Merge v5.0-rc7 into drm-next

Backmerging for nouveau and imx that needed some fixes for next pulls.

Signed-off-by: Dave Airlie <airlied@redhat.com>

7 files changed, 111 insertions, 38 deletions

diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index 9c106e47e640..bc64b810e0d5 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -2799,6 +2799,7 @@ static int init_broadwell_mmio_info(struct intel_gvt *gvt)
 	MMIO_DFH(_MMIO(0xe2a0), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(_MMIO(0xe2b0), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(_MMIO(0xe2c0), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(_MMIO(0x21f0), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/i915/gvt/hypercall.h b/drivers/gpu/drm/i915/gvt/hypercall.h
index 5e01cc8d9b16..4862fb12778e 100644
--- a/drivers/gpu/drm/i915/gvt/hypercall.h
+++ b/drivers/gpu/drm/i915/gvt/hypercall.h
@@ -47,7 +47,7 @@ struct intel_gvt_mpt {
 	int (*host_init)(struct device *dev, void *gvt, const void *ops);
 	void (*host_exit)(struct device *dev);
 	int (*attach_vgpu)(void *vgpu, unsigned long *handle);
-	void (*detach_vgpu)(unsigned long handle);
+	void (*detach_vgpu)(void *vgpu);
 	int (*inject_msi)(unsigned long handle, u32 addr, u16 data);
 	unsigned long (*from_virt_to_mfn)(void *p);
 	int (*enable_page_track)(unsigned long handle, u64 gfn);
diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c
index 63eef86a2a85..d5fcc447d22f 100644
--- a/drivers/gpu/drm/i915/gvt/kvmgt.c
+++ b/drivers/gpu/drm/i915/gvt/kvmgt.c
@@ -1147,7 +1147,7 @@ static int intel_vgpu_mmap(struct mdev_device *mdev, struct vm_area_struct *vma)
 {
 	unsigned int index;
 	u64 virtaddr;
-	unsigned long req_size, pgoff = 0;
+	unsigned long req_size, pgoff, req_start;
 	pgprot_t pg_prot;
 	struct intel_vgpu *vgpu = mdev_get_drvdata(mdev);
 
@@ -1165,7 +1165,17 @@ static int intel_vgpu_mmap(struct mdev_device *mdev, struct vm_area_struct *vma)
 	pg_prot = vma->vm_page_prot;
 	virtaddr = vma->vm_start;
 	req_size = vma->vm_end - vma->vm_start;
-	pgoff = vgpu_aperture_pa_base(vgpu) >> PAGE_SHIFT;
+	pgoff = vma->vm_pgoff &
+		((1U << (VFIO_PCI_OFFSET_SHIFT - PAGE_SHIFT)) - 1);
+	req_start = pgoff << PAGE_SHIFT;
+
+	if (!intel_vgpu_in_aperture(vgpu, req_start))
+		return -EINVAL;
+	if (req_start + req_size >
+	    vgpu_aperture_offset(vgpu) + vgpu_aperture_sz(vgpu))
+		return -EINVAL;
+
+	pgoff = (gvt_aperture_pa_base(vgpu->gvt) >> PAGE_SHIFT) + pgoff;
 
 	return remap_pfn_range(vma, virtaddr, pgoff, req_size, pg_prot);
 }
@@ -1813,9 +1823,21 @@ static int kvmgt_attach_vgpu(void *vgpu, unsigned long *handle)
 	return 0;
 }
 
-static void kvmgt_detach_vgpu(unsigned long handle)
+static void kvmgt_detach_vgpu(void *p_vgpu)
 {
-	/* nothing to do here */
+	int i;
+	struct intel_vgpu *vgpu = (struct intel_vgpu *)p_vgpu;
+
+	if (!vgpu->vdev.region)
+		return;
+
+	for (i = 0; i < vgpu->vdev.num_regions; i++)
+		if (vgpu->vdev.region[i].ops->release)
+			vgpu->vdev.region[i].ops->release(vgpu,
+					&vgpu->vdev.region[i]);
+	vgpu->vdev.num_regions = 0;
+	kfree(vgpu->vdev.region);
+	vgpu->vdev.region = NULL;
 }
 
 static int kvmgt_inject_msi(unsigned long handle, u32 addr, u16 data)
diff --git a/drivers/gpu/drm/i915/gvt/mpt.h b/drivers/gpu/drm/i915/gvt/mpt.h
index 5d8b8f228d8f..0f9440128123 100644
--- a/drivers/gpu/drm/i915/gvt/mpt.h
+++ b/drivers/gpu/drm/i915/gvt/mpt.h
@@ -99,7 +99,7 @@ static inline void intel_gvt_hypervisor_detach_vgpu(struct intel_vgpu *vgpu)
 	if (!intel_gvt_host.mpt->detach_vgpu)
 		return;
 
-	intel_gvt_host.mpt->detach_vgpu(vgpu->handle);
+	intel_gvt_host.mpt->detach_vgpu(vgpu);
 }
 
 #define MSI_CAP_CONTROL(offset) (offset + 2)
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index b7957eefb976..1bb8f936fdaa 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -333,6 +333,9 @@ static void release_shadow_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 
 	i915_gem_object_unpin_map(wa_ctx->indirect_ctx.obj);
 	i915_gem_object_put(wa_ctx->indirect_ctx.obj);
+
+	wa_ctx->indirect_ctx.obj = NULL;
+	wa_ctx->indirect_ctx.shadow_va = NULL;
 }
 
 static int set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
@@ -357,6 +360,33 @@ static int set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
 	return 0;
 }
 
+static int
+intel_gvt_workload_req_alloc(struct intel_vgpu_workload *workload)
+{
+	struct intel_vgpu *vgpu = workload->vgpu;
+	struct intel_vgpu_submission *s = &vgpu->submission;
+	struct i915_gem_context *shadow_ctx = s->shadow_ctx;
+	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
+	struct intel_engine_cs *engine = dev_priv->engine[workload->ring_id];
+	struct i915_request *rq;
+	int ret = 0;
+
+	lockdep_assert_held(&dev_priv->drm.struct_mutex);
+
+	if (workload->req)
+		goto out;
+
+	rq = i915_request_alloc(engine, shadow_ctx);
+	if (IS_ERR(rq)) {
+		gvt_vgpu_err("fail to allocate gem request\n");
+		ret = PTR_ERR(rq);
+		goto out;
+	}
+	workload->req = i915_request_get(rq);
+out:
+	return ret;
+}
+
 /**
  * intel_gvt_scan_and_shadow_workload - audit the workload by scanning and
  * shadow it as well, include ringbuffer,wa_ctx and ctx.
@@ -373,12 +403,11 @@ int intel_gvt_scan_and_shadow_workload(struct intel_vgpu_workload *workload)
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
 	struct intel_engine_cs *engine = dev_priv->engine[workload->ring_id];
 	struct intel_context *ce;
-	struct i915_request *rq;
 	int ret;
 
 	lockdep_assert_held(&dev_priv->drm.struct_mutex);
 
-	if (workload->req)
+	if (workload->shadow)
 		return 0;
 
 	ret = set_context_ppgtt_from_shadow(workload, shadow_ctx);
@@ -418,22 +447,8 @@ int intel_gvt_scan_and_shadow_workload(struct intel_vgpu_workload *workload)
 			goto err_shadow;
 	}
 
-	rq = i915_request_alloc(engine, shadow_ctx);
-	if (IS_ERR(rq)) {
-		gvt_vgpu_err("fail to allocate gem request\n");
-		ret = PTR_ERR(rq);
-		goto err_shadow;
-	}
-	workload->req = i915_request_get(rq);
-
-	ret = populate_shadow_context(workload);
-	if (ret)
-		goto err_req;
-
+	workload->shadow = true;
 	return 0;
-err_req:
-	rq = fetch_and_zero(&workload->req);
-	i915_request_put(rq);
 err_shadow:
 	release_shadow_wa_ctx(&workload->wa_ctx);
 err_unpin:
@@ -672,23 +687,31 @@ static int dispatch_workload(struct intel_vgpu_workload *workload)
 	mutex_lock(&vgpu->vgpu_lock);
 	mutex_lock(&dev_priv->drm.struct_mutex);
 
+	ret = intel_gvt_workload_req_alloc(workload);
+	if (ret)
+		goto err_req;
+
 	ret = intel_gvt_scan_and_shadow_workload(workload);
 	if (ret)
 		goto out;
 
-	ret = prepare_workload(workload);
+	ret = populate_shadow_context(workload);
+	if (ret) {
+		release_shadow_wa_ctx(&workload->wa_ctx);
+		goto out;
+	}
 
+	ret = prepare_workload(workload);
 out:
-	if (ret)
-		workload->status = ret;
-
 	if (!IS_ERR_OR_NULL(workload->req)) {
 		gvt_dbg_sched("ring id %d submit workload to i915 %p\n",
 				ring_id, workload->req);
 		i915_request_add(workload->req);
 		workload->dispatched = true;
 	}
-
+err_req:
+	if (ret)
+		workload->status = ret;
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 	mutex_unlock(&vgpu->vgpu_lock);
 	return ret;
@@ -892,11 +915,6 @@ static void complete_current_workload(struct intel_gvt *gvt, int ring_id)
 
 	list_del_init(&workload->list);
 
-	if (!workload->status) {
-		release_shadow_batch_buffer(workload);
-		release_shadow_wa_ctx(&workload->wa_ctx);
-	}
-
 	if (workload->status || (vgpu->resetting_eng & ENGINE_MASK(ring_id))) {
 		/* if workload->status is not successful means HW GPU
 		 * has occurred GPU hang or something wrong with i915/GVT,
@@ -1262,6 +1280,9 @@ void intel_vgpu_destroy_workload(struct intel_vgpu_workload *workload)
 {
 	struct intel_vgpu_submission *s = &workload->vgpu->submission;
 
+	release_shadow_batch_buffer(workload);
+	release_shadow_wa_ctx(&workload->wa_ctx);
+
 	if (workload->shadow_mm)
 		intel_vgpu_mm_put(workload->shadow_mm);
 
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.h b/drivers/gpu/drm/i915/gvt/scheduler.h
index 1e9eec6a32fe..0635b2c4bed7 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.h
+++ b/drivers/gpu/drm/i915/gvt/scheduler.h
@@ -83,6 +83,7 @@ struct intel_vgpu_workload {
 	struct i915_request *req;
 	/* if this workload has been dispatched to i915? */
 	bool dispatched;
+	bool shadow;      /* if workload has done shadow of guest request */
 	int status;
 
 	struct intel_vgpu_mm *shadow_mm;
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index 30ae96c5c97c..5e00ee9270b5 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -54,7 +54,12 @@
 struct opregion_header {
 	u8 signature[16];
 	u32 size;
-	u32 opregion_ver;
+	struct {
+		u8 rsvd;
+		u8 revision;
+		u8 minor;
+		u8 major;
+	}  __packed over;
 	u8 bios_ver[32];
 	u8 vbios_ver[16];
 	u8 driver_ver[16];
@@ -118,7 +123,8 @@ struct opregion_asle {
 	u64 fdss;
 	u32 fdsp;
 	u32 stat;
-	u64 rvda;	/* Physical address of raw vbt data */
+	u64 rvda;	/* Physical (2.0) or relative from opregion (2.1+)
+			 * address of raw VBT data. */
 	u32 rvds;	/* Size of raw vbt data */
 	u8 rsvd[58];
 } __packed;
@@ -924,6 +930,11 @@ int intel_opregion_setup(struct drm_i915_private *dev_priv)
 	opregion->header = base;
 	opregion->lid_state = base + ACPI_CLID;
 
+	DRM_DEBUG_DRIVER("ACPI OpRegion version %u.%u.%u\n",
+			 opregion->header->over.major,
+			 opregion->header->over.minor,
+			 opregion->header->over.revision);
+
 	mboxes = opregion->header->mboxes;
 	if (mboxes & MBOX_ACPI) {
 		DRM_DEBUG_DRIVER("Public ACPI methods supported\n");
@@ -952,11 +963,26 @@ int intel_opregion_setup(struct drm_i915_private *dev_priv)
 	if (dmi_check_system(intel_no_opregion_vbt))
 		goto out;
 
-	if (opregion->header->opregion_ver >= 2 && opregion->asle &&
+	if (opregion->header->over.major >= 2 && opregion->asle &&
 	    opregion->asle->rvda && opregion->asle->rvds) {
-		opregion->rvda = memremap(opregion->asle->rvda,
-					  opregion->asle->rvds,
+		resource_size_t rvda = opregion->asle->rvda;
+
+		/*
+		 * opregion 2.0: rvda is the physical VBT address.
+		 *
+		 * opregion 2.1+: rvda is unsigned, relative offset from
+		 * opregion base, and should never point within opregion.
+		 */
+		if (opregion->header->over.major > 2 ||
+		    opregion->header->over.minor >= 1) {
+			WARN_ON(rvda < OPREGION_SIZE);
+
+			rvda += asls;
+		}
+
+		opregion->rvda = memremap(rvda, opregion->asle->rvds,
 					  MEMREMAP_WB);
+
 		vbt = opregion->rvda;
 		vbt_size = opregion->asle->rvds;
 		if (intel_bios_is_valid_vbt(vbt, vbt_size)) {
@@ -966,6 +992,8 @@ int intel_opregion_setup(struct drm_i915_private *dev_priv)
 			goto out;
 		} else {
 			DRM_DEBUG_KMS("Invalid VBT in ACPI OpRegion (RVDA)\n");
+			memunmap(opregion->rvda);
+			opregion->rvda = NULL;
 		}
 	}