@ static struct mock_request *first_request(struct mock_engine *engine)
 					link);
 }
 
+static void advance(struct mock_engine *engine,
+		    struct mock_request *request)
+{
+	list_del_init(&request->link);
+	mock_seqno_advance(&engine->base, request->base.global_seqno);
+}
+
 static void hw_delay_complete(struct timer_list *t)
 {
 	struct mock_engine *engine = from_timer(engine, t, hw_delay);
@@ -39,15 +46,23 @@ static void hw_delay_complete(struct timer_list *t)
 
 	spin_lock(&engine->hw_lock);
 
-	request = first_request(engine);
-	if (request) {
-		list_del_init(&request->link);
-		mock_seqno_advance(&engine->base, request->base.global_seqno);
-	}
-
+	/* Timer fired, first request is complete */
 	request = first_request(engine);
 	if (request)
-		mod_timer(&engine->hw_delay, jiffies + request->delay);
+		advance(engine, request);
+
+	/*
+	 * Also immediately signal any subsequent 0-delay requests, but
+	 * requeue the timer for the next delayed request.
+	 */
+	while ((request = first_request(engine))) {
+		if (request->delay) {
+			mod_timer(&engine->hw_delay, jiffies + request->delay);
+			break;
+		}
+
+		advance(engine, request);
+	}
 
 	spin_unlock(&engine->hw_lock);
 }
@@ -98,16 +113,22 @@ static void mock_submit_request(struct drm_i915_gem_request *request)
 
 	spin_lock_irq(&engine->hw_lock);
 	list_add_tail(&mock->link, &engine->hw_queue);
-	if (mock->link.prev == &engine->hw_queue)
-		mod_timer(&engine->hw_delay, jiffies + mock->delay);
+	if (mock->link.prev == &engine->hw_queue) {
+		if (mock->delay)
+			mod_timer(&engine->hw_delay, jiffies + mock->delay);
+		else
+			advance(engine, mock);
+	}
 	spin_unlock_irq(&engine->hw_lock);
 }
 
 static struct intel_ring *mock_ring(struct intel_engine_cs *engine)
 {
-	const unsigned long sz = roundup_pow_of_two(sizeof(struct intel_ring));
+	const unsigned long sz = PAGE_SIZE / 2;
 	struct intel_ring *ring;
 
+	BUILD_BUG_ON(MIN_SPACE_FOR_ADD_REQUEST > sz);
+
 	ring = kzalloc(sizeof(*ring) + sz, GFP_KERNEL);
 	if (!ring)
 		return NULL;
diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
index 04eb936..80f152a 100644
--- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c
+++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
@@ -179,8 +179,8 @@ struct drm_i915_private *mock_gem_device(void)
 		I915_GTT_PAGE_SIZE_64K |
 		I915_GTT_PAGE_SIZE_2M;
 
-	spin_lock_init(&i915->mm.object_stat_lock);
 	mock_uncore_init(i915);
+	i915_gem_init__mm(i915);
 
 	init_waitqueue_head(&i915->gpu_error.wait_queue);
 	init_waitqueue_head(&i915->gpu_error.reset_queue);
@@ -189,11 +189,6 @@ struct drm_i915_private *mock_gem_device(void)
 	if (!i915->wq)
 		goto put_device;
 
-	INIT_WORK(&i915->mm.free_work, __i915_gem_free_work);
-	init_llist_head(&i915->mm.free_list);
-	INIT_LIST_HEAD(&i915->mm.unbound_list);
-	INIT_LIST_HEAD(&i915->mm.bound_list);
-
 	mock_init_contexts(i915);
 
 	INIT_DELAYED_WORK(&i915->gt.retire_work, mock_retire_work_handler);