summaryrefslogtreecommitdiffstats
path: root/drivers/infiniband/hw/mthca
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/infiniband/hw/mthca')
-rw-r--r--drivers/infiniband/hw/mthca/mthca_cmd.c9
-rw-r--r--drivers/infiniband/hw/mthca/mthca_dev.h2
-rw-r--r--drivers/infiniband/hw/mthca/mthca_main.c102
-rw-r--r--drivers/infiniband/hw/mthca/mthca_qp.c221
-rw-r--r--drivers/infiniband/hw/mthca/mthca_srq.c36
-rw-r--r--drivers/infiniband/hw/mthca/mthca_wqe.h15
6 files changed, 190 insertions, 195 deletions
diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.c b/drivers/infiniband/hw/mthca/mthca_cmd.c
index f40558d7647..6966f943f44 100644
--- a/drivers/infiniband/hw/mthca/mthca_cmd.c
+++ b/drivers/infiniband/hw/mthca/mthca_cmd.c
@@ -290,6 +290,12 @@ static int mthca_cmd_post(struct mthca_dev *dev,
err = mthca_cmd_post_hcr(dev, in_param, out_param, in_modifier,
op_modifier, op, token, event);
+ /*
+ * Make sure that our HCR writes don't get mixed in with
+ * writes from another CPU starting a FW command.
+ */
+ mmiowb();
+
mutex_unlock(&dev->cmd.hcr_mutex);
return err;
}
@@ -357,8 +363,6 @@ void mthca_cmd_event(struct mthca_dev *dev,
context->status = status;
context->out_param = out_param;
- context->token += dev->cmd.token_mask + 1;
-
complete(&context->done);
}
@@ -380,6 +384,7 @@ static int mthca_cmd_wait(struct mthca_dev *dev,
spin_lock(&dev->cmd.context_lock);
BUG_ON(dev->cmd.free_head < 0);
context = &dev->cmd.context[dev->cmd.free_head];
+ context->token += dev->cmd.token_mask + 1;
dev->cmd.free_head = context->next;
spin_unlock(&dev->cmd.context_lock);
diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h
index 9bae3cc6060..15aa32eb78b 100644
--- a/drivers/infiniband/hw/mthca/mthca_dev.h
+++ b/drivers/infiniband/hw/mthca/mthca_dev.h
@@ -83,7 +83,7 @@ enum {
MTHCA_QP_CONTEXT_SIZE = 0x200,
MTHCA_RDB_ENTRY_SIZE = 0x20,
MTHCA_AV_SIZE = 0x20,
- MTHCA_MGM_ENTRY_SIZE = 0x40,
+ MTHCA_MGM_ENTRY_SIZE = 0x100,
/* Arbel FW gives us these, but we need them for Tavor */
MTHCA_MPT_ENTRY_SIZE = 0x40,
diff --git a/drivers/infiniband/hw/mthca/mthca_main.c b/drivers/infiniband/hw/mthca/mthca_main.c
index aa563e61de6..60de6f93869 100644
--- a/drivers/infiniband/hw/mthca/mthca_main.c
+++ b/drivers/infiniband/hw/mthca/mthca_main.c
@@ -61,13 +61,13 @@ MODULE_PARM_DESC(debug_level, "Enable debug tracing if > 0");
#ifdef CONFIG_PCI_MSI
-static int msi_x = 0;
+static int msi_x = 1;
module_param(msi_x, int, 0444);
MODULE_PARM_DESC(msi_x, "attempt to use MSI-X if nonzero");
static int msi = 0;
module_param(msi, int, 0444);
-MODULE_PARM_DESC(msi, "attempt to use MSI if nonzero");
+MODULE_PARM_DESC(msi, "attempt to use MSI if nonzero (deprecated, use MSI-X instead)");
#else /* CONFIG_PCI_MSI */
@@ -137,40 +137,23 @@ static const char mthca_version[] __devinitdata =
static int mthca_tune_pci(struct mthca_dev *mdev)
{
- int cap;
- u16 val;
-
if (!tune_pci)
return 0;
/* First try to max out Read Byte Count */
- cap = pci_find_capability(mdev->pdev, PCI_CAP_ID_PCIX);
- if (cap) {
- if (pci_read_config_word(mdev->pdev, cap + PCI_X_CMD, &val)) {
- mthca_err(mdev, "Couldn't read PCI-X command register, "
- "aborting.\n");
- return -ENODEV;
- }
- val = (val & ~PCI_X_CMD_MAX_READ) | (3 << 2);
- if (pci_write_config_word(mdev->pdev, cap + PCI_X_CMD, val)) {
- mthca_err(mdev, "Couldn't write PCI-X command register, "
- "aborting.\n");
+ if (pci_find_capability(mdev->pdev, PCI_CAP_ID_PCIX)) {
+ if (pcix_set_mmrbc(mdev->pdev, pcix_get_max_mmrbc(mdev->pdev))) {
+ mthca_err(mdev, "Couldn't set PCI-X max read count, "
+ "aborting.\n");
return -ENODEV;
}
} else if (!(mdev->mthca_flags & MTHCA_FLAG_PCIE))
mthca_info(mdev, "No PCI-X capability, not setting RBC.\n");
- cap = pci_find_capability(mdev->pdev, PCI_CAP_ID_EXP);
- if (cap) {
- if (pci_read_config_word(mdev->pdev, cap + PCI_EXP_DEVCTL, &val)) {
- mthca_err(mdev, "Couldn't read PCI Express device control "
- "register, aborting.\n");
- return -ENODEV;
- }
- val = (val & ~PCI_EXP_DEVCTL_READRQ) | (5 << 12);
- if (pci_write_config_word(mdev->pdev, cap + PCI_EXP_DEVCTL, val)) {
- mthca_err(mdev, "Couldn't write PCI Express device control "
- "register, aborting.\n");
+ if (pci_find_capability(mdev->pdev, PCI_CAP_ID_EXP)) {
+ if (pcie_set_readrq(mdev->pdev, 4096)) {
+ mthca_err(mdev, "Couldn't write PCI Express read request, "
+ "aborting.\n");
return -ENODEV;
}
} else if (mdev->mthca_flags & MTHCA_FLAG_PCIE)
@@ -833,14 +816,19 @@ static int mthca_setup_hca(struct mthca_dev *dev)
err = mthca_NOP(dev, &status);
if (err || status) {
- mthca_err(dev, "NOP command failed to generate interrupt (IRQ %d), aborting.\n",
- dev->mthca_flags & MTHCA_FLAG_MSI_X ?
- dev->eq_table.eq[MTHCA_EQ_CMD].msi_x_vector :
- dev->pdev->irq);
- if (dev->mthca_flags & (MTHCA_FLAG_MSI | MTHCA_FLAG_MSI_X))
- mthca_err(dev, "Try again with MSI/MSI-X disabled.\n");
- else
+ if (dev->mthca_flags & (MTHCA_FLAG_MSI | MTHCA_FLAG_MSI_X)) {
+ mthca_warn(dev, "NOP command failed to generate interrupt "
+ "(IRQ %d).\n",
+ dev->mthca_flags & MTHCA_FLAG_MSI_X ?
+ dev->eq_table.eq[MTHCA_EQ_CMD].msi_x_vector :
+ dev->pdev->irq);
+ mthca_warn(dev, "Trying again with MSI/MSI-X disabled.\n");
+ } else {
+ mthca_err(dev, "NOP command failed to generate interrupt "
+ "(IRQ %d), aborting.\n",
+ dev->pdev->irq);
mthca_err(dev, "BIOS or ACPI interrupt routing problem?\n");
+ }
goto err_cmd_poll;
}
@@ -1115,12 +1103,6 @@ static int __mthca_init_one(struct pci_dev *pdev, int hca_type)
goto err_free_dev;
}
- if (msi_x && !mthca_enable_msi_x(mdev))
- mdev->mthca_flags |= MTHCA_FLAG_MSI_X;
- if (msi && !(mdev->mthca_flags & MTHCA_FLAG_MSI_X) &&
- !pci_enable_msi(pdev))
- mdev->mthca_flags |= MTHCA_FLAG_MSI;
-
if (mthca_cmd_init(mdev)) {
mthca_err(mdev, "Failed to init command interface, aborting.\n");
goto err_free_dev;
@@ -1135,7 +1117,7 @@ static int __mthca_init_one(struct pci_dev *pdev, int hca_type)
goto err_cmd;
if (mdev->fw_ver < mthca_hca_table[hca_type].latest_fw) {
- mthca_warn(mdev, "HCA FW version %d.%d.%3d is old (%d.%d.%3d is current).\n",
+ mthca_warn(mdev, "HCA FW version %d.%d.%03d is old (%d.%d.%03d is current).\n",
(int) (mdev->fw_ver >> 32), (int) (mdev->fw_ver >> 16) & 0xffff,
(int) (mdev->fw_ver & 0xffff),
(int) (mthca_hca_table[hca_type].latest_fw >> 32),
@@ -1144,7 +1126,35 @@ static int __mthca_init_one(struct pci_dev *pdev, int hca_type)
mthca_warn(mdev, "If you have problems, try updating your HCA FW.\n");
}
+ if (msi_x && !mthca_enable_msi_x(mdev))
+ mdev->mthca_flags |= MTHCA_FLAG_MSI_X;
+ else if (msi) {
+ static int warned;
+
+ if (!warned) {
+ printk(KERN_WARNING PFX "WARNING: MSI support will be "
+ "removed from the ib_mthca driver in January 2008.\n");
+ printk(KERN_WARNING " If you are using MSI and cannot "
+ "switch to MSI-X, please tell "
+ "<general@lists.openfabrics.org>.\n");
+ ++warned;
+ }
+
+ if (!pci_enable_msi(pdev))
+ mdev->mthca_flags |= MTHCA_FLAG_MSI;
+ }
+
err = mthca_setup_hca(mdev);
+ if (err == -EBUSY && (mdev->mthca_flags & (MTHCA_FLAG_MSI | MTHCA_FLAG_MSI_X))) {
+ if (mdev->mthca_flags & MTHCA_FLAG_MSI_X)
+ pci_disable_msix(pdev);
+ if (mdev->mthca_flags & MTHCA_FLAG_MSI)
+ pci_disable_msi(pdev);
+ mdev->mthca_flags &= ~(MTHCA_FLAG_MSI_X | MTHCA_FLAG_MSI);
+
+ err = mthca_setup_hca(mdev);
+ }
+
if (err)
goto err_close;
@@ -1180,17 +1190,17 @@ err_cleanup:
mthca_cleanup_uar_table(mdev);
err_close:
+ if (mdev->mthca_flags & MTHCA_FLAG_MSI_X)
+ pci_disable_msix(pdev);
+ if (mdev->mthca_flags & MTHCA_FLAG_MSI)
+ pci_disable_msi(pdev);
+
mthca_close_hca(mdev);
err_cmd:
mthca_cmd_cleanup(mdev);
err_free_dev:
- if (mdev->mthca_flags & MTHCA_FLAG_MSI_X)
- pci_disable_msix(pdev);
- if (mdev->mthca_flags & MTHCA_FLAG_MSI)
- pci_disable_msi(pdev);
-
ib_dealloc_device(&mdev->ib_dev);
err_free_res:
diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c
index 11f1d99db40..df01b2026a6 100644
--- a/drivers/infiniband/hw/mthca/mthca_qp.c
+++ b/drivers/infiniband/hw/mthca/mthca_qp.c
@@ -1578,6 +1578,45 @@ static inline int mthca_wq_overflow(struct mthca_wq *wq, int nreq,
return cur + nreq >= wq->max;
}
+static __always_inline void set_raddr_seg(struct mthca_raddr_seg *rseg,
+ u64 remote_addr, u32 rkey)
+{
+ rseg->raddr = cpu_to_be64(remote_addr);
+ rseg->rkey = cpu_to_be32(rkey);
+ rseg->reserved = 0;
+}
+
+static __always_inline void set_atomic_seg(struct mthca_atomic_seg *aseg,
+ struct ib_send_wr *wr)
+{
+ if (wr->opcode == IB_WR_ATOMIC_CMP_AND_SWP) {
+ aseg->swap_add = cpu_to_be64(wr->wr.atomic.swap);
+ aseg->compare = cpu_to_be64(wr->wr.atomic.compare_add);
+ } else {
+ aseg->swap_add = cpu_to_be64(wr->wr.atomic.compare_add);
+ aseg->compare = 0;
+ }
+
+}
+
+static void set_tavor_ud_seg(struct mthca_tavor_ud_seg *useg,
+ struct ib_send_wr *wr)
+{
+ useg->lkey = cpu_to_be32(to_mah(wr->wr.ud.ah)->key);
+ useg->av_addr = cpu_to_be64(to_mah(wr->wr.ud.ah)->avdma);
+ useg->dqpn = cpu_to_be32(wr->wr.ud.remote_qpn);
+ useg->qkey = cpu_to_be32(wr->wr.ud.remote_qkey);
+
+}
+
+static void set_arbel_ud_seg(struct mthca_arbel_ud_seg *useg,
+ struct ib_send_wr *wr)
+{
+ memcpy(useg->av, to_mah(wr->wr.ud.ah)->av, MTHCA_AV_SIZE);
+ useg->dqpn = cpu_to_be32(wr->wr.ud.remote_qpn);
+ useg->qkey = cpu_to_be32(wr->wr.ud.remote_qkey);
+}
+
int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
struct ib_send_wr **bad_wr)
{
@@ -1590,8 +1629,15 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
int nreq;
int i;
int size;
- int size0 = 0;
- u32 f0 = 0;
+ /*
+ * f0 and size0 are only used if nreq != 0, and they will
+ * always be initialized the first time through the main loop
+ * before nreq is incremented. So nreq cannot become non-zero
+ * without initializing f0 and size0, and they are in fact
+ * never used uninitialized.
+ */
+ int uninitialized_var(size0);
+ u32 uninitialized_var(f0);
int ind;
u8 op0 = 0;
@@ -1636,25 +1682,11 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
switch (wr->opcode) {
case IB_WR_ATOMIC_CMP_AND_SWP:
case IB_WR_ATOMIC_FETCH_AND_ADD:
- ((struct mthca_raddr_seg *) wqe)->raddr =
- cpu_to_be64(wr->wr.atomic.remote_addr);
- ((struct mthca_raddr_seg *) wqe)->rkey =
- cpu_to_be32(wr->wr.atomic.rkey);
- ((struct mthca_raddr_seg *) wqe)->reserved = 0;
-
+ set_raddr_seg(wqe, wr->wr.atomic.remote_addr,
+ wr->wr.atomic.rkey);
wqe += sizeof (struct mthca_raddr_seg);
- if (wr->opcode == IB_WR_ATOMIC_CMP_AND_SWP) {
- ((struct mthca_atomic_seg *) wqe)->swap_add =
- cpu_to_be64(wr->wr.atomic.swap);
- ((struct mthca_atomic_seg *) wqe)->compare =
- cpu_to_be64(wr->wr.atomic.compare_add);
- } else {
- ((struct mthca_atomic_seg *) wqe)->swap_add =
- cpu_to_be64(wr->wr.atomic.compare_add);
- ((struct mthca_atomic_seg *) wqe)->compare = 0;
- }
-
+ set_atomic_seg(wqe, wr);
wqe += sizeof (struct mthca_atomic_seg);
size += (sizeof (struct mthca_raddr_seg) +
sizeof (struct mthca_atomic_seg)) / 16;
@@ -1663,12 +1695,9 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
case IB_WR_RDMA_WRITE:
case IB_WR_RDMA_WRITE_WITH_IMM:
case IB_WR_RDMA_READ:
- ((struct mthca_raddr_seg *) wqe)->raddr =
- cpu_to_be64(wr->wr.rdma.remote_addr);
- ((struct mthca_raddr_seg *) wqe)->rkey =
- cpu_to_be32(wr->wr.rdma.rkey);
- ((struct mthca_raddr_seg *) wqe)->reserved = 0;
- wqe += sizeof (struct mthca_raddr_seg);
+ set_raddr_seg(wqe, wr->wr.rdma.remote_addr,
+ wr->wr.rdma.rkey);
+ wqe += sizeof (struct mthca_raddr_seg);
size += sizeof (struct mthca_raddr_seg) / 16;
break;
@@ -1683,12 +1712,9 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
switch (wr->opcode) {
case IB_WR_RDMA_WRITE:
case IB_WR_RDMA_WRITE_WITH_IMM:
- ((struct mthca_raddr_seg *) wqe)->raddr =
- cpu_to_be64(wr->wr.rdma.remote_addr);
- ((struct mthca_raddr_seg *) wqe)->rkey =
- cpu_to_be32(wr->wr.rdma.rkey);
- ((struct mthca_raddr_seg *) wqe)->reserved = 0;
- wqe += sizeof (struct mthca_raddr_seg);
+ set_raddr_seg(wqe, wr->wr.rdma.remote_addr,
+ wr->wr.rdma.rkey);
+ wqe += sizeof (struct mthca_raddr_seg);
size += sizeof (struct mthca_raddr_seg) / 16;
break;
@@ -1700,16 +1726,8 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
break;
case UD:
- ((struct mthca_tavor_ud_seg *) wqe)->lkey =
- cpu_to_be32(to_mah(wr->wr.ud.ah)->key);
- ((struct mthca_tavor_ud_seg *) wqe)->av_addr =
- cpu_to_be64(to_mah(wr->wr.ud.ah)->avdma);
- ((struct mthca_tavor_ud_seg *) wqe)->dqpn =
- cpu_to_be32(wr->wr.ud.remote_qpn);
- ((struct mthca_tavor_ud_seg *) wqe)->qkey =
- cpu_to_be32(wr->wr.ud.remote_qkey);
-
- wqe += sizeof (struct mthca_tavor_ud_seg);
+ set_tavor_ud_seg(wqe, wr);
+ wqe += sizeof (struct mthca_tavor_ud_seg);
size += sizeof (struct mthca_tavor_ud_seg) / 16;
break;
@@ -1734,13 +1752,8 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
}
for (i = 0; i < wr->num_sge; ++i) {
- ((struct mthca_data_seg *) wqe)->byte_count =
- cpu_to_be32(wr->sg_list[i].length);
- ((struct mthca_data_seg *) wqe)->lkey =
- cpu_to_be32(wr->sg_list[i].lkey);
- ((struct mthca_data_seg *) wqe)->addr =
- cpu_to_be64(wr->sg_list[i].addr);
- wqe += sizeof (struct mthca_data_seg);
+ mthca_set_data_seg(wqe, wr->sg_list + i);
+ wqe += sizeof (struct mthca_data_seg);
size += sizeof (struct mthca_data_seg) / 16;
}
@@ -1768,11 +1781,11 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
mthca_opcode[wr->opcode]);
wmb();
((struct mthca_next_seg *) prev_wqe)->ee_nds =
- cpu_to_be32((size0 ? 0 : MTHCA_NEXT_DBD) | size |
+ cpu_to_be32((nreq ? 0 : MTHCA_NEXT_DBD) | size |
((wr->send_flags & IB_SEND_FENCE) ?
MTHCA_NEXT_FENCE : 0));
- if (!size0) {
+ if (!nreq) {
size0 = size;
op0 = mthca_opcode[wr->opcode];
f0 = wr->send_flags & IB_SEND_FENCE ?
@@ -1822,7 +1835,14 @@ int mthca_tavor_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr,
int nreq;
int i;
int size;
- int size0 = 0;
+ /*
+ * size0 is only used if nreq != 0, and it will always be
+ * initialized the first time through the main loop before
+ * nreq is incremented. So nreq cannot become non-zero
+ * without initializing size0, and it is in fact never used
+ * uninitialized.
+ */
+ int uninitialized_var(size0);
int ind;
void *wqe;
void *prev_wqe;
@@ -1863,13 +1883,8 @@ int mthca_tavor_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr,
}
for (i = 0; i < wr->num_sge; ++i) {
- ((struct mthca_data_seg *) wqe)->byte_count =
- cpu_to_be32(wr->sg_list[i].length);
- ((struct mthca_data_seg *) wqe)->lkey =
- cpu_to_be32(wr->sg_list[i].lkey);
- ((struct mthca_data_seg *) wqe)->addr =
- cpu_to_be64(wr->sg_list[i].addr);
- wqe += sizeof (struct mthca_data_seg);
+ mthca_set_data_seg(wqe, wr->sg_list + i);
+ wqe += sizeof (struct mthca_data_seg);
size += sizeof (struct mthca_data_seg) / 16;
}
@@ -1881,7 +1896,7 @@ int mthca_tavor_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr,
((struct mthca_next_seg *) prev_wqe)->ee_nds =
cpu_to_be32(MTHCA_NEXT_DBD | size);
- if (!size0)
+ if (!nreq)
size0 = size;
++ind;
@@ -1903,7 +1918,6 @@ int mthca_tavor_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr,
qp->rq.next_ind = ind;
qp->rq.head += MTHCA_TAVOR_MAX_WQES_PER_RECV_DB;
- size0 = 0;
}
}
@@ -1945,8 +1959,15 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
int nreq;
int i;
int size;
- int size0 = 0;
- u32 f0 = 0;
+ /*
+ * f0 and size0 are only used if nreq != 0, and they will
+ * always be initialized the first time through the main loop
+ * before nreq is incremented. So nreq cannot become non-zero
+ * without initializing f0 and size0, and they are in fact
+ * never used uninitialized.
+ */
+ int uninitialized_var(size0);
+ u32 uninitialized_var(f0);
int ind;
u8 op0 = 0;
@@ -1966,7 +1987,6 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
doorbell[1] = cpu_to_be32((qp->qpn << 8) | size0);
qp->sq.head += MTHCA_ARBEL_MAX_WQES_PER_SEND_DB;
- size0 = 0;
/*
* Make sure that descriptors are written before
@@ -2017,26 +2037,12 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
switch (wr->opcode) {
case IB_WR_ATOMIC_CMP_AND_SWP:
case IB_WR_ATOMIC_FETCH_AND_ADD:
- ((struct mthca_raddr_seg *) wqe)->raddr =
- cpu_to_be64(wr->wr.atomic.remote_addr);
- ((struct mthca_raddr_seg *) wqe)->rkey =
- cpu_to_be32(wr->wr.atomic.rkey);
- ((struct mthca_raddr_seg *) wqe)->reserved = 0;
-
+ set_raddr_seg(wqe, wr->wr.atomic.remote_addr,
+ wr->wr.atomic.rkey);
wqe += sizeof (struct mthca_raddr_seg);
- if (wr->opcode == IB_WR_ATOMIC_CMP_AND_SWP) {
- ((struct mthca_atomic_seg *) wqe)->swap_add =
- cpu_to_be64(wr->wr.atomic.swap);
- ((struct mthca_atomic_seg *) wqe)->compare =
- cpu_to_be64(wr->wr.atomic.compare_add);
- } else {
- ((struct mthca_atomic_seg *) wqe)->swap_add =
- cpu_to_be64(wr->wr.atomic.compare_add);
- ((struct mthca_atomic_seg *) wqe)->compare = 0;
- }
-
- wqe += sizeof (struct mthca_atomic_seg);
+ set_atomic_seg(wqe, wr);
+ wqe += sizeof (struct mthca_atomic_seg);
size += (sizeof (struct mthca_raddr_seg) +
sizeof (struct mthca_atomic_seg)) / 16;
break;
@@ -2044,12 +2050,9 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
case IB_WR_RDMA_READ:
case IB_WR_RDMA_WRITE:
case IB_WR_RDMA_WRITE_WITH_IMM:
- ((struct mthca_raddr_seg *) wqe)->raddr =
- cpu_to_be64(wr->wr.rdma.remote_addr);
- ((struct mthca_raddr_seg *) wqe)->rkey =
- cpu_to_be32(wr->wr.rdma.rkey);
- ((struct mthca_raddr_seg *) wqe)->reserved = 0;
- wqe += sizeof (struct mthca_raddr_seg);
+ set_raddr_seg(wqe, wr->wr.rdma.remote_addr,
+ wr->wr.rdma.rkey);
+ wqe += sizeof (struct mthca_raddr_seg);
size += sizeof (struct mthca_raddr_seg) / 16;
break;
@@ -2064,12 +2067,9 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
switch (wr->opcode) {
case IB_WR_RDMA_WRITE:
case IB_WR_RDMA_WRITE_WITH_IMM:
- ((struct mthca_raddr_seg *) wqe)->raddr =
- cpu_to_be64(wr->wr.rdma.remote_addr);
- ((struct mthca_raddr_seg *) wqe)->rkey =
- cpu_to_be32(wr->wr.rdma.rkey);
- ((struct mthca_raddr_seg *) wqe)->reserved = 0;
- wqe += sizeof (struct mthca_raddr_seg);
+ set_raddr_seg(wqe, wr->wr.rdma.remote_addr,
+ wr->wr.rdma.rkey);
+ wqe += sizeof (struct mthca_raddr_seg);
size += sizeof (struct mthca_raddr_seg) / 16;
break;
@@ -2081,14 +2081,8 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
break;
case UD:
- memcpy(((struct mthca_arbel_ud_seg *) wqe)->av,
- to_mah(wr->wr.ud.ah)->av, MTHCA_AV_SIZE);
- ((struct mthca_arbel_ud_seg *) wqe)->dqpn =
- cpu_to_be32(wr->wr.ud.remote_qpn);
- ((struct mthca_arbel_ud_seg *) wqe)->qkey =
- cpu_to_be32(wr->wr.ud.remote_qkey);
-
- wqe += sizeof (struct mthca_arbel_ud_seg);
+ set_arbel_ud_seg(wqe, wr);
+ wqe += sizeof (struct mthca_arbel_ud_seg);
size += sizeof (struct mthca_arbel_ud_seg) / 16;
break;
@@ -2113,13 +2107,8 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
}
for (i = 0; i < wr->num_sge; ++i) {
- ((struct mthca_data_seg *) wqe)->byte_count =
- cpu_to_be32(wr->sg_list[i].length);
- ((struct mthca_data_seg *) wqe)->lkey =
- cpu_to_be32(wr->sg_list[i].lkey);
- ((struct mthca_data_seg *) wqe)->addr =
- cpu_to_be64(wr->sg_list[i].addr);
- wqe += sizeof (struct mthca_data_seg);
+ mthca_set_data_seg(wqe, wr->sg_list + i);
+ wqe += sizeof (struct mthca_data_seg);
size += sizeof (struct mthca_data_seg) / 16;
}
@@ -2151,7 +2140,7 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
((wr->send_flags & IB_SEND_FENCE) ?
MTHCA_NEXT_FENCE : 0));
- if (!size0) {
+ if (!nreq) {
size0 = size;
op0 = mthca_opcode[wr->opcode];
f0 = wr->send_flags & IB_SEND_FENCE ?
@@ -2241,20 +2230,12 @@ int mthca_arbel_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr,
}
for (i = 0; i < wr->num_sge; ++i) {
- ((struct mthca_data_seg *) wqe)->byte_count =
- cpu_to_be32(wr->sg_list[i].length);
- ((struct mthca_data_seg *) wqe)->lkey =
- cpu_to_be32(wr->sg_list[i].lkey);
- ((struct mthca_data_seg *) wqe)->addr =
- cpu_to_be64(wr->sg_list[i].addr);
+ mthca_set_data_seg(wqe, wr->sg_list + i);
wqe += sizeof (struct mthca_data_seg);
}
- if (i < qp->rq.max_gs) {
- ((struct mthca_data_seg *) wqe)->byte_count = 0;
- ((struct mthca_data_seg *) wqe)->lkey = cpu_to_be32(MTHCA_INVAL_LKEY);
- ((struct mthca_data_seg *) wqe)->addr = 0;
- }
+ if (i < qp->rq.max_gs)
+ mthca_set_data_seg_inval(wqe);
qp->wrid[ind] = wr->wr_id;
diff --git a/drivers/infiniband/hw/mthca/mthca_srq.c b/drivers/infiniband/hw/mthca/mthca_srq.c
index b8f05a52667..3f58c11a62b 100644
--- a/drivers/infiniband/hw/mthca/mthca_srq.c
+++ b/drivers/infiniband/hw/mthca/mthca_srq.c
@@ -509,7 +509,7 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr,
for (nreq = 0; wr; wr = wr->next) {
ind = srq->first_free;
- if (ind < 0) {
+ if (unlikely(ind < 0)) {
mthca_err(dev, "SRQ %06x full\n", srq->srqn);
err = -ENOMEM;
*bad_wr = wr;
@@ -519,7 +519,7 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr,
wqe = get_wqe(srq, ind);
next_ind = *wqe_to_link(wqe);
- if (next_ind < 0) {
+ if (unlikely(next_ind < 0)) {
mthca_err(dev, "SRQ %06x full\n", srq->srqn);
err = -ENOMEM;
*bad_wr = wr;
@@ -543,20 +543,12 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr,
}
for (i = 0; i < wr->num_sge; ++i) {
- ((struct mthca_data_seg *) wqe)->byte_count =
- cpu_to_be32(wr->sg_list[i].length);
- ((struct mthca_data_seg *) wqe)->lkey =
- cpu_to_be32(wr->sg_list[i].lkey);
- ((struct mthca_data_seg *) wqe)->addr =
- cpu_to_be64(wr->sg_list[i].addr);
+ mthca_set_data_seg(wqe, wr->sg_list + i);
wqe += sizeof (struct mthca_data_seg);
}
- if (i < srq->max_gs) {
- ((struct mthca_data_seg *) wqe)->byte_count = 0;
- ((struct mthca_data_seg *) wqe)->lkey = cpu_to_be32(MTHCA_INVAL_LKEY);
- ((struct mthca_data_seg *) wqe)->addr = 0;
- }
+ if (i < srq->max_gs)
+ mthca_set_data_seg_inval(wqe);
((struct mthca_next_seg *) prev_wqe)->nda_op =
cpu_to_be32((ind << srq->wqe_shift) | 1);
@@ -631,7 +623,7 @@ int mthca_arbel_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr,
for (nreq = 0; wr; ++nreq, wr = wr->next) {
ind = srq->first_free;
- if (ind < 0) {
+ if (unlikely(ind < 0)) {
mthca_err(dev, "SRQ %06x full\n", srq->srqn);
err = -ENOMEM;
*bad_wr = wr;
@@ -641,7 +633,7 @@ int mthca_arbel_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr,
wqe = get_wqe(srq, ind);
next_ind = *wqe_to_link(wqe);
- if (next_ind < 0) {
+ if (unlikely(next_ind < 0)) {
mthca_err(dev, "SRQ %06x full\n", srq->srqn);
err = -ENOMEM;
*bad_wr = wr;
@@ -662,20 +654,12 @@ int mthca_arbel_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr,
}
for (i = 0; i < wr->num_sge; ++i) {
- ((struct mthca_data_seg *) wqe)->byte_count =
- cpu_to_be32(wr->sg_list[i].length);
- ((struct mthca_data_seg *) wqe)->lkey =
- cpu_to_be32(wr->sg_list[i].lkey);
- ((struct mthca_data_seg *) wqe)->addr =
- cpu_to_be64(wr->sg_list[i].addr);
+ mthca_set_data_seg(wqe, wr->sg_list + i);
wqe += sizeof (struct mthca_data_seg);
}
- if (i < srq->max_gs) {
- ((struct mthca_data_seg *) wqe)->byte_count = 0;
- ((struct mthca_data_seg *) wqe)->lkey = cpu_to_be32(MTHCA_INVAL_LKEY);
- ((struct mthca_data_seg *) wqe)->addr = 0;
- }
+ if (i < srq->max_gs)
+ mthca_set_data_seg_inval(wqe);
srq->wrid[ind] = wr->wr_id;
srq->first_free = next_ind;
diff --git a/drivers/infiniband/hw/mthca/mthca_wqe.h b/drivers/infiniband/hw/mthca/mthca_wqe.h
index e7d2c1e8619..f6a66fe78e4 100644
--- a/drivers/infiniband/hw/mthca/mthca_wqe.h
+++ b/drivers/infiniband/hw/mthca/mthca_wqe.h
@@ -113,4 +113,19 @@ struct mthca_mlx_seg {
__be16 vcrc;
};
+static __always_inline void mthca_set_data_seg(struct mthca_data_seg *dseg,
+ struct ib_sge *sg)
+{
+ dseg->byte_count = cpu_to_be32(sg->length);
+ dseg->lkey = cpu_to_be32(sg->lkey);
+ dseg->addr = cpu_to_be64(sg->addr);
+}
+
+static __always_inline void mthca_set_data_seg_inval(struct mthca_data_seg *dseg)
+{
+ dseg->byte_count = 0;
+ dseg->lkey = cpu_to_be32(MTHCA_INVAL_LKEY);
+ dseg->addr = 0;
+}
+
#endif /* MTHCA_WQE_H */
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-01-06 12:53:59 +0000