uverbs_cmd.c

来自「linux2.6.16版本」· C语言 代码 · 共 1,834 行 · 第 1/3 页

	uobj->async_events_reported = 0;
	INIT_LIST_HEAD(&uobj->comp_list);
	INIT_LIST_HEAD(&uobj->async_list);

	cq = file->device->ib_dev->create_cq(file->device->ib_dev, cmd.cqe,
					     file->ucontext, &udata);
	if (IS_ERR(cq)) {
		ret = PTR_ERR(cq);
		goto err;
	}

	cq->device        = file->device->ib_dev;
	cq->uobject       = &uobj->uobject;
	cq->comp_handler  = ib_uverbs_comp_handler;
	cq->event_handler = ib_uverbs_cq_event_handler;
	cq->cq_context    = ev_file;
	atomic_set(&cq->usecnt, 0);

	mutex_lock(&ib_uverbs_idr_mutex);

retry:
	if (!idr_pre_get(&ib_uverbs_cq_idr, GFP_KERNEL)) {
		ret = -ENOMEM;
		goto err_up;
	}

	ret = idr_get_new(&ib_uverbs_cq_idr, cq, &uobj->uobject.id);

	if (ret == -EAGAIN)
		goto retry;
	if (ret)
		goto err_up;

	memset(&resp, 0, sizeof resp);
	resp.cq_handle = uobj->uobject.id;
	resp.cqe       = cq->cqe;

	if (copy_to_user((void __user *) (unsigned long) cmd.response,
			 &resp, sizeof resp)) {
		ret = -EFAULT;
		goto err_idr;
	}

	mutex_lock(&file->mutex);
	list_add_tail(&uobj->uobject.list, &file->ucontext->cq_list);
	mutex_unlock(&file->mutex);

	mutex_unlock(&ib_uverbs_idr_mutex);

	return in_len;

err_idr:
	idr_remove(&ib_uverbs_cq_idr, uobj->uobject.id);

err_up:
	mutex_unlock(&ib_uverbs_idr_mutex);
	ib_destroy_cq(cq);

err:
	if (ev_file)
		ib_uverbs_release_ucq(file, ev_file, uobj);
	kfree(uobj);
	return ret;
}

ssize_t ib_uverbs_poll_cq(struct ib_uverbs_file *file,
			  const char __user *buf, int in_len,
			  int out_len)
{
	struct ib_uverbs_poll_cq       cmd;
	struct ib_uverbs_poll_cq_resp *resp;
	struct ib_cq                  *cq;
	struct ib_wc                  *wc;
	int                            ret = 0;
	int                            i;
	int                            rsize;

	if (copy_from_user(&cmd, buf, sizeof cmd))
		return -EFAULT;

	wc = kmalloc(cmd.ne * sizeof *wc, GFP_KERNEL);
	if (!wc)
		return -ENOMEM;

	rsize = sizeof *resp + cmd.ne * sizeof(struct ib_uverbs_wc);
	resp = kmalloc(rsize, GFP_KERNEL);
	if (!resp) {
		ret = -ENOMEM;
		goto out_wc;
	}

	mutex_lock(&ib_uverbs_idr_mutex);
	cq = idr_find(&ib_uverbs_cq_idr, cmd.cq_handle);
	if (!cq || cq->uobject->context != file->ucontext) {
		ret = -EINVAL;
		goto out;
	}

	resp->count = ib_poll_cq(cq, cmd.ne, wc);

	for (i = 0; i < resp->count; i++) {
		resp->wc[i].wr_id 	   = wc[i].wr_id;
		resp->wc[i].status 	   = wc[i].status;
		resp->wc[i].opcode 	   = wc[i].opcode;
		resp->wc[i].vendor_err 	   = wc[i].vendor_err;
		resp->wc[i].byte_len 	   = wc[i].byte_len;
		resp->wc[i].imm_data 	   = (__u32 __force) wc[i].imm_data;
		resp->wc[i].qp_num 	   = wc[i].qp_num;
		resp->wc[i].src_qp 	   = wc[i].src_qp;
		resp->wc[i].wc_flags 	   = wc[i].wc_flags;
		resp->wc[i].pkey_index 	   = wc[i].pkey_index;
		resp->wc[i].slid 	   = wc[i].slid;
		resp->wc[i].sl 		   = wc[i].sl;
		resp->wc[i].dlid_path_bits = wc[i].dlid_path_bits;
		resp->wc[i].port_num 	   = wc[i].port_num;
	}

	if (copy_to_user((void __user *) (unsigned long) cmd.response, resp, rsize))
		ret = -EFAULT;

out:
	mutex_unlock(&ib_uverbs_idr_mutex);
	kfree(resp);

out_wc:
	kfree(wc);
	return ret ? ret : in_len;
}

ssize_t ib_uverbs_req_notify_cq(struct ib_uverbs_file *file,
				const char __user *buf, int in_len,
				int out_len)
{
	struct ib_uverbs_req_notify_cq cmd;
	struct ib_cq                  *cq;
	int                            ret = -EINVAL;

	if (copy_from_user(&cmd, buf, sizeof cmd))
		return -EFAULT;

	mutex_lock(&ib_uverbs_idr_mutex);
	cq = idr_find(&ib_uverbs_cq_idr, cmd.cq_handle);
	if (cq && cq->uobject->context == file->ucontext) {
		ib_req_notify_cq(cq, cmd.solicited_only ?
					IB_CQ_SOLICITED : IB_CQ_NEXT_COMP);
		ret = in_len;
	}
	mutex_unlock(&ib_uverbs_idr_mutex);

	return ret;
}

ssize_t ib_uverbs_destroy_cq(struct ib_uverbs_file *file,
			     const char __user *buf, int in_len,
			     int out_len)
{
	struct ib_uverbs_destroy_cq      cmd;
	struct ib_uverbs_destroy_cq_resp resp;
	struct ib_cq               	*cq;
	struct ib_ucq_object        	*uobj;
	struct ib_uverbs_event_file	*ev_file;
	u64				 user_handle;
	int                        	 ret = -EINVAL;

	if (copy_from_user(&cmd, buf, sizeof cmd))
		return -EFAULT;

	memset(&resp, 0, sizeof resp);

	mutex_lock(&ib_uverbs_idr_mutex);

	cq = idr_find(&ib_uverbs_cq_idr, cmd.cq_handle);
	if (!cq || cq->uobject->context != file->ucontext)
		goto out;

	user_handle = cq->uobject->user_handle;
	uobj        = container_of(cq->uobject, struct ib_ucq_object, uobject);
	ev_file     = cq->cq_context;

	ret = ib_destroy_cq(cq);
	if (ret)
		goto out;

	idr_remove(&ib_uverbs_cq_idr, cmd.cq_handle);

	mutex_lock(&file->mutex);
	list_del(&uobj->uobject.list);
	mutex_unlock(&file->mutex);

	ib_uverbs_release_ucq(file, ev_file, uobj);

	resp.comp_events_reported  = uobj->comp_events_reported;
	resp.async_events_reported = uobj->async_events_reported;

	kfree(uobj);

	if (copy_to_user((void __user *) (unsigned long) cmd.response,
			 &resp, sizeof resp))
		ret = -EFAULT;

out:
	mutex_unlock(&ib_uverbs_idr_mutex);

	return ret ? ret : in_len;
}

ssize_t ib_uverbs_create_qp(struct ib_uverbs_file *file,
			    const char __user *buf, int in_len,
			    int out_len)
{
	struct ib_uverbs_create_qp      cmd;
	struct ib_uverbs_create_qp_resp resp;
	struct ib_udata                 udata;
	struct ib_uqp_object           *uobj;
	struct ib_pd                   *pd;
	struct ib_cq                   *scq, *rcq;
	struct ib_srq                  *srq;
	struct ib_qp                   *qp;
	struct ib_qp_init_attr          attr;
	int ret;

	if (out_len < sizeof resp)
		return -ENOSPC;

	if (copy_from_user(&cmd, buf, sizeof cmd))
		return -EFAULT;

	INIT_UDATA(&udata, buf + sizeof cmd,
		   (unsigned long) cmd.response + sizeof resp,
		   in_len - sizeof cmd, out_len - sizeof resp);

	uobj = kmalloc(sizeof *uobj, GFP_KERNEL);
	if (!uobj)
		return -ENOMEM;

	mutex_lock(&ib_uverbs_idr_mutex);

	pd  = idr_find(&ib_uverbs_pd_idr, cmd.pd_handle);
	scq = idr_find(&ib_uverbs_cq_idr, cmd.send_cq_handle);
	rcq = idr_find(&ib_uverbs_cq_idr, cmd.recv_cq_handle);
	srq = cmd.is_srq ? idr_find(&ib_uverbs_srq_idr, cmd.srq_handle) : NULL;

	if (!pd  || pd->uobject->context  != file->ucontext ||
	    !scq || scq->uobject->context != file->ucontext ||
	    !rcq || rcq->uobject->context != file->ucontext ||
	    (cmd.is_srq && (!srq || srq->uobject->context != file->ucontext))) {
		ret = -EINVAL;
		goto err_up;
	}

	attr.event_handler = ib_uverbs_qp_event_handler;
	attr.qp_context    = file;
	attr.send_cq       = scq;
	attr.recv_cq       = rcq;
	attr.srq           = srq;
	attr.sq_sig_type   = cmd.sq_sig_all ? IB_SIGNAL_ALL_WR : IB_SIGNAL_REQ_WR;
	attr.qp_type       = cmd.qp_type;

	attr.cap.max_send_wr     = cmd.max_send_wr;
	attr.cap.max_recv_wr     = cmd.max_recv_wr;
	attr.cap.max_send_sge    = cmd.max_send_sge;
	attr.cap.max_recv_sge    = cmd.max_recv_sge;
	attr.cap.max_inline_data = cmd.max_inline_data;

	uobj->uevent.uobject.user_handle = cmd.user_handle;
	uobj->uevent.uobject.context     = file->ucontext;
	uobj->uevent.events_reported     = 0;
	INIT_LIST_HEAD(&uobj->uevent.event_list);
	INIT_LIST_HEAD(&uobj->mcast_list);

	qp = pd->device->create_qp(pd, &attr, &udata);
	if (IS_ERR(qp)) {
		ret = PTR_ERR(qp);
		goto err_up;
	}

	qp->device     	  = pd->device;
	qp->pd         	  = pd;
	qp->send_cq    	  = attr.send_cq;
	qp->recv_cq    	  = attr.recv_cq;
	qp->srq	       	  = attr.srq;
	qp->uobject       = &uobj->uevent.uobject;
	qp->event_handler = attr.event_handler;
	qp->qp_context    = attr.qp_context;
	qp->qp_type	  = attr.qp_type;
	atomic_inc(&pd->usecnt);
	atomic_inc(&attr.send_cq->usecnt);
	atomic_inc(&attr.recv_cq->usecnt);
	if (attr.srq)
		atomic_inc(&attr.srq->usecnt);

	memset(&resp, 0, sizeof resp);
	resp.qpn = qp->qp_num;

retry:
	if (!idr_pre_get(&ib_uverbs_qp_idr, GFP_KERNEL)) {
		ret = -ENOMEM;
		goto err_destroy;
	}

	ret = idr_get_new(&ib_uverbs_qp_idr, qp, &uobj->uevent.uobject.id);

	if (ret == -EAGAIN)
		goto retry;
	if (ret)
		goto err_destroy;

	resp.qp_handle       = uobj->uevent.uobject.id;
	resp.max_recv_sge    = attr.cap.max_recv_sge;
	resp.max_send_sge    = attr.cap.max_send_sge;
	resp.max_recv_wr     = attr.cap.max_recv_wr;
	resp.max_send_wr     = attr.cap.max_send_wr;
	resp.max_inline_data = attr.cap.max_inline_data;

	if (copy_to_user((void __user *) (unsigned long) cmd.response,
			 &resp, sizeof resp)) {
		ret = -EFAULT;
		goto err_idr;
	}

	mutex_lock(&file->mutex);
	list_add_tail(&uobj->uevent.uobject.list, &file->ucontext->qp_list);
	mutex_unlock(&file->mutex);

	mutex_unlock(&ib_uverbs_idr_mutex);

	return in_len;

err_idr:
	idr_remove(&ib_uverbs_qp_idr, uobj->uevent.uobject.id);

err_destroy:
	ib_destroy_qp(qp);
	atomic_dec(&pd->usecnt);
	atomic_dec(&attr.send_cq->usecnt);
	atomic_dec(&attr.recv_cq->usecnt);
	if (attr.srq)
		atomic_dec(&attr.srq->usecnt);

err_up:
	mutex_unlock(&ib_uverbs_idr_mutex);

	kfree(uobj);
	return ret;
}

ssize_t ib_uverbs_modify_qp(struct ib_uverbs_file *file,
			    const char __user *buf, int in_len,
			    int out_len)
{
	struct ib_uverbs_modify_qp cmd;
	struct ib_qp              *qp;
	struct ib_qp_attr         *attr;
	int                        ret;

	if (copy_from_user(&cmd, buf, sizeof cmd))
		return -EFAULT;

	attr = kmalloc(sizeof *attr, GFP_KERNEL);
	if (!attr)
		return -ENOMEM;

	mutex_lock(&ib_uverbs_idr_mutex);

	qp = idr_find(&ib_uverbs_qp_idr, cmd.qp_handle);
	if (!qp || qp->uobject->context != file->ucontext) {
		ret = -EINVAL;
		goto out;
	}

	attr->qp_state 		  = cmd.qp_state;
	attr->cur_qp_state 	  = cmd.cur_qp_state;
	attr->path_mtu 		  = cmd.path_mtu;
	attr->path_mig_state 	  = cmd.path_mig_state;
	attr->qkey 		  = cmd.qkey;
	attr->rq_psn 		  = cmd.rq_psn;
	attr->sq_psn 		  = cmd.sq_psn;
	attr->dest_qp_num 	  = cmd.dest_qp_num;
	attr->qp_access_flags 	  = cmd.qp_access_flags;
	attr->pkey_index 	  = cmd.pkey_index;
	attr->alt_pkey_index 	  = cmd.pkey_index;
	attr->en_sqd_async_notify = cmd.en_sqd_async_notify;
	attr->max_rd_atomic 	  = cmd.max_rd_atomic;
	attr->max_dest_rd_atomic  = cmd.max_dest_rd_atomic;
	attr->min_rnr_timer 	  = cmd.min_rnr_timer;
	attr->port_num 		  = cmd.port_num;
	attr->timeout 		  = cmd.timeout;
	attr->retry_cnt 	  = cmd.retry_cnt;
	attr->rnr_retry 	  = cmd.rnr_retry;
	attr->alt_port_num 	  = cmd.alt_port_num;
	attr->alt_timeout 	  = cmd.alt_timeout;

	memcpy(attr->ah_attr.grh.dgid.raw, cmd.dest.dgid, 16);
	attr->ah_attr.grh.flow_label        = cmd.dest.flow_label;
	attr->ah_attr.grh.sgid_index        = cmd.dest.sgid_index;
	attr->ah_attr.grh.hop_limit         = cmd.dest.hop_limit;
	attr->ah_attr.grh.traffic_class     = cmd.dest.traffic_class;
	attr->ah_attr.dlid 	    	    = cmd.dest.dlid;
	attr->ah_attr.sl   	    	    = cmd.dest.sl;
	attr->ah_attr.src_path_bits 	    = cmd.dest.src_path_bits;
	attr->ah_attr.static_rate   	    = cmd.dest.static_rate;
	attr->ah_attr.ah_flags 	    	    = cmd.dest.is_global ? IB_AH_GRH : 0;
	attr->ah_attr.port_num 	    	    = cmd.dest.port_num;

	memcpy(attr->alt_ah_attr.grh.dgid.raw, cmd.alt_dest.dgid, 16);
	attr->alt_ah_attr.grh.flow_label    = cmd.alt_dest.flow_label;
	attr->alt_ah_attr.grh.sgid_index    = cmd.alt_dest.sgid_index;
	attr->alt_ah_attr.grh.hop_limit     = cmd.alt_dest.hop_limit;
	attr->alt_ah_attr.grh.traffic_class = cmd.alt_dest.traffic_class;
	attr->alt_ah_attr.dlid 	    	    = cmd.alt_dest.dlid;
	attr->alt_ah_attr.sl   	    	    = cmd.alt_dest.sl;
	attr->alt_ah_attr.src_path_bits     = cmd.alt_dest.src_path_bits;
	attr->alt_ah_attr.static_rate       = cmd.alt_dest.static_rate;
	attr->alt_ah_attr.ah_flags 	    = cmd.alt_dest.is_global ? IB_AH_GRH : 0;
	attr->alt_ah_attr.port_num 	    = cmd.alt_dest.port_num;

	ret = ib_modify_qp(qp, attr, cmd.attr_mask);
	if (ret)
		goto out;

	ret = in_len;

out:
	mutex_unlock(&ib_uverbs_idr_mutex);
	kfree(attr);

	return ret;
}

ssize_t ib_uverbs_destroy_qp(struct ib_uverbs_file *file,
			     const char __user *buf, int in_len,
			     int out_len)
{
	struct ib_uverbs_destroy_qp      cmd;
	struct ib_uverbs_destroy_qp_resp resp;
	struct ib_qp               	*qp;
	struct ib_uqp_object        	*uobj;
	int                        	 ret = -EINVAL;

	if (copy_from_user(&cmd, buf, sizeof cmd))
		return -EFAULT;

	memset(&resp, 0, sizeof resp);

	mutex_lock(&ib_uverbs_idr_mutex);

	qp = idr_find(&ib_uverbs_qp_idr, cmd.qp_handle);
	if (!qp || qp->uobject->context != file->ucontext)
		goto out;

	uobj = container_of(qp->uobject, struct ib_uqp_object, uevent.uobject);

	if (!list_empty(&uobj->mcast_list)) {
		ret = -EBUSY;
		goto out;
	}

	ret = ib_destroy_qp(qp);
	if (ret)
		goto out;

	idr_remove(&ib_uverbs_qp_idr, cmd.qp_handle);

	mutex_lock(&file->mutex);
	list_del(&uobj->uevent.uobject.list);
	mutex_unlock(&file->mutex);

	ib_uverbs_release_uevent(file, &uobj->uevent);

	resp.events_reported = uobj->uevent.events_reported;

	kfree(uobj);

	if (copy_to_user((void __user *) (unsigned long) cmd.response,
			 &resp, sizeof resp))
		ret = -EFAULT;

out:
	mutex_unlock(&ib_uverbs_idr_mutex);

	return ret ? ret : in_len;
}

ssize_t ib_uverbs_post_send(struct ib_uverbs_file *file,
                            const char __user *buf, int in_len,
                            int out_len)
{
	struct ib_uverbs_post_send      cmd;
	struct ib_uverbs_post_send_resp resp;
	struct ib_uverbs_send_wr       *user_wr;
	struct ib_send_wr              *wr = NULL, *last, *next, *bad_wr;
	struct ib_qp                   *qp;
	int                             i, sg_ind;
	ssize_t                         ret = -EINVAL;

	if (copy_from_user(&cmd, buf, sizeof cmd))
		return -EFAULT;

	if (in_len < sizeof cmd + cmd.wqe_size * cmd.wr_count +
	    cmd.sge_count * sizeof (struct ib_uverbs_sge))
		return -EINVAL;

	if (cmd.wqe_size < sizeof (struct ib_uverbs_send_wr))
		return -EINVAL;

	user_wr = kmalloc(cmd.wqe_size, GFP_KERNEL);
	if (!user_wr)
		return -ENOMEM;

	mutex_lock(&ib_uverbs_idr_mutex);

	qp = idr_find(&ib_uverbs_qp_idr, cmd.qp_handle);
	if (!qp || qp->uobject->context != file->ucontext)
		goto out;

	sg_ind = 0;
	last = NULL;
	for (i = 0; i < cmd.wr_count; ++i) {
		if (copy_from_user(user_wr,
				   buf + sizeof cmd + i * cmd.wqe_size,
				   cmd.wqe_size)) {
			ret = -EFAULT;
			goto out;
		}

		if (user_wr->num_sge + sg_ind > cmd.sge_count) {
			ret = -EINVAL;
			goto out;
		}

		next = kmalloc(ALIGN(sizeof *next, sizeof (struct ib_sge)) +
			       user_wr->num_sge * sizeof (struct ib_sge),
			       GFP_KERNEL);
		if (!next) {
			ret = -ENOMEM;
			goto out;
		}

		if (!last)
			wr = next;
		else
			last->next = next;
		last = next;

		next->next       = NULL;
		next->wr_id      = user_wr->wr_id;
		next->num_sge    = user_wr->num_sge;
		next->opcode     = user_wr->opcode;
		next->send_flags = user_wr->send_flags;
		next->imm_data   = (__be32 __force) user_wr->imm_data;

		if (qp->qp_type == IB_QPT_UD) {
			next->wr.ud.ah = idr_find(&ib_uverbs_ah_idr,
						  user_wr->wr.ud.ah);
			if (!next->wr.ud.ah) {
				ret = -EINVAL;
				goto out;
			}
			next->wr.ud.remote_qpn  = user_wr->wr.ud.remote_qpn;
			next->wr.ud.remote_qkey = user_wr->wr.ud.remote_qkey;
		} else {
			switch (next->opcode) {
			case IB_WR_RDMA_WRITE:
			case IB_WR_RDMA_WRITE_WITH_IMM:
			case IB_WR_RDMA_READ:
				next->wr.rdma.remote_addr =
					user_wr->wr.rdma.remote_addr;
				next->wr.rdma.rkey        =
					user_wr->wr.rdma.rkey;
				break;
			case IB_WR_ATOMIC_CMP_AND_SWP:
			case IB_WR_ATOMIC_FETCH_AND_ADD:
				next->wr.atomic.remote_addr =
					user_wr->wr.atomic.remote_addr;
				next->wr.atomic.compare_add =
					user_wr->wr.atomic.compare_add;
				next->wr.atomic.swap = user_wr->wr.atomic.swap;
				next->wr.atomic.rkey = user_wr->wr.atomic.rkey;
				break;
			default:
				break;
			}
		}

		if (next->num_sge) {
			next->sg_list = (void *) next +
				ALIGN(sizeof *next, sizeof (struct ib_sge));
			if (copy_from_user(next->sg_list,
					   buf + sizeof cmd +
					   cmd.wr_count * cmd.wqe_size +
					   sg_ind * sizeof (struct ib_sge),
					   next->num_sge * sizeof (struct ib_sge))) {
				ret = -EFAULT;
				goto out;
			}
			sg_ind += next->num_sge;
		} else
			next->sg_list = NULL;
	}

	resp.bad_wr = 0;
	ret = qp->device->post_send(qp, wr, &bad_wr);
	if (ret)
		for (next = wr; next; next = next->next) {
			++resp.bad_wr;
			if (next == bad_wr)
				break;
		}

	if (copy_to_user((void __user *) (unsigned long) cmd.response,
			 &resp, sizeof resp))
		ret = -EFAULT;