Commit b7c0e47d authored by Dave Airlie's avatar Dave Airlie
Browse files

Merge tag 'drm-vc4-next-2016-11-16' of https://github.com/anholt/linux into drm-next 

This pull request brings in fragment shader threading and ETC1 support
for vc4.
parents 25bfe018 c778cc5d

drivers/gpu/drm/vc4/vc4_drv.c

+8 −3
Original line number Diff line number Diff line
@@ -61,23 +61,28 @@ static int vc4_get_param_ioctl(struct drm_device *dev, void *data, 
		if (ret < 0)

			return ret;

		args->value = V3D_READ(V3D_IDENT0);

		pm_runtime_put(&vc4->v3d->pdev->dev);

		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);

		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);

		break;

	case DRM_VC4_PARAM_V3D_IDENT1:

		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);

		if (ret < 0)

			return ret;

		args->value = V3D_READ(V3D_IDENT1);

		pm_runtime_put(&vc4->v3d->pdev->dev);

		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);

		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);

		break;

	case DRM_VC4_PARAM_V3D_IDENT2:

		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);

		if (ret < 0)

			return ret;

		args->value = V3D_READ(V3D_IDENT2);

		pm_runtime_put(&vc4->v3d->pdev->dev);

		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);

		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);

		break;

	case DRM_VC4_PARAM_SUPPORTS_BRANCHES:

	case DRM_VC4_PARAM_SUPPORTS_ETC1:

	case DRM_VC4_PARAM_SUPPORTS_THREADED_FS:

		args->value = true;

		break;

	default:

drivers/gpu/drm/vc4/vc4_drv.h

+2 −0
Original line number Diff line number Diff line
@@ -381,6 +381,8 @@ struct vc4_validated_shader_info { 


	uint32_t num_uniform_addr_offsets;

	uint32_t *uniform_addr_offsets;



	bool is_threaded;

};



/**

drivers/gpu/drm/vc4/vc4_gem.c

+9 −6
Original line number Diff line number Diff line
@@ -544,14 +544,15 @@ vc4_cl_lookup_bos(struct drm_device *dev, 


	handles = drm_malloc_ab(exec->bo_count, sizeof(uint32_t));

	if (!handles) {

		ret = -ENOMEM;

		DRM_ERROR("Failed to allocate incoming GEM handles\n");

		goto fail;

	}



	ret = copy_from_user(handles,

	if (copy_from_user(handles,

			   (void __user *)(uintptr_t)args->bo_handles,

			     exec->bo_count * sizeof(uint32_t));

	if (ret) {

			   exec->bo_count * sizeof(uint32_t))) {

		ret = -EFAULT;

		DRM_ERROR("Failed to copy in GEM handles\n");

		goto fail;

	}
@@ -708,8 +709,10 @@ vc4_complete_exec(struct drm_device *dev, struct vc4_exec_info *exec) 
	}



	mutex_lock(&vc4->power_lock);

	if (--vc4->power_refcount == 0)

		pm_runtime_put(&vc4->v3d->pdev->dev);

	if (--vc4->power_refcount == 0) {

		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);

		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);

	}

	mutex_unlock(&vc4->power_lock);



	kfree(exec);

drivers/gpu/drm/vc4/vc4_v3d.c

+2 −0
Original line number Diff line number Diff line
@@ -222,6 +222,8 @@ static int vc4_v3d_bind(struct device *dev, struct device *master, void *data) 
		return ret;

	}



	pm_runtime_use_autosuspend(dev);

	pm_runtime_set_autosuspend_delay(dev, 40); /* a little over 2 frames. */

	pm_runtime_enable(dev);



	return 0;

drivers/gpu/drm/vc4/vc4_validate.c

+19 −5
Original line number Diff line number Diff line
@@ -644,6 +644,13 @@ reloc_tex(struct vc4_exec_info *exec, 
		cpp = 1;

		break;

	case VC4_TEXTURE_TYPE_ETC1:

		/* ETC1 is arranged as 64-bit blocks, where each block is 4x4

		 * pixels.

		 */

		cpp = 8;

		width = (width + 3) >> 2;

		height = (height + 3) >> 2;

		break;

	case VC4_TEXTURE_TYPE_BW1:

	case VC4_TEXTURE_TYPE_A4:

	case VC4_TEXTURE_TYPE_A1:
@@ -782,11 +789,6 @@ validate_gl_shader_rec(struct drm_device *dev, 
	exec->shader_rec_v += roundup(packet_size, 16);

	exec->shader_rec_size -= packet_size;



	if (!(*(uint16_t *)pkt_u & VC4_SHADER_FLAG_FS_SINGLE_THREAD)) {

		DRM_ERROR("Multi-threaded fragment shaders not supported.\n");

		return -EINVAL;

	}



	for (i = 0; i < shader_reloc_count; i++) {

		if (src_handles[i] > exec->bo_count) {

			DRM_ERROR("Shader handle %d too big\n", src_handles[i]);
@@ -803,6 +805,18 @@ validate_gl_shader_rec(struct drm_device *dev, 
			return -EINVAL;

	}



	if (((*(uint16_t *)pkt_u & VC4_SHADER_FLAG_FS_SINGLE_THREAD) == 0) !=

	    to_vc4_bo(&bo[0]->base)->validated_shader->is_threaded) {

		DRM_ERROR("Thread mode of CL and FS do not match\n");

		return -EINVAL;

	}



	if (to_vc4_bo(&bo[1]->base)->validated_shader->is_threaded ||

	    to_vc4_bo(&bo[2]->base)->validated_shader->is_threaded) {

		DRM_ERROR("cs and vs cannot be threaded\n");

		return -EINVAL;

	}



	for (i = 0; i < shader_reloc_count; i++) {

		struct vc4_validated_shader_info *validated_shader;

		uint32_t o = shader_reloc_offsets[i];