drm/msm: Count how many times iova memory is pinned

Add a reference count to track how many times a particular
chunk of iova memory is pinned (mapped) in the iomu and
add msm_gem_unpin_iova to give up references.

It is important to note that msm_gem_unpin_iova replaces
msm_gem_put_iova because the new implicit behavior
that an assigned iova in a given vma is now valid for the
life of the buffer and what we are really focusing on is
the use of that iova.

For now the unmappings are lazy; once the reference counts
go to zero they *COULD* be unmapped dynamically but that
will require an outside force such as a shrinker or
mm_notifiers.  For now, we're just focusing on getting
the counting right and setting ourselves up to be ready
for the future.
Signed-off-by: Jordan Crouse <jcrouse@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

drivers/gpu/drm/msm/adreno/a5xx_debugfs.c

@@ -130,15 +130,13 @@ reset_set(void *data, u64 val)
 	adreno_gpu->fw[ADRENO_FW_PFP] = NULL;
 
 	if (a5xx_gpu->pm4_bo) {
-		if (a5xx_gpu->pm4_iova)
-			msm_gem_put_iova(a5xx_gpu->pm4_bo, gpu->aspace);
+		msm_gem_unpin_iova(a5xx_gpu->pm4_bo, gpu->aspace);
 		drm_gem_object_put(a5xx_gpu->pm4_bo);
 		a5xx_gpu->pm4_bo = NULL;
 	}
 
 	if (a5xx_gpu->pfp_bo) {
-		if (a5xx_gpu->pfp_iova)
-			msm_gem_put_iova(a5xx_gpu->pfp_bo, gpu->aspace);
+		msm_gem_unpin_iova(a5xx_gpu->pfp_bo, gpu->aspace);
 		drm_gem_object_put(a5xx_gpu->pfp_bo);
 		a5xx_gpu->pfp_bo = NULL;
 	}

drivers/gpu/drm/msm/adreno/a5xx_gpu.c

@@ -840,20 +840,17 @@ static void a5xx_destroy(struct msm_gpu *gpu)
 	a5xx_preempt_fini(gpu);
 
 	if (a5xx_gpu->pm4_bo) {
-		if (a5xx_gpu->pm4_iova)
-			msm_gem_put_iova(a5xx_gpu->pm4_bo, gpu->aspace);
+		msm_gem_unpin_iova(a5xx_gpu->pm4_bo, gpu->aspace);
 		drm_gem_object_put_unlocked(a5xx_gpu->pm4_bo);
 	}
 
 	if (a5xx_gpu->pfp_bo) {
-		if (a5xx_gpu->pfp_iova)
-			msm_gem_put_iova(a5xx_gpu->pfp_bo, gpu->aspace);
+		msm_gem_unpin_iova(a5xx_gpu->pfp_bo, gpu->aspace);
 		drm_gem_object_put_unlocked(a5xx_gpu->pfp_bo);
 	}
 
 	if (a5xx_gpu->gpmu_bo) {
-		if (a5xx_gpu->gpmu_iova)
-			msm_gem_put_iova(a5xx_gpu->gpmu_bo, gpu->aspace);
+		msm_gem_unpin_iova(a5xx_gpu->gpmu_bo, gpu->aspace);
 		drm_gem_object_put_unlocked(a5xx_gpu->gpmu_bo);
 	}
 

drivers/gpu/drm/msm/adreno/a6xx_gpu.c

@@ -736,8 +736,7 @@ static void a6xx_destroy(struct msm_gpu *gpu)
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
 
 	if (a6xx_gpu->sqe_bo) {
-		if (a6xx_gpu->sqe_iova)
-			msm_gem_put_iova(a6xx_gpu->sqe_bo, gpu->aspace);
+		msm_gem_unpin_iova(a6xx_gpu->sqe_bo, gpu->aspace);
 		drm_gem_object_put_unlocked(a6xx_gpu->sqe_bo);
 	}
 

drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c

@@ -152,7 +152,7 @@ static int a6xx_crashdumper_run(struct msm_gpu *gpu,
 static void a6xx_crashdumper_free(struct msm_gpu *gpu,
 		struct a6xx_crashdumper *dumper)
 {
-	msm_gem_put_iova(dumper->bo, gpu->aspace);
+	msm_gem_unpin_iova(dumper->bo, gpu->aspace);
 	msm_gem_put_vaddr(dumper->bo);
 
 	drm_gem_object_unreference(dumper->bo);

drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c

@@ -128,7 +128,7 @@ static void unref_cursor_worker(struct drm_flip_work *work, void *val)
 	struct mdp4_kms *mdp4_kms = get_kms(&mdp4_crtc->base);
 	struct msm_kms *kms = &mdp4_kms->base.base;
 
-	msm_gem_put_iova(val, kms->aspace);
+	msm_gem_unpin_iova(val, kms->aspace);
 	drm_gem_object_put_unlocked(val);
 }
 

drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c

@@ -165,7 +165,7 @@ static void mdp4_destroy(struct msm_kms *kms)
 	struct msm_gem_address_space *aspace = kms->aspace;
 
 	if (mdp4_kms->blank_cursor_iova)
-		msm_gem_put_iova(mdp4_kms->blank_cursor_bo, kms->aspace);
+		msm_gem_unpin_iova(mdp4_kms->blank_cursor_bo, kms->aspace);
 	drm_gem_object_put_unlocked(mdp4_kms->blank_cursor_bo);
 
 	if (aspace) {

drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c

@@ -173,7 +173,7 @@ static void unref_cursor_worker(struct drm_flip_work *work, void *val)
 	struct mdp5_kms *mdp5_kms = get_kms(&mdp5_crtc->base);
 	struct msm_kms *kms = &mdp5_kms->base.base;
 
-	msm_gem_put_iova(val, kms->aspace);
+	msm_gem_unpin_iova(val, kms->aspace);
 	drm_gem_object_put_unlocked(val);
 }
 

drivers/gpu/drm/msm/dsi/dsi_host.c

@@ -1118,7 +1118,7 @@ static void dsi_tx_buf_free(struct msm_dsi_host *msm_host)
 
 	priv = dev->dev_private;
 	if (msm_host->tx_gem_obj) {
-		msm_gem_put_iova(msm_host->tx_gem_obj, priv->kms->aspace);
+		msm_gem_unpin_iova(msm_host->tx_gem_obj, priv->kms->aspace);
 		drm_gem_object_put_unlocked(msm_host->tx_gem_obj);
 		msm_host->tx_gem_obj = NULL;
 	}

drivers/gpu/drm/msm/msm_drv.h

@@ -243,10 +243,14 @@ void msm_atomic_state_free(struct drm_atomic_state *state);
 
 int msm_gem_init_vma(struct msm_gem_address_space *aspace,
 		struct msm_gem_vma *vma, int npages);
+void msm_gem_purge_vma(struct msm_gem_address_space *aspace,
+		struct msm_gem_vma *vma);
 void msm_gem_unmap_vma(struct msm_gem_address_space *aspace,
 		struct msm_gem_vma *vma);
 int msm_gem_map_vma(struct msm_gem_address_space *aspace,
 		struct msm_gem_vma *vma, struct sg_table *sgt, int npages);
+void msm_gem_close_vma(struct msm_gem_address_space *aspace,
+		struct msm_gem_vma *vma);
 
 void msm_gem_address_space_put(struct msm_gem_address_space *aspace);
 
@@ -275,10 +279,10 @@ int msm_gem_get_and_pin_iova(struct drm_gem_object *obj,
 		struct msm_gem_address_space *aspace, uint64_t *iova);
 uint64_t msm_gem_iova(struct drm_gem_object *obj,
 		struct msm_gem_address_space *aspace);
+void msm_gem_unpin_iova(struct drm_gem_object *obj,
+		struct msm_gem_address_space *aspace);
 struct page **msm_gem_get_pages(struct drm_gem_object *obj);
 void msm_gem_put_pages(struct drm_gem_object *obj);
-void msm_gem_put_iova(struct drm_gem_object *obj,
-		struct msm_gem_address_space *aspace);
 int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev,
 		struct drm_mode_create_dumb *args);
 int msm_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev,

drivers/gpu/drm/msm/msm_fb.c

@@ -81,7 +81,7 @@ void msm_framebuffer_cleanup(struct drm_framebuffer *fb,
 	int i, n = fb->format->num_planes;
 
 	for (i = 0; i < n; i++)
-		msm_gem_put_iova(fb->obj[i], aspace);
+		msm_gem_unpin_iova(fb->obj[i], aspace);
 }
 
 uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb,

drivers/gpu/drm/msm/msm_gem.c

@@ -352,7 +352,8 @@ put_iova(struct drm_gem_object *obj)
 	WARN_ON(!mutex_is_locked(&msm_obj->lock));
 
 	list_for_each_entry_safe(vma, tmp, &msm_obj->vmas, list) {
-		msm_gem_unmap_vma(vma->aspace, vma);
+		msm_gem_purge_vma(vma->aspace, vma);
+		msm_gem_close_vma(vma->aspace, vma);
 		del_vma(vma);
 	}
 }
@@ -430,7 +431,10 @@ int msm_gem_get_and_pin_iova(struct drm_gem_object *obj,
 	return ret;
 }
 
-/* Get an iova but don't pin the memory behind it */
+/*
+ * Get an iova but don't pin it. Doesn't need a put because iovas are currently
+ * valid for the life of the object
+ */
 int msm_gem_get_iova(struct drm_gem_object *obj,
 		struct msm_gem_address_space *aspace, uint64_t *iova)
 {
@@ -444,7 +448,6 @@ int msm_gem_get_iova(struct drm_gem_object *obj,
 	return ret;
 }
 
-
 /* get iova without taking a reference, used in places where you have
  * already done a 'msm_gem_get_and_pin_iova' or 'msm_gem_get_iova'
  */
@@ -462,15 +465,24 @@ uint64_t msm_gem_iova(struct drm_gem_object *obj,
 	return vma ? vma->iova : 0;
 }
 
-void msm_gem_put_iova(struct drm_gem_object *obj,
+/*
+ * Unpin a iova by updating the reference counts. The memory isn't actually
+ * purged until something else (shrinker, mm_notifier, destroy, etc) decides
+ * to get rid of it
+ */
+void msm_gem_unpin_iova(struct drm_gem_object *obj,
 		struct msm_gem_address_space *aspace)
 {
-	// XXX TODO ..
-	// NOTE: probably don't need a _locked() version.. we wouldn't
-	// normally unmap here, but instead just mark that it could be
-	// unmapped (if the iova refcnt drops to zero), but then later
-	// if another _get_iova_locked() fails we can start unmapping
-	// things that are no longer needed..
+	struct msm_gem_object *msm_obj = to_msm_bo(obj);
+	struct msm_gem_vma *vma;
+
+	mutex_lock(&msm_obj->lock);
+	vma = lookup_vma(obj, aspace);
+
+	if (!WARN_ON(!vma))
+		msm_gem_unmap_vma(aspace, vma);
+
+	mutex_unlock(&msm_obj->lock);
 }
 
 int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev,
@@ -789,8 +801,9 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m)
 		seq_puts(m, "      vmas:");
 
 		list_for_each_entry(vma, &msm_obj->vmas, list)
-			seq_printf(m, " [%s: %08llx,%s]", vma->aspace->name,
-				vma->iova, vma->mapped ? "mapped" : "unmapped");
+			seq_printf(m, " [%s: %08llx,%s,inuse=%d]", vma->aspace->name,
+				vma->iova, vma->mapped ? "mapped" : "unmapped",
+				vma->inuse);
 
 		seq_puts(m, "\n");
 	}
@@ -1093,7 +1106,7 @@ static void *_msm_gem_kernel_new(struct drm_device *dev, uint32_t size,
 
 	vaddr = msm_gem_get_vaddr(obj);
 	if (IS_ERR(vaddr)) {
-		msm_gem_put_iova(obj, aspace);
+		msm_gem_unpin_iova(obj, aspace);
 		ret = PTR_ERR(vaddr);
 		goto err;
 	}
@@ -1133,7 +1146,7 @@ void msm_gem_kernel_put(struct drm_gem_object *bo,
 		return;
 
 	msm_gem_put_vaddr(bo);
-	msm_gem_put_iova(bo, aspace);
+	msm_gem_unpin_iova(bo, aspace);
 
 	if (locked)
 		drm_gem_object_put(bo);

drivers/gpu/drm/msm/msm_gem.h

@@ -42,6 +42,7 @@ struct msm_gem_vma {
 	struct msm_gem_address_space *aspace;
 	struct list_head list;    /* node in msm_gem_object::vmas */
 	bool mapped;
+	int inuse;
 };
 
 struct msm_gem_object {

drivers/gpu/drm/msm/msm_gem_submit.c

@@ -167,7 +167,7 @@ static void submit_unlock_unpin_bo(struct msm_gem_submit *submit,
 	struct msm_gem_object *msm_obj = submit->bos[i].obj;
 
 	if (submit->bos[i].flags & BO_PINNED)
-		msm_gem_put_iova(&msm_obj->base, submit->gpu->aspace);
+		msm_gem_unpin_iova(&msm_obj->base, submit->gpu->aspace);
 
 	if (submit->bos[i].flags & BO_LOCKED)
 		ww_mutex_unlock(&msm_obj->resv->lock);

drivers/gpu/drm/msm/msm_gem_vma.c

@@ -38,26 +38,32 @@ void msm_gem_address_space_put(struct msm_gem_address_space *aspace)
 		kref_put(&aspace->kref, msm_gem_address_space_destroy);
 }
 
-void
-msm_gem_unmap_vma(struct msm_gem_address_space *aspace,
+/* Actually unmap memory for the vma */
+void msm_gem_purge_vma(struct msm_gem_address_space *aspace,
 		struct msm_gem_vma *vma)
 {
-	if (!aspace || !vma->iova)
+	unsigned size = vma->node.size << PAGE_SHIFT;
+
+	/* Print a message if we try to purge a vma in use */
+	if (WARN_ON(vma->inuse > 0))
 		return;
 
-	if (aspace->mmu) {
-		unsigned size = vma->node.size << PAGE_SHIFT;
-		aspace->mmu->funcs->unmap(aspace->mmu, vma->iova, size);
-	}
+	/* Don't do anything if the memory isn't mapped */
+	if (!vma->mapped)
+		return;
 
-	spin_lock(&aspace->lock);
-	drm_mm_remove_node(&vma->node);
-	spin_unlock(&aspace->lock);
+	if (aspace->mmu)
+		aspace->mmu->funcs->unmap(aspace->mmu, vma->iova, size);
 
-	vma->iova = 0;
 	vma->mapped = false;
+}
 
-	msm_gem_address_space_put(aspace);
+/* Remove reference counts for the mapping */
+void msm_gem_unmap_vma(struct msm_gem_address_space *aspace,
+		struct msm_gem_vma *vma)
+{
+	if (!WARN_ON(!vma->iova))
+		vma->inuse--;
 }
 
 int
@@ -70,6 +76,9 @@ msm_gem_map_vma(struct msm_gem_address_space *aspace,
 	if (WARN_ON(!vma->iova))
 		return -EINVAL;
 
+	/* Increase the usage counter */
+	vma->inuse++;
+
 	if (vma->mapped)
 		return 0;
 
@@ -85,6 +94,23 @@ msm_gem_map_vma(struct msm_gem_address_space *aspace,
 	return ret;
 }
 
+/* Close an iova.  Warn if it is still in use */
+void msm_gem_close_vma(struct msm_gem_address_space *aspace,
+		struct msm_gem_vma *vma)
+{
+	if (WARN_ON(vma->inuse > 0 || vma->mapped))
+		return;
+
+	spin_lock(&aspace->lock);
+	if (vma->iova)
+		drm_mm_remove_node(&vma->node);
+	spin_unlock(&aspace->lock);
+
+	vma->iova = 0;
+
+	msm_gem_address_space_put(aspace);
+}
+
 /* Initialize a new vma and allocate an iova for it */
 int msm_gem_init_vma(struct msm_gem_address_space *aspace,
 		struct msm_gem_vma *vma, int npages)
@@ -109,6 +135,7 @@ int msm_gem_init_vma(struct msm_gem_address_space *aspace,
 	return 0;
 }
 
+
 struct msm_gem_address_space *
 msm_gem_address_space_create(struct device *dev, struct iommu_domain *domain,
 		const char *name)

drivers/gpu/drm/msm/msm_gpu.c

@@ -689,7 +689,7 @@ static void retire_submit(struct msm_gpu *gpu, struct msm_ringbuffer *ring,
 		struct msm_gem_object *msm_obj = submit->bos[i].obj;
 		/* move to inactive: */
 		msm_gem_move_to_inactive(&msm_obj->base);
-		msm_gem_put_iova(&msm_obj->base, gpu->aspace);
+		msm_gem_unpin_iova(&msm_obj->base, gpu->aspace);
 		drm_gem_object_put(&msm_obj->base);
 	}