drm/nouveau: always map device

The next commit removes the nvif rd/wr methods from nvif_device, which
were probably a bad idea, and mostly intended as a fallback if ioremap()
failed (or wasn't available, as was the case in some tools I once used).

The nv04 KMS driver already mapped the device, because it's mostly been
kept alive on life-support for many years and still directly bashes PRI
a lot for modesetting.

Post-nv50, I tried pretty hard to keep PRI accesses out of the DRM code,
but there's still a few random places where we do, and those were using
the rd/wr paths prior to this commit.

This allocates and maps a new nvif_device (which will replace the usage
of nouveau_drm.master.device later on), and replicates this pointer to
all other possible users.

This will be cleaned up by the end of another patch series, after it's
been made safe to do so.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20240726043828.58966-25-bskeggs@nvidia.com

drivers/gpu/drm/nouveau/dispnv04/disp.c

@@ -189,7 +189,6 @@ static void
 nv04_display_destroy(struct drm_device *dev)
 {
 	struct nv04_display *disp = nv04_display(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_encoder *encoder;
 	struct nouveau_crtc *nv_crtc;
 
@@ -206,8 +205,6 @@ nv04_display_destroy(struct drm_device *dev)
 
 	nouveau_display(dev)->priv = NULL;
 	vfree(disp);
-
-	nvif_object_unmap(&drm->client.device.object);
 }
 
 int
@@ -229,8 +226,6 @@ nv04_display_create(struct drm_device *dev)
 
 	disp->drm = drm;
 
-	nvif_object_map(&drm->client.device.object, NULL, 0);
-
 	nouveau_display(dev)->priv = disp;
 	nouveau_display(dev)->dtor = nv04_display_destroy;
 	nouveau_display(dev)->init = nv04_display_init;

drivers/gpu/drm/nouveau/include/nvif/device.h

@@ -20,6 +20,7 @@ struct nvif_device {
 
 int  nvif_device_ctor(struct nvif_client *, const char *name, struct nvif_device *);
 void nvif_device_dtor(struct nvif_device *);
+int  nvif_device_map(struct nvif_device *);
 u64  nvif_device_time(struct nvif_device *);
 
 /*XXX*/

drivers/gpu/drm/nouveau/nouveau_drm.c

@@ -206,6 +206,7 @@ nouveau_cli_fini(struct nouveau_cli *cli)
 	nouveau_vmm_fini(&cli->svm);
 	nouveau_vmm_fini(&cli->vmm);
 	nvif_mmu_dtor(&cli->mmu);
+	cli->device.object.map.ptr = NULL;
 	nvif_device_dtor(&cli->device);
 	if (cli != &cli->drm->master) {
 		mutex_lock(&cli->drm->master.lock);
@@ -267,6 +268,8 @@ nouveau_cli_init(struct nouveau_drm *drm, const char *sname,
 		goto done;
 	}
 
+	cli->device.object.map.ptr = drm->device.object.map.ptr;
+
 	ret = nvif_mclass(&cli->device.object, mmus);
 	if (ret < 0) {
 		NV_PRINTK(err, cli, "No supported MMU class\n");
@@ -715,6 +718,7 @@ nouveau_drm_device_del(struct nouveau_drm *drm)
 	if (drm->dev)
 		drm_dev_put(drm->dev);
 
+	nvif_device_dtor(&drm->device);
 	nvif_client_dtor(&drm->master.base);
 	nvif_parent_dtor(&drm->parent);
 
@@ -751,6 +755,18 @@ nouveau_drm_device_new(const struct drm_driver *drm_driver, struct device *paren
 	if (ret)
 		goto done;
 
+	ret = nvif_device_ctor(&drm->master.base, "drmDevice", &drm->device);
+	if (ret) {
+		NV_ERROR(drm, "Device allocation failed: %d\n", ret);
+		goto done;
+	}
+
+	ret = nvif_device_map(&drm->device);
+	if (ret) {
+		NV_ERROR(drm, "Failed to map PRI: %d\n", ret);
+		goto done;
+	}
+
 done:
 	if (ret) {
 		nouveau_drm_device_del(drm);

drivers/gpu/drm/nouveau/nouveau_drv.h

@@ -203,6 +203,8 @@ u_memcpya(uint64_t user, unsigned int nmemb, unsigned int size)
 struct nouveau_drm {
 	struct nvkm_device *nvkm;
 	struct nvif_parent parent;
+	struct nvif_device device;
+
 	struct nouveau_cli master;
 	struct nouveau_cli client;
 	struct drm_device *dev;

drivers/gpu/drm/nouveau/nvif/device.c

@@ -38,6 +38,12 @@ nvif_device_time(struct nvif_device *device)
 	return device->user.func->time(&device->user);
 }
 
+int
+nvif_device_map(struct nvif_device *device)
+{
+	return nvif_object_map(&device->object, NULL, 0);
+}
+
 void
 nvif_device_dtor(struct nvif_device *device)
 {