[ewg] [PATCH 2/2 v3] IB/mlx4: Add send with invalidate support
Eli Cohen
eli at dev.mellanox.co.il
Sun Jun 1 08:05:08 PDT 2008
>From f463e695702406987a0365ebbaae0a6cf344fd23 Mon Sep 17 00:00:00 2001
From: Eli Cohen <eli at mellanox.co.il>
Date: Sun, 11 May 2008 15:02:04 +0300
Subject:
Add send with invalidate support to mlx4.
Signed-off-by: Eli Cohen <eli at mellanox.co.il>
---
Changes since v2:
use query dev cap to check device capabilities regarding send with invalidate.
drivers/infiniband/hw/mlx4/cq.c | 8 ++++++++
drivers/infiniband/hw/mlx4/main.c | 2 ++
drivers/infiniband/hw/mlx4/qp.c | 22 +++++++++++++++++-----
drivers/net/mlx4/fw.c | 9 +--------
drivers/net/mlx4/fw.h | 2 +-
drivers/net/mlx4/mr.c | 6 ++++--
include/linux/mlx4/device.h | 7 ++++++-
7 files changed, 39 insertions(+), 17 deletions(-)
diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 4521319..291e856 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -637,6 +637,7 @@ repoll:
case MLX4_OPCODE_SEND_IMM:
wc->wc_flags |= IB_WC_WITH_IMM;
case MLX4_OPCODE_SEND:
+ case MLX4_OPCODE_SEND_INVAL:
wc->opcode = IB_WC_SEND;
break;
case MLX4_OPCODE_RDMA_READ:
@@ -676,6 +677,13 @@ repoll:
wc->wc_flags = IB_WC_WITH_IMM;
wc->imm_data = cqe->immed_rss_invalid;
break;
+ case MLX4_RECV_OPCODE_SEND_INVAL:
+ wc->opcode = IB_WC_RECV;
+ wc->wc_flags = IB_WC_WITH_INVALIDATE;
+ /*
+ * TBD: maybe we should just call this ieth_val
+ */
+ wc->imm_data = be32_to_cpu(cqe->immed_rss_invalid);
}
wc->slid = be16_to_cpu(cqe->rlid);
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 4d61e32..a574050 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -103,6 +103,8 @@ static int mlx4_ib_query_device(struct ib_device *ibdev,
props->device_cap_flags |= IB_DEVICE_UD_IP_CSUM;
if (dev->dev->caps.max_gso_sz)
props->device_cap_flags |= IB_DEVICE_UD_TSO;
+ if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_RI)
+ props->device_cap_flags |= IB_DEVICE_SEND_W_INV;
props->vendor_id = be32_to_cpup((__be32 *) (out_mad->data + 36)) &
0xffffff;
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index a80df22..293b57e 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -78,6 +78,7 @@ static const __be32 mlx4_ib_opcode[] = {
[IB_WR_RDMA_READ] = __constant_cpu_to_be32(MLX4_OPCODE_RDMA_READ),
[IB_WR_ATOMIC_CMP_AND_SWP] = __constant_cpu_to_be32(MLX4_OPCODE_ATOMIC_CS),
[IB_WR_ATOMIC_FETCH_AND_ADD] = __constant_cpu_to_be32(MLX4_OPCODE_ATOMIC_FA),
+ [IB_WR_SEND_WITH_INV] = __constant_cpu_to_be32(MLX4_OPCODE_SEND_INVAL),
};
static struct mlx4_ib_sqp *to_msqp(struct mlx4_ib_qp *mqp)
@@ -1447,6 +1448,21 @@ static int build_lso_seg(struct mlx4_lso_seg *wqe, struct ib_send_wr *wr,
return 0;
}
+static __be32 get_ieth(struct ib_send_wr *wr)
+{
+ switch (wr->opcode) {
+ case IB_WR_SEND_WITH_IMM:
+ case IB_WR_RDMA_WRITE_WITH_IMM:
+ return wr->ex.imm_data;
+
+ case IB_WR_SEND_WITH_INV:
+ return cpu_to_be32(wr->ex.invalidate_rkey);
+
+ default:
+ return 0;
+ }
+}
+
int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
struct ib_send_wr **bad_wr)
{
@@ -1493,11 +1509,7 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
MLX4_WQE_CTRL_TCP_UDP_CSUM) : 0) |
qp->sq_signal_bits;
- if (wr->opcode == IB_WR_SEND_WITH_IMM ||
- wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM)
- ctrl->imm = wr->ex.imm_data;
- else
- ctrl->imm = 0;
+ ctrl->imm = get_ieth(wr);
wqe += sizeof *ctrl;
size = sizeof *ctrl / 16;
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index d82f275..02bff1d 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -170,7 +170,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
#define QUERY_DEV_CAP_C_MPT_ENTRY_SZ_OFFSET 0x8e
#define QUERY_DEV_CAP_MTT_ENTRY_SZ_OFFSET 0x90
#define QUERY_DEV_CAP_D_MPT_ENTRY_SZ_OFFSET 0x92
-#define QUERY_DEV_CAP_BMME_FLAGS_OFFSET 0x97
+#define QUERY_DEV_CAP_BMME_FLAGS_OFFSET 0x94
#define QUERY_DEV_CAP_RSVD_LKEY_OFFSET 0x98
#define QUERY_DEV_CAP_MAX_ICM_SZ_OFFSET 0xa0
@@ -345,13 +345,6 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
}
}
- if (dev_cap->bmme_flags & 1)
- mlx4_dbg(dev, "Base MM extensions: yes "
- "(flags %d, rsvd L_Key %08x)\n",
- dev_cap->bmme_flags, dev_cap->reserved_lkey);
- else
- mlx4_dbg(dev, "Base MM extensions: no\n");
-
/*
* Each UAR has 4 EQ doorbells; so if a UAR is reserved, then
* we can't use any EQs whose doorbell falls on that page,
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index 306cb9b..11544d5 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -93,7 +93,7 @@ struct mlx4_dev_cap {
int cmpt_entry_sz;
int mtt_entry_sz;
int resize_srq;
- u8 bmme_flags;
+ u32 bmme_flags;
u32 reserved_lkey;
u64 max_icm_sz;
int max_gso_sz;
diff --git a/drivers/net/mlx4/mr.c b/drivers/net/mlx4/mr.c
index 03a9abc..e78f53d 100644
--- a/drivers/net/mlx4/mr.c
+++ b/drivers/net/mlx4/mr.c
@@ -47,7 +47,7 @@ struct mlx4_mpt_entry {
__be32 flags;
__be32 qpn;
__be32 key;
- __be32 pd;
+ __be32 pd_flags;
__be64 start;
__be64 length;
__be32 lkey;
@@ -71,6 +71,8 @@ struct mlx4_mpt_entry {
#define MLX4_MPT_STATUS_SW 0xF0
#define MLX4_MPT_STATUS_HW 0x00
+#define MLX4_MPT_FLAG_EN_INV 0x3000000
+
static u32 mlx4_buddy_alloc(struct mlx4_buddy *buddy, int order)
{
int o;
@@ -320,7 +322,7 @@ int mlx4_mr_enable(struct mlx4_dev *dev, struct mlx4_mr *mr)
mr->access);
mpt_entry->key = cpu_to_be32(key_to_hw_index(mr->key));
- mpt_entry->pd = cpu_to_be32(mr->pd);
+ mpt_entry->pd_flags = cpu_to_be32(mr->pd | MLX4_MPT_FLAG_EN_INV);
mpt_entry->start = cpu_to_be64(mr->iova);
mpt_entry->length = cpu_to_be64(mr->size);
mpt_entry->entity_size = cpu_to_be32(mr->mtt.page_shift);
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index a744383..b7eccca 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -65,7 +65,12 @@ enum {
MLX4_DEV_CAP_FLAG_ATOMIC = 1 << 18,
MLX4_DEV_CAP_FLAG_RAW_MCAST = 1 << 19,
MLX4_DEV_CAP_FLAG_UD_AV_PORT = 1 << 20,
- MLX4_DEV_CAP_FLAG_UD_MCAST = 1 << 21
+ MLX4_DEV_CAP_FLAG_UD_MCAST = 1 << 21,
+ MLX4_DEV_CAP_FLAG_LI = 1 << 6,
+ MLX4_DEV_CAP_FLAG_RI = 1 << 7,
+ MLX4_DEV_CAP_FLAG_TYPE2WIN = 1 << 9,
+ MLX4_DEV_CAP_FLAG_RLKEY = 1 << 10,
+ MLX4_DEV_CAP_FLAG_FRWR = 1 << 11
};
enum mlx4_event {
--
1.5.5.1
More information about the ewg
mailing list