drivers/gpu/drm/nouveau/nouveau_chan.c - kernel/msm-4.19 - Gitiles

 /*
  * Copyright 2012 Red Hat Inc.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
  * to deal in the Software without restriction, including without limitation
  * the rights to use, copy, modify, merge, publish, distribute, sublicense,
  * and/or sell copies of the Software, and to permit persons to whom the
  * Software is furnished to do so, subject to the following conditions:
  *
  * The above copyright notice and this permission notice shall be included in
  * all copies or substantial portions of the Software.
  *
  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
  * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
  * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  *
  * Authors: Ben Skeggs
  */

 #include <core/object.h>
 #include <core/client.h>
 #include <core/device.h>
 #include <core/class.h>

 #include <subdev/fb.h>
 #include <subdev/vm.h>
 #include <subdev/instmem.h>

 #include <engine/software.h>

 #include "nouveau_drm.h"
 #include "nouveau_dma.h"
 #include "nouveau_bo.h"
 #include "nouveau_chan.h"
 #include "nouveau_fence.h"
 #include "nouveau_abi16.h"

 MODULE_PARM_DESC(vram_pushbuf, "Create DMA push buffers in VRAM");
 static int nouveau_vram_pushbuf;
 module_param_named(vram_pushbuf, nouveau_vram_pushbuf, int, 0400);

 int
 nouveau_channel_idle(struct nouveau_channel *chan)
 {
 	struct nouveau_drm *drm = chan->drm;
 	struct nouveau_fence *fence = NULL;
 	int ret;

 	ret = nouveau_fence_new(chan, &fence);
 	if (!ret) {
 		ret = nouveau_fence_wait(fence, false, false);
 		nouveau_fence_unref(&fence);
 	}

 	if (ret)
 		NV_ERROR(drm, "failed to idle channel 0x%08x\n", chan->handle);
 	return ret;
 }

 void
 nouveau_channel_del(struct nouveau_channel **pchan)
 {
 	struct nouveau_channel *chan = *pchan;
 	if (chan) {
 		struct nouveau_object *client = nv_object(chan->cli);
 		if (chan->fence) {
 			nouveau_channel_idle(chan);
 			nouveau_fence(chan->drm)->context_del(chan);
 		}
 		nouveau_object_del(client, NVDRM_DEVICE, chan->handle);
 		nouveau_object_del(client, NVDRM_DEVICE, chan->push.handle);
 		nouveau_bo_vma_del(chan->push.buffer, &chan->push.vma);
 		nouveau_bo_unmap(chan->push.buffer);
 		nouveau_bo_ref(NULL, &chan->push.buffer);
 		kfree(chan);
 	}
 	*pchan = NULL;
 }

 static int
 nouveau_channel_prep(struct nouveau_drm *drm, struct nouveau_cli *cli,
 		     u32 parent, u32 handle, u32 size,
 		     struct nouveau_channel **pchan)
 {
 	struct nouveau_device *device = nv_device(drm->device);
 	struct nouveau_instmem *imem = nouveau_instmem(device);
 	struct nouveau_vmmgr *vmm = nouveau_vmmgr(device);
 	struct nouveau_fb *pfb = nouveau_fb(device);
 	struct nouveau_client *client = &cli->base;
 	struct nv_dma_class args = {};
 	struct nouveau_channel *chan;
 	struct nouveau_object *push;
 	u32 target;
 	int ret;

 	chan = *pchan = kzalloc(sizeof(*chan), GFP_KERNEL);
 	if (!chan)
 		return -ENOMEM;

 	chan->cli = cli;
 	chan->drm = drm;
 	chan->handle = handle;

 	/* allocate memory for dma push buffer */
 	target = TTM_PL_FLAG_TT;
 	if (nouveau_vram_pushbuf)
 		target = TTM_PL_FLAG_VRAM;

 	ret = nouveau_bo_new(drm->dev, size, 0, target, 0, 0, NULL,
 			    &chan->push.buffer);
 	if (ret == 0) {
 		ret = nouveau_bo_pin(chan->push.buffer, target);
 		if (ret == 0)
 			ret = nouveau_bo_map(chan->push.buffer);
 	}

 	if (ret) {
 		nouveau_channel_del(pchan);
 		return ret;
 	}

 	/* create dma object covering the *entire* memory space that the
 	 * pushbuf lives in, this is because the GEM code requires that
 	 * we be able to call out to other (indirect) push buffers
 	 */
 	chan->push.vma.offset = chan->push.buffer->bo.offset;
 	chan->push.handle = NVDRM_PUSH | (handle & 0xffff);

 	if (device->card_type >= NV_50) {
 		ret = nouveau_bo_vma_add(chan->push.buffer, client->vm,
 					&chan->push.vma);
 		if (ret) {
 			nouveau_channel_del(pchan);
 			return ret;
 		}

 		args.flags = NV_DMA_TARGET_VM | NV_DMA_ACCESS_VM;
 		args.start = 0;
 		args.limit = client->vm->vmm->limit - 1;
 	} else
 	if (chan->push.buffer->bo.mem.mem_type == TTM_PL_VRAM) {
 		u64 limit = pfb->ram.size - imem->reserved - 1;
 		if (device->card_type == NV_04) {
 			/* nv04 vram pushbuf hack, retarget to its location in
 			 * the framebuffer bar rather than direct vram access..
 			 * nfi why this exists, it came from the -nv ddx.
 			 */
 			args.flags = NV_DMA_TARGET_PCI | NV_DMA_ACCESS_RDWR;
 			args.start = pci_resource_start(device->pdev, 1);
 			args.limit = args.start + limit;
 		} else {
 			args.flags = NV_DMA_TARGET_VRAM | NV_DMA_ACCESS_RDWR;
 			args.start = 0;
 			args.limit = limit;
 		}
 	} else {
 		if (chan->drm->agp.stat == ENABLED) {
 			args.flags = NV_DMA_TARGET_AGP | NV_DMA_ACCESS_RDWR;
 			args.start = chan->drm->agp.base;
 			args.limit = chan->drm->agp.base +
 				     chan->drm->agp.size - 1;
 		} else {
 			args.flags = NV_DMA_TARGET_VM | NV_DMA_ACCESS_RDWR;
 			args.start = 0;
 			args.limit = vmm->limit - 1;
 		}
 	}

 	ret = nouveau_object_new(nv_object(chan->cli), parent,
 				 chan->push.handle, 0x0002,
 				 &args, sizeof(args), &push);
 	if (ret) {
 		nouveau_channel_del(pchan);
 		return ret;
 	}

 	return 0;
 }

 int
 nouveau_channel_ind(struct nouveau_drm *drm, struct nouveau_cli *cli,
 		    u32 parent, u32 handle, struct nouveau_channel **pchan)
 {
 	static const u16 oclasses[] = { 0xa06f, 0x906f, 0x826f, 0x506f, 0 };
 	const u16 *oclass = oclasses;
 	struct nve0_channel_ind_class args;
 	struct nouveau_channel *chan;
 	int ret;

 	/* allocate dma push buffer */
 	ret = nouveau_channel_prep(drm, cli, parent, handle, 0x12000, &chan);
 	*pchan = chan;
 	if (ret)
 		return ret;

 	/* create channel object */
 	args.pushbuf = chan->push.handle;
 	args.ioffset = 0x10000 + chan->push.vma.offset;
 	args.ilength = 0x02000;
 	args.engine  = NVE0_CHANNEL_IND_ENGINE_GR;

 	do {
 		ret = nouveau_object_new(nv_object(cli), parent, handle,
 					 *oclass++, &args, sizeof(args),
 					 &chan->object);
 		if (ret == 0)
 			return ret;
 	} while (*oclass);

 	nouveau_channel_del(pchan);
 	return ret;
 }

 static int
 nouveau_channel_dma(struct nouveau_drm *drm, struct nouveau_cli *cli,
 		    u32 parent, u32 handle, struct nouveau_channel **pchan)
 {
 	static const u16 oclasses[] = { 0x006e, 0 };
 	const u16 *oclass = oclasses;
 	struct nv_channel_dma_class args;
 	struct nouveau_channel *chan;
 	int ret;

 	/* allocate dma push buffer */
 	ret = nouveau_channel_prep(drm, cli, parent, handle, 0x10000, &chan);
 	*pchan = chan;
 	if (ret)
 		return ret;

 	/* create channel object */
 	args.pushbuf = chan->push.handle;
 	args.offset = chan->push.vma.offset;

 	do {
 		ret = nouveau_object_new(nv_object(cli), parent, handle,
 					 *oclass++, &args, sizeof(args),
 					 &chan->object);
 		if (ret == 0)
 			return ret;
 	} while (ret && *oclass);

 	nouveau_channel_del(pchan);
 	return ret;
 }

 static int
 nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
 {
 	struct nouveau_client *client = nv_client(chan->cli);
 	struct nouveau_device *device = nv_device(chan->drm->device);
 	struct nouveau_instmem *imem = nouveau_instmem(device);
 	struct nouveau_vmmgr *vmm = nouveau_vmmgr(device);
 	struct nouveau_fb *pfb = nouveau_fb(device);
 	struct nouveau_software_chan *swch;
 	struct nouveau_object *object;
 	struct nv_dma_class args;
 	int ret, i;

 	chan->vram = vram;
 	chan->gart = gart;

 	/* allocate dma objects to cover all allowed vram, and gart */
 	if (device->card_type < NV_C0) {
 		if (device->card_type >= NV_50) {
 			args.flags = NV_DMA_TARGET_VM | NV_DMA_ACCESS_VM;
 			args.start = 0;
 			args.limit = client->vm->vmm->limit - 1;
 		} else {
 			args.flags = NV_DMA_TARGET_VRAM | NV_DMA_ACCESS_RDWR;
 			args.start = 0;
 			args.limit = pfb->ram.size - imem->reserved - 1;
 		}

 		ret = nouveau_object_new(nv_object(client), chan->handle, vram,
 					 0x003d, &args, sizeof(args), &object);
 		if (ret)
 			return ret;

 		if (device->card_type >= NV_50) {
 			args.flags = NV_DMA_TARGET_VM | NV_DMA_ACCESS_VM;
 			args.start = 0;
 			args.limit = client->vm->vmm->limit - 1;
 		} else
 		if (chan->drm->agp.stat == ENABLED) {
 			args.flags = NV_DMA_TARGET_AGP | NV_DMA_ACCESS_RDWR;
 			args.start = chan->drm->agp.base;
 			args.limit = chan->drm->agp.base +
 				     chan->drm->agp.size - 1;
 		} else {
 			args.flags = NV_DMA_TARGET_VM | NV_DMA_ACCESS_RDWR;
 			args.start = 0;
 			args.limit = vmm->limit - 1;
 		}

 		ret = nouveau_object_new(nv_object(client), chan->handle, gart,
 					 0x003d, &args, sizeof(args), &object);
 		if (ret)
 			return ret;
 	}

 	/* initialise dma tracking parameters */
 	switch (nv_hclass(chan->object) & 0xffff) {
 	case 0x006e:
 		chan->user_put = 0x40;
 		chan->user_get = 0x44;
 		chan->dma.max = (0x10000 / 4) - 2;
 		break;
 	default:
 		chan->user_put = 0x40;
 		chan->user_get = 0x44;
 		chan->user_get_hi = 0x60;
 		chan->dma.ib_base =  0x10000 / 4;
 		chan->dma.ib_max  = (0x02000 / 8) - 1;
 		chan->dma.ib_put  = 0;
 		chan->dma.ib_free = chan->dma.ib_max - chan->dma.ib_put;
 		chan->dma.max = chan->dma.ib_base;
 		break;
 	}

 	chan->dma.put = 0;
 	chan->dma.cur = chan->dma.put;
 	chan->dma.free = chan->dma.max - chan->dma.cur;

 	ret = RING_SPACE(chan, NOUVEAU_DMA_SKIPS);
 	if (ret)
 		return ret;

 	for (i = 0; i < NOUVEAU_DMA_SKIPS; i++)
 		OUT_RING(chan, 0x00000000);

 	/* allocate software object class (used for fences on <= nv05, and
 	 * to signal flip completion), bind it to a subchannel.
 	 */
 	ret = nouveau_object_new(nv_object(client), chan->handle,
 				 NvSw, nouveau_abi16_swclass(chan->drm),
 				 NULL, 0, &object);
 	if (ret)
 		return ret;

 	swch = (void *)object->parent;
 	swch->flip = nouveau_flip_complete;
 	swch->flip_data = chan;

 	if (device->card_type < NV_C0) {
 		ret = RING_SPACE(chan, 2);
 		if (ret)
 			return ret;

 		BEGIN_NV04(chan, NvSubSw, 0x0000, 1);
 		OUT_RING  (chan, NvSw);
 		FIRE_RING (chan);
 	}

 	/* initialise synchronisation */
 	return nouveau_fence(chan->drm)->context_new(chan);
 }

 int
 nouveau_channel_new(struct nouveau_drm *drm, struct nouveau_cli *cli,
 		    u32 parent, u32 handle, u32 vram, u32 gart,
 		    struct nouveau_channel **pchan)
 {
 	int ret;

 	ret = nouveau_channel_ind(drm, cli, parent, handle, pchan);
 	if (ret) {
 		NV_DEBUG(drm, "ib channel create, %d\n", ret);
 		ret = nouveau_channel_dma(drm, cli, parent, handle, pchan);
 		if (ret) {
 			NV_DEBUG(drm, "dma channel create, %d\n", ret);
 			return ret;
 		}
 	}

 	ret = nouveau_channel_init(*pchan, vram, gart);
 	if (ret) {
 		NV_ERROR(drm, "channel failed to initialise, %d\n", ret);
 		nouveau_channel_del(pchan);
 		return ret;
 	}

 	return 0;
 }
	/*
	* Copyright 2012 Red Hat Inc.
	*
	* Permission is hereby granted, free of charge, to any person obtaining a
	* copy of this software and associated documentation files (the "Software"),
	* to deal in the Software without restriction, including without limitation
	* the rights to use, copy, modify, merge, publish, distribute, sublicense,
	* and/or sell copies of the Software, and to permit persons to whom the
	* Software is furnished to do so, subject to the following conditions:
	*
	* The above copyright notice and this permission notice shall be included in
	* all copies or substantial portions of the Software.
	*
	* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
	* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
	* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
	* THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
	* OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
	* ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
	* OTHER DEALINGS IN THE SOFTWARE.
	*
	* Authors: Ben Skeggs
	*/

	#include <core/object.h>
	#include <core/client.h>
	#include <core/device.h>
	#include <core/class.h>

	#include <subdev/fb.h>
	#include <subdev/vm.h>
	#include <subdev/instmem.h>

	#include <engine/software.h>

	#include "nouveau_drm.h"
	#include "nouveau_dma.h"
	#include "nouveau_bo.h"
	#include "nouveau_chan.h"
	#include "nouveau_fence.h"
	#include "nouveau_abi16.h"

	MODULE_PARM_DESC(vram_pushbuf, "Create DMA push buffers in VRAM");
	static int nouveau_vram_pushbuf;
	module_param_named(vram_pushbuf, nouveau_vram_pushbuf, int, 0400);

	int
	nouveau_channel_idle(struct nouveau_channel *chan)
	{
	struct nouveau_drm *drm = chan->drm;
	struct nouveau_fence *fence = NULL;
	int ret;

	ret = nouveau_fence_new(chan, &fence);
	if (!ret) {
	ret = nouveau_fence_wait(fence, false, false);
	nouveau_fence_unref(&fence);
	}

	if (ret)
	NV_ERROR(drm, "failed to idle channel 0x%08x\n", chan->handle);
	return ret;
	}

	void
	nouveau_channel_del(struct nouveau_channel **pchan)
	{
	struct nouveau_channel chan = pchan;
	if (chan) {
	struct nouveau_object *client = nv_object(chan->cli);
	if (chan->fence) {
	nouveau_channel_idle(chan);
	nouveau_fence(chan->drm)->context_del(chan);
	}
	nouveau_object_del(client, NVDRM_DEVICE, chan->handle);
	nouveau_object_del(client, NVDRM_DEVICE, chan->push.handle);
	nouveau_bo_vma_del(chan->push.buffer, &chan->push.vma);
	nouveau_bo_unmap(chan->push.buffer);
	nouveau_bo_ref(NULL, &chan->push.buffer);
	kfree(chan);
	}
	*pchan = NULL;
	}

	static int
	nouveau_channel_prep(struct nouveau_drm drm, struct nouveau_cli cli,
	u32 parent, u32 handle, u32 size,
	struct nouveau_channel **pchan)
	{
	struct nouveau_device *device = nv_device(drm->device);
	struct nouveau_instmem *imem = nouveau_instmem(device);
	struct nouveau_vmmgr *vmm = nouveau_vmmgr(device);
	struct nouveau_fb *pfb = nouveau_fb(device);
	struct nouveau_client *client = &cli->base;
	struct nv_dma_class args = {};
	struct nouveau_channel *chan;
	struct nouveau_object *push;
	u32 target;
	int ret;

	chan = pchan = kzalloc(sizeof(chan), GFP_KERNEL);
	if (!chan)
	return -ENOMEM;

	chan->cli = cli;
	chan->drm = drm;
	chan->handle = handle;

	/* allocate memory for dma push buffer */
	target = TTM_PL_FLAG_TT;
	if (nouveau_vram_pushbuf)
	target = TTM_PL_FLAG_VRAM;

	ret = nouveau_bo_new(drm->dev, size, 0, target, 0, 0, NULL,
	&chan->push.buffer);
	if (ret == 0) {
	ret = nouveau_bo_pin(chan->push.buffer, target);
	if (ret == 0)
	ret = nouveau_bo_map(chan->push.buffer);
	}

	if (ret) {
	nouveau_channel_del(pchan);
	return ret;
	}

	/* create dma object covering the entire memory space that the
	* pushbuf lives in, this is because the GEM code requires that
	* we be able to call out to other (indirect) push buffers
	*/
	chan->push.vma.offset = chan->push.buffer->bo.offset;
	chan->push.handle = NVDRM_PUSH \| (handle & 0xffff);

	if (device->card_type >= NV_50) {
	ret = nouveau_bo_vma_add(chan->push.buffer, client->vm,
	&chan->push.vma);
	if (ret) {
	nouveau_channel_del(pchan);
	return ret;
	}

	args.flags = NV_DMA_TARGET_VM \| NV_DMA_ACCESS_VM;
	args.start = 0;
	args.limit = client->vm->vmm->limit - 1;
	} else
	if (chan->push.buffer->bo.mem.mem_type == TTM_PL_VRAM) {
	u64 limit = pfb->ram.size - imem->reserved - 1;
	if (device->card_type == NV_04) {
	/* nv04 vram pushbuf hack, retarget to its location in
	* the framebuffer bar rather than direct vram access..
	* nfi why this exists, it came from the -nv ddx.
	*/
	args.flags = NV_DMA_TARGET_PCI \| NV_DMA_ACCESS_RDWR;
	args.start = pci_resource_start(device->pdev, 1);
	args.limit = args.start + limit;
	} else {
	args.flags = NV_DMA_TARGET_VRAM \| NV_DMA_ACCESS_RDWR;
	args.start = 0;
	args.limit = limit;
	}
	} else {
	if (chan->drm->agp.stat == ENABLED) {
	args.flags = NV_DMA_TARGET_AGP \| NV_DMA_ACCESS_RDWR;
	args.start = chan->drm->agp.base;
	args.limit = chan->drm->agp.base +
	chan->drm->agp.size - 1;
	} else {
	args.flags = NV_DMA_TARGET_VM \| NV_DMA_ACCESS_RDWR;
	args.start = 0;
	args.limit = vmm->limit - 1;
	}
	}

	ret = nouveau_object_new(nv_object(chan->cli), parent,
	chan->push.handle, 0x0002,
	&args, sizeof(args), &push);
	if (ret) {
	nouveau_channel_del(pchan);
	return ret;
	}

	return 0;
	}

	int
	nouveau_channel_ind(struct nouveau_drm drm, struct nouveau_cli cli,
	u32 parent, u32 handle, struct nouveau_channel **pchan)
	{
	static const u16 oclasses[] = { 0xa06f, 0x906f, 0x826f, 0x506f, 0 };
	const u16 *oclass = oclasses;
	struct nve0_channel_ind_class args;
	struct nouveau_channel *chan;
	int ret;

	/* allocate dma push buffer */
	ret = nouveau_channel_prep(drm, cli, parent, handle, 0x12000, &chan);
	*pchan = chan;
	if (ret)
	return ret;

	/* create channel object */
	args.pushbuf = chan->push.handle;
	args.ioffset = 0x10000 + chan->push.vma.offset;
	args.ilength = 0x02000;
	args.engine = NVE0_CHANNEL_IND_ENGINE_GR;

	do {
	ret = nouveau_object_new(nv_object(cli), parent, handle,
	*oclass++, &args, sizeof(args),
	&chan->object);
	if (ret == 0)
	return ret;
	} while (*oclass);

	nouveau_channel_del(pchan);
	return ret;
	}

	static int
	nouveau_channel_dma(struct nouveau_drm drm, struct nouveau_cli cli,
	u32 parent, u32 handle, struct nouveau_channel **pchan)
	{
	static const u16 oclasses[] = { 0x006e, 0 };
	const u16 *oclass = oclasses;
	struct nv_channel_dma_class args;
	struct nouveau_channel *chan;
	int ret;

	/* allocate dma push buffer */
	ret = nouveau_channel_prep(drm, cli, parent, handle, 0x10000, &chan);
	*pchan = chan;
	if (ret)
	return ret;

	/* create channel object */
	args.pushbuf = chan->push.handle;
	args.offset = chan->push.vma.offset;

	do {
	ret = nouveau_object_new(nv_object(cli), parent, handle,
	*oclass++, &args, sizeof(args),
	&chan->object);
	if (ret == 0)
	return ret;
	} while (ret && *oclass);

	nouveau_channel_del(pchan);
	return ret;
	}

	static int
	nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
	{
	struct nouveau_client *client = nv_client(chan->cli);
	struct nouveau_device *device = nv_device(chan->drm->device);
	struct nouveau_instmem *imem = nouveau_instmem(device);
	struct nouveau_vmmgr *vmm = nouveau_vmmgr(device);
	struct nouveau_fb *pfb = nouveau_fb(device);
	struct nouveau_software_chan *swch;
	struct nouveau_object *object;
	struct nv_dma_class args;
	int ret, i;

	chan->vram = vram;
	chan->gart = gart;

	/* allocate dma objects to cover all allowed vram, and gart */
	if (device->card_type < NV_C0) {
	if (device->card_type >= NV_50) {
	args.flags = NV_DMA_TARGET_VM \| NV_DMA_ACCESS_VM;
	args.start = 0;
	args.limit = client->vm->vmm->limit - 1;
	} else {
	args.flags = NV_DMA_TARGET_VRAM \| NV_DMA_ACCESS_RDWR;
	args.start = 0;
	args.limit = pfb->ram.size - imem->reserved - 1;
	}

	ret = nouveau_object_new(nv_object(client), chan->handle, vram,
	0x003d, &args, sizeof(args), &object);
	if (ret)
	return ret;

	if (device->card_type >= NV_50) {
	args.flags = NV_DMA_TARGET_VM \| NV_DMA_ACCESS_VM;
	args.start = 0;
	args.limit = client->vm->vmm->limit - 1;
	} else
	if (chan->drm->agp.stat == ENABLED) {
	args.flags = NV_DMA_TARGET_AGP \| NV_DMA_ACCESS_RDWR;
	args.start = chan->drm->agp.base;
	args.limit = chan->drm->agp.base +
	chan->drm->agp.size - 1;
	} else {
	args.flags = NV_DMA_TARGET_VM \| NV_DMA_ACCESS_RDWR;
	args.start = 0;
	args.limit = vmm->limit - 1;
	}

	ret = nouveau_object_new(nv_object(client), chan->handle, gart,
	0x003d, &args, sizeof(args), &object);
	if (ret)
	return ret;
	}

	/* initialise dma tracking parameters */
	switch (nv_hclass(chan->object) & 0xffff) {
	case 0x006e:
	chan->user_put = 0x40;
	chan->user_get = 0x44;
	chan->dma.max = (0x10000 / 4) - 2;
	break;
	default:
	chan->user_put = 0x40;
	chan->user_get = 0x44;
	chan->user_get_hi = 0x60;
	chan->dma.ib_base = 0x10000 / 4;
	chan->dma.ib_max = (0x02000 / 8) - 1;
	chan->dma.ib_put = 0;
	chan->dma.ib_free = chan->dma.ib_max - chan->dma.ib_put;
	chan->dma.max = chan->dma.ib_base;
	break;
	}

	chan->dma.put = 0;
	chan->dma.cur = chan->dma.put;
	chan->dma.free = chan->dma.max - chan->dma.cur;

	ret = RING_SPACE(chan, NOUVEAU_DMA_SKIPS);
	if (ret)
	return ret;

	for (i = 0; i < NOUVEAU_DMA_SKIPS; i++)
	OUT_RING(chan, 0x00000000);

	/* allocate software object class (used for fences on <= nv05, and
	* to signal flip completion), bind it to a subchannel.
	*/
	ret = nouveau_object_new(nv_object(client), chan->handle,
	NvSw, nouveau_abi16_swclass(chan->drm),
	NULL, 0, &object);
	if (ret)
	return ret;

	swch = (void *)object->parent;
	swch->flip = nouveau_flip_complete;
	swch->flip_data = chan;

	if (device->card_type < NV_C0) {
	ret = RING_SPACE(chan, 2);
	if (ret)
	return ret;

	BEGIN_NV04(chan, NvSubSw, 0x0000, 1);
	OUT_RING (chan, NvSw);
	FIRE_RING (chan);
	}

	/* initialise synchronisation */
	return nouveau_fence(chan->drm)->context_new(chan);
	}

	int
	nouveau_channel_new(struct nouveau_drm drm, struct nouveau_cli cli,
	u32 parent, u32 handle, u32 vram, u32 gart,
	struct nouveau_channel **pchan)
	{
	int ret;

	ret = nouveau_channel_ind(drm, cli, parent, handle, pchan);
	if (ret) {
	NV_DEBUG(drm, "ib channel create, %d\n", ret);
	ret = nouveau_channel_dma(drm, cli, parent, handle, pchan);
	if (ret) {
	NV_DEBUG(drm, "dma channel create, %d\n", ret);
	return ret;
	}
	}

	ret = nouveau_channel_init(*pchan, vram, gart);
	if (ret) {
	NV_ERROR(drm, "channel failed to initialise, %d\n", ret);
	nouveau_channel_del(pchan);
	return ret;
	}

	return 0;
	}