#include <linux/log2.h>
#include <linux/slab.h>
+#include <linux/netdevice.h>
#include <rdma/ib_cache.h>
#include <rdma/ib_pack.h>
+#include <rdma/ib_addr.h>
#include <linux/mlx4/qp.h>
enum {
MLX4_IB_DEFAULT_SCHED_QUEUE = 0x83,
- MLX4_IB_DEFAULT_QP0_SCHED_QUEUE = 0x3f
+ MLX4_IB_DEFAULT_QP0_SCHED_QUEUE = 0x3f,
+ MLX4_IB_LINK_TYPE_IB = 0,
+ MLX4_IB_LINK_TYPE_ETH = 1
};
enum {
/*
- * Largest possible UD header: send with GRH and immediate data.
+ * Largest possible UD header: send with GRH and immediate
+ * data plus 18 bytes for an Ethernet header with VLAN/802.1Q
+ * tag. (LRH would only use 8 bytes, so Ethernet is the
+ * biggest case)
*/
- MLX4_IB_UD_HEADER_SIZE = 72,
+ MLX4_IB_UD_HEADER_SIZE = 82,
MLX4_IB_LSO_HEADER_SPARE = 128,
};
+enum {
+ MLX4_IB_IBOE_ETHERTYPE = 0x8915
+};
+
struct mlx4_ib_sqp {
struct mlx4_ib_qp qp;
int pkey_index;
mutex_init(&qp->mutex);
spin_lock_init(&qp->sq.lock);
spin_lock_init(&qp->rq.lock);
+ INIT_LIST_HEAD(&qp->gid_list);
qp->state = IB_QPS_RESET;
if (init_attr->sq_sig_type == IB_SIGNAL_ALL_WR)
}
}
+static void del_gid_entries(struct mlx4_ib_qp *qp)
+{
+ struct mlx4_ib_gid_entry *ge, *tmp;
+
+ list_for_each_entry_safe(ge, tmp, &qp->gid_list, list) {
+ list_del(&ge->list);
+ kfree(ge);
+ }
+}
+
static void destroy_qp_common(struct mlx4_ib_dev *dev, struct mlx4_ib_qp *qp,
int is_user)
{
if (!qp->ibqp.srq)
mlx4_db_free(dev->dev, &qp->db);
}
+
+ del_gid_entries(qp);
}
struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
static int mlx4_set_path(struct mlx4_ib_dev *dev, const struct ib_ah_attr *ah,
struct mlx4_qp_path *path, u8 port)
{
+ int err;
+ int is_eth = rdma_port_get_link_layer(&dev->ib_dev, port) ==
+ IB_LINK_LAYER_ETHERNET;
+ u8 mac[6];
+ int is_mcast;
+ u16 vlan_tag;
+ int vidx;
+
path->grh_mylmc = ah->src_path_bits & 0x7f;
path->rlid = cpu_to_be16(ah->dlid);
if (ah->static_rate) {
memcpy(path->rgid, ah->grh.dgid.raw, 16);
}
- path->sched_queue = MLX4_IB_DEFAULT_SCHED_QUEUE |
- ((port - 1) << 6) | ((ah->sl & 0xf) << 2);
+ if (is_eth) {
+ path->sched_queue = MLX4_IB_DEFAULT_SCHED_QUEUE |
+ ((port - 1) << 6) | ((ah->sl & 7) << 3) | ((ah->sl & 8) >> 1);
+
+ if (!(ah->ah_flags & IB_AH_GRH))
+ return -1;
+
+ err = mlx4_ib_resolve_grh(dev, ah, mac, &is_mcast, port);
+ if (err)
+ return err;
+
+ memcpy(path->dmac, mac, 6);
+ path->ackto = MLX4_IB_LINK_TYPE_ETH;
+ /* use index 0 into MAC table for IBoE */
+ path->grh_mylmc &= 0x80;
+
+ vlan_tag = rdma_get_vlan_id(&dev->iboe.gid_table[port - 1][ah->grh.sgid_index]);
+ if (vlan_tag < 0x1000) {
+ if (mlx4_find_cached_vlan(dev->dev, port, vlan_tag, &vidx))
+ return -ENOENT;
+
+ path->vlan_index = vidx;
+ path->fl = 1 << 6;
+ }
+ } else
+ path->sched_queue = MLX4_IB_DEFAULT_SCHED_QUEUE |
+ ((port - 1) << 6) | ((ah->sl & 0xf) << 2);
return 0;
}
+static void update_mcg_macs(struct mlx4_ib_dev *dev, struct mlx4_ib_qp *qp)
+{
+ struct mlx4_ib_gid_entry *ge, *tmp;
+
+ list_for_each_entry_safe(ge, tmp, &qp->gid_list, list) {
+ if (!ge->added && mlx4_ib_add_mc(dev, qp, &ge->gid)) {
+ ge->added = 1;
+ ge->port = qp->port;
+ }
+ }
+}
+
static int __mlx4_ib_modify_qp(struct ib_qp *ibqp,
const struct ib_qp_attr *attr, int attr_mask,
enum ib_qp_state cur_state, enum ib_qp_state new_state)
}
if (attr_mask & IB_QP_TIMEOUT) {
- context->pri_path.ackto = attr->timeout << 3;
+ context->pri_path.ackto |= attr->timeout << 3;
optpar |= MLX4_QP_OPTPAR_ACK_TIMEOUT;
}
qp->atomic_rd_en = attr->qp_access_flags;
if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC)
qp->resp_depth = attr->max_dest_rd_atomic;
- if (attr_mask & IB_QP_PORT)
+ if (attr_mask & IB_QP_PORT) {
qp->port = attr->port_num;
+ update_mcg_macs(dev, qp);
+ }
if (attr_mask & IB_QP_ALT_PATH)
qp->alt_port = attr->alt_port_num;
struct mlx4_wqe_mlx_seg *mlx = wqe;
struct mlx4_wqe_inline_seg *inl = wqe + sizeof *mlx;
struct mlx4_ib_ah *ah = to_mah(wr->wr.ud.ah);
+ union ib_gid sgid;
u16 pkey;
int send_size;
int header_size;
int spc;
int i;
+ int is_eth;
+ int is_vlan = 0;
+ int is_grh;
+ u16 vlan;
send_size = 0;
for (i = 0; i < wr->num_sge; ++i)
send_size += wr->sg_list[i].length;
- ib_ud_header_init(send_size, mlx4_ib_ah_grh_present(ah), 0, &sqp->ud_header);
+ is_eth = rdma_port_get_link_layer(sqp->qp.ibqp.device, sqp->qp.port) == IB_LINK_LAYER_ETHERNET;
+ is_grh = mlx4_ib_ah_grh_present(ah);
+ if (is_eth) {
+ ib_get_cached_gid(ib_dev, be32_to_cpu(ah->av.ib.port_pd) >> 24,
+ ah->av.ib.gid_index, &sgid);
+ vlan = rdma_get_vlan_id(&sgid);
+ is_vlan = vlan < 0x1000;
+ }
+ ib_ud_header_init(send_size, !is_eth, is_eth, is_vlan, is_grh, 0, &sqp->ud_header);
+
+ if (!is_eth) {
+ sqp->ud_header.lrh.service_level =
+ be32_to_cpu(ah->av.ib.sl_tclass_flowlabel) >> 28;
+ sqp->ud_header.lrh.destination_lid = ah->av.ib.dlid;
+ sqp->ud_header.lrh.source_lid = cpu_to_be16(ah->av.ib.g_slid & 0x7f);
+ }
- sqp->ud_header.lrh.service_level =
- be32_to_cpu(ah->av.sl_tclass_flowlabel) >> 28;
- sqp->ud_header.lrh.destination_lid = ah->av.dlid;
- sqp->ud_header.lrh.source_lid = cpu_to_be16(ah->av.g_slid & 0x7f);
- if (mlx4_ib_ah_grh_present(ah)) {
+ if (is_grh) {
sqp->ud_header.grh.traffic_class =
- (be32_to_cpu(ah->av.sl_tclass_flowlabel) >> 20) & 0xff;
+ (be32_to_cpu(ah->av.ib.sl_tclass_flowlabel) >> 20) & 0xff;
sqp->ud_header.grh.flow_label =
- ah->av.sl_tclass_flowlabel & cpu_to_be32(0xfffff);
- sqp->ud_header.grh.hop_limit = ah->av.hop_limit;
- ib_get_cached_gid(ib_dev, be32_to_cpu(ah->av.port_pd) >> 24,
- ah->av.gid_index, &sqp->ud_header.grh.source_gid);
+ ah->av.ib.sl_tclass_flowlabel & cpu_to_be32(0xfffff);
+ sqp->ud_header.grh.hop_limit = ah->av.ib.hop_limit;
+ ib_get_cached_gid(ib_dev, be32_to_cpu(ah->av.ib.port_pd) >> 24,
+ ah->av.ib.gid_index, &sqp->ud_header.grh.source_gid);
memcpy(sqp->ud_header.grh.destination_gid.raw,
- ah->av.dgid, 16);
+ ah->av.ib.dgid, 16);
}
mlx->flags &= cpu_to_be32(MLX4_WQE_CTRL_CQ_UPDATE);
- mlx->flags |= cpu_to_be32((!sqp->qp.ibqp.qp_num ? MLX4_WQE_MLX_VL15 : 0) |
- (sqp->ud_header.lrh.destination_lid ==
- IB_LID_PERMISSIVE ? MLX4_WQE_MLX_SLR : 0) |
- (sqp->ud_header.lrh.service_level << 8));
- mlx->rlid = sqp->ud_header.lrh.destination_lid;
+
+ if (!is_eth) {
+ mlx->flags |= cpu_to_be32((!sqp->qp.ibqp.qp_num ? MLX4_WQE_MLX_VL15 : 0) |
+ (sqp->ud_header.lrh.destination_lid ==
+ IB_LID_PERMISSIVE ? MLX4_WQE_MLX_SLR : 0) |
+ (sqp->ud_header.lrh.service_level << 8));
+ mlx->rlid = sqp->ud_header.lrh.destination_lid;
+ }
switch (wr->opcode) {
case IB_WR_SEND:
return -EINVAL;
}
- sqp->ud_header.lrh.virtual_lane = !sqp->qp.ibqp.qp_num ? 15 : 0;
- if (sqp->ud_header.lrh.destination_lid == IB_LID_PERMISSIVE)
- sqp->ud_header.lrh.source_lid = IB_LID_PERMISSIVE;
+ if (is_eth) {
+ u8 *smac;
+
+ memcpy(sqp->ud_header.eth.dmac_h, ah->av.eth.mac, 6);
+ /* FIXME: cache smac value? */
+ smac = to_mdev(sqp->qp.ibqp.device)->iboe.netdevs[sqp->qp.port - 1]->dev_addr;
+ memcpy(sqp->ud_header.eth.smac_h, smac, 6);
+ if (!memcmp(sqp->ud_header.eth.smac_h, sqp->ud_header.eth.dmac_h, 6))
+ mlx->flags |= cpu_to_be32(MLX4_WQE_CTRL_FORCE_LOOPBACK);
+ if (!is_vlan) {
+ sqp->ud_header.eth.type = cpu_to_be16(MLX4_IB_IBOE_ETHERTYPE);
+ } else {
+ u16 pcp;
+
+ sqp->ud_header.vlan.type = cpu_to_be16(MLX4_IB_IBOE_ETHERTYPE);
+ pcp = (be32_to_cpu(ah->av.ib.sl_tclass_flowlabel) >> 27 & 3) << 13;
+ sqp->ud_header.vlan.tag = cpu_to_be16(vlan | pcp);
+ }
+ } else {
+ sqp->ud_header.lrh.virtual_lane = !sqp->qp.ibqp.qp_num ? 15 : 0;
+ if (sqp->ud_header.lrh.destination_lid == IB_LID_PERMISSIVE)
+ sqp->ud_header.lrh.source_lid = IB_LID_PERMISSIVE;
+ }
sqp->ud_header.bth.solicited_event = !!(wr->send_flags & IB_SEND_SOLICITED);
if (!sqp->qp.ibqp.qp_num)
ib_get_cached_pkey(ib_dev, sqp->qp.port, sqp->pkey_index, &pkey);
}
static void set_datagram_seg(struct mlx4_wqe_datagram_seg *dseg,
- struct ib_send_wr *wr)
+ struct ib_send_wr *wr, __be16 *vlan)
{
memcpy(dseg->av, &to_mah(wr->wr.ud.ah)->av, sizeof (struct mlx4_av));
dseg->dqpn = cpu_to_be32(wr->wr.ud.remote_qpn);
dseg->qkey = cpu_to_be32(wr->wr.ud.remote_qkey);
+ dseg->vlan = to_mah(wr->wr.ud.ah)->av.eth.vlan;
+ memcpy(dseg->mac, to_mah(wr->wr.ud.ah)->av.eth.mac, 6);
+ *vlan = dseg->vlan;
}
static void set_mlx_icrc_seg(void *dseg)
__be32 uninitialized_var(lso_hdr_sz);
__be32 blh;
int i;
+ __be16 vlan = cpu_to_be16(0xffff);
spin_lock_irqsave(&qp->sq.lock, flags);
break;
case IB_QPT_UD:
- set_datagram_seg(wqe, wr);
+ set_datagram_seg(wqe, wr, &vlan);
wqe += sizeof (struct mlx4_wqe_datagram_seg);
size += sizeof (struct mlx4_wqe_datagram_seg) / 16;
ctrl->fence_size = (wr->send_flags & IB_SEND_FENCE ?
MLX4_WQE_CTRL_FENCE : 0) | size;
+ if (be16_to_cpu(vlan) < 0x1000) {
+ ctrl->ins_vlan = 1 << 6;
+ ctrl->vlan_tag = vlan;
+ }
+
/*
* Make sure descriptor is fully written before
* setting ownership bit (because HW can start
return ib_flags;
}
-static void to_ib_ah_attr(struct mlx4_dev *dev, struct ib_ah_attr *ib_ah_attr,
+static void to_ib_ah_attr(struct mlx4_ib_dev *ibdev, struct ib_ah_attr *ib_ah_attr,
struct mlx4_qp_path *path)
{
+ struct mlx4_dev *dev = ibdev->dev;
+ int is_eth;
+
memset(ib_ah_attr, 0, sizeof *ib_ah_attr);
ib_ah_attr->port_num = path->sched_queue & 0x40 ? 2 : 1;
if (ib_ah_attr->port_num == 0 || ib_ah_attr->port_num > dev->caps.num_ports)
return;
+ is_eth = rdma_port_get_link_layer(&ibdev->ib_dev, ib_ah_attr->port_num) ==
+ IB_LINK_LAYER_ETHERNET;
+ if (is_eth)
+ ib_ah_attr->sl = ((path->sched_queue >> 3) & 0x7) |
+ ((path->sched_queue & 4) << 1);
+ else
+ ib_ah_attr->sl = (path->sched_queue >> 2) & 0xf;
+
ib_ah_attr->dlid = be16_to_cpu(path->rlid);
- ib_ah_attr->sl = (path->sched_queue >> 2) & 0xf;
ib_ah_attr->src_path_bits = path->grh_mylmc & 0x7f;
ib_ah_attr->static_rate = path->static_rate ? path->static_rate - 5 : 0;
ib_ah_attr->ah_flags = (path->grh_mylmc & (1 << 7)) ? IB_AH_GRH : 0;
to_ib_qp_access_flags(be32_to_cpu(context.params2));
if (qp->ibqp.qp_type == IB_QPT_RC || qp->ibqp.qp_type == IB_QPT_UC) {
- to_ib_ah_attr(dev->dev, &qp_attr->ah_attr, &context.pri_path);
- to_ib_ah_attr(dev->dev, &qp_attr->alt_ah_attr, &context.alt_path);
+ to_ib_ah_attr(dev, &qp_attr->ah_attr, &context.pri_path);
+ to_ib_ah_attr(dev, &qp_attr->alt_ah_attr, &context.alt_path);
qp_attr->alt_pkey_index = context.alt_path.pkey_index & 0x7f;
qp_attr->alt_port_num = qp_attr->alt_ah_attr.port_num;
}
projects / mv-sheeva.git / blobdiff