From 501be6c1c72417eab05e7413671a38ea991a8ebc Mon Sep 17 00:00:00 2001
From: Thierry Reding <treding@nvidia.com>
Date: Wed, 25 Mar 2020 21:16:03 +0100
Subject: drm/tegra: Fix SMMU support on Tegra124 and Tegra210

When testing whether or not to enable the use of the SMMU, consult the
supported DMA mask rather than the actually configured DMA mask, since
the latter might already have been restricted.

Fixes: 2d9384ff9177 ("drm/tegra: Relax IOMMU usage criteria on old Tegra")
Tested-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
---
 drivers/gpu/drm/tegra/drm.c |  3 ++-
 drivers/gpu/host1x/dev.c    | 13 +++++++++++++
 include/linux/host1x.h      |  3 +++
 3 files changed, 18 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index bd268028fb3d..583cd6e0ae27 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -1039,6 +1039,7 @@ void tegra_drm_free(struct tegra_drm *tegra, size_t size, void *virt,
 
 static bool host1x_drm_wants_iommu(struct host1x_device *dev)
 {
+	struct host1x *host1x = dev_get_drvdata(dev->dev.parent);
 	struct iommu_domain *domain;
 
 	/*
@@ -1076,7 +1077,7 @@ static bool host1x_drm_wants_iommu(struct host1x_device *dev)
 	 * sufficient and whether or not the host1x is attached to an IOMMU
 	 * doesn't matter.
 	 */
-	if (!domain && dma_get_mask(dev->dev.parent) <= DMA_BIT_MASK(32))
+	if (!domain && host1x_get_dma_mask(host1x) <= DMA_BIT_MASK(32))
 		return true;
 
 	return domain != NULL;
diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c
index 388bcc2889aa..40a4b9f8b861 100644
--- a/drivers/gpu/host1x/dev.c
+++ b/drivers/gpu/host1x/dev.c
@@ -502,6 +502,19 @@ static void __exit tegra_host1x_exit(void)
 }
 module_exit(tegra_host1x_exit);
 
+/**
+ * host1x_get_dma_mask() - query the supported DMA mask for host1x
+ * @host1x: host1x instance
+ *
+ * Note that this returns the supported DMA mask for host1x, which can be
+ * different from the applicable DMA mask under certain circumstances.
+ */
+u64 host1x_get_dma_mask(struct host1x *host1x)
+{
+	return host1x->info->dma_mask;
+}
+EXPORT_SYMBOL(host1x_get_dma_mask);
+
 MODULE_AUTHOR("Thierry Reding <thierry.reding@avionic-design.de>");
 MODULE_AUTHOR("Terje Bergstrom <tbergstrom@nvidia.com>");
 MODULE_DESCRIPTION("Host1x driver for Tegra products");
diff --git a/include/linux/host1x.h b/include/linux/host1x.h
index 62d216ff1097..c230b4e70d75 100644
--- a/include/linux/host1x.h
+++ b/include/linux/host1x.h
@@ -17,9 +17,12 @@ enum host1x_class {
 	HOST1X_CLASS_GR3D = 0x60,
 };
 
+struct host1x;
 struct host1x_client;
 struct iommu_group;
 
+u64 host1x_get_dma_mask(struct host1x *host1x);
+
 /**
  * struct host1x_client_ops - host1x client operations
  * @init: host1x client initialization code
-- 
cgit v1.2.3-59-g8ed1b


From 4010e729349fcab69183f338fe3743df17a473a0 Mon Sep 17 00:00:00 2001
From: Thierry Reding <treding@nvidia.com>
Date: Wed, 25 Mar 2020 21:16:04 +0100
Subject: gpu: host1x: Use SMMU on Tegra124 and Tegra210

Tegra124 and Tegra210 support addressing more than 32 bits of physical
memory. However, since their host1x does not support the wide GATHER
opcode, they should use the SMMU if at all possible to ensure that all
the system memory can be used for command buffers, irrespective of
whether or not the host1x firewall is enabled.

Tested-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
---
 drivers/gpu/host1x/dev.c | 46 ++++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 42 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c
index 40a4b9f8b861..d24344e91922 100644
--- a/drivers/gpu/host1x/dev.c
+++ b/drivers/gpu/host1x/dev.c
@@ -192,17 +192,55 @@ static void host1x_setup_sid_table(struct host1x *host)
 	}
 }
 
+static bool host1x_wants_iommu(struct host1x *host1x)
+{
+	/*
+	 * If we support addressing a maximum of 32 bits of physical memory
+	 * and if the host1x firewall is enabled, there's no need to enable
+	 * IOMMU support. This can happen for example on Tegra20, Tegra30
+	 * and Tegra114.
+	 *
+	 * Tegra124 and later can address up to 34 bits of physical memory and
+	 * many platforms come equipped with more than 2 GiB of system memory,
+	 * which requires crossing the 4 GiB boundary. But there's a catch: on
+	 * SoCs before Tegra186 (i.e. Tegra124 and Tegra210), the host1x can
+	 * only address up to 32 bits of memory in GATHER opcodes, which means
+	 * that command buffers need to either be in the first 2 GiB of system
+	 * memory (which could quickly lead to memory exhaustion), or command
+	 * buffers need to be treated differently from other buffers (which is
+	 * not possible with the current ABI).
+	 *
+	 * A third option is to use the IOMMU in these cases to make sure all
+	 * buffers will be mapped into a 32-bit IOVA space that host1x can
+	 * address. This allows all of the system memory to be used and works
+	 * within the limitations of the host1x on these SoCs.
+	 *
+	 * In summary, default to enable IOMMU on Tegra124 and later. For any
+	 * of the earlier SoCs, only use the IOMMU for additional safety when
+	 * the host1x firewall is disabled.
+	 */
+	if (host1x->info->dma_mask <= DMA_BIT_MASK(32)) {
+		if (IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL))
+			return false;
+	}
+
+	return true;
+}
+
 static struct iommu_domain *host1x_iommu_attach(struct host1x *host)
 {
 	struct iommu_domain *domain = iommu_get_domain_for_dev(host->dev);
 	int err;
 
 	/*
-	 * If the host1x firewall is enabled, there's no need to enable IOMMU
-	 * support. Similarly, if host1x is already attached to an IOMMU (via
-	 * the DMA API), don't try to attach again.
+	 * We may not always want to enable IOMMU support (for example if the
+	 * host1x firewall is already enabled and we don't support addressing
+	 * more than 32 bits of physical memory), so check for that first.
+	 *
+	 * Similarly, if host1x is already attached to an IOMMU (via the DMA
+	 * API), don't try to attach again.
 	 */
-	if (IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL) || domain)
+	if (!host1x_wants_iommu(host) || domain)
 		return domain;
 
 	host->group = iommu_group_get(host->dev);
-- 
cgit v1.2.3-59-g8ed1b


From 30523408c0232d4f33ca0719004e5bffaf04220c Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Tue, 28 Apr 2020 10:02:55 +0100
Subject: drm/i915: Avoid dereferencing a dead context

Once the intel_context is closed, the GEM context may be freed and so
the link from intel_context.gem_context is invalid.

<3>[  219.782944] BUG: KASAN: use-after-free in intel_engine_coredump_alloc+0x1bc3/0x2250 [i915]
<3>[  219.782996] Read of size 8 at addr ffff8881d7dff0b8 by task kworker/0:1/12

<4>[  219.783052] CPU: 0 PID: 12 Comm: kworker/0:1 Tainted: G     U            5.7.0-rc2-g1f3ffd7683d54-kasan_118+ #1
<4>[  219.783055] Hardware name: System manufacturer System Product Name/Z170 PRO GAMING, BIOS 3402 04/26/2017
<4>[  219.783105] Workqueue: events heartbeat [i915]
<4>[  219.783109] Call Trace:
<4>[  219.783113]  <IRQ>
<4>[  219.783119]  dump_stack+0x96/0xdb
<4>[  219.783177]  ? intel_engine_coredump_alloc+0x1bc3/0x2250 [i915]
<4>[  219.783182]  print_address_description.constprop.6+0x16/0x310
<4>[  219.783239]  ? intel_engine_coredump_alloc+0x1bc3/0x2250 [i915]
<4>[  219.783295]  ? intel_engine_coredump_alloc+0x1bc3/0x2250 [i915]
<4>[  219.783300]  __kasan_report+0x137/0x190
<4>[  219.783359]  ? intel_engine_coredump_alloc+0x1bc3/0x2250 [i915]
<4>[  219.783366]  kasan_report+0x32/0x50
<4>[  219.783426]  intel_engine_coredump_alloc+0x1bc3/0x2250 [i915]
<4>[  219.783481]  execlists_reset+0x39c/0x13d0 [i915]
<4>[  219.783494]  ? mark_held_locks+0x9e/0xe0
<4>[  219.783546]  ? execlists_hold+0xfc0/0xfc0 [i915]
<4>[  219.783551]  ? lockdep_hardirqs_on+0x348/0x5f0
<4>[  219.783557]  ? _raw_spin_unlock_irqrestore+0x34/0x60
<4>[  219.783606]  ? execlists_submission_tasklet+0x118/0x3a0 [i915]
<4>[  219.783615]  tasklet_action_common.isra.14+0x13b/0x410
<4>[  219.783623]  ? __do_softirq+0x1e4/0x9a7
<4>[  219.783630]  __do_softirq+0x226/0x9a7
<4>[  219.783643]  do_softirq_own_stack+0x2a/0x40
<4>[  219.783647]  </IRQ>
<4>[  219.783692]  ? heartbeat+0x3e2/0x10f0 [i915]
<4>[  219.783696]  do_softirq.part.13+0x49/0x50
<4>[  219.783700]  __local_bh_enable_ip+0x1a2/0x1e0
<4>[  219.783748]  heartbeat+0x409/0x10f0 [i915]
<4>[  219.783801]  ? __live_idle_pulse+0x9f0/0x9f0 [i915]
<4>[  219.783806]  ? lock_acquire+0x1ac/0x8a0
<4>[  219.783811]  ? process_one_work+0x811/0x1870
<4>[  219.783827]  ? rcu_read_lock_sched_held+0x9c/0xd0
<4>[  219.783832]  ? rcu_read_lock_bh_held+0xb0/0xb0
<4>[  219.783836]  ? _raw_spin_unlock_irq+0x1f/0x40
<4>[  219.783845]  process_one_work+0x8ca/0x1870
<4>[  219.783848]  ? lock_acquire+0x1ac/0x8a0
<4>[  219.783852]  ? worker_thread+0x1d0/0xb80
<4>[  219.783864]  ? pwq_dec_nr_in_flight+0x2c0/0x2c0
<4>[  219.783870]  ? do_raw_spin_lock+0x129/0x290
<4>[  219.783886]  worker_thread+0x82/0xb80
<4>[  219.783895]  ? __kthread_parkme+0xaf/0x1b0
<4>[  219.783902]  ? process_one_work+0x1870/0x1870
<4>[  219.783906]  kthread+0x34e/0x420
<4>[  219.783911]  ? kthread_create_on_node+0xc0/0xc0
<4>[  219.783918]  ret_from_fork+0x3a/0x50

<3>[  219.783950] Allocated by task 1264:
<4>[  219.783975]  save_stack+0x19/0x40
<4>[  219.783978]  __kasan_kmalloc.constprop.3+0xa0/0xd0
<4>[  219.784029]  i915_gem_create_context+0xa2/0xab8 [i915]
<4>[  219.784081]  i915_gem_context_create_ioctl+0x1fa/0x450 [i915]
<4>[  219.784085]  drm_ioctl_kernel+0x1d8/0x270
<4>[  219.784088]  drm_ioctl+0x676/0x930
<4>[  219.784092]  ksys_ioctl+0xb7/0xe0
<4>[  219.784096]  __x64_sys_ioctl+0x6a/0xb0
<4>[  219.784100]  do_syscall_64+0x94/0x530
<4>[  219.784103]  entry_SYSCALL_64_after_hwframe+0x49/0xb3

<3>[  219.784120] Freed by task 12:
<4>[  219.784141]  save_stack+0x19/0x40
<4>[  219.784145]  __kasan_slab_free+0x130/0x180
<4>[  219.784148]  kmem_cache_free_bulk+0x1bd/0x500
<4>[  219.784152]  kfree_rcu_work+0x1d8/0x890
<4>[  219.784155]  process_one_work+0x8ca/0x1870
<4>[  219.784158]  worker_thread+0x82/0xb80
<4>[  219.784162]  kthread+0x34e/0x420
<4>[  219.784165]  ret_from_fork+0x3a/0x50

Fixes: 2e46a2a0b014 ("drm/i915: Use explicit flag to mark unreachable intel_context")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200428090255.10035-1-chris@chris-wilson.co.uk
(cherry picked from commit 24aac336ff78eb55aedda043ed982c61dc3ac49a)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/i915_gpu_error.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 2a4cd0ba5464..5c8e51d2ba5b 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -1207,8 +1207,6 @@ static void engine_record_registers(struct intel_engine_coredump *ee)
 static void record_request(const struct i915_request *request,
 			   struct i915_request_coredump *erq)
 {
-	const struct i915_gem_context *ctx;
-
 	erq->flags = request->fence.flags;
 	erq->context = request->fence.context;
 	erq->seqno = request->fence.seqno;
@@ -1219,9 +1217,13 @@ static void record_request(const struct i915_request *request,
 
 	erq->pid = 0;
 	rcu_read_lock();
-	ctx = rcu_dereference(request->context->gem_context);
-	if (ctx)
-		erq->pid = pid_nr(ctx->pid);
+	if (!intel_context_is_closed(request->context)) {
+		const struct i915_gem_context *ctx;
+
+		ctx = rcu_dereference(request->context->gem_context);
+		if (ctx)
+			erq->pid = pid_nr(ctx->pid);
+	}
 	rcu_read_unlock();
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From fe5a708267911d55cce42910d93e303924b088fd Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Fri, 1 May 2020 13:22:49 +0100
Subject: drm/i915/gt: Make timeslicing an explicit engine property

In order to allow userspace to rely on timeslicing to reorder their
batches, we must support preemption of those user batches. Declare
timeslicing as an explicit property that is a combination of having the
kernel support and HW support.

Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Fixes: 8ee36e048c98 ("drm/i915/execlists: Minimalistic timeslicing")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200501122249.12417-1-chris@chris-wilson.co.uk
(cherry picked from commit a211da9c771bf97395a3ced83a3aa383372b13a7)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_engine.h       |  9 ---------
 drivers/gpu/drm/i915/gt/intel_engine_types.h | 18 ++++++++++++++----
 drivers/gpu/drm/i915/gt/intel_lrc.c          |  5 ++++-
 3 files changed, 18 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_engine.h b/drivers/gpu/drm/i915/gt/intel_engine.h
index b469de0dd9b6..a1aa0d3e8be1 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine.h
@@ -333,13 +333,4 @@ intel_engine_has_preempt_reset(const struct intel_engine_cs *engine)
 	return intel_engine_has_preemption(engine);
 }
 
-static inline bool
-intel_engine_has_timeslices(const struct intel_engine_cs *engine)
-{
-	if (!IS_ACTIVE(CONFIG_DRM_I915_TIMESLICE_DURATION))
-		return false;
-
-	return intel_engine_has_semaphores(engine);
-}
-
 #endif /* _INTEL_RINGBUFFER_H_ */
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h
index 80cdde712842..bf4d13e08887 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h
@@ -483,10 +483,11 @@ struct intel_engine_cs {
 #define I915_ENGINE_SUPPORTS_STATS   BIT(1)
 #define I915_ENGINE_HAS_PREEMPTION   BIT(2)
 #define I915_ENGINE_HAS_SEMAPHORES   BIT(3)
-#define I915_ENGINE_NEEDS_BREADCRUMB_TASKLET BIT(4)
-#define I915_ENGINE_IS_VIRTUAL       BIT(5)
-#define I915_ENGINE_HAS_RELATIVE_MMIO BIT(6)
-#define I915_ENGINE_REQUIRES_CMD_PARSER BIT(7)
+#define I915_ENGINE_HAS_TIMESLICES   BIT(4)
+#define I915_ENGINE_NEEDS_BREADCRUMB_TASKLET BIT(5)
+#define I915_ENGINE_IS_VIRTUAL       BIT(6)
+#define I915_ENGINE_HAS_RELATIVE_MMIO BIT(7)
+#define I915_ENGINE_REQUIRES_CMD_PARSER BIT(8)
 	unsigned int flags;
 
 	/*
@@ -584,6 +585,15 @@ intel_engine_has_semaphores(const struct intel_engine_cs *engine)
 	return engine->flags & I915_ENGINE_HAS_SEMAPHORES;
 }
 
+static inline bool
+intel_engine_has_timeslices(const struct intel_engine_cs *engine)
+{
+	if (!IS_ACTIVE(CONFIG_DRM_I915_TIMESLICE_DURATION))
+		return false;
+
+	return engine->flags & I915_ENGINE_HAS_TIMESLICES;
+}
+
 static inline bool
 intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine)
 {
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index 683014e7bc51..017fd5bf9c69 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -4369,8 +4369,11 @@ void intel_execlists_set_default_submission(struct intel_engine_cs *engine)
 	engine->flags |= I915_ENGINE_SUPPORTS_STATS;
 	if (!intel_vgpu_active(engine->i915)) {
 		engine->flags |= I915_ENGINE_HAS_SEMAPHORES;
-		if (HAS_LOGICAL_RING_PREEMPTION(engine->i915))
+		if (HAS_LOGICAL_RING_PREEMPTION(engine->i915)) {
 			engine->flags |= I915_ENGINE_HAS_PREEMPTION;
+			if (IS_ACTIVE(CONFIG_DRM_I915_TIMESLICE_DURATION))
+				engine->flags |= I915_ENGINE_HAS_TIMESLICES;
+		}
 	}
 
 	if (INTEL_GEN(engine->i915) >= 12)
-- 
cgit v1.2.3-59-g8ed1b


From 421abe200321a2c907ede1a6208c558284ba0b75 Mon Sep 17 00:00:00 2001
From: Sultan Alsawaf <sultan@kerneltoast.com>
Date: Thu, 30 Apr 2020 14:46:54 -0700
Subject: drm/i915: Don't enable WaIncreaseLatencyIPCEnabled when IPC is
 disabled
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In commit 5a7d202b1574, a logical AND was erroneously changed to an OR,
causing WaIncreaseLatencyIPCEnabled to be enabled unconditionally for
kabylake and coffeelake, even when IPC is disabled. Fix the logic so
that WaIncreaseLatencyIPCEnabled is only used when IPC is enabled.

Fixes: 5a7d202b1574 ("drm/i915: Drop WaIncreaseLatencyIPCEnabled/1140 for cnl")
Cc: stable@vger.kernel.org # 5.3.x+
Signed-off-by: Sultan Alsawaf <sultan@kerneltoast.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200430214654.51314-1-sultan@kerneltoast.com
(cherry picked from commit 690d22dafa88b82453516387b475664047a6bd14)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/intel_pm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 8375054ba27d..a52986a9e7a6 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -4992,7 +4992,7 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state,
 	 * WaIncreaseLatencyIPCEnabled: kbl,cfl
 	 * Display WA #1141: kbl,cfl
 	 */
-	if ((IS_KABYLAKE(dev_priv) || IS_COFFEELAKE(dev_priv)) ||
+	if ((IS_KABYLAKE(dev_priv) || IS_COFFEELAKE(dev_priv)) &&
 	    dev_priv->ipc_enabled)
 		latency += 4;
 
-- 
cgit v1.2.3-59-g8ed1b


From 82152d424b6cb6fc1ede7d03d69c04e786688740 Mon Sep 17 00:00:00 2001
From: Peter Jones <pjones@redhat.com>
Date: Fri, 6 Jul 2018 15:04:24 -0400
Subject: Make the "Reducing compressed framebufer size" message be
 DRM_INFO_ONCE()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This was sort of annoying me:

random:~$ dmesg | tail -1
[523884.039227] [drm] Reducing the compressed framebuffer size. This may lead to less power savings than a non-reduced-size. Try to increase stolen memory size if available in BIOS.
random:~$ dmesg | grep -c "Reducing the compressed"
47

This patch makes it DRM_INFO_ONCE() just like the similar message
farther down in that function is pr_info_once().

Cc: stable@vger.kernel.org
Signed-off-by: Peter Jones <pjones@redhat.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1745
Link: https://patchwork.freedesktop.org/patch/msgid/20180706190424.29194-1-pjones@redhat.com
[vsyrjala: Rebase due to per-device logging]
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
(cherry picked from commit 6b7fc6a3e6af4ff5773949d0fed70d8e7f68d5ce)
[Rodrigo: port back to DRM_INFO_ONCE]
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/display/intel_fbc.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index 2e5d835a9eaa..c125ca9ab9b3 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -485,8 +485,7 @@ static int intel_fbc_alloc_cfb(struct drm_i915_private *dev_priv,
 	if (!ret)
 		goto err_llb;
 	else if (ret > 1) {
-		DRM_INFO("Reducing the compressed framebuffer size. This may lead to less power savings than a non-reduced-size. Try to increase stolen memory size if available in BIOS.\n");
-
+		DRM_INFO_ONCE("Reducing the compressed framebuffer size. This may lead to less power savings than a non-reduced-size. Try to increase stolen memory size if available in BIOS.\n");
 	}
 
 	fbc->threshold = ret;
-- 
cgit v1.2.3-59-g8ed1b


From 4457a9db2bdec2360ddb15242341696108167886 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Mon, 4 May 2020 10:58:28 +0300
Subject: drm/i915/tgl+: Fix interrupt handling for DP AUX transactions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Unmask/enable AUX interrupts on all ports on TGL+. So far the interrupts
worked only on port A, which meant each transaction on other ports took
10ms.

Cc: <stable@vger.kernel.org> # v5.4+
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200504075828.20348-1-imre.deak@intel.com
(cherry picked from commit 054318c7e35f1d7d06b216143fff5f32405047ee)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/i915_irq.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index d91557d842dc..8a2b83807ffc 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -3361,7 +3361,7 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	u32 de_pipe_masked = gen8_de_pipe_fault_mask(dev_priv) |
 		GEN8_PIPE_CDCLK_CRC_DONE;
 	u32 de_pipe_enables;
-	u32 de_port_masked = GEN8_AUX_CHANNEL_A;
+	u32 de_port_masked = gen8_de_port_aux_mask(dev_priv);
 	u32 de_port_enables;
 	u32 de_misc_masked = GEN8_DE_EDP_PSR;
 	enum pipe pipe;
@@ -3369,18 +3369,8 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	if (INTEL_GEN(dev_priv) <= 10)
 		de_misc_masked |= GEN8_DE_MISC_GSE;
 
-	if (INTEL_GEN(dev_priv) >= 9) {
-		de_port_masked |= GEN9_AUX_CHANNEL_B | GEN9_AUX_CHANNEL_C |
-				  GEN9_AUX_CHANNEL_D;
-		if (IS_GEN9_LP(dev_priv))
-			de_port_masked |= BXT_DE_PORT_GMBUS;
-	}
-
-	if (INTEL_GEN(dev_priv) >= 11)
-		de_port_masked |= ICL_AUX_CHANNEL_E;
-
-	if (IS_CNL_WITH_PORT_F(dev_priv) || INTEL_GEN(dev_priv) >= 11)
-		de_port_masked |= CNL_AUX_CHANNEL_F;
+	if (IS_GEN9_LP(dev_priv))
+		de_port_masked |= BXT_DE_PORT_GMBUS;
 
 	de_pipe_enables = de_pipe_masked | GEN8_PIPE_VBLANK |
 					   GEN8_PIPE_FIFO_UNDERRUN;
-- 
cgit v1.2.3-59-g8ed1b


From af23facc38c26ac188b6adac2cae2dafaca0c82d Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Fri, 3 Apr 2020 13:01:50 +0100
Subject: drm/i915: Check current i915_vma.pin_count status first on unbind

Do an early rejection of a i915_vma_unbind() attempt if the i915_vma is
currently pinned, without waiting to see if the inflight operations may
unpin it. We see this problem with the shrinker trying to unbind the
active vma from inside its bind worker:

<6> [472.618968] Workqueue: events_unbound fence_work [i915]
<4> [472.618970] Call Trace:
<4> [472.618974]  ? __schedule+0x2e5/0x810
<4> [472.618978]  schedule+0x37/0xe0
<4> [472.618982]  schedule_preempt_disabled+0xf/0x20
<4> [472.618984]  __mutex_lock+0x281/0x9c0
<4> [472.618987]  ? mark_held_locks+0x49/0x70
<4> [472.618989]  ? _raw_spin_unlock_irqrestore+0x47/0x60
<4> [472.619038]  ? i915_vma_unbind+0xae/0x110 [i915]
<4> [472.619084]  ? i915_vma_unbind+0xae/0x110 [i915]
<4> [472.619122]  i915_vma_unbind+0xae/0x110 [i915]
<4> [472.619165]  i915_gem_object_unbind+0x1dc/0x400 [i915]
<4> [472.619208]  i915_gem_shrink+0x328/0x660 [i915]
<4> [472.619250]  ? i915_gem_shrink_all+0x38/0x60 [i915]
<4> [472.619282]  i915_gem_shrink_all+0x38/0x60 [i915]
<4> [472.619325]  vm_alloc_page.constprop.25+0x1aa/0x240 [i915]
<4> [472.619330]  ? rcu_read_lock_sched_held+0x4d/0x80
<4> [472.619363]  ? __alloc_pd+0xb/0x30 [i915]
<4> [472.619366]  ? module_assert_mutex_or_preempt+0xf/0x30
<4> [472.619368]  ? __module_address+0x23/0xe0
<4> [472.619371]  ? is_module_address+0x26/0x40
<4> [472.619374]  ? static_obj+0x34/0x50
<4> [472.619376]  ? lockdep_init_map+0x4d/0x1e0
<4> [472.619407]  setup_page_dma+0xd/0x90 [i915]
<4> [472.619437]  alloc_pd+0x29/0x50 [i915]
<4> [472.619470]  __gen8_ppgtt_alloc+0x443/0x6b0 [i915]
<4> [472.619503]  gen8_ppgtt_alloc+0xd7/0x300 [i915]
<4> [472.619535]  ppgtt_bind_vma+0x2a/0xe0 [i915]
<4> [472.619577]  __vma_bind+0x26/0x40 [i915]
<4> [472.619611]  fence_work+0x1c/0x90 [i915]
<4> [472.619617]  process_one_work+0x26a/0x620

Fixes: 2850748ef876 ("drm/i915: Pull i915_vma_pin under the vm->mutex")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200403120150.17091-1-chris@chris-wilson.co.uk
(cherry picked from commit 614654abe847a42fc75d7eb5096e46f796a438b6)
(cherry picked from commit dd086cf516d9bea3878abb267f62ccc53acd764b)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/i915_vma.c | 25 +++++++++----------------
 1 file changed, 9 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index 82e3bc280622..2cd7a7e87c0a 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -1228,18 +1228,6 @@ int __i915_vma_unbind(struct i915_vma *vma)
 
 	lockdep_assert_held(&vma->vm->mutex);
 
-	/*
-	 * First wait upon any activity as retiring the request may
-	 * have side-effects such as unpinning or even unbinding this vma.
-	 *
-	 * XXX Actually waiting under the vm->mutex is a hinderance and
-	 * should be pipelined wherever possible. In cases where that is
-	 * unavoidable, we should lift the wait to before the mutex.
-	 */
-	ret = i915_vma_sync(vma);
-	if (ret)
-		return ret;
-
 	if (i915_vma_is_pinned(vma)) {
 		vma_print_allocator(vma, "is pinned");
 		return -EAGAIN;
@@ -1313,15 +1301,20 @@ int i915_vma_unbind(struct i915_vma *vma)
 	if (!drm_mm_node_allocated(&vma->node))
 		return 0;
 
-	if (i915_vma_is_bound(vma, I915_VMA_GLOBAL_BIND))
-		/* XXX not always required: nop_clear_range */
-		wakeref = intel_runtime_pm_get(&vm->i915->runtime_pm);
-
 	/* Optimistic wait before taking the mutex */
 	err = i915_vma_sync(vma);
 	if (err)
 		goto out_rpm;
 
+	if (i915_vma_is_pinned(vma)) {
+		vma_print_allocator(vma, "is pinned");
+		return -EAGAIN;
+	}
+
+	if (i915_vma_is_bound(vma, I915_VMA_GLOBAL_BIND))
+		/* XXX not always required: nop_clear_range */
+		wakeref = intel_runtime_pm_get(&vm->i915->runtime_pm);
+
 	err = mutex_lock_interruptible(&vm->mutex);
 	if (err)
 		goto out_rpm;
-- 
cgit v1.2.3-59-g8ed1b


From 220dcfc1485bb79e300f08a28b475869feccbb53 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Tue, 7 Apr 2020 14:08:11 +0100
Subject: drm/i915/gt: Yield the timeslice if caught waiting on a user
 semaphore

If we find ourselves waiting on a MI_SEMAPHORE_WAIT, either within the
user batch or in our own preamble, the engine raises a
GT_WAIT_ON_SEMAPHORE interrupt. We can unmask that interrupt and so
respond to a semaphore wait by yielding the timeslice, if we have
another context to yield to!

The only real complication is that the interrupt is only generated for
the start of the semaphore wait, and is asynchronous to our
process_csb() -- that is, we may not have registered the timeslice before
we see the interrupt. To ensure we don't miss a potential semaphore
blocking forward progress (e.g. selftests/live_timeslice_preempt) we mark
the interrupt and apply it to the next timeslice regardless of whether it
was active at the time.

v2: We use semaphores in preempt-to-busy, within the timeslicing
implementation itself! Ergo, when we do insert a preemption due to an
expired timeslice, the new context may start with the missed semaphore
flagged by the retired context and be yielded, ad infinitum. To avoid
this, read the context id at the time of the semaphore interrupt and
only yield if that context is still active.

Fixes: 8ee36e048c98 ("drm/i915/execlists: Minimalistic timeslicing")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200407130811.17321-1-chris@chris-wilson.co.uk
(cherry picked from commit c4e8ba7390346a77ffe33ec3f210bc62e0b6c8c6)
(cherry picked from commit cd60e4ac4738a6921592c4f7baf87f9a3499f0e2)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_engine_cs.c    |  6 ++++
 drivers/gpu/drm/i915/gt/intel_engine_types.h |  9 ++++++
 drivers/gpu/drm/i915/gt/intel_gt_irq.c       | 15 ++++++++--
 drivers/gpu/drm/i915/gt/intel_lrc.c          | 41 ++++++++++++++++++++++++----
 drivers/gpu/drm/i915/gt/selftest_lrc.c       | 34 ++++++++++++-----------
 drivers/gpu/drm/i915/i915_reg.h              |  1 +
 6 files changed, 82 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
index 3aa8a652c16d..883a9b7fe88d 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
@@ -1295,6 +1295,12 @@ static void intel_engine_print_registers(struct intel_engine_cs *engine,
 
 	if (engine->id == RENDER_CLASS && IS_GEN_RANGE(dev_priv, 4, 7))
 		drm_printf(m, "\tCCID: 0x%08x\n", ENGINE_READ(engine, CCID));
+	if (HAS_EXECLISTS(dev_priv)) {
+		drm_printf(m, "\tEL_STAT_HI: 0x%08x\n",
+			   ENGINE_READ(engine, RING_EXECLIST_STATUS_HI));
+		drm_printf(m, "\tEL_STAT_LO: 0x%08x\n",
+			   ENGINE_READ(engine, RING_EXECLIST_STATUS_LO));
+	}
 	drm_printf(m, "\tRING_START: 0x%08x\n",
 		   ENGINE_READ(engine, RING_START));
 	drm_printf(m, "\tRING_HEAD:  0x%08x\n",
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h
index bf4d13e08887..763f8f530ded 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h
@@ -156,6 +156,15 @@ struct intel_engine_execlists {
 	 */
 	struct i915_priolist default_priolist;
 
+	/**
+	 * @yield: CCID at the time of the last semaphore-wait interrupt.
+	 *
+	 * Instead of leaving a semaphore busy-spinning on an engine, we would
+	 * like to switch to another ready context, i.e. yielding the semaphore
+	 * timeslice.
+	 */
+	u32 yield;
+
 	/**
 	 * @error_interrupt: CS Master EIR
 	 *
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
index f0e7fd95165a..0cc7dd54f4f9 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_irq.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
@@ -39,6 +39,15 @@ cs_irq_handler(struct intel_engine_cs *engine, u32 iir)
 		}
 	}
 
+	if (iir & GT_WAIT_SEMAPHORE_INTERRUPT) {
+		WRITE_ONCE(engine->execlists.yield,
+			   ENGINE_READ_FW(engine, RING_EXECLIST_STATUS_HI));
+		ENGINE_TRACE(engine, "semaphore yield: %08x\n",
+			     engine->execlists.yield);
+		if (del_timer(&engine->execlists.timer))
+			tasklet = true;
+	}
+
 	if (iir & GT_CONTEXT_SWITCH_INTERRUPT)
 		tasklet = true;
 
@@ -228,7 +237,8 @@ void gen11_gt_irq_postinstall(struct intel_gt *gt)
 	const u32 irqs =
 		GT_CS_MASTER_ERROR_INTERRUPT |
 		GT_RENDER_USER_INTERRUPT |
-		GT_CONTEXT_SWITCH_INTERRUPT;
+		GT_CONTEXT_SWITCH_INTERRUPT |
+		GT_WAIT_SEMAPHORE_INTERRUPT;
 	struct intel_uncore *uncore = gt->uncore;
 	const u32 dmask = irqs << 16 | irqs;
 	const u32 smask = irqs << 16;
@@ -366,7 +376,8 @@ void gen8_gt_irq_postinstall(struct intel_gt *gt)
 	const u32 irqs =
 		GT_CS_MASTER_ERROR_INTERRUPT |
 		GT_RENDER_USER_INTERRUPT |
-		GT_CONTEXT_SWITCH_INTERRUPT;
+		GT_CONTEXT_SWITCH_INTERRUPT |
+		GT_WAIT_SEMAPHORE_INTERRUPT;
 	const u32 gt_interrupts[] = {
 		irqs << GEN8_RCS_IRQ_SHIFT | irqs << GEN8_BCS_IRQ_SHIFT,
 		irqs << GEN8_VCS0_IRQ_SHIFT | irqs << GEN8_VCS1_IRQ_SHIFT,
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index 017fd5bf9c69..a21b962c736f 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -1754,7 +1754,8 @@ static void defer_active(struct intel_engine_cs *engine)
 }
 
 static bool
-need_timeslice(struct intel_engine_cs *engine, const struct i915_request *rq)
+need_timeslice(const struct intel_engine_cs *engine,
+	       const struct i915_request *rq)
 {
 	int hint;
 
@@ -1768,6 +1769,32 @@ need_timeslice(struct intel_engine_cs *engine, const struct i915_request *rq)
 	return hint >= effective_prio(rq);
 }
 
+static bool
+timeslice_yield(const struct intel_engine_execlists *el,
+		const struct i915_request *rq)
+{
+	/*
+	 * Once bitten, forever smitten!
+	 *
+	 * If the active context ever busy-waited on a semaphore,
+	 * it will be treated as a hog until the end of its timeslice (i.e.
+	 * until it is scheduled out and replaced by a new submission,
+	 * possibly even its own lite-restore). The HW only sends an interrupt
+	 * on the first miss, and we do know if that semaphore has been
+	 * signaled, or even if it is now stuck on another semaphore. Play
+	 * safe, yield if it might be stuck -- it will be given a fresh
+	 * timeslice in the near future.
+	 */
+	return upper_32_bits(rq->context->lrc_desc) == READ_ONCE(el->yield);
+}
+
+static bool
+timeslice_expired(const struct intel_engine_execlists *el,
+		  const struct i915_request *rq)
+{
+	return timer_expired(&el->timer) || timeslice_yield(el, rq);
+}
+
 static int
 switch_prio(struct intel_engine_cs *engine, const struct i915_request *rq)
 {
@@ -1783,8 +1810,7 @@ timeslice(const struct intel_engine_cs *engine)
 	return READ_ONCE(engine->props.timeslice_duration_ms);
 }
 
-static unsigned long
-active_timeslice(const struct intel_engine_cs *engine)
+static unsigned long active_timeslice(const struct intel_engine_cs *engine)
 {
 	const struct intel_engine_execlists *execlists = &engine->execlists;
 	const struct i915_request *rq = *execlists->active;
@@ -1946,13 +1972,14 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
 
 			last = NULL;
 		} else if (need_timeslice(engine, last) &&
-			   timer_expired(&engine->execlists.timer)) {
+			   timeslice_expired(execlists, last)) {
 			ENGINE_TRACE(engine,
-				     "expired last=%llx:%lld, prio=%d, hint=%d\n",
+				     "expired last=%llx:%lld, prio=%d, hint=%d, yield?=%s\n",
 				     last->fence.context,
 				     last->fence.seqno,
 				     last->sched.attr.priority,
-				     execlists->queue_priority_hint);
+				     execlists->queue_priority_hint,
+				     yesno(timeslice_yield(execlists, last)));
 
 			ring_set_paused(engine, 1);
 			defer_active(engine);
@@ -2213,6 +2240,7 @@ done:
 		}
 		clear_ports(port + 1, last_port - port);
 
+		WRITE_ONCE(execlists->yield, -1);
 		execlists_submit_ports(engine);
 		set_preempt_timeout(engine, *active);
 	} else {
@@ -4452,6 +4480,7 @@ logical_ring_default_irqs(struct intel_engine_cs *engine)
 	engine->irq_enable_mask = GT_RENDER_USER_INTERRUPT << shift;
 	engine->irq_keep_mask = GT_CONTEXT_SWITCH_INTERRUPT << shift;
 	engine->irq_keep_mask |= GT_CS_MASTER_ERROR_INTERRUPT << shift;
+	engine->irq_keep_mask |= GT_WAIT_SEMAPHORE_INTERRUPT << shift;
 }
 
 static void rcs_submission_override(struct intel_engine_cs *engine)
diff --git a/drivers/gpu/drm/i915/gt/selftest_lrc.c b/drivers/gpu/drm/i915/gt/selftest_lrc.c
index 6f06ba750a0a..f95ae15ce865 100644
--- a/drivers/gpu/drm/i915/gt/selftest_lrc.c
+++ b/drivers/gpu/drm/i915/gt/selftest_lrc.c
@@ -929,7 +929,7 @@ create_rewinder(struct intel_context *ce,
 			goto err;
 	}
 
-	cs = intel_ring_begin(rq, 10);
+	cs = intel_ring_begin(rq, 14);
 	if (IS_ERR(cs)) {
 		err = PTR_ERR(cs);
 		goto err;
@@ -941,8 +941,8 @@ create_rewinder(struct intel_context *ce,
 	*cs++ = MI_SEMAPHORE_WAIT |
 		MI_SEMAPHORE_GLOBAL_GTT |
 		MI_SEMAPHORE_POLL |
-		MI_SEMAPHORE_SAD_NEQ_SDD;
-	*cs++ = 0;
+		MI_SEMAPHORE_SAD_GTE_SDD;
+	*cs++ = idx;
 	*cs++ = offset;
 	*cs++ = 0;
 
@@ -951,6 +951,11 @@ create_rewinder(struct intel_context *ce,
 	*cs++ = offset + idx * sizeof(u32);
 	*cs++ = 0;
 
+	*cs++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT;
+	*cs++ = offset;
+	*cs++ = 0;
+	*cs++ = idx + 1;
+
 	intel_ring_advance(rq, cs);
 
 	rq->sched.attr.priority = I915_PRIORITY_MASK;
@@ -984,7 +989,7 @@ static int live_timeslice_rewind(void *arg)
 
 	for_each_engine(engine, gt, id) {
 		enum { A1, A2, B1 };
-		enum { X = 1, Y, Z };
+		enum { X = 1, Z, Y };
 		struct i915_request *rq[3] = {};
 		struct intel_context *ce;
 		unsigned long heartbeat;
@@ -1017,13 +1022,13 @@ static int live_timeslice_rewind(void *arg)
 			goto err;
 		}
 
-		rq[0] = create_rewinder(ce, NULL, slot, 1);
+		rq[0] = create_rewinder(ce, NULL, slot, X);
 		if (IS_ERR(rq[0])) {
 			intel_context_put(ce);
 			goto err;
 		}
 
-		rq[1] = create_rewinder(ce, NULL, slot, 2);
+		rq[1] = create_rewinder(ce, NULL, slot, Y);
 		intel_context_put(ce);
 		if (IS_ERR(rq[1]))
 			goto err;
@@ -1041,7 +1046,7 @@ static int live_timeslice_rewind(void *arg)
 			goto err;
 		}
 
-		rq[2] = create_rewinder(ce, rq[0], slot, 3);
+		rq[2] = create_rewinder(ce, rq[0], slot, Z);
 		intel_context_put(ce);
 		if (IS_ERR(rq[2]))
 			goto err;
@@ -1055,15 +1060,12 @@ static int live_timeslice_rewind(void *arg)
 		GEM_BUG_ON(!timer_pending(&engine->execlists.timer));
 
 		/* ELSP[] = { { A:rq1, A:rq2 }, { B:rq1 } } */
-		GEM_BUG_ON(!i915_request_is_active(rq[A1]));
-		GEM_BUG_ON(!i915_request_is_active(rq[A2]));
-		GEM_BUG_ON(!i915_request_is_active(rq[B1]));
-
-		/* Wait for the timeslice to kick in */
-		del_timer(&engine->execlists.timer);
-		tasklet_hi_schedule(&engine->execlists.tasklet);
-		intel_engine_flush_submission(engine);
-
+		if (i915_request_is_active(rq[A2])) { /* semaphore yielded! */
+			/* Wait for the timeslice to kick in */
+			del_timer(&engine->execlists.timer);
+			tasklet_hi_schedule(&engine->execlists.tasklet);
+			intel_engine_flush_submission(engine);
+		}
 		/* -> ELSP[] = { { A:rq1 }, { B:rq1 } } */
 		GEM_BUG_ON(!i915_request_is_active(rq[A1]));
 		GEM_BUG_ON(!i915_request_is_active(rq[B1]));
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index e0c6021fdaf9..6e12000c4b6b 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -3094,6 +3094,7 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define GT_BSD_CS_ERROR_INTERRUPT		(1 << 15)
 #define GT_BSD_USER_INTERRUPT			(1 << 12)
 #define GT_RENDER_L3_PARITY_ERROR_INTERRUPT_S1	(1 << 11) /* hsw+; rsvd on snb, ivb, vlv */
+#define GT_WAIT_SEMAPHORE_INTERRUPT		REG_BIT(11) /* bdw+ */
 #define GT_CONTEXT_SWITCH_INTERRUPT		(1 <<  8)
 #define GT_RENDER_L3_PARITY_ERROR_INTERRUPT	(1 <<  5) /* !snb */
 #define GT_RENDER_PIPECTL_NOTIFY_INTERRUPT	(1 <<  4)
-- 
cgit v1.2.3-59-g8ed1b


From 47bf7b7a7151bad568b9523d14477a353a450066 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Mon, 20 Apr 2020 13:53:55 +0100
Subject: drm/i915/gem: Remove object_is_locked assertion from
 unpin_from_display_plane

Since moving the obj->vma.list to a spin_lock, and the vm->bound_list to
its vm->mutex, along with tracking shrinkable status under its own
spinlock, we no long require the object to be locked by the caller.

This is fortunate as it appears we can be called with the lock along an
error path in flipping:

<4> [139.942851] WARN_ON(debug_locks && !lock_is_held(&(&((obj)->base.resv)->lock.base)->dep_map))
<4> [139.943242] WARNING: CPU: 0 PID: 1203 at drivers/gpu/drm/i915/gem/i915_gem_domain.c:405 i915_gem_object_unpin_from_display_plane+0x70/0x130 [i915]
<4> [139.943263] Modules linked in: snd_hda_intel i915 vgem snd_hda_codec_realtek snd_hda_codec_generic coretemp snd_intel_dspcfg snd_hda_codec snd_hwdep snd_hda_core r8169 lpc_ich snd_pcm realtek prime_numbers [last unloaded: i915]
<4> [139.943347] CPU: 0 PID: 1203 Comm: kms_flip Tainted: G     U            5.6.0-gd0fda5c2cf3f1-drmtip_474+ #1
<4> [139.943363] Hardware name:  /D510MO, BIOS MOPNV10J.86A.0311.2010.0802.2346 08/02/2010
<4> [139.943589] RIP: 0010:i915_gem_object_unpin_from_display_plane+0x70/0x130 [i915]
<4> [139.943589] Code: 85 28 01 00 00 be ff ff ff ff 48 8d 78 60 e8 d7 9b f0 e2 85 c0 75 b9 48 c7 c6 50 b9 38 c0 48 c7 c7 e9 48 3c c0 e8 20 d4 e9 e2 <0f> 0b eb a2 48 c7 c1 08 bb 38 c0 ba 0a 01 00 00 48 c7 c6 88 a3 35
<4> [139.943589] RSP: 0018:ffffb774c0603b48 EFLAGS: 00010282
<4> [139.943589] RAX: 0000000000000000 RBX: ffff9a142fa36e80 RCX: 0000000000000006
<4> [139.943589] RDX: 000000000000160d RSI: ffff9a142c1a88f8 RDI: ffffffffa434a64d
<4> [139.943589] RBP: ffff9a1410a513c0 R08: ffff9a142c1a88f8 R09: 0000000000000000
<4> [139.943589] R10: 0000000000000000 R11: 0000000000000000 R12: ffff9a1436ee94b8
<4> [139.943589] R13: 0000000000000001 R14: 00000000ffffffff R15: ffff9a1410960000
<4> [139.943589] FS:  00007fc73a744e40(0000) GS:ffff9a143da00000(0000) knlGS:0000000000000000
<4> [139.943589] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<4> [139.943589] CR2: 00007fc73997e098 CR3: 000000002f5fe000 CR4: 00000000000006f0
<4> [139.943589] Call Trace:
<4> [139.943589]  intel_pin_and_fence_fb_obj+0x1c9/0x1f0 [i915]
<4> [139.943589]  intel_plane_pin_fb+0x3f/0xd0 [i915]
<4> [139.943589]  intel_prepare_plane_fb+0x13b/0x5c0 [i915]
<4> [139.943589]  drm_atomic_helper_prepare_planes+0x85/0x110
<4> [139.943589]  intel_atomic_commit+0xda/0x390 [i915]
<4> [139.943589]  drm_atomic_helper_page_flip+0x9c/0xd0
<4> [139.943589]  ? drm_event_reserve_init+0x46/0x60
<4> [139.943589]  drm_mode_page_flip_ioctl+0x587/0x5d0

This completes the symmetry lost in commit 8b1c78e06e61 ("drm/i915: Avoid
calling i915_gem_object_unbind holding object lock").

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1743
Fixes: 8b1c78e06e61 ("drm/i915: Avoid calling i915_gem_object_unbind holding object lock")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Andi Shyti <andi.shyti@intel.com>
Cc: <stable@vger.kernel.org> # v5.6+
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200420125356.26614-1-chris@chris-wilson.co.uk
(cherry picked from commit a95f3ac21d64d62c746f836598d1467d5837fa28)
(cherry picked from commit 2208b85fa1766ee4821a9435d548578b67090531)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/gem/i915_gem_domain.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
index 0cc40e77bbd2..4f96c8788a2e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
@@ -368,7 +368,6 @@ static void i915_gem_object_bump_inactive_ggtt(struct drm_i915_gem_object *obj)
 	struct drm_i915_private *i915 = to_i915(obj->base.dev);
 	struct i915_vma *vma;
 
-	GEM_BUG_ON(!i915_gem_object_has_pinned_pages(obj));
 	if (!atomic_read(&obj->bind_count))
 		return;
 
@@ -400,12 +399,8 @@ static void i915_gem_object_bump_inactive_ggtt(struct drm_i915_gem_object *obj)
 void
 i915_gem_object_unpin_from_display_plane(struct i915_vma *vma)
 {
-	struct drm_i915_gem_object *obj = vma->obj;
-
-	assert_object_held(obj);
-
 	/* Bump the LRU to try and avoid premature eviction whilst flipping  */
-	i915_gem_object_bump_inactive_ggtt(obj);
+	i915_gem_object_bump_inactive_ggtt(vma->obj);
 
 	i915_vma_unpin(vma);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 53b2622e774681943230fb9f31096512fff99fe7 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Tue, 28 Apr 2020 19:47:49 +0100
Subject: drm/i915/execlists: Avoid reusing the same logical CCID

The bspec is confusing on the nature of the upper 32bits of the LRC
descriptor. Once upon a time, it said that it uses the upper 32b to
decide if it should perform a lite-restore, and so we must ensure that
each unique context submitted to HW is given a unique CCID [for the
duration of it being on the HW]. Currently, this is achieved by using
a small circular tag, and assigning every context submitted to HW a
new id. However, this tag is being cleared on repinning an inflight
context such that we end up re-using the 0 tag for multiple contexts.

To avoid accidentally clearing the CCID in the upper 32bits of the LRC
descriptor, split the descriptor into two dwords so we can update the
GGTT address separately from the CCID.

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1796
Fixes: 2935ed5339c4 ("drm/i915: Remove logical HW ID")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: <stable@vger.kernel.org> # v5.5+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200428184751.11257-1-chris@chris-wilson.co.uk
(cherry picked from commit 2632f174a2e1a5fd40a70404fa8ccfd0b1f79ebd)
(cherry picked from commit a4b70fcc587860f4b972f68217d8ebebe295ec15)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_context_types.h     |  8 +++-
 drivers/gpu/drm/i915/gt/intel_engine_types.h      |  5 +++
 drivers/gpu/drm/i915/gt/intel_lrc.c               | 50 ++++++++++-------------
 drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c |  2 +-
 drivers/gpu/drm/i915/gvt/scheduler.c              |  4 +-
 drivers/gpu/drm/i915/i915_perf.c                  |  3 +-
 6 files changed, 37 insertions(+), 35 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_context_types.h b/drivers/gpu/drm/i915/gt/intel_context_types.h
index 07cb83a0d017..ca0d4f4f3615 100644
--- a/drivers/gpu/drm/i915/gt/intel_context_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_context_types.h
@@ -69,7 +69,13 @@ struct intel_context {
 #define CONTEXT_NOPREEMPT		7
 
 	u32 *lrc_reg_state;
-	u64 lrc_desc;
+	union {
+		struct {
+			u32 lrca;
+			u32 ccid;
+		};
+		u64 desc;
+	} lrc;
 	u32 tag; /* cookie passed to HW to track this context on submission */
 
 	/* Time on GPU as tracked by the hw. */
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h
index 763f8f530ded..8dd210a2c340 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h
@@ -156,6 +156,11 @@ struct intel_engine_execlists {
 	 */
 	struct i915_priolist default_priolist;
 
+	/**
+	 * @ccid: identifier for contexts submitted to this engine
+	 */
+	u32 ccid;
+
 	/**
 	 * @yield: CCID at the time of the last semaphore-wait interrupt.
 	 *
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index a21b962c736f..e8b02f84aa3d 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -456,10 +456,10 @@ assert_priority_queue(const struct i915_request *prev,
  * engine info, SW context ID and SW counter need to form a unique number
  * (Context ID) per lrc.
  */
-static u64
+static u32
 lrc_descriptor(struct intel_context *ce, struct intel_engine_cs *engine)
 {
-	u64 desc;
+	u32 desc;
 
 	desc = INTEL_LEGACY_32B_CONTEXT;
 	if (i915_vm_is_4lvl(ce->vm))
@@ -470,21 +470,7 @@ lrc_descriptor(struct intel_context *ce, struct intel_engine_cs *engine)
 	if (IS_GEN(engine->i915, 8))
 		desc |= GEN8_CTX_L3LLC_COHERENT;
 
-	desc |= i915_ggtt_offset(ce->state); /* bits 12-31 */
-	/*
-	 * The following 32bits are copied into the OA reports (dword 2).
-	 * Consider updating oa_get_render_ctx_id in i915_perf.c when changing
-	 * anything below.
-	 */
-	if (INTEL_GEN(engine->i915) >= 11) {
-		desc |= (u64)engine->instance << GEN11_ENGINE_INSTANCE_SHIFT;
-								/* bits 48-53 */
-
-		desc |= (u64)engine->class << GEN11_ENGINE_CLASS_SHIFT;
-								/* bits 61-63 */
-	}
-
-	return desc;
+	return i915_ggtt_offset(ce->state) | desc;
 }
 
 static inline unsigned int dword_in_page(void *addr)
@@ -1192,7 +1178,7 @@ static void reset_active(struct i915_request *rq,
 	__execlists_update_reg_state(ce, engine, head);
 
 	/* We've switched away, so this should be a no-op, but intent matters */
-	ce->lrc_desc |= CTX_DESC_FORCE_RESTORE;
+	ce->lrc.desc |= CTX_DESC_FORCE_RESTORE;
 }
 
 static u32 intel_context_get_runtime(const struct intel_context *ce)
@@ -1251,18 +1237,19 @@ __execlists_schedule_in(struct i915_request *rq)
 	if (IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM))
 		execlists_check_context(ce, engine);
 
-	ce->lrc_desc &= ~GENMASK_ULL(47, 37);
 	if (ce->tag) {
 		/* Use a fixed tag for OA and friends */
-		ce->lrc_desc |= (u64)ce->tag << 32;
+		ce->lrc.ccid = ce->tag;
 	} else {
 		/* We don't need a strict matching tag, just different values */
-		ce->lrc_desc |=
-			(u64)(++engine->context_tag % NUM_CONTEXT_TAG) <<
-			GEN11_SW_CTX_ID_SHIFT;
+		ce->lrc.ccid =
+			(++engine->context_tag % NUM_CONTEXT_TAG) <<
+			(GEN11_SW_CTX_ID_SHIFT - 32);
 		BUILD_BUG_ON(NUM_CONTEXT_TAG > GEN12_MAX_CONTEXT_HW_ID);
 	}
 
+	ce->lrc.ccid |= engine->execlists.ccid;
+
 	__intel_gt_pm_get(engine->gt);
 	execlists_context_status_change(rq, INTEL_CONTEXT_SCHEDULE_IN);
 	intel_engine_context_in(engine);
@@ -1361,7 +1348,7 @@ execlists_schedule_out(struct i915_request *rq)
 static u64 execlists_update_context(struct i915_request *rq)
 {
 	struct intel_context *ce = rq->context;
-	u64 desc = ce->lrc_desc;
+	u64 desc = ce->lrc.desc;
 	u32 tail, prev;
 
 	/*
@@ -1400,7 +1387,7 @@ static u64 execlists_update_context(struct i915_request *rq)
 	 */
 	wmb();
 
-	ce->lrc_desc &= ~CTX_DESC_FORCE_RESTORE;
+	ce->lrc.desc &= ~CTX_DESC_FORCE_RESTORE;
 	return desc;
 }
 
@@ -1785,7 +1772,7 @@ timeslice_yield(const struct intel_engine_execlists *el,
 	 * safe, yield if it might be stuck -- it will be given a fresh
 	 * timeslice in the near future.
 	 */
-	return upper_32_bits(rq->context->lrc_desc) == READ_ONCE(el->yield);
+	return rq->context->lrc.ccid == READ_ONCE(el->yield);
 }
 
 static bool
@@ -3071,7 +3058,7 @@ __execlists_context_pin(struct intel_context *ce,
 	if (IS_ERR(vaddr))
 		return PTR_ERR(vaddr);
 
-	ce->lrc_desc = lrc_descriptor(ce, engine) | CTX_DESC_FORCE_RESTORE;
+	ce->lrc.lrca = lrc_descriptor(ce, engine) | CTX_DESC_FORCE_RESTORE;
 	ce->lrc_reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE;
 	__execlists_update_reg_state(ce, engine, ce->ring->tail);
 
@@ -3100,7 +3087,7 @@ static void execlists_context_reset(struct intel_context *ce)
 				 ce, ce->engine, ce->ring, true);
 	__execlists_update_reg_state(ce, ce->engine, ce->ring->tail);
 
-	ce->lrc_desc |= CTX_DESC_FORCE_RESTORE;
+	ce->lrc.desc |= CTX_DESC_FORCE_RESTORE;
 }
 
 static const struct intel_context_ops execlists_context_ops = {
@@ -3781,7 +3768,7 @@ out_replay:
 		     head, ce->ring->tail);
 	__execlists_reset_reg_state(ce, engine);
 	__execlists_update_reg_state(ce, engine, head);
-	ce->lrc_desc |= CTX_DESC_FORCE_RESTORE; /* paranoid: GPU was reset! */
+	ce->lrc.desc |= CTX_DESC_FORCE_RESTORE; /* paranoid: GPU was reset! */
 
 unwind:
 	/* Push back any incomplete requests for replay after the reset. */
@@ -4548,6 +4535,11 @@ int intel_execlists_submission_setup(struct intel_engine_cs *engine)
 	else
 		execlists->csb_size = GEN11_CSB_ENTRIES;
 
+	if (INTEL_GEN(engine->i915) >= 11) {
+		execlists->ccid |= engine->instance << (GEN11_ENGINE_INSTANCE_SHIFT - 32);
+		execlists->ccid |= engine->class << (GEN11_ENGINE_CLASS_SHIFT - 32);
+	}
+
 	reset_csb_pointers(engine);
 
 	/* Finally, take ownership and responsibility for cleanup! */
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
index fe7778c28d2d..aa6d56e25a10 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
@@ -217,7 +217,7 @@ static void guc_wq_item_append(struct intel_guc *guc,
 static void guc_add_request(struct intel_guc *guc, struct i915_request *rq)
 {
 	struct intel_engine_cs *engine = rq->engine;
-	u32 ctx_desc = lower_32_bits(rq->context->lrc_desc);
+	u32 ctx_desc = rq->context->lrc.ccid;
 	u32 ring_tail = intel_ring_set_tail(rq->ring, rq->tail) / sizeof(u64);
 
 	guc_wq_item_append(guc, engine->guc_id, ctx_desc,
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index cb11c3184085..6eb6710b35e9 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -290,7 +290,7 @@ static void
 shadow_context_descriptor_update(struct intel_context *ce,
 				 struct intel_vgpu_workload *workload)
 {
-	u64 desc = ce->lrc_desc;
+	u64 desc = ce->lrc.desc;
 
 	/*
 	 * Update bits 0-11 of the context descriptor which includes flags
@@ -300,7 +300,7 @@ shadow_context_descriptor_update(struct intel_context *ce,
 	desc |= (u64)workload->ctx_desc.addressing_mode <<
 		GEN8_CTX_ADDRESSING_MODE_SHIFT;
 
-	ce->lrc_desc = desc;
+	ce->lrc.desc = desc;
 }
 
 static int copy_workload_to_ring_buffer(struct intel_vgpu_workload *workload)
diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index 66a46e41d5ef..b5030192be3e 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -1310,8 +1310,7 @@ static int oa_get_render_ctx_id(struct i915_perf_stream *stream)
 			 * dropped by GuC. They won't be part of the context
 			 * ID in the OA reports, so squash those lower bits.
 			 */
-			stream->specific_ctx_id =
-				lower_32_bits(ce->lrc_desc) >> 12;
+			stream->specific_ctx_id = ce->lrc.lrca >> 12;
 
 			/*
 			 * GuC uses the top bit to signal proxy submission, so
-- 
cgit v1.2.3-59-g8ed1b


From 1bc6a60143a4f9264cc6e09ceb9919f4e813a872 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Tue, 28 Apr 2020 19:47:50 +0100
Subject: drm/i915/execlists: Track inflight CCID

The presumption is that by using a circular counter that is twice as
large as the maximum ELSP submission, we would never reuse the same CCID
for two inflight contexts.

However, if we continually preempt an active context such that it always
remains inflight, it can be resubmitted with an arbitrary number of
paired contexts. As each of its paired contexts will use a new CCID,
eventually it will wrap and submit two ELSP with the same CCID.

Rather than use a simple circular counter, switch over to a small bitmap
of inflight ids so we can avoid reusing one that is still potentially
active.

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1796
Fixes: 2935ed5339c4 ("drm/i915: Remove logical HW ID")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: <stable@vger.kernel.org> # v5.5+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200428184751.11257-2-chris@chris-wilson.co.uk
(cherry picked from commit 5c4a53e3b1cbc38d0906e382f1037290658759bb)
(cherry picked from commit 134711240307d5586ae8e828d2699db70a8b74f2)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_engine_types.h |  3 +--
 drivers/gpu/drm/i915/gt/intel_lrc.c          | 29 +++++++++++++++++++++-------
 drivers/gpu/drm/i915/i915_perf.c             |  3 +--
 drivers/gpu/drm/i915/selftests/i915_vma.c    |  2 +-
 4 files changed, 25 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h
index 8dd210a2c340..0be674ae1cf6 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h
@@ -309,8 +309,7 @@ struct intel_engine_cs {
 	u32 context_size;
 	u32 mmio_base;
 
-	unsigned int context_tag;
-#define NUM_CONTEXT_TAG roundup_pow_of_two(2 * EXECLIST_MAX_PORTS)
+	unsigned long context_tag;
 
 	struct rb_node uabi_node;
 
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index e8b02f84aa3d..77420372d813 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -1239,13 +1239,17 @@ __execlists_schedule_in(struct i915_request *rq)
 
 	if (ce->tag) {
 		/* Use a fixed tag for OA and friends */
+		GEM_BUG_ON(ce->tag <= BITS_PER_LONG);
 		ce->lrc.ccid = ce->tag;
 	} else {
 		/* We don't need a strict matching tag, just different values */
-		ce->lrc.ccid =
-			(++engine->context_tag % NUM_CONTEXT_TAG) <<
-			(GEN11_SW_CTX_ID_SHIFT - 32);
-		BUILD_BUG_ON(NUM_CONTEXT_TAG > GEN12_MAX_CONTEXT_HW_ID);
+		unsigned int tag = ffs(engine->context_tag);
+
+		GEM_BUG_ON(tag == 0 || tag >= BITS_PER_LONG);
+		clear_bit(tag - 1, &engine->context_tag);
+		ce->lrc.ccid = tag << (GEN11_SW_CTX_ID_SHIFT - 32);
+
+		BUILD_BUG_ON(BITS_PER_LONG > GEN12_MAX_CONTEXT_HW_ID);
 	}
 
 	ce->lrc.ccid |= engine->execlists.ccid;
@@ -1289,7 +1293,8 @@ static void kick_siblings(struct i915_request *rq, struct intel_context *ce)
 
 static inline void
 __execlists_schedule_out(struct i915_request *rq,
-			 struct intel_engine_cs * const engine)
+			 struct intel_engine_cs * const engine,
+			 unsigned int ccid)
 {
 	struct intel_context * const ce = rq->context;
 
@@ -1307,6 +1312,14 @@ __execlists_schedule_out(struct i915_request *rq,
 	    i915_request_completed(rq))
 		intel_engine_add_retire(engine, ce->timeline);
 
+	ccid >>= GEN11_SW_CTX_ID_SHIFT - 32;
+	ccid &= GEN12_MAX_CONTEXT_HW_ID;
+	if (ccid < BITS_PER_LONG) {
+		GEM_BUG_ON(ccid == 0);
+		GEM_BUG_ON(test_bit(ccid - 1, &engine->context_tag));
+		set_bit(ccid - 1, &engine->context_tag);
+	}
+
 	intel_context_update_runtime(ce);
 	intel_engine_context_out(engine);
 	execlists_context_status_change(rq, INTEL_CONTEXT_SCHEDULE_OUT);
@@ -1332,15 +1345,17 @@ execlists_schedule_out(struct i915_request *rq)
 {
 	struct intel_context * const ce = rq->context;
 	struct intel_engine_cs *cur, *old;
+	u32 ccid;
 
 	trace_i915_request_out(rq);
 
+	ccid = rq->context->lrc.ccid;
 	old = READ_ONCE(ce->inflight);
 	do
 		cur = ptr_unmask_bits(old, 2) ? ptr_dec(old) : NULL;
 	while (!try_cmpxchg(&ce->inflight, &old, cur));
 	if (!cur)
-		__execlists_schedule_out(rq, old);
+		__execlists_schedule_out(rq, old, ccid);
 
 	i915_request_put(rq);
 }
@@ -3556,7 +3571,7 @@ static void enable_execlists(struct intel_engine_cs *engine)
 
 	enable_error_interrupt(engine);
 
-	engine->context_tag = 0;
+	engine->context_tag = GENMASK(BITS_PER_LONG - 2, 0);
 }
 
 static bool unexpected_starting_state(struct intel_engine_cs *engine)
diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index b5030192be3e..cf2c01f17da8 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -1327,11 +1327,10 @@ static int oa_get_render_ctx_id(struct i915_perf_stream *stream)
 			((1U << GEN11_SW_CTX_ID_WIDTH) - 1) << (GEN11_SW_CTX_ID_SHIFT - 32);
 		/*
 		 * Pick an unused context id
-		 * 0 - (NUM_CONTEXT_TAG - 1) are used by other contexts
+		 * 0 - BITS_PER_LONG are used by other contexts
 		 * GEN12_MAX_CONTEXT_HW_ID (0x7ff) is used by idle context
 		 */
 		stream->specific_ctx_id = (GEN12_MAX_CONTEXT_HW_ID - 1) << (GEN11_SW_CTX_ID_SHIFT - 32);
-		BUILD_BUG_ON((GEN12_MAX_CONTEXT_HW_ID - 1) < NUM_CONTEXT_TAG);
 		break;
 	}
 
diff --git a/drivers/gpu/drm/i915/selftests/i915_vma.c b/drivers/gpu/drm/i915/selftests/i915_vma.c
index 58b5f40a07dd..af89c7fc8f59 100644
--- a/drivers/gpu/drm/i915/selftests/i915_vma.c
+++ b/drivers/gpu/drm/i915/selftests/i915_vma.c
@@ -173,7 +173,7 @@ static int igt_vma_create(void *arg)
 		}
 
 		nc = 0;
-		for_each_prime_number(num_ctx, 2 * NUM_CONTEXT_TAG) {
+		for_each_prime_number(num_ctx, 2 * BITS_PER_LONG) {
 			for (; nc < num_ctx; nc++) {
 				ctx = mock_context(i915, "mock");
 				if (!ctx)
-- 
cgit v1.2.3-59-g8ed1b


From ee79be181aee2f59522fdf81c580aaed5753e3af Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Thu, 30 Apr 2020 11:15:19 +0800
Subject: drm/amdgpu: disable MGCG/MGLS also on gfx CG ungate

Otherwise, MGCG/MGLS will be left enabled.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index f92c158d89a1..0c98f705da0e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4273,7 +4273,7 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 		/* ===  CGCG /CGLS for GFX 3D Only === */
 		gfx_v10_0_update_3d_clock_gating(adev, enable);
 		/* ===  MGCG + MGLS === */
-		/* gfx_v10_0_update_medium_grain_clock_gating(adev, enable); */
+		gfx_v10_0_update_medium_grain_clock_gating(adev, enable);
 	}
 
 	if (adev->cg_flags &
-- 
cgit v1.2.3-59-g8ed1b


From 1fe48ec08d9f2e26d893a6c05bd6c99a3490f9ef Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Thu, 30 Apr 2020 11:24:02 +0800
Subject: drm/amdgpu: drop unnecessary cancel_delayed_work_sync on PG ungate

As this is already properly handled in amdgpu_gfx_off_ctrl(). In fact,
this unnecessary cancel_delayed_work_sync may leave a small time window
for race condition and is dangerous.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c |  6 +-----
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c  | 12 +++---------
 2 files changed, 4 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 0c98f705da0e..d35ac7407c8e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4353,11 +4353,7 @@ static int gfx_v10_0_set_powergating_state(void *handle,
 	switch (adev->asic_type) {
 	case CHIP_NAVI10:
 	case CHIP_NAVI14:
-		if (!enable) {
-			amdgpu_gfx_off_ctrl(adev, false);
-			cancel_delayed_work_sync(&adev->gfx.gfx_off_delay_work);
-		} else
-			amdgpu_gfx_off_ctrl(adev, true);
+		amdgpu_gfx_off_ctrl(adev, enable);
 		break;
 	default:
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 0c390485bc10..0127f1ce16e6 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -5025,10 +5025,9 @@ static int gfx_v9_0_set_powergating_state(void *handle,
 	switch (adev->asic_type) {
 	case CHIP_RAVEN:
 	case CHIP_RENOIR:
-		if (!enable) {
+		if (!enable)
 			amdgpu_gfx_off_ctrl(adev, false);
-			cancel_delayed_work_sync(&adev->gfx.gfx_off_delay_work);
-		}
+
 		if (adev->pg_flags & AMD_PG_SUPPORT_RLC_SMU_HS) {
 			gfx_v9_0_enable_sck_slow_down_on_power_up(adev, true);
 			gfx_v9_0_enable_sck_slow_down_on_power_down(adev, true);
@@ -5052,12 +5051,7 @@ static int gfx_v9_0_set_powergating_state(void *handle,
 			amdgpu_gfx_off_ctrl(adev, true);
 		break;
 	case CHIP_VEGA12:
-		if (!enable) {
-			amdgpu_gfx_off_ctrl(adev, false);
-			cancel_delayed_work_sync(&adev->gfx.gfx_off_delay_work);
-		} else {
-			amdgpu_gfx_off_ctrl(adev, true);
-		}
+		amdgpu_gfx_off_ctrl(adev, enable);
 		break;
 	default:
 		break;
-- 
cgit v1.2.3-59-g8ed1b


From f4fcfa4282c1a1bf51475ebb0ffda623eebf1191 Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Thu, 30 Apr 2020 14:38:39 +0800
Subject: drm/amd/powerplay: perform PG ungate prior to CG ungate

Since gfxoff should be disabled first before trying to access those
GC registers.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/powerplay/amd_powerplay.c | 6 +++---
 drivers/gpu/drm/amd/powerplay/amdgpu_smu.c    | 6 +++---
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
index e4e5a53b2b4e..8e2acb4df860 100644
--- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
+++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
@@ -319,12 +319,12 @@ static void pp_dpm_en_umd_pstate(struct pp_hwmgr  *hwmgr,
 		if (*level & profile_mode_mask) {
 			hwmgr->saved_dpm_level = hwmgr->dpm_level;
 			hwmgr->en_umd_pstate = true;
-			amdgpu_device_ip_set_clockgating_state(hwmgr->adev,
-						AMD_IP_BLOCK_TYPE_GFX,
-						AMD_CG_STATE_UNGATE);
 			amdgpu_device_ip_set_powergating_state(hwmgr->adev,
 					AMD_IP_BLOCK_TYPE_GFX,
 					AMD_PG_STATE_UNGATE);
+			amdgpu_device_ip_set_clockgating_state(hwmgr->adev,
+						AMD_IP_BLOCK_TYPE_GFX,
+						AMD_CG_STATE_UNGATE);
 		}
 	} else {
 		/* exit umd pstate, restore level, enable gfx cg*/
diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
index e8b27fab6aa1..09fa685b811b 100644
--- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
@@ -1744,12 +1744,12 @@ static int smu_enable_umd_pstate(void *handle,
 		if (*level & profile_mode_mask) {
 			smu_dpm_ctx->saved_dpm_level = smu_dpm_ctx->dpm_level;
 			smu_dpm_ctx->enable_umd_pstate = true;
-			amdgpu_device_ip_set_clockgating_state(smu->adev,
-							       AMD_IP_BLOCK_TYPE_GFX,
-							       AMD_CG_STATE_UNGATE);
 			amdgpu_device_ip_set_powergating_state(smu->adev,
 							       AMD_IP_BLOCK_TYPE_GFX,
 							       AMD_PG_STATE_UNGATE);
+			amdgpu_device_ip_set_clockgating_state(smu->adev,
+							       AMD_IP_BLOCK_TYPE_GFX,
+							       AMD_CG_STATE_UNGATE);
 		}
 	} else {
 		/* exit umd pstate, restore level, enable gfx cg*/
-- 
cgit v1.2.3-59-g8ed1b


From a6aacb2b26e85aa619cf0c6f98d0ca77314cd2a1 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Tue, 5 May 2020 09:42:26 -0400
Subject: drm/amdgpu: force fbdev into vram
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We set the fb smem pointer to the offset into the BAR, so keep
the fbdev bo in vram.

Bug: https://bugzilla.kernel.org/show_bug.cgi?id=207581
Fixes: 6c8d74caa2fa33 ("drm/amdgpu: Enable scatter gather display support")
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
index 9ae7b61f696a..25ddb482466a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
@@ -133,8 +133,7 @@ static int amdgpufb_create_pinned_object(struct amdgpu_fbdev *rfbdev,
 	u32 cpp;
 	u64 flags = AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
 			       AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS     |
-			       AMDGPU_GEM_CREATE_VRAM_CLEARED 	     |
-			       AMDGPU_GEM_CREATE_CPU_GTT_USWC;
+			       AMDGPU_GEM_CREATE_VRAM_CLEARED;
 
 	info = drm_get_format_info(adev->ddev, mode_cmd);
 	cpp = info->cpp[0];
-- 
cgit v1.2.3-59-g8ed1b


From 39b3128d7ffd44e400e581e6f49e88cb42bef9a1 Mon Sep 17 00:00:00 2001
From: Felix Kuehling <Felix.Kuehling@amd.com>
Date: Tue, 5 May 2020 14:02:43 -0400
Subject: drm/amdgpu: Use GEM obj reference for KFD BOs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Releasing the AMDGPU BO ref directly leads to problems when BOs were
exported as DMA bufs. Releasing the GEM reference makes sure that the
AMDGPU/TTM BO is not freed too early.

Also take a GEM reference when importing BOs from DMABufs to keep
references to imported BOs balances properly.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Tested-by: Alex Sierra <alex.sierra@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Sierra <alex.sierra@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 9dff792c9290..6a5b91d23fd9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1343,7 +1343,7 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu(
 	}
 
 	/* Free the BO*/
-	amdgpu_bo_unref(&mem->bo);
+	drm_gem_object_put_unlocked(&mem->bo->tbo.base);
 	mutex_destroy(&mem->lock);
 	kfree(mem);
 
@@ -1688,7 +1688,8 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd,
 		| KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE
 		| KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE;
 
-	(*mem)->bo = amdgpu_bo_ref(bo);
+	drm_gem_object_get(&bo->tbo.base);
+	(*mem)->bo = bo;
 	(*mem)->va = va;
 	(*mem)->domain = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ?
 		AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT;
-- 
cgit v1.2.3-59-g8ed1b


From b2b6290a23986a5c88384887b8a589a3c4ebe292 Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Thu, 7 May 2020 18:17:55 +0800
Subject: drm/amdgpu: enable hibernate support on Navi1X

BACO is needed to support hibernate on Navi1X.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h        | 1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c    | 2 ++
 drivers/gpu/drm/amd/powerplay/amdgpu_smu.c | 2 +-
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 2992a49ad4a5..8ac1581a6b53 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -945,6 +945,7 @@ struct amdgpu_device {
 
 	/* s3/s4 mask */
 	bool                            in_suspend;
+	bool				in_hibernate;
 
 	/* record last mm index being written through WREG32*/
 	unsigned long last_mm_index;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 466bfe541e45..a735d79a717b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -1181,7 +1181,9 @@ static int amdgpu_pmops_freeze(struct device *dev)
 	struct amdgpu_device *adev = drm_dev->dev_private;
 	int r;
 
+	adev->in_hibernate = true;
 	r = amdgpu_device_suspend(drm_dev, true);
+	adev->in_hibernate = false;
 	if (r)
 		return r;
 	return amdgpu_asic_reset(adev);
diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
index 09fa685b811b..e77046931e4c 100644
--- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
@@ -1476,7 +1476,7 @@ static int smu_disable_dpm(struct smu_context *smu)
 	bool use_baco = !smu->is_apu &&
 		((adev->in_gpu_reset &&
 		  (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO)) ||
-		 (adev->in_runpm && amdgpu_asic_supports_baco(adev)));
+		 ((adev->in_runpm || adev->in_hibernate) && amdgpu_asic_supports_baco(adev)));
 
 	ret = smu_get_smc_version(smu, NULL, &smu_version);
 	if (ret) {
-- 
cgit v1.2.3-59-g8ed1b


From bff1a6112b09a810d8a1c0c0ea306e58810d4f0b Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 8 May 2020 14:33:09 -0400
Subject: drm/amdgpu: implement soft_recovery for gfx10
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Same as gfx9.  This allows us to kill the waves for hung
shaders.

Acked-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index d35ac7407c8e..0e0daf0021b6 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4914,6 +4914,19 @@ static void gfx_v10_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring,
 							   ref, mask);
 }
 
+static void gfx_v10_0_ring_soft_recovery(struct amdgpu_ring *ring,
+					 unsigned vmid)
+{
+	struct amdgpu_device *adev = ring->adev;
+	uint32_t value = 0;
+
+	value = REG_SET_FIELD(value, SQ_CMD, CMD, 0x03);
+	value = REG_SET_FIELD(value, SQ_CMD, MODE, 0x01);
+	value = REG_SET_FIELD(value, SQ_CMD, CHECK_VMID, 1);
+	value = REG_SET_FIELD(value, SQ_CMD, VM_ID, vmid);
+	WREG32_SOC15(GC, 0, mmSQ_CMD, value);
+}
+
 static void
 gfx_v10_0_set_gfx_eop_interrupt_state(struct amdgpu_device *adev,
 				      uint32_t me, uint32_t pipe,
@@ -5305,6 +5318,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = {
 	.emit_wreg = gfx_v10_0_ring_emit_wreg,
 	.emit_reg_wait = gfx_v10_0_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = gfx_v10_0_ring_emit_reg_write_reg_wait,
+	.soft_recovery = gfx_v10_0_ring_soft_recovery,
 };
 
 static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_compute = {
-- 
cgit v1.2.3-59-g8ed1b


From 2346ef47e871536cf4e36b77859bfdeb39b49024 Mon Sep 17 00:00:00 2001
From: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Date: Wed, 6 May 2020 15:47:54 -0400
Subject: drm/amd/display: Fix vblank and pageflip event handling for FreeSync

[Why]
We're sending the drm vblank event a frame too early in the case where
the pageflip happens close to VUPDATE and ends up blocking the signal.

The implementation in DM was previously correct *before* we started
sending vblank events from VSTARTUP unconditionally to handle cases
where HUBP was off, OTG was ON and userspace was still requesting some
DRM planes enabled. As part of that patch series we dropped VUPDATE
since it was deemed close enough to VSTARTUP, but there's a key
difference betweeen VSTARTUP and VUPDATE - the VUPDATE signal can be
blocked if we're holding the pipe lock.

There was a fix recently to revert the unconditional behavior for the
DCN VSTARTUP vblank event since it was sending the pageflip event on
the wrong frame - once again, due to blocking VUPDATE and having the
address start scanning out two frames later.

The problem with this fix is it didn't update the logic that calls
drm_crtc_handle_vblank(), so the timestamps are totally bogus now.

[How]
Essentially reverts most of the original VSTARTUP series but retains
the behavior to send back events when active planes == 0.

Some refactoring/cleanup was done to not have duplicated code in both
the handlers.

Fixes: 16f17eda8bad ("drm/amd/display: Send vblank and user events at vsartup for DCN")
Fixes: 3a2ce8d66a4b ("drm/amd/display: Disable VUpdate interrupt for DCN hardware")
Fixes: 2b5aed9ac3f7 ("drm/amd/display: Fix pageflip event race condition for DCN.")

Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-and-Tested-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Leo Li <sunpeng.li@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org # 5.6.x
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 137 +++++++++-------------
 1 file changed, 55 insertions(+), 82 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 9c83c1303f08..c3df6ef9f101 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -441,7 +441,7 @@ static void dm_vupdate_high_irq(void *interrupt_params)
 
 /**
  * dm_crtc_high_irq() - Handles CRTC interrupt
- * @interrupt_params: ignored
+ * @interrupt_params: used for determining the CRTC instance
  *
  * Handles the CRTC/VSYNC interrupt by notfying DRM's VBLANK
  * event handler.
@@ -455,70 +455,6 @@ static void dm_crtc_high_irq(void *interrupt_params)
 	unsigned long flags;
 
 	acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VBLANK);
-
-	if (acrtc) {
-		acrtc_state = to_dm_crtc_state(acrtc->base.state);
-
-		DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d\n",
-			      acrtc->crtc_id,
-			      amdgpu_dm_vrr_active(acrtc_state));
-
-		/* Core vblank handling at start of front-porch is only possible
-		 * in non-vrr mode, as only there vblank timestamping will give
-		 * valid results while done in front-porch. Otherwise defer it
-		 * to dm_vupdate_high_irq after end of front-porch.
-		 */
-		if (!amdgpu_dm_vrr_active(acrtc_state))
-			drm_crtc_handle_vblank(&acrtc->base);
-
-		/* Following stuff must happen at start of vblank, for crc
-		 * computation and below-the-range btr support in vrr mode.
-		 */
-		amdgpu_dm_crtc_handle_crc_irq(&acrtc->base);
-
-		if (acrtc_state->stream && adev->family >= AMDGPU_FAMILY_AI &&
-		    acrtc_state->vrr_params.supported &&
-		    acrtc_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE) {
-			spin_lock_irqsave(&adev->ddev->event_lock, flags);
-			mod_freesync_handle_v_update(
-				adev->dm.freesync_module,
-				acrtc_state->stream,
-				&acrtc_state->vrr_params);
-
-			dc_stream_adjust_vmin_vmax(
-				adev->dm.dc,
-				acrtc_state->stream,
-				&acrtc_state->vrr_params.adjust);
-			spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
-		}
-	}
-}
-
-#if defined(CONFIG_DRM_AMD_DC_DCN)
-/**
- * dm_dcn_crtc_high_irq() - Handles VStartup interrupt for DCN generation ASICs
- * @interrupt params - interrupt parameters
- *
- * Notify DRM's vblank event handler at VSTARTUP
- *
- * Unlike DCE hardware, we trigger the handler at VSTARTUP. at which:
- * * We are close enough to VUPDATE - the point of no return for hw
- * * We are in the fixed portion of variable front porch when vrr is enabled
- * * We are before VUPDATE, where double-buffered vrr registers are swapped
- *
- * It is therefore the correct place to signal vblank, send user flip events,
- * and update VRR.
- */
-static void dm_dcn_crtc_high_irq(void *interrupt_params)
-{
-	struct common_irq_params *irq_params = interrupt_params;
-	struct amdgpu_device *adev = irq_params->adev;
-	struct amdgpu_crtc *acrtc;
-	struct dm_crtc_state *acrtc_state;
-	unsigned long flags;
-
-	acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VBLANK);
-
 	if (!acrtc)
 		return;
 
@@ -528,22 +464,35 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params)
 			 amdgpu_dm_vrr_active(acrtc_state),
 			 acrtc_state->active_planes);
 
+	/**
+	 * Core vblank handling at start of front-porch is only possible
+	 * in non-vrr mode, as only there vblank timestamping will give
+	 * valid results while done in front-porch. Otherwise defer it
+	 * to dm_vupdate_high_irq after end of front-porch.
+	 */
+	if (!amdgpu_dm_vrr_active(acrtc_state))
+		drm_crtc_handle_vblank(&acrtc->base);
+
+	/**
+	 * Following stuff must happen at start of vblank, for crc
+	 * computation and below-the-range btr support in vrr mode.
+	 */
 	amdgpu_dm_crtc_handle_crc_irq(&acrtc->base);
-	drm_crtc_handle_vblank(&acrtc->base);
+
+	/* BTR updates need to happen before VUPDATE on Vega and above. */
+	if (adev->family < AMDGPU_FAMILY_AI)
+		return;
 
 	spin_lock_irqsave(&adev->ddev->event_lock, flags);
 
-	if (acrtc_state->vrr_params.supported &&
+	if (acrtc_state->stream && acrtc_state->vrr_params.supported &&
 	    acrtc_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE) {
-		mod_freesync_handle_v_update(
-		adev->dm.freesync_module,
-		acrtc_state->stream,
-		&acrtc_state->vrr_params);
+		mod_freesync_handle_v_update(adev->dm.freesync_module,
+					     acrtc_state->stream,
+					     &acrtc_state->vrr_params);
 
-		dc_stream_adjust_vmin_vmax(
-			adev->dm.dc,
-			acrtc_state->stream,
-			&acrtc_state->vrr_params.adjust);
+		dc_stream_adjust_vmin_vmax(adev->dm.dc, acrtc_state->stream,
+					   &acrtc_state->vrr_params.adjust);
 	}
 
 	/*
@@ -556,7 +505,8 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params)
 	 * avoid race conditions between flip programming and completion,
 	 * which could cause too early flip completion events.
 	 */
-	if (acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED &&
+	if (adev->family >= AMDGPU_FAMILY_RV &&
+	    acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED &&
 	    acrtc_state->active_planes == 0) {
 		if (acrtc->event) {
 			drm_crtc_send_vblank_event(&acrtc->base, acrtc->event);
@@ -568,7 +518,6 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params)
 
 	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
 }
-#endif
 
 static int dm_set_clockgating_state(void *handle,
 		  enum amd_clockgating_state state)
@@ -2445,8 +2394,36 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 		c_irq_params->adev = adev;
 		c_irq_params->irq_src = int_params.irq_source;
 
+		amdgpu_dm_irq_register_interrupt(
+			adev, &int_params, dm_crtc_high_irq, c_irq_params);
+	}
+
+	/* Use VUPDATE_NO_LOCK interrupt on DCN, which seems to correspond to
+	 * the regular VUPDATE interrupt on DCE. We want DC_IRQ_SOURCE_VUPDATEx
+	 * to trigger at end of each vblank, regardless of state of the lock,
+	 * matching DCE behaviour.
+	 */
+	for (i = DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT;
+	     i <= DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT + adev->mode_info.num_crtc - 1;
+	     i++) {
+		r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DCE, i, &adev->vupdate_irq);
+
+		if (r) {
+			DRM_ERROR("Failed to add vupdate irq id!\n");
+			return r;
+		}
+
+		int_params.int_context = INTERRUPT_HIGH_IRQ_CONTEXT;
+		int_params.irq_source =
+			dc_interrupt_to_irq_source(dc, i, 0);
+
+		c_irq_params = &adev->dm.vupdate_params[int_params.irq_source - DC_IRQ_SOURCE_VUPDATE1];
+
+		c_irq_params->adev = adev;
+		c_irq_params->irq_src = int_params.irq_source;
+
 		amdgpu_dm_irq_register_interrupt(adev, &int_params,
-				dm_dcn_crtc_high_irq, c_irq_params);
+				dm_vupdate_high_irq, c_irq_params);
 	}
 
 	/* Use GRPH_PFLIP interrupt */
@@ -4453,10 +4430,6 @@ static inline int dm_set_vupdate_irq(struct drm_crtc *crtc, bool enable)
 	struct amdgpu_device *adev = crtc->dev->dev_private;
 	int rc;
 
-	/* Do not set vupdate for DCN hardware */
-	if (adev->family > AMDGPU_FAMILY_AI)
-		return 0;
-
 	irq_source = IRQ_TYPE_VUPDATE + acrtc->otg_inst;
 
 	rc = dc_interrupt_set(adev->dm.dc, irq_source, enable) ? 0 : -EBUSY;
-- 
cgit v1.2.3-59-g8ed1b


From 626bf90fe03fa080d8df06bb0397c95c53ae8e27 Mon Sep 17 00:00:00 2001
From: Simon Ser <contact@emersion.fr>
Date: Mon, 30 Mar 2020 09:23:21 +0000
Subject: drm/amd/display: add basic atomic check for cursor plane

This patch adds a basic cursor check when an atomic test-only commit is
performed. The position and size of the cursor plane is checked.

This should fix user-space relying on atomic checks to assign buffers to
planes.

Signed-off-by: Simon Ser <contact@emersion.fr>
Reported-by: Roman Gilg <subdiff@gmail.com>
References: https://github.com/emersion/libliftoff/issues/46
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Harry Wentland <hwentlan@amd.com>
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 26 +++++++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index c3df6ef9f101..28e651b173ab 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -7855,6 +7855,7 @@ static int dm_update_plane_state(struct dc *dc,
 	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
 	struct dm_crtc_state *dm_new_crtc_state, *dm_old_crtc_state;
 	struct dm_plane_state *dm_new_plane_state, *dm_old_plane_state;
+	struct amdgpu_crtc *new_acrtc;
 	bool needs_reset;
 	int ret = 0;
 
@@ -7864,9 +7865,30 @@ static int dm_update_plane_state(struct dc *dc,
 	dm_new_plane_state = to_dm_plane_state(new_plane_state);
 	dm_old_plane_state = to_dm_plane_state(old_plane_state);
 
-	/*TODO Implement atomic check for cursor plane */
-	if (plane->type == DRM_PLANE_TYPE_CURSOR)
+	/*TODO Implement better atomic check for cursor plane */
+	if (plane->type == DRM_PLANE_TYPE_CURSOR) {
+		if (!enable || !new_plane_crtc ||
+			drm_atomic_plane_disabling(plane->state, new_plane_state))
+			return 0;
+
+		new_acrtc = to_amdgpu_crtc(new_plane_crtc);
+
+		if ((new_plane_state->crtc_w > new_acrtc->max_cursor_width) ||
+			(new_plane_state->crtc_h > new_acrtc->max_cursor_height)) {
+			DRM_DEBUG_ATOMIC("Bad cursor size %d x %d\n",
+							 new_plane_state->crtc_w, new_plane_state->crtc_h);
+			return -EINVAL;
+		}
+
+		if (new_plane_state->crtc_x <= -new_acrtc->max_cursor_width ||
+			new_plane_state->crtc_y <= -new_acrtc->max_cursor_height) {
+			DRM_DEBUG_ATOMIC("Bad cursor position %d, %d\n",
+							 new_plane_state->crtc_x, new_plane_state->crtc_y);
+			return -EINVAL;
+		}
+
 		return 0;
+	}
 
 	needs_reset = should_reset_plane(state, plane, old_plane_state,
 					 new_plane_state);
-- 
cgit v1.2.3-59-g8ed1b


From f965b68188ab59a40a421ced1b05a2fea638465c Mon Sep 17 00:00:00 2001
From: Colin Xu <colin.xu@intel.com>
Date: Fri, 8 May 2020 14:05:06 +0800
Subject: drm/i915/gvt: Init DPLL/DDI vreg for virtual display instead of
 inheritance.

Init value of some display vregs rea inherited from host pregs. When
host display in different status, i.e. all monitors unpluged, different
display configurations, etc., GVT virtual display setup don't consistent
thus may lead to guest driver consider display goes malfunctional.

The added init vreg values are based on PRMs and fixed by calcuation
from current configuration (only PIPE_A) and the virtual EDID.

Fixes: 04d348ae3f0a ("drm/i915/gvt: vGPU display virtualization")
Acked-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Signed-off-by: Colin Xu <colin.xu@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20200508060506.216250-1-colin.xu@intel.com
---
 drivers/gpu/drm/i915/gvt/display.c | 49 ++++++++++++++++++++++++++++++++++----
 1 file changed, 44 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/gvt/display.c b/drivers/gpu/drm/i915/gvt/display.c
index a83df2f84eb9..a1696e9ce4b6 100644
--- a/drivers/gpu/drm/i915/gvt/display.c
+++ b/drivers/gpu/drm/i915/gvt/display.c
@@ -208,14 +208,41 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu)
 				SKL_FUSE_PG_DIST_STATUS(SKL_PG0) |
 				SKL_FUSE_PG_DIST_STATUS(SKL_PG1) |
 				SKL_FUSE_PG_DIST_STATUS(SKL_PG2);
-		vgpu_vreg_t(vgpu, LCPLL1_CTL) |=
-				LCPLL_PLL_ENABLE |
-				LCPLL_PLL_LOCK;
-		vgpu_vreg_t(vgpu, LCPLL2_CTL) |= LCPLL_PLL_ENABLE;
-
+		/*
+		 * Only 1 PIPE enabled in current vGPU display and PIPE_A is
+		 *  tied to TRANSCODER_A in HW, so it's safe to assume PIPE_A,
+		 *   TRANSCODER_A can be enabled. PORT_x depends on the input of
+		 *   setup_virtual_dp_monitor, we can bind DPLL0 to any PORT_x
+		 *   so we fixed to DPLL0 here.
+		 * Setup DPLL0: DP link clk 1620 MHz, non SSC, DP Mode
+		 */
+		vgpu_vreg_t(vgpu, DPLL_CTRL1) =
+			DPLL_CTRL1_OVERRIDE(DPLL_ID_SKL_DPLL0);
+		vgpu_vreg_t(vgpu, DPLL_CTRL1) |=
+			DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1620, DPLL_ID_SKL_DPLL0);
+		vgpu_vreg_t(vgpu, LCPLL1_CTL) =
+			LCPLL_PLL_ENABLE | LCPLL_PLL_LOCK;
+		vgpu_vreg_t(vgpu, DPLL_STATUS) = DPLL_LOCK(DPLL_ID_SKL_DPLL0);
+		/*
+		 * Golden M/N are calculated based on:
+		 *   24 bpp, 4 lanes, 154000 pixel clk (from virtual EDID),
+		 *   DP link clk 1620 MHz and non-constant_n.
+		 * TODO: calculate DP link symbol clk and stream clk m/n.
+		 */
+		vgpu_vreg_t(vgpu, PIPE_DATA_M1(TRANSCODER_A)) = 63 << TU_SIZE_SHIFT;
+		vgpu_vreg_t(vgpu, PIPE_DATA_M1(TRANSCODER_A)) |= 0x5b425e;
+		vgpu_vreg_t(vgpu, PIPE_DATA_N1(TRANSCODER_A)) = 0x800000;
+		vgpu_vreg_t(vgpu, PIPE_LINK_M1(TRANSCODER_A)) = 0x3cd6e;
+		vgpu_vreg_t(vgpu, PIPE_LINK_N1(TRANSCODER_A)) = 0x80000;
 	}
 
 	if (intel_vgpu_has_monitor_on_port(vgpu, PORT_B)) {
+		vgpu_vreg_t(vgpu, DPLL_CTRL2) &=
+			~DPLL_CTRL2_DDI_CLK_OFF(PORT_B);
+		vgpu_vreg_t(vgpu, DPLL_CTRL2) |=
+			DPLL_CTRL2_DDI_CLK_SEL(DPLL_ID_SKL_DPLL0, PORT_B);
+		vgpu_vreg_t(vgpu, DPLL_CTRL2) |=
+			DPLL_CTRL2_DDI_SEL_OVERRIDE(PORT_B);
 		vgpu_vreg_t(vgpu, SFUSE_STRAP) |= SFUSE_STRAP_DDIB_DETECTED;
 		vgpu_vreg_t(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) &=
 			~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK |
@@ -236,6 +263,12 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu)
 	}
 
 	if (intel_vgpu_has_monitor_on_port(vgpu, PORT_C)) {
+		vgpu_vreg_t(vgpu, DPLL_CTRL2) &=
+			~DPLL_CTRL2_DDI_CLK_OFF(PORT_C);
+		vgpu_vreg_t(vgpu, DPLL_CTRL2) |=
+			DPLL_CTRL2_DDI_CLK_SEL(DPLL_ID_SKL_DPLL0, PORT_C);
+		vgpu_vreg_t(vgpu, DPLL_CTRL2) |=
+			DPLL_CTRL2_DDI_SEL_OVERRIDE(PORT_C);
 		vgpu_vreg_t(vgpu, SDEISR) |= SDE_PORTC_HOTPLUG_CPT;
 		vgpu_vreg_t(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) &=
 			~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK |
@@ -256,6 +289,12 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu)
 	}
 
 	if (intel_vgpu_has_monitor_on_port(vgpu, PORT_D)) {
+		vgpu_vreg_t(vgpu, DPLL_CTRL2) &=
+			~DPLL_CTRL2_DDI_CLK_OFF(PORT_D);
+		vgpu_vreg_t(vgpu, DPLL_CTRL2) |=
+			DPLL_CTRL2_DDI_CLK_SEL(DPLL_ID_SKL_DPLL0, PORT_D);
+		vgpu_vreg_t(vgpu, DPLL_CTRL2) |=
+			DPLL_CTRL2_DDI_SEL_OVERRIDE(PORT_D);
 		vgpu_vreg_t(vgpu, SDEISR) |= SDE_PORTD_HOTPLUG_CPT;
 		vgpu_vreg_t(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) &=
 			~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK |
-- 
cgit v1.2.3-59-g8ed1b


From 72a7a9925e2beea09b109dffb3384c9bf920d9da Mon Sep 17 00:00:00 2001
From: Zhenyu Wang <zhenyuw@linux.intel.com>
Date: Wed, 6 May 2020 17:59:18 +0800
Subject: drm/i915/gvt: Fix kernel oops for 3-level ppgtt guest

As i915 won't allocate extra PDP for current default PML4 table,
so for 3-level ppgtt guest, we would hit kernel pointer access
failure on extra PDP pointers. So this trys to bypass that now.
It won't impact real shadow PPGTT setup, so guest context still
works.

This is verified on 4.15 guest kernel with i915.enable_ppgtt=1
to force on old aliasing ppgtt behavior.

Fixes: 4f15665ccbba ("drm/i915: Add ppgtt to GVT GEM context")
Reviewed-by: Xiong Zhang <xiong.y.zhang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20200506095918.124913-1-zhenyuw@linux.intel.com
---
 drivers/gpu/drm/i915/gvt/scheduler.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index cb11c3184085..0f1d6998cf9c 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -379,7 +379,11 @@ static void set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
 		for (i = 0; i < GVT_RING_CTX_NR_PDPS; i++) {
 			struct i915_page_directory * const pd =
 				i915_pd_entry(ppgtt->pd, i);
-
+			/* skip now as current i915 ppgtt alloc won't allocate
+			   top level pdp for non 4-level table, won't impact
+			   shadow ppgtt. */
+			if (!pd)
+				break;
 			px_dma(pd) = mm->ppgtt_mm.shadow_pdps[i];
 		}
 	}
-- 
cgit v1.2.3-59-g8ed1b


From bc850943486887e3859597a266767f95db90aa72 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Wed, 6 May 2020 17:21:36 +0100
Subject: drm/i915: Propagate error from completed fences

We need to preserve fatal errors from fences that are being terminated
as we hook them up.

Fixes: ef4688497512 ("drm/i915: Propagate fence errors")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200506162136.3325-1-chris@chris-wilson.co.uk
(cherry picked from commit 24fe5f2ab2478053d50a3bc629ada895903a5cbc)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/i915_request.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index c0df71d7d0ff..182a2995f674 100644
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -1017,8 +1017,10 @@ i915_request_await_request(struct i915_request *to, struct i915_request *from)
 	GEM_BUG_ON(to == from);
 	GEM_BUG_ON(to->timeline == from->timeline);
 
-	if (i915_request_completed(from))
+	if (i915_request_completed(from)) {
+		i915_sw_fence_set_error_once(&to->submit, from->fence.error);
 		return 0;
+	}
 
 	if (to->engine->schedule) {
 		ret = i915_sched_node_add_dependency(&to->sched, &from->sched);
-- 
cgit v1.2.3-59-g8ed1b


From a9d094dcf7845af85f82adcad9f793e51e4d14c8 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 7 May 2020 16:51:09 +0100
Subject: drm/i915: Mark concurrent submissions with a weak-dependency

We recorded the dependencies for WAIT_FOR_SUBMIT in order that we could
correctly perform priority inheritance from the parallel branches to the
common trunk. However, for the purpose of timeslicing and reset
handling, the dependency is weak -- as we the pair of requests are
allowed to run in parallel and not in strict succession.

The real significance though is that this allows us to rearrange
groups of WAIT_FOR_SUBMIT linked requests along the single engine, and
so can resolve user level inter-batch scheduling dependencies from user
semaphores.

Fixes: c81471f5e95c ("drm/i915: Copy across scheduler behaviour flags across submit fences")
Testcase: igt/gem_exec_fence/submit
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: <stable@vger.kernel.org> # v5.6+
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200507155109.8892-1-chris@chris-wilson.co.uk
(cherry picked from commit 6b6cd2ebd8d071e55998e32b648bb8081f7f02bb)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_lrc.c         | 3 +++
 drivers/gpu/drm/i915/i915_request.c         | 8 ++++++--
 drivers/gpu/drm/i915/i915_scheduler.c       | 6 +++---
 drivers/gpu/drm/i915/i915_scheduler.h       | 3 ++-
 drivers/gpu/drm/i915/i915_scheduler_types.h | 1 +
 5 files changed, 15 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index 77420372d813..2dfaddb8811e 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -1721,6 +1721,9 @@ static void defer_request(struct i915_request *rq, struct list_head * const pl)
 			struct i915_request *w =
 				container_of(p->waiter, typeof(*w), sched);
 
+			if (p->flags & I915_DEPENDENCY_WEAK)
+				continue;
+
 			/* Leave semaphores spinning on the other engines */
 			if (w->engine != rq->engine)
 				continue;
diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index 182a2995f674..e2b78db685ea 100644
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -1023,7 +1023,9 @@ i915_request_await_request(struct i915_request *to, struct i915_request *from)
 	}
 
 	if (to->engine->schedule) {
-		ret = i915_sched_node_add_dependency(&to->sched, &from->sched);
+		ret = i915_sched_node_add_dependency(&to->sched,
+						     &from->sched,
+						     I915_DEPENDENCY_EXTERNAL);
 		if (ret < 0)
 			return ret;
 	}
@@ -1185,7 +1187,9 @@ __i915_request_await_execution(struct i915_request *to,
 
 	/* Couple the dependency tree for PI on this exposed to->fence */
 	if (to->engine->schedule) {
-		err = i915_sched_node_add_dependency(&to->sched, &from->sched);
+		err = i915_sched_node_add_dependency(&to->sched,
+						     &from->sched,
+						     I915_DEPENDENCY_WEAK);
 		if (err < 0)
 			return err;
 	}
diff --git a/drivers/gpu/drm/i915/i915_scheduler.c b/drivers/gpu/drm/i915/i915_scheduler.c
index 68b06a7ba667..f0a9e8958ca0 100644
--- a/drivers/gpu/drm/i915/i915_scheduler.c
+++ b/drivers/gpu/drm/i915/i915_scheduler.c
@@ -456,7 +456,8 @@ bool __i915_sched_node_add_dependency(struct i915_sched_node *node,
 }
 
 int i915_sched_node_add_dependency(struct i915_sched_node *node,
-				   struct i915_sched_node *signal)
+				   struct i915_sched_node *signal,
+				   unsigned long flags)
 {
 	struct i915_dependency *dep;
 
@@ -465,8 +466,7 @@ int i915_sched_node_add_dependency(struct i915_sched_node *node,
 		return -ENOMEM;
 
 	if (!__i915_sched_node_add_dependency(node, signal, dep,
-					      I915_DEPENDENCY_EXTERNAL |
-					      I915_DEPENDENCY_ALLOC))
+					      flags | I915_DEPENDENCY_ALLOC))
 		i915_dependency_free(dep);
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/i915_scheduler.h b/drivers/gpu/drm/i915/i915_scheduler.h
index d1dc4efef77b..6f0bf00fc569 100644
--- a/drivers/gpu/drm/i915/i915_scheduler.h
+++ b/drivers/gpu/drm/i915/i915_scheduler.h
@@ -34,7 +34,8 @@ bool __i915_sched_node_add_dependency(struct i915_sched_node *node,
 				      unsigned long flags);
 
 int i915_sched_node_add_dependency(struct i915_sched_node *node,
-				   struct i915_sched_node *signal);
+				   struct i915_sched_node *signal,
+				   unsigned long flags);
 
 void i915_sched_node_fini(struct i915_sched_node *node);
 
diff --git a/drivers/gpu/drm/i915/i915_scheduler_types.h b/drivers/gpu/drm/i915/i915_scheduler_types.h
index d18e70550054..7186875088a0 100644
--- a/drivers/gpu/drm/i915/i915_scheduler_types.h
+++ b/drivers/gpu/drm/i915/i915_scheduler_types.h
@@ -78,6 +78,7 @@ struct i915_dependency {
 	unsigned long flags;
 #define I915_DEPENDENCY_ALLOC		BIT(0)
 #define I915_DEPENDENCY_EXTERNAL	BIT(1)
+#define I915_DEPENDENCY_WEAK		BIT(2)
 };
 
 #endif /* _I915_SCHEDULER_TYPES_H_ */
-- 
cgit v1.2.3-59-g8ed1b


From 975f543e7522e17b8a4bf34d7daeac44819aee5a Mon Sep 17 00:00:00 2001
From: Tom St Denis <tom.stdenis@amd.com>
Date: Thu, 7 May 2020 08:35:40 -0400
Subject: drm/amd/amdgpu: add raven1 part to the gfxoff quirk list

On my raven1 system (rev c6) with VBIOS 113-RAVEN-114 GFXOFF is
not stable (resulting in large block tiling noise in some applications).

Disabling GFXOFF via the quirk list fixes the problems for me.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 0127f1ce16e6..d2d9dce68c2f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -1236,6 +1236,8 @@ static const struct amdgpu_gfxoff_quirk amdgpu_gfxoff_quirk_list[] = {
 	{ 0x1002, 0x15dd, 0x1002, 0x15dd, 0xc8 },
 	/* https://bugzilla.kernel.org/show_bug.cgi?id=207171 */
 	{ 0x1002, 0x15dd, 0x103c, 0x83e7, 0xd3 },
+	/* GFXOFF is unstable on C6 parts with a VBIOS 113-RAVEN-114 */
+	{ 0x1002, 0x15dd, 0x1002, 0x15dd, 0xc6 },
 	{ 0, 0, 0, 0, 0 },
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 650e723cecf2738dee828564396f3239829aba83 Mon Sep 17 00:00:00 2001
From: "Leo (Hanghong) Ma" <hanghong.ma@amd.com>
Date: Fri, 8 May 2020 14:18:07 -0400
Subject: drm/amd/amdgpu: Update update_config() logic

[Why]
For MST case: when update_config is called to disable a stream,
this clears the settings for all the streams on that link.
We should only clear the settings for the stream that was disabled.

[How]
Clear the settings after the call to remove display is called.

Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Leo (Hanghong) Ma <hanghong.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
index 78e1c11d4ae5..dcf84a61de37 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
@@ -398,15 +398,15 @@ static void update_config(void *handle, struct cp_psp_stream_config *config)
 	struct mod_hdcp_display *display = &hdcp_work[link_index].display;
 	struct mod_hdcp_link *link = &hdcp_work[link_index].link;
 
-	memset(display, 0, sizeof(*display));
-	memset(link, 0, sizeof(*link));
-
-	display->index = aconnector->base.index;
-
 	if (config->dpms_off) {
 		hdcp_remove_display(hdcp_work, link_index, aconnector);
 		return;
 	}
+
+	memset(display, 0, sizeof(*display));
+	memset(link, 0, sizeof(*link));
+
+	display->index = aconnector->base.index;
 	display->state = MOD_HDCP_DISPLAY_ACTIVE;
 
 	if (aconnector->dc_sink != NULL)
-- 
cgit v1.2.3-59-g8ed1b


From c54a8f1f329197d83d941ad84c4aa38bf282cbbd Mon Sep 17 00:00:00 2001
From: Bernard Zhao <bernard@vivo.com>
Date: Tue, 28 Apr 2020 06:17:47 -0700
Subject: drm/meson: pm resume add return errno branch

pm_resump api did not handle drm_mode_config_helper_resume error.
This change add handle to return drm_mode_config_helper_resume`s
error number. This code logic is aligned with api pm_suspend.
After this change, the code maybe a bit readable.

Signed-off-by: Bernard Zhao <bernard@vivo.com>
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200428131747.2099-1-bernard@vivo.com
---
 drivers/gpu/drm/meson/meson_drv.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c
index b5f5eb7b4bb9..8c2e1b47e81a 100644
--- a/drivers/gpu/drm/meson/meson_drv.c
+++ b/drivers/gpu/drm/meson/meson_drv.c
@@ -412,9 +412,7 @@ static int __maybe_unused meson_drv_pm_resume(struct device *dev)
 	if (priv->afbcd.ops)
 		priv->afbcd.ops->init(priv);
 
-	drm_mode_config_helper_resume(priv->drm);
-
-	return 0;
+	return drm_mode_config_helper_resume(priv->drm);
 }
 
 static int compare_of(struct device *dev, void *data)
-- 
cgit v1.2.3-59-g8ed1b


From 955da9d77435acac066139e9d7f7723ce7204a1d Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Sat, 9 May 2020 12:52:17 +0100
Subject: drm/i915: Handle idling during i915_gem_evict_something busy loops

i915_gem_evict_something() is charged with finding a slot within the GTT
that we may reuse. Since our goal is not to stall, we first look for a
slot that only overlaps idle vma. To this end, on the first pass we move
any active vma to the end of the search list. However, we only stopped
moving active vma after we see the first active vma twice. If during the
search, that first active vma completed, we would not notice and keep on
extending the search list.

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1746
Fixes: 2850748ef876 ("drm/i915: Pull i915_vma_pin under the vm->mutex")
Fixes: b1e3177bd1d8 ("drm/i915: Coordinate i915_active with its own mutex")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: <stable@vger.kernel.org> # v5.5+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200509115217.26853-1-chris@chris-wilson.co.uk
(cherry picked from commit 73e28cc40bf00b5d168cb8f5cff1ae63e9097446)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/i915_gem_evict.c | 26 ++++++++++++--------------
 1 file changed, 12 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c
index 4518b9b35c3d..02ad1acd117c 100644
--- a/drivers/gpu/drm/i915/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/i915_gem_evict.c
@@ -128,6 +128,13 @@ search_again:
 	active = NULL;
 	INIT_LIST_HEAD(&eviction_list);
 	list_for_each_entry_safe(vma, next, &vm->bound_list, vm_link) {
+		if (vma == active) { /* now seen this vma twice */
+			if (flags & PIN_NONBLOCK)
+				break;
+
+			active = ERR_PTR(-EAGAIN);
+		}
+
 		/*
 		 * We keep this list in a rough least-recently scanned order
 		 * of active elements (inactive elements are cheap to reap).
@@ -143,21 +150,12 @@ search_again:
 		 * To notice when we complete one full cycle, we record the
 		 * first active element seen, before moving it to the tail.
 		 */
-		if (i915_vma_is_active(vma)) {
-			if (vma == active) {
-				if (flags & PIN_NONBLOCK)
-					break;
-
-				active = ERR_PTR(-EAGAIN);
-			}
-
-			if (active != ERR_PTR(-EAGAIN)) {
-				if (!active)
-					active = vma;
+		if (active != ERR_PTR(-EAGAIN) && i915_vma_is_active(vma)) {
+			if (!active)
+				active = vma;
 
-				list_move_tail(&vma->vm_link, &vm->bound_list);
-				continue;
-			}
+			list_move_tail(&vma->vm_link, &vm->bound_list);
+			continue;
 		}
 
 		if (mark_free(&scan, vma, flags, &eviction_list))
-- 
cgit v1.2.3-59-g8ed1b