leandrof/linux
1
0
Fork 0
forked from Minki/linux
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Pull request for 5.3-rc5

Browse Source 
- Fix siw buffer mapping issue
 - Fix siw 32/64 casting issues
 - Fix a KASAN access issue in bnxt_re
 - Fix several memory leaks (hfi1, mlx4)
 - Fix a NULL deref in cma_cleanup
 - Fixes for UMR memory support in mlx5 (4 patch series)
 - Fix namespace check for restrack
 - Fixes for counter support
 - Fixes for hfi1 TID processing (5 patch series)
 - Fix potential NULL deref in siw
 - Fix memory page calculations in mlx5
 
 Signed-off-by: Doug Ledford <dledford@redhat.com>
 -----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCAAdFiEErmsb2hIrI7QmWxJ0uCajMw5XL90FAl1gLY8ACgkQuCajMw5X
 L930yxAAsPMWG9B8TYT80M/+4iA0SP2o9WqJ6VjWt5j8ArcjWKHb9aepmSxMViUq
 T+W+ZLqk9tCfqEv88Z4T4iXBhwlzqzz6Xj5goohL2L4sOViit+YoNdx9stfW3yNh
 DPOfoxwIehxMiy00OGViQ1F/nC4KeyTYtMtoPgnYeB/7Jqzc20ipkZNopi6MIufn
 xSAwwaatzvj00nB1b+DC1eu9IzLWBjvzMmhPI9GBpgYTC6if43Q6PBwp6+hdag0K
 jNMHvO2BvtjtMBiZsFtaO3wu2gKIgR5CMxVMsQ5wnhZMDE3kQI7Ilrl0za0RdwMa
 +XMjn7mzExUjeNy2kXIxl3i9oH5s/iSmA1bqmcsG1q0dRmg7eI0iaB+ZTonpzMOB
 +7oTeKnklR0Q6vQ+rKu24vEND86l4eAWcOhqFat2jzCWOYMgRsnq0NCxzvG6NduF
 MdBNDkOIN+SmC+/9tMIwuUerZSrXGm2C6x4T6+YPYgxDOf+iKFm6g6lkGDrNCLsR
 g7qDRCWxlhAOqjlnFuH2T56IEmlFPC8RDA9sICxUnjH295ucBfO7w5fGRX9eCc0N
 0WrAlKbE9/4Hu7B2RN3slwJf2WpmE3ceOdEHSvfY2lje1SVXzYAl8o1bus8tSg3G
 YfWKHkCwA+h8X81lS3KHmlJ9eJnPdUwwBmOrcPHR2aQAw7uboPw=
 =bLpd
 -----END PGP SIGNATURE-----

Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma

Pull rdma fixes from Doug Ledford:
 "No beating around the bush: this is a monster pull request for an -rc5
  kernel. Intel hit me with a series of fixes for TID processing.
  Mellanox hit me with a series for their UMR memory support.

  And we had one fix for siw that fixes the 32bit build warnings and
  because of the number of casts that had to be changed to properly
  silence the warnings, that one patch alone is a full 40% of the LOC of
  this entire pull request. Given that this is the initial release
  kernel for siw, I'm trying to fix anything in it that we can, so that
  adds to the impetus to take fixes for it like this one.

  I had to do a rebase early in the week. Jason had thought he put a
  patch on the rc queue that he needed to be there so he could base some
  work off of it, and it had actually not been placed there. So he asked
  me (on Tuesday) to fix that up before pushing my wip branch to the
  official rc branch. I did, and that's why the early patches look like
  they were all committed at the same time on Tuesday. That bunch had
  been in my queue prior.

  The various patches all pass my test for being legitimate fixes and
  not attempts to slide new features or development into a late rc.
  Well, they were all fixes with the exception of a couple clean up
  patches people wrote for making the fixes they also wrote better (like
  a cleanup patch to move UMR checking into a function so that the
  remaining UMR fix patches can reference that function), so I left
  those in place too.

  My apologies for the LOC count and the number of patches here, it's
  just how the cards fell this cycle.

  Summary:

   - Fix siw buffer mapping issue

   - Fix siw 32/64 casting issues

   - Fix a KASAN access issue in bnxt_re

   - Fix several memory leaks (hfi1, mlx4)

   - Fix a NULL deref in cma_cleanup

   - Fixes for UMR memory support in mlx5 (4 patch series)

   - Fix namespace check for restrack

   - Fixes for counter support

   - Fixes for hfi1 TID processing (5 patch series)

   - Fix potential NULL deref in siw

   - Fix memory page calculations in mlx5"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (21 commits)
  RDMA/siw: Fix 64/32bit pointer inconsistency
  RDMA/siw: Fix SGL mapping issues
  RDMA/bnxt_re: Fix stack-out-of-bounds in bnxt_qplib_rcfw_send_message
  infiniband: hfi1: fix memory leaks
  infiniband: hfi1: fix a memory leak bug
  IB/mlx4: Fix memory leaks
  RDMA/cma: fix null-ptr-deref Read in cma_cleanup
  IB/mlx5: Block MR WR if UMR is not possible
  IB/mlx5: Fix MR re-registration flow to use UMR properly
  IB/mlx5: Report and handle ODP support properly
  IB/mlx5: Consolidate use_umr checks into single function
  RDMA/restrack: Rewrite PID namespace check to be reliable
  RDMA/counters: Properly implement PID checks
  IB/core: Fix NULL pointer dereference when bind QP to counter
  IB/hfi1: Drop stale TID RDMA packets that cause TIDErr
  IB/hfi1: Add additional checks when handling TID RDMA WRITE DATA packet
  IB/hfi1: Add additional checks when handling TID RDMA READ RESP packet
  IB/hfi1: Unsafe PSN checking for TID RDMA READ Resp packet
  IB/hfi1: Drop stale TID RDMA packets
  RDMA/siw: Fix potential NULL de-ref
  ...
This commit is contained in:
Linus Torvalds 2019-08-23 14:53:09 -07:00
commit 9140d8bdd4
26 changed files with 248 additions and 239 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
drivers/infiniband/core/cma.c
View File

@ -4724,10 +4724,14 @@ static int __init cma_init(void)
if (ret)
goto err;
cma_configfs_init();
ret = cma_configfs_init();
if (ret)
goto err_ib;
return 0;
err_ib:
ib_unregister_client(&cma_client);
err:
unregister_netdevice_notifier(&cma_nb);
ib_sa_unregister_client(&sa_client);

10
drivers/infiniband/core/counters.c
View File

@ -149,13 +149,11 @@ static bool auto_mode_match(struct ib_qp *qp, struct rdma_counter *counter,
struct auto_mode_param *param = &counter->mode.param;
bool match = true;
if (rdma_is_kernel_res(&counter->res) != rdma_is_kernel_res(&qp->res))
if (!rdma_is_visible_in_pid_ns(&qp->res))
return false;
/* Ensure that counter belong to right PID */
if (!rdma_is_kernel_res(&counter->res) &&
!rdma_is_kernel_res(&qp->res) &&
(task_pid_vnr(counter->res.task) != current->pid))
/* Ensure that counter belongs to the right PID */
if (task_pid_nr(counter->res.task) != task_pid_nr(qp->res.task))
return false;
if (auto_mask & RDMA_COUNTER_MASK_QP_TYPE)
@ -424,7 +422,7 @@ static struct ib_qp *rdma_counter_get_qp(struct ib_device *dev, u32 qp_num)
return qp;
err:
rdma_restrack_put(&qp->res);
rdma_restrack_put(res);
return NULL;
}

3
drivers/infiniband/core/nldev.c
View File

@ -382,8 +382,7 @@ static int fill_res_info(struct sk_buff *msg, struct ib_device *device)
for (i = 0; i < RDMA_RESTRACK_MAX; i++) {
if (!names[i])
continue;
curr = rdma_restrack_count(device, i,
task_active_pid_ns(current));
curr = rdma_restrack_count(device, i);
ret = fill_res_info_entry(msg, names[i], curr);
if (ret)
goto err;

15
drivers/infiniband/core/restrack.c
View File

@ -107,10 +107,8 @@ void rdma_restrack_clean(struct ib_device *dev)
* rdma_restrack_count() - the current usage of specific object
* @dev: IB device
* @type: actual type of object to operate
* @ns: PID namespace
*/
int rdma_restrack_count(struct ib_device *dev, enum rdma_restrack_type type,
struct pid_namespace *ns)
int rdma_restrack_count(struct ib_device *dev, enum rdma_restrack_type type)
{
struct rdma_restrack_root *rt = &dev->res[type];
struct rdma_restrack_entry *e;
@ -119,10 +117,9 @@ int rdma_restrack_count(struct ib_device *dev, enum rdma_restrack_type type,
xa_lock(&rt->xa);
xas_for_each(&xas, e, U32_MAX) {
if (ns == &init_pid_ns ||
(!rdma_is_kernel_res(e) &&
ns == task_active_pid_ns(e->task)))
cnt++;
if (!rdma_is_visible_in_pid_ns(e))
continue;
cnt++;
}
xa_unlock(&rt->xa);
return cnt;
@ -360,5 +357,7 @@ bool rdma_is_visible_in_pid_ns(struct rdma_restrack_entry *res)
*/
if (rdma_is_kernel_res(res))
return task_active_pid_ns(current) == &init_pid_ns;
return task_active_pid_ns(current) == task_active_pid_ns(res->task);
/* PID 0 means that resource is not found in current namespace */
return task_pid_vnr(res->task);
}

7
drivers/infiniband/core/umem.c
View File

@ -379,14 +379,9 @@ EXPORT_SYMBOL(ib_umem_release);
int ib_umem_page_count(struct ib_umem *umem)
{
int i;
int n;
int i, n = 0;
struct scatterlist *sg;
if (umem->is_odp)
return ib_umem_num_pages(umem);
n = 0;
for_each_sg(umem->sg_head.sgl, sg, umem->nmap, i)
n += sg_dma_len(sg) >> PAGE_SHIFT;

8
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
View File

@ -136,6 +136,13 @@ static int __send_message(struct bnxt_qplib_rcfw *rcfw, struct cmdq_base *req,
spin_unlock_irqrestore(&cmdq->lock, flags);
return -EBUSY;
}
size = req->cmd_size;
/* change the cmd_size to the number of 16byte cmdq unit.
* req->cmd_size is modified here
*/
bnxt_qplib_set_cmd_slots(req);
memset(resp, 0, sizeof(*resp));
crsqe->resp = (struct creq_qp_event *)resp;
crsqe->resp->cookie = req->cookie;
@ -150,7 +157,6 @@ static int __send_message(struct bnxt_qplib_rcfw *rcfw, struct cmdq_base *req,
cmdq_ptr = (struct bnxt_qplib_cmdqe **)cmdq->pbl_ptr;
preq = (u8 *)req;
size = req->cmd_size * BNXT_QPLIB_CMDQE_UNITS;
do {
/* Locate the next cmdq slot */
sw_prod = HWQ_CMP(cmdq->prod, cmdq);

11
drivers/infiniband/hw/bnxt_re/qplib_rcfw.h
View File

@ -55,9 +55,7 @@
do { \
memset(&(req), 0, sizeof((req))); \
(req).opcode = CMDQ_BASE_OPCODE_##CMD; \
(req).cmd_size = (sizeof((req)) + \
BNXT_QPLIB_CMDQE_UNITS - 1) / \
BNXT_QPLIB_CMDQE_UNITS; \
(req).cmd_size = sizeof((req)); \
(req).flags = cpu_to_le16(cmd_flags); \
} while (0)
@ -95,6 +93,13 @@ static inline u32 bnxt_qplib_cmdqe_cnt_per_pg(u32 depth)
BNXT_QPLIB_CMDQE_UNITS);
}
/* Set the cmd_size to a factor of CMDQE unit */
static inline void bnxt_qplib_set_cmd_slots(struct cmdq_base *req)
{
req->cmd_size = (req->cmd_size + BNXT_QPLIB_CMDQE_UNITS - 1) /
BNXT_QPLIB_CMDQE_UNITS;
}
#define MAX_CMDQ_IDX(depth) ((depth) - 1)
static inline u32 bnxt_qplib_max_cmdq_idx_per_pg(u32 depth)

12
drivers/infiniband/hw/hfi1/fault.c
View File

@ -141,12 +141,14 @@ static ssize_t fault_opcodes_write(struct file *file, const char __user *buf,
if (!data)
return -ENOMEM;
copy = min(len, datalen - 1);
if (copy_from_user(data, buf, copy))
return -EFAULT;
if (copy_from_user(data, buf, copy)) {
ret = -EFAULT;
goto free_data;
}
ret = debugfs_file_get(file->f_path.dentry);
if (unlikely(ret))
return ret;
goto free_data;
ptr = data;
token = ptr;
for (ptr = data; *ptr; ptr = end + 1, token = ptr) {
@ -195,6 +197,7 @@ static ssize_t fault_opcodes_write(struct file *file, const char __user *buf,
ret = len;
debugfs_file_put(file->f_path.dentry);
free_data:
kfree(data);
return ret;
}
@ -214,7 +217,7 @@ static ssize_t fault_opcodes_read(struct file *file, char __user *buf,
return -ENOMEM;
ret = debugfs_file_get(file->f_path.dentry);
if (unlikely(ret))
return ret;
goto free_data;
bit = find_first_bit(fault->opcodes, bitsize);
while (bit < bitsize) {
zero = find_next_zero_bit(fault->opcodes, bitsize, bit);
@ -232,6 +235,7 @@ static ssize_t fault_opcodes_read(struct file *file, char __user *buf,
data[size - 1] = '\n';
data[size] = '\0';
ret = simple_read_from_buffer(buf, len, pos, data, size);
free_data:
kfree(data);
return ret;
}

76
drivers/infiniband/hw/hfi1/tid_rdma.c
View File

@ -2574,18 +2574,9 @@ void hfi1_kern_read_tid_flow_free(struct rvt_qp *qp)
hfi1_kern_clear_hw_flow(priv->rcd, qp);
}
static bool tid_rdma_tid_err(struct hfi1_ctxtdata *rcd,
struct hfi1_packet *packet, u8 rcv_type,
u8 opcode)
static bool tid_rdma_tid_err(struct hfi1_packet *packet, u8 rcv_type)
{
struct rvt_qp *qp = packet->qp;
struct hfi1_qp_priv *qpriv = qp->priv;
u32 ipsn;
struct ib_other_headers *ohdr = packet->ohdr;
struct rvt_ack_entry *e;
struct tid_rdma_request *req;
struct rvt_dev_info *rdi = ib_to_rvt(qp->ibqp.device);
u32 i;
if (rcv_type >= RHF_RCV_TYPE_IB)
goto done;
@ -2602,41 +2593,9 @@ static bool tid_rdma_tid_err(struct hfi1_ctxtdata *rcd,
if (rcv_type == RHF_RCV_TYPE_EAGER) {
hfi1_restart_rc(qp, qp->s_last_psn + 1, 1);
hfi1_schedule_send(qp);
goto done_unlock;
}
/*
* For TID READ response, error out QP after freeing the tid
* resources.
*/
if (opcode == TID_OP(READ_RESP)) {
ipsn = mask_psn(be32_to_cpu(ohdr->u.tid_rdma.r_rsp.verbs_psn));
if (cmp_psn(ipsn, qp->s_last_psn) > 0 &&
cmp_psn(ipsn, qp->s_psn) < 0) {
hfi1_kern_read_tid_flow_free(qp);
spin_unlock(&qp->s_lock);
rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR);
goto done;
}
goto done_unlock;
}
/*
* Error out the qp for TID RDMA WRITE
*/
hfi1_kern_clear_hw_flow(qpriv->rcd, qp);
for (i = 0; i < rvt_max_atomic(rdi); i++) {
e = &qp->s_ack_queue[i];
if (e->opcode == TID_OP(WRITE_REQ)) {
req = ack_to_tid_req(e);
hfi1_kern_exp_rcv_clear_all(req);
}
}
spin_unlock(&qp->s_lock);
rvt_rc_error(qp, IB_WC_LOC_LEN_ERR);
goto done;
done_unlock:
/* Since no payload is delivered, just drop the packet */
spin_unlock(&qp->s_lock);
done:
return true;
@ -2687,12 +2646,12 @@ static bool handle_read_kdeth_eflags(struct hfi1_ctxtdata *rcd,
u32 fpsn;
lockdep_assert_held(&qp->r_lock);
spin_lock(&qp->s_lock);
/* If the psn is out of valid range, drop the packet */
if (cmp_psn(ibpsn, qp->s_last_psn) < 0 ||
cmp_psn(ibpsn, qp->s_psn) > 0)
return ret;
goto s_unlock;
spin_lock(&qp->s_lock);
/*
* Note that NAKs implicitly ACK outstanding SEND and RDMA write
* requests and implicitly NAK RDMA read and atomic requests issued
@ -2740,9 +2699,12 @@ static bool handle_read_kdeth_eflags(struct hfi1_ctxtdata *rcd,
wqe = do_rc_completion(qp, wqe, ibp);
if (qp->s_acked == qp->s_tail)
break;
goto s_unlock;
}
if (qp->s_acked == qp->s_tail)
goto s_unlock;
/* Handle the eflags for the request */
if (wqe->wr.opcode != IB_WR_TID_RDMA_READ)
goto s_unlock;
@ -2922,7 +2884,7 @@ bool hfi1_handle_kdeth_eflags(struct hfi1_ctxtdata *rcd,
if (lnh == HFI1_LRH_GRH)
goto r_unlock;
if (tid_rdma_tid_err(rcd, packet, rcv_type, opcode))
if (tid_rdma_tid_err(packet, rcv_type))
goto r_unlock;
}
@ -2942,8 +2904,15 @@ bool hfi1_handle_kdeth_eflags(struct hfi1_ctxtdata *rcd,
*/
spin_lock(&qp->s_lock);
qpriv = qp->priv;
if (qpriv->r_tid_tail == HFI1_QP_WQE_INVALID ||
qpriv->r_tid_tail == qpriv->r_tid_head)
goto unlock;
e = &qp->s_ack_queue[qpriv->r_tid_tail];
if (e->opcode != TID_OP(WRITE_REQ))
goto unlock;
req = ack_to_tid_req(e);
if (req->comp_seg == req->cur_seg)
goto unlock;
flow = &req->flows[req->clear_tail];
trace_hfi1_eflags_err_write(qp, rcv_type, rte, psn);
trace_hfi1_rsp_handle_kdeth_eflags(qp, psn);
@ -4509,7 +4478,7 @@ void hfi1_rc_rcv_tid_rdma_ack(struct hfi1_packet *packet)
struct rvt_swqe *wqe;
struct tid_rdma_request *req;
struct tid_rdma_flow *flow;
u32 aeth, psn, req_psn, ack_psn, resync_psn, ack_kpsn;
u32 aeth, psn, req_psn, ack_psn, flpsn, resync_psn, ack_kpsn;
unsigned long flags;
u16 fidx;
@ -4538,6 +4507,9 @@ void hfi1_rc_rcv_tid_rdma_ack(struct hfi1_packet *packet)
ack_kpsn--;
}
if (unlikely(qp->s_acked == qp->s_tail))
goto ack_op_err;
wqe = rvt_get_swqe_ptr(qp, qp->s_acked);
if (wqe->wr.opcode != IB_WR_TID_RDMA_WRITE)
@ -4550,7 +4522,8 @@ void hfi1_rc_rcv_tid_rdma_ack(struct hfi1_packet *packet)
trace_hfi1_tid_flow_rcv_tid_ack(qp, req->acked_tail, flow);
/* Drop stale ACK/NAK */
if (cmp_psn(psn, full_flow_psn(flow, flow->flow_state.spsn)) < 0)
if (cmp_psn(psn, full_flow_psn(flow, flow->flow_state.spsn)) < 0 ||
cmp_psn(req_psn, flow->flow_state.resp_ib_psn) < 0)
goto ack_op_err;
while (cmp_psn(ack_kpsn,
@ -4712,7 +4685,12 @@ done:
switch ((aeth >> IB_AETH_CREDIT_SHIFT) &
IB_AETH_CREDIT_MASK) {
case 0: /* PSN sequence error */
if (!req->flows)
break;
flow = &req->flows[req->acked_tail];
flpsn = full_flow_psn(flow, flow->flow_state.lpsn);
if (cmp_psn(psn, flpsn) > 0)
break;
trace_hfi1_tid_flow_rcv_tid_ack(qp, req->acked_tail,
flow);
req->r_ack_psn = mask_psn(be32_to_cpu(ohdr->bth[2]));

4
drivers/infiniband/hw/mlx4/mad.c
View File

@ -1677,8 +1677,6 @@ tx_err:
tx_buf_size, DMA_TO_DEVICE);
kfree(tun_qp->tx_ring[i].buf.addr);
}
kfree(tun_qp->tx_ring);
tun_qp->tx_ring = NULL;
i = MLX4_NUM_TUNNEL_BUFS;
err:
while (i > 0) {
@ -1687,6 +1685,8 @@ err:
rx_buf_size, DMA_FROM_DEVICE);
kfree(tun_qp->ring[i].addr);
}
kfree(tun_qp->tx_ring);
tun_qp->tx_ring = NULL;
kfree(tun_qp->ring);
tun_qp->ring = NULL;
return -ENOMEM;

6
drivers/infiniband/hw/mlx5/main.c
View File

@ -1023,7 +1023,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
props->timestamp_mask = 0x7FFFFFFFFFFFFFFFULL;
if (IS_ENABLED(CONFIG_INFINIBAND_ON_DEMAND_PAGING)) {
if (MLX5_CAP_GEN(mdev, pg))
if (dev->odp_caps.general_caps & IB_ODP_SUPPORT)
props->device_cap_flags |= IB_DEVICE_ON_DEMAND_PAGING;
props->odp_caps = dev->odp_caps;
}
@ -6139,6 +6139,8 @@ static int mlx5_ib_stage_init_init(struct mlx5_ib_dev *dev)
dev->port[i].roce.last_port_state = IB_PORT_DOWN;
}
mlx5_ib_internal_fill_odp_caps(dev);
err = mlx5_ib_init_multiport_master(dev);
if (err)
return err;
@ -6563,8 +6565,6 @@ static void mlx5_ib_stage_dev_res_cleanup(struct mlx5_ib_dev *dev)
static int mlx5_ib_stage_odp_init(struct mlx5_ib_dev *dev)
{
mlx5_ib_internal_fill_odp_caps(dev);
return mlx5_ib_odp_init_one(dev);
}

5
drivers/infiniband/hw/mlx5/mem.c
View File

@ -57,9 +57,10 @@ void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr,
int entry;
if (umem->is_odp) {
unsigned int page_shift = to_ib_umem_odp(umem)->page_shift;
struct ib_umem_odp *odp = to_ib_umem_odp(umem);
unsigned int page_shift = odp->page_shift;
*ncont = ib_umem_page_count(umem);
*ncont = ib_umem_odp_num_pages(odp);
*count = *ncont << (page_shift - PAGE_SHIFT);
*shift = page_shift;
if (order)

14
drivers/infiniband/hw/mlx5/mlx5_ib.h
View File

@ -1475,4 +1475,18 @@ int bfregn_to_uar_index(struct mlx5_ib_dev *dev,
bool dyn_bfreg);
int mlx5_ib_qp_set_counter(struct ib_qp *qp, struct rdma_counter *counter);
static inline bool mlx5_ib_can_use_umr(struct mlx5_ib_dev *dev,
bool do_modify_atomic)
{
if (MLX5_CAP_GEN(dev->mdev, umr_modify_entity_size_disabled))
return false;
if (do_modify_atomic &&
MLX5_CAP_GEN(dev->mdev, atomic) &&
MLX5_CAP_GEN(dev->mdev, umr_modify_atomic_disabled))
return false;
return true;
}
#endif /* MLX5_IB_H */

7
drivers/infiniband/hw/mlx5/mr.c
View File

@ -1293,9 +1293,7 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
if (err < 0)
return ERR_PTR(err);
use_umr = !MLX5_CAP_GEN(dev->mdev, umr_modify_entity_size_disabled) &&
(!MLX5_CAP_GEN(dev->mdev, umr_modify_atomic_disabled) ||
!MLX5_CAP_GEN(dev->mdev, atomic));
use_umr = mlx5_ib_can_use_umr(dev, true);
if (order <= mr_cache_max_order(dev) && use_umr) {
mr = alloc_mr_from_cache(pd, umem, virt_addr, length, ncont,
@ -1448,7 +1446,8 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start,
goto err;
}
if (flags & IB_MR_REREG_TRANS && !use_umr_mtt_update(mr, addr, len)) {
if (!mlx5_ib_can_use_umr(dev, true) ||
(flags & IB_MR_REREG_TRANS && !use_umr_mtt_update(mr, addr, len))) {
/*
* UMR can't be used - MKey needs to be replaced.
*/

17
drivers/infiniband/hw/mlx5/odp.c
View File

@ -301,7 +301,8 @@ void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev)
memset(caps, 0, sizeof(*caps));
if (!MLX5_CAP_GEN(dev->mdev, pg))
if (!MLX5_CAP_GEN(dev->mdev, pg) ||
!mlx5_ib_can_use_umr(dev, true))
return;
caps->general_caps = IB_ODP_SUPPORT;
@ -355,7 +356,8 @@ void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev)
if (MLX5_CAP_GEN(dev->mdev, fixed_buffer_size) &&
MLX5_CAP_GEN(dev->mdev, null_mkey) &&
MLX5_CAP_GEN(dev->mdev, umr_extended_translation_offset))
MLX5_CAP_GEN(dev->mdev, umr_extended_translation_offset) &&
!MLX5_CAP_GEN(dev->mdev, umr_indirect_mkey_disabled))
caps->general_caps |= IB_ODP_SUPPORT_IMPLICIT;
return;
@ -1622,8 +1624,10 @@ int mlx5_ib_odp_init_one(struct mlx5_ib_dev *dev)
{
int ret = 0;
if (dev->odp_caps.general_caps & IB_ODP_SUPPORT)
ib_set_device_ops(&dev->ib_dev, &mlx5_ib_dev_odp_ops);
if (!(dev->odp_caps.general_caps & IB_ODP_SUPPORT))
return ret;
ib_set_device_ops(&dev->ib_dev, &mlx5_ib_dev_odp_ops);
if (dev->odp_caps.general_caps & IB_ODP_SUPPORT_IMPLICIT) {
ret = mlx5_cmd_null_mkey(dev->mdev, &dev->null_mkey);
@ -1633,9 +1637,6 @@ int mlx5_ib_odp_init_one(struct mlx5_ib_dev *dev)
}
}
if (!MLX5_CAP_GEN(dev->mdev, pg))
return ret;
ret = mlx5_ib_create_pf_eq(dev, &dev->odp_pf_eq);
return ret;
@ -1643,7 +1644,7 @@ int mlx5_ib_odp_init_one(struct mlx5_ib_dev *dev)
void mlx5_ib_odp_cleanup_one(struct mlx5_ib_dev *dev)
{
if (!MLX5_CAP_GEN(dev->mdev, pg))
if (!(dev->odp_caps.general_caps & IB_ODP_SUPPORT))
return;
mlx5_ib_destroy_pf_eq(dev, &dev->odp_pf_eq);

24
drivers/infiniband/hw/mlx5/qp.c
View File

@ -4162,7 +4162,7 @@ static u64 get_xlt_octo(u64 bytes)
MLX5_IB_UMR_OCTOWORD;
}
static __be64 frwr_mkey_mask(void)
static __be64 frwr_mkey_mask(bool atomic)
{
u64 result;
@ -4175,10 +4175,12 @@ static __be64 frwr_mkey_mask(void)
MLX5_MKEY_MASK_LW |
MLX5_MKEY_MASK_RR |
MLX5_MKEY_MASK_RW |
MLX5_MKEY_MASK_A |
MLX5_MKEY_MASK_SMALL_FENCE |
MLX5_MKEY_MASK_FREE;
if (atomic)
result |= MLX5_MKEY_MASK_A;
return cpu_to_be64(result);
}
@ -4204,7 +4206,7 @@ static __be64 sig_mkey_mask(void)
}
static void set_reg_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr,
struct mlx5_ib_mr *mr, u8 flags)
struct mlx5_ib_mr *mr, u8 flags, bool atomic)
{
int size = (mr->ndescs + mr->meta_ndescs) * mr->desc_size;
@ -4212,7 +4214,7 @@ static void set_reg_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr,
umr->flags = flags;
umr->xlt_octowords = cpu_to_be16(get_xlt_octo(size));
umr->mkey_mask = frwr_mkey_mask();
umr->mkey_mask = frwr_mkey_mask(atomic);
}
static void set_linv_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr)
@ -4811,10 +4813,22 @@ static int set_reg_wr(struct mlx5_ib_qp *qp,
{
struct mlx5_ib_mr *mr = to_mmr(wr->mr);
struct mlx5_ib_pd *pd = to_mpd(qp->ibqp.pd);
struct mlx5_ib_dev *dev = to_mdev(pd->ibpd.device);
int mr_list_size = (mr->ndescs + mr->meta_ndescs) * mr->desc_size;
bool umr_inline = mr_list_size <= MLX5_IB_SQ_UMR_INLINE_THRESHOLD;
bool atomic = wr->access & IB_ACCESS_REMOTE_ATOMIC;
u8 flags = 0;
if (!mlx5_ib_can_use_umr(dev, atomic)) {
mlx5_ib_warn(to_mdev(qp->ibqp.device),
"Fast update of %s for MR is disabled\n",
(MLX5_CAP_GEN(dev->mdev,
umr_modify_entity_size_disabled)) ?
"entity size" :
"atomic access");
return -EINVAL;
}
if (unlikely(wr->wr.send_flags & IB_SEND_INLINE)) {
mlx5_ib_warn(to_mdev(qp->ibqp.device),
"Invalid IB_SEND_INLINE send flag\n");
@ -4826,7 +4840,7 @@ static int set_reg_wr(struct mlx5_ib_qp *qp,
if (umr_inline)
flags |= MLX5_UMR_INLINE;
set_reg_umr_seg(*seg, mr, flags);
set_reg_umr_seg(*seg, mr, flags, atomic);
*seg += sizeof(struct mlx5_wqe_umr_ctrl_seg);
*size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16;
handle_post_send_edge(&qp->sq, seg, *size, cur_edge);

8
drivers/infiniband/sw/siw/siw.h
View File

@ -138,9 +138,9 @@ struct siw_umem {
};
struct siw_pble {
u64 addr; /* Address of assigned user buffer */
u64 size; /* Size of this entry */
u64 pbl_off; /* Total offset from start of PBL */
dma_addr_t addr; /* Address of assigned buffer */
unsigned int size; /* Size of this entry */
unsigned long pbl_off; /* Total offset from start of PBL */
};
struct siw_pbl {
@ -734,7 +734,7 @@ static inline void siw_crc_skb(struct siw_rx_stream *srx, unsigned int len)
"MEM[0x%08x] %s: " fmt, mem->stag, __func__, ##__VA_ARGS__)
#define siw_dbg_cep(cep, fmt, ...) \
ibdev_dbg(&cep->sdev->base_dev, "CEP[0x%p] %s: " fmt, \
ibdev_dbg(&cep->sdev->base_dev, "CEP[0x%pK] %s: " fmt, \
cep, __func__, ##__VA_ARGS__)
void siw_cq_flush(struct siw_cq *cq);

82
drivers/infiniband/sw/siw/siw_cm.c
View File

@ -355,8 +355,8 @@ static int siw_cm_upcall(struct siw_cep *cep, enum iw_cm_event_type reason,
getname_local(cep->sock, &event.local_addr);
getname_peer(cep->sock, &event.remote_addr);
}
siw_dbg_cep(cep, "[QP %u]: id 0x%p, reason=%d, status=%d\n",
cep->qp ? qp_id(cep->qp) : -1, id, reason, status);
siw_dbg_cep(cep, "[QP %u]: reason=%d, status=%d\n",
cep->qp ? qp_id(cep->qp) : UINT_MAX, reason, status);
return id->event_handler(id, &event);
}
@ -947,8 +947,6 @@ static void siw_accept_newconn(struct siw_cep *cep)
siw_cep_get(new_cep);
new_s->sk->sk_user_data = new_cep;
siw_dbg_cep(cep, "listen socket 0x%p, new 0x%p\n", s, new_s);
if (siw_tcp_nagle == false) {
int val = 1;
@ -1011,7 +1009,8 @@ static void siw_cm_work_handler(struct work_struct *w)
cep = work->cep;
siw_dbg_cep(cep, "[QP %u]: work type: %d, state %d\n",
cep->qp ? qp_id(cep->qp) : -1, work->type, cep->state);
cep->qp ? qp_id(cep->qp) : UINT_MAX,
work->type, cep->state);
siw_cep_set_inuse(cep);
@ -1145,9 +1144,9 @@ static void siw_cm_work_handler(struct work_struct *w)
}
if (release_cep) {
siw_dbg_cep(cep,
"release: timer=%s, QP[%u], id 0x%p\n",
"release: timer=%s, QP[%u]\n",
cep->mpa_timer ? "y" : "n",
cep->qp ? qp_id(cep->qp) : -1, cep->cm_id);
cep->qp ? qp_id(cep->qp) : UINT_MAX);
siw_cancel_mpatimer(cep);
@ -1211,8 +1210,8 @@ int siw_cm_queue_work(struct siw_cep *cep, enum siw_work_type type)
else
delay = MPAREP_TIMEOUT;
}
siw_dbg_cep(cep, "[QP %u]: work type: %d, work 0x%p, timeout %lu\n",
cep->qp ? qp_id(cep->qp) : -1, type, work, delay);
siw_dbg_cep(cep, "[QP %u]: work type: %d, timeout %lu\n",
cep->qp ? qp_id(cep->qp) : -1, type, delay);
queue_delayed_work(siw_cm_wq, &work->work, delay);
@ -1376,16 +1375,16 @@ int siw_connect(struct iw_cm_id *id, struct iw_cm_conn_param *params)
}
if (v4)
siw_dbg_qp(qp,
"id 0x%p, pd_len %d, laddr %pI4 %d, raddr %pI4 %d\n",
id, pd_len,
"pd_len %d, laddr %pI4 %d, raddr %pI4 %d\n",
pd_len,
&((struct sockaddr_in *)(laddr))->sin_addr,
ntohs(((struct sockaddr_in *)(laddr))->sin_port),
&((struct sockaddr_in *)(raddr))->sin_addr,
ntohs(((struct sockaddr_in *)(raddr))->sin_port));
else
siw_dbg_qp(qp,
"id 0x%p, pd_len %d, laddr %pI6 %d, raddr %pI6 %d\n",
id, pd_len,
"pd_len %d, laddr %pI6 %d, raddr %pI6 %d\n",
pd_len,
&((struct sockaddr_in6 *)(laddr))->sin6_addr,
ntohs(((struct sockaddr_in6 *)(laddr))->sin6_port),
&((struct sockaddr_in6 *)(raddr))->sin6_addr,
@ -1508,14 +1507,13 @@ int siw_connect(struct iw_cm_id *id, struct iw_cm_conn_param *params)
if (rv >= 0) {
rv = siw_cm_queue_work(cep, SIW_CM_WORK_MPATIMEOUT);
if (!rv) {
siw_dbg_cep(cep, "id 0x%p, [QP %u]: exit\n", id,
qp_id(qp));
siw_dbg_cep(cep, "[QP %u]: exit\n", qp_id(qp));
siw_cep_set_free(cep);
return 0;
}
}
error:
siw_dbg_qp(qp, "failed: %d\n", rv);
siw_dbg(id->device, "failed: %d\n", rv);
if (cep) {
siw_socket_disassoc(s);
@ -1540,7 +1538,8 @@ error:
} else if (s) {
sock_release(s);
}
siw_qp_put(qp);
if (qp)
siw_qp_put(qp);
return rv;
}
@ -1580,7 +1579,7 @@ int siw_accept(struct iw_cm_id *id, struct iw_cm_conn_param *params)
siw_cancel_mpatimer(cep);
if (cep->state != SIW_EPSTATE_RECVD_MPAREQ) {
siw_dbg_cep(cep, "id 0x%p: out of state\n", id);
siw_dbg_cep(cep, "out of state\n");
siw_cep_set_free(cep);
siw_cep_put(cep);
@ -1601,7 +1600,7 @@ int siw_accept(struct iw_cm_id *id, struct iw_cm_conn_param *params)
up_write(&qp->state_lock);
goto error;
}
siw_dbg_cep(cep, "id 0x%p\n", id);
siw_dbg_cep(cep, "[QP %d]\n", params->qpn);
if (try_gso && cep->mpa.hdr.params.bits & MPA_RR_FLAG_GSO_EXP) {
siw_dbg_cep(cep, "peer allows GSO on TX\n");
@ -1611,8 +1610,8 @@ int siw_accept(struct iw_cm_id *id, struct iw_cm_conn_param *params)
params->ird > sdev->attrs.max_ird) {
siw_dbg_cep(
cep,
"id 0x%p, [QP %u]: ord %d (max %d), ird %d (max %d)\n",
id, qp_id(qp), params->ord, sdev->attrs.max_ord,
"[QP %u]: ord %d (max %d), ird %d (max %d)\n",
qp_id(qp), params->ord, sdev->attrs.max_ord,
params->ird, sdev->attrs.max_ird);
rv = -EINVAL;
up_write(&qp->state_lock);
@ -1624,8 +1623,8 @@ int siw_accept(struct iw_cm_id *id, struct iw_cm_conn_param *params)
if (params->private_data_len > max_priv_data) {
siw_dbg_cep(
cep,
"id 0x%p, [QP %u]: private data length: %d (max %d)\n",
id, qp_id(qp), params->private_data_len, max_priv_data);
"[QP %u]: private data length: %d (max %d)\n",
qp_id(qp), params->private_data_len, max_priv_data);
rv = -EINVAL;
up_write(&qp->state_lock);
goto error;
@ -1679,7 +1678,7 @@ int siw_accept(struct iw_cm_id *id, struct iw_cm_conn_param *params)
qp_attrs.flags = SIW_MPA_CRC;
qp_attrs.state = SIW_QP_STATE_RTS;
siw_dbg_cep(cep, "id 0x%p, [QP%u]: moving to rts\n", id, qp_id(qp));
siw_dbg_cep(cep, "[QP%u]: moving to rts\n", qp_id(qp));
/* Associate QP with CEP */
siw_cep_get(cep);
@ -1700,8 +1699,8 @@ int siw_accept(struct iw_cm_id *id, struct iw_cm_conn_param *params)
if (rv)
goto error;
siw_dbg_cep(cep, "id 0x%p, [QP %u]: send mpa reply, %d byte pdata\n",
id, qp_id(qp), params->private_data_len);
siw_dbg_cep(cep, "[QP %u]: send mpa reply, %d byte pdata\n",
qp_id(qp), params->private_data_len);
rv = siw_send_mpareqrep(cep, params->private_data,
params->private_data_len);
@ -1759,14 +1758,14 @@ int siw_reject(struct iw_cm_id *id, const void *pdata, u8 pd_len)
siw_cancel_mpatimer(cep);
if (cep->state != SIW_EPSTATE_RECVD_MPAREQ) {
siw_dbg_cep(cep, "id 0x%p: out of state\n", id);
siw_dbg_cep(cep, "out of state\n");
siw_cep_set_free(cep);
siw_cep_put(cep); /* put last reference */
return -ECONNRESET;
}
siw_dbg_cep(cep, "id 0x%p, cep->state %d, pd_len %d\n", id, cep->state,
siw_dbg_cep(cep, "cep->state %d, pd_len %d\n", cep->state,
pd_len);
if (__mpa_rr_revision(cep->mpa.hdr.params.bits) >= MPA_REVISION_1) {
@ -1804,14 +1803,14 @@ static int siw_listen_address(struct iw_cm_id *id, int backlog,
rv = kernel_setsockopt(s, SOL_SOCKET, SO_REUSEADDR, (char *)&s_val,
sizeof(s_val));
if (rv) {
siw_dbg(id->device, "id 0x%p: setsockopt error: %d\n", id, rv);
siw_dbg(id->device, "setsockopt error: %d\n", rv);
goto error;
}
rv = s->ops->bind(s, laddr, addr_family == AF_INET ?
sizeof(struct sockaddr_in) :
sizeof(struct sockaddr_in6));
if (rv) {
siw_dbg(id->device, "id 0x%p: socket bind error: %d\n", id, rv);
siw_dbg(id->device, "socket bind error: %d\n", rv);
goto error;
}
cep = siw_cep_alloc(sdev);
@ -1824,13 +1823,13 @@ static int siw_listen_address(struct iw_cm_id *id, int backlog,
rv = siw_cm_alloc_work(cep, backlog);
if (rv) {
siw_dbg(id->device,
"id 0x%p: alloc_work error %d, backlog %d\n", id,
"alloc_work error %d, backlog %d\n",
rv, backlog);
goto error;
}
rv = s->ops->listen(s, backlog);
if (rv) {
siw_dbg(id->device, "id 0x%p: listen error %d\n", id, rv);
siw_dbg(id->device, "listen error %d\n", rv);
goto error;
}
cep->cm_id = id;
@ -1914,8 +1913,7 @@ static void siw_drop_listeners(struct iw_cm_id *id)
list_del(p);
siw_dbg_cep(cep, "id 0x%p: drop cep, state %d\n", id,
cep->state);
siw_dbg_cep(cep, "drop cep, state %d\n", cep->state);
siw_cep_set_inuse(cep);
@ -1952,7 +1950,7 @@ int siw_create_listen(struct iw_cm_id *id, int backlog)
struct net_device *dev = to_siw_dev(id->device)->netdev;
int rv = 0, listeners = 0;
siw_dbg(id->device, "id 0x%p: backlog %d\n", id, backlog);
siw_dbg(id->device, "backlog %d\n", backlog);
/*
* For each attached address of the interface, create a
@ -1968,8 +1966,8 @@ int siw_create_listen(struct iw_cm_id *id, int backlog)
s_raddr = (struct sockaddr_in *)&id->remote_addr;
siw_dbg(id->device,
"id 0x%p: laddr %pI4:%d, raddr %pI4:%d\n",
id, &s_laddr.sin_addr, ntohs(s_laddr.sin_port),
"laddr %pI4:%d, raddr %pI4:%d\n",
&s_laddr.sin_addr, ntohs(s_laddr.sin_port),
&s_raddr->sin_addr, ntohs(s_raddr->sin_port));
rtnl_lock();
@ -1994,8 +1992,8 @@ int siw_create_listen(struct iw_cm_id *id, int backlog)
*s_raddr = &to_sockaddr_in6(id->remote_addr);
siw_dbg(id->device,
"id 0x%p: laddr %pI6:%d, raddr %pI6:%d\n",
id, &s_laddr->sin6_addr, ntohs(s_laddr->sin6_port),
"laddr %pI6:%d, raddr %pI6:%d\n",
&s_laddr->sin6_addr, ntohs(s_laddr->sin6_port),
&s_raddr->sin6_addr, ntohs(s_raddr->sin6_port));
read_lock_bh(&in6_dev->lock);
@ -2028,17 +2026,15 @@ int siw_create_listen(struct iw_cm_id *id, int backlog)
else if (!rv)
rv = -EINVAL;
siw_dbg(id->device, "id 0x%p: %s\n", id, rv ? "FAIL" : "OK");
siw_dbg(id->device, "%s\n", rv ? "FAIL" : "OK");
return rv;
}
int siw_destroy_listen(struct iw_cm_id *id)
{
siw_dbg(id->device, "id 0x%p\n", id);
if (!id->provider_data) {
siw_dbg(id->device, "id 0x%p: no cep(s)\n", id);
siw_dbg(id->device, "no cep(s)\n");
return 0;
}
siw_drop_listeners(id);

5
drivers/infiniband/sw/siw/siw_cq.c
View File

@ -71,9 +71,10 @@ int siw_reap_cqe(struct siw_cq *cq, struct ib_wc *wc)
wc->wc_flags = IB_WC_WITH_INVALIDATE;
}
wc->qp = cqe->base_qp;
siw_dbg_cq(cq, "idx %u, type %d, flags %2x, id 0x%p\n",
siw_dbg_cq(cq,
"idx %u, type %d, flags %2x, id 0x%pK\n",
cq->cq_get % cq->num_cqe, cqe->opcode,
cqe->flags, (void *)cqe->id);
cqe->flags, (void *)(uintptr_t)cqe->id);
}
WRITE_ONCE(cqe->flags, 0);
cq->cq_get++;

14
drivers/infiniband/sw/siw/siw_mem.c
View File

@ -197,12 +197,12 @@ int siw_check_mem(struct ib_pd *pd, struct siw_mem *mem, u64 addr,
*/
if (addr < mem->va || addr + len > mem->va + mem->len) {
siw_dbg_pd(pd, "MEM interval len %d\n", len);
siw_dbg_pd(pd, "[0x%016llx, 0x%016llx] out of bounds\n",
(unsigned long long)addr,
(unsigned long long)(addr + len));
siw_dbg_pd(pd, "[0x%016llx, 0x%016llx] STag=0x%08x\n",
(unsigned long long)mem->va,
(unsigned long long)(mem->va + mem->len),
siw_dbg_pd(pd, "[0x%pK, 0x%pK] out of bounds\n",
(void *)(uintptr_t)addr,
(void *)(uintptr_t)(addr + len));
siw_dbg_pd(pd, "[0x%pK, 0x%pK] STag=0x%08x\n",
(void *)(uintptr_t)mem->va,
(void *)(uintptr_t)(mem->va + mem->len),
mem->stag);
return -E_BASE_BOUNDS;
@ -330,7 +330,7 @@ out:
* Optionally, provides remaining len within current element, and
* current PBL index for later resume at same element.
*/
u64 siw_pbl_get_buffer(struct siw_pbl *pbl, u64 off, int *len, int *idx)
dma_addr_t siw_pbl_get_buffer(struct siw_pbl *pbl, u64 off, int *len, int *idx)
{
int i = idx ? *idx : 0;

2
drivers/infiniband/sw/siw/siw_mem.h
View File

@ -9,7 +9,7 @@
struct siw_umem *siw_umem_get(u64 start, u64 len, bool writable);
void siw_umem_release(struct siw_umem *umem, bool dirty);
struct siw_pbl *siw_pbl_alloc(u32 num_buf);
u64 siw_pbl_get_buffer(struct siw_pbl *pbl, u64 off, int *len, int *idx);
dma_addr_t siw_pbl_get_buffer(struct siw_pbl *pbl, u64 off, int *len, int *idx);
struct siw_mem *siw_mem_id2obj(struct siw_device *sdev, int stag_index);
int siw_mem_add(struct siw_device *sdev, struct siw_mem *m);
int siw_invalidate_stag(struct ib_pd *pd, u32 stag);

2
drivers/infiniband/sw/siw/siw_qp.c
View File

@ -949,7 +949,7 @@ skip_irq:
rv = -EINVAL;
goto out;
}
wqe->sqe.sge[0].laddr = (u64)&wqe->sqe.sge[1];
wqe->sqe.sge[0].laddr = (uintptr_t)&wqe->sqe.sge[1];
wqe->sqe.sge[0].lkey = 0;
wqe->sqe.num_sge = 1;
}

26
drivers/infiniband/sw/siw/siw_qp_rx.c
View File

@ -38,9 +38,10 @@ static int siw_rx_umem(struct siw_rx_stream *srx, struct siw_umem *umem,
p = siw_get_upage(umem, dest_addr);
if (unlikely(!p)) {
pr_warn("siw: %s: [QP %u]: bogus addr: %p, %p\n",
pr_warn("siw: %s: [QP %u]: bogus addr: %pK, %pK\n",
__func__, qp_id(rx_qp(srx)),
(void *)dest_addr, (void *)umem->fp_addr);
(void *)(uintptr_t)dest_addr,
(void *)(uintptr_t)umem->fp_addr);
/* siw internal error */
srx->skb_copied += copied;
srx->skb_new -= copied;
@ -50,7 +51,7 @@ static int siw_rx_umem(struct siw_rx_stream *srx, struct siw_umem *umem,
pg_off = dest_addr & ~PAGE_MASK;
bytes = min(len, (int)PAGE_SIZE - pg_off);
siw_dbg_qp(rx_qp(srx), "page %p, bytes=%u\n", p, bytes);
siw_dbg_qp(rx_qp(srx), "page %pK, bytes=%u\n", p, bytes);
dest = kmap_atomic(p);
rv = skb_copy_bits(srx->skb, srx->skb_offset, dest + pg_off,
@ -104,11 +105,11 @@ static int siw_rx_kva(struct siw_rx_stream *srx, void *kva, int len)
{
int rv;
siw_dbg_qp(rx_qp(srx), "kva: 0x%p, len: %u\n", kva, len);
siw_dbg_qp(rx_qp(srx), "kva: 0x%pK, len: %u\n", kva, len);
rv = skb_copy_bits(srx->skb, srx->skb_offset, kva, len);
if (unlikely(rv)) {
pr_warn("siw: [QP %u]: %s, len %d, kva 0x%p, rv %d\n",
pr_warn("siw: [QP %u]: %s, len %d, kva 0x%pK, rv %d\n",
qp_id(rx_qp(srx)), __func__, len, kva, rv);
return rv;
@ -132,7 +133,7 @@ static int siw_rx_pbl(struct siw_rx_stream *srx, int *pbl_idx,
while (len) {
int bytes;
u64 buf_addr =
dma_addr_t buf_addr =
siw_pbl_get_buffer(pbl, offset, &bytes, pbl_idx);
if (!buf_addr)
break;
@ -485,8 +486,8 @@ int siw_proc_send(struct siw_qp *qp)
mem_p = *mem;
if (mem_p->mem_obj == NULL)
rv = siw_rx_kva(srx,
(void *)(sge->laddr + frx->sge_off),
sge_bytes);
(void *)(uintptr_t)(sge->laddr + frx->sge_off),
sge_bytes);
else if (!mem_p->is_pbl)
rv = siw_rx_umem(srx, mem_p->umem,
sge->laddr + frx->sge_off, sge_bytes);
@ -598,8 +599,8 @@ int siw_proc_write(struct siw_qp *qp)
if (mem->mem_obj == NULL)
rv = siw_rx_kva(srx,
(void *)(srx->ddp_to + srx->fpdu_part_rcvd),
bytes);
(void *)(uintptr_t)(srx->ddp_to + srx->fpdu_part_rcvd),
bytes);
else if (!mem->is_pbl)
rv = siw_rx_umem(srx, mem->umem,
srx->ddp_to + srx->fpdu_part_rcvd, bytes);
@ -841,8 +842,9 @@ int siw_proc_rresp(struct siw_qp *qp)
bytes = min(srx->fpdu_part_rem, srx->skb_new);
if (mem_p->mem_obj == NULL)
rv = siw_rx_kva(srx, (void *)(sge->laddr + wqe->processed),
bytes);
rv = siw_rx_kva(srx,
(void *)(uintptr_t)(sge->laddr + wqe->processed),
bytes);
else if (!mem_p->is_pbl)
rv = siw_rx_umem(srx, mem_p->umem, sge->laddr + wqe->processed,
bytes);

80
drivers/infiniband/sw/siw/siw_qp_tx.c
View File

@ -26,7 +26,7 @@ static struct page *siw_get_pblpage(struct siw_mem *mem, u64 addr, int *idx)
{
struct siw_pbl *pbl = mem->pbl;
u64 offset = addr - mem->va;
u64 paddr = siw_pbl_get_buffer(pbl, offset, NULL, idx);
dma_addr_t paddr = siw_pbl_get_buffer(pbl, offset, NULL, idx);
if (paddr)
return virt_to_page(paddr);
@ -37,7 +37,7 @@ static struct page *siw_get_pblpage(struct siw_mem *mem, u64 addr, int *idx)
/*
* Copy short payload at provided destination payload address
*/
static int siw_try_1seg(struct siw_iwarp_tx *c_tx, u64 paddr)
static int siw_try_1seg(struct siw_iwarp_tx *c_tx, void *paddr)
{
struct siw_wqe *wqe = &c_tx->wqe_active;
struct siw_sge *sge = &wqe->sqe.sge[0];
@ -50,16 +50,16 @@ static int siw_try_1seg(struct siw_iwarp_tx *c_tx, u64 paddr)
return 0;
if (tx_flags(wqe) & SIW_WQE_INLINE) {
memcpy((void *)paddr, &wqe->sqe.sge[1], bytes);
memcpy(paddr, &wqe->sqe.sge[1], bytes);
} else {
struct siw_mem *mem = wqe->mem[0];
if (!mem->mem_obj) {
/* Kernel client using kva */
memcpy((void *)paddr, (void *)sge->laddr, bytes);
memcpy(paddr,
(const void *)(uintptr_t)sge->laddr, bytes);
} else if (c_tx->in_syscall) {
if (copy_from_user((void *)paddr,
(const void __user *)sge->laddr,
if (copy_from_user(paddr, u64_to_user_ptr(sge->laddr),
bytes))
return -EFAULT;
} else {
@ -79,12 +79,12 @@ static int siw_try_1seg(struct siw_iwarp_tx *c_tx, u64 paddr)
buffer = kmap_atomic(p);
if (likely(PAGE_SIZE - off >= bytes)) {
memcpy((void *)paddr, buffer + off, bytes);
memcpy(paddr, buffer + off, bytes);
kunmap_atomic(buffer);
} else {
unsigned long part = bytes - (PAGE_SIZE - off);
memcpy((void *)paddr, buffer + off, part);
memcpy(paddr, buffer + off, part);
kunmap_atomic(buffer);
if (!mem->is_pbl)
@ -98,7 +98,7 @@ static int siw_try_1seg(struct siw_iwarp_tx *c_tx, u64 paddr)
return -EFAULT;
buffer = kmap_atomic(p);
memcpy((void *)(paddr + part), buffer,
memcpy(paddr + part, buffer,
bytes - part);
kunmap_atomic(buffer);
}
@ -166,7 +166,7 @@ static int siw_qp_prepare_tx(struct siw_iwarp_tx *c_tx)
c_tx->ctrl_len = sizeof(struct iwarp_send);
crc = (char *)&c_tx->pkt.send_pkt.crc;
data = siw_try_1seg(c_tx, (u64)crc);
data = siw_try_1seg(c_tx, crc);
break;
case SIW_OP_SEND_REMOTE_INV:
@ -189,7 +189,7 @@ static int siw_qp_prepare_tx(struct siw_iwarp_tx *c_tx)
c_tx->ctrl_len = sizeof(struct iwarp_send_inv);
crc = (char *)&c_tx->pkt.send_pkt.crc;
data = siw_try_1seg(c_tx, (u64)crc);
data = siw_try_1seg(c_tx, crc);
break;
case SIW_OP_WRITE:
@ -201,7 +201,7 @@ static int siw_qp_prepare_tx(struct siw_iwarp_tx *c_tx)
c_tx->ctrl_len = sizeof(struct iwarp_rdma_write);
crc = (char *)&c_tx->pkt.write_pkt.crc;
data = siw_try_1seg(c_tx, (u64)crc);
data = siw_try_1seg(c_tx, crc);
break;
case SIW_OP_READ_RESPONSE:
@ -216,7 +216,7 @@ static int siw_qp_prepare_tx(struct siw_iwarp_tx *c_tx)
c_tx->ctrl_len = sizeof(struct iwarp_rdma_rresp);
crc = (char *)&c_tx->pkt.write_pkt.crc;
data = siw_try_1seg(c_tx, (u64)crc);
data = siw_try_1seg(c_tx, crc);
break;
default:
@ -398,15 +398,13 @@ static int siw_0copy_tx(struct socket *s, struct page **page,
#define MAX_TRAILER (MPA_CRC_SIZE + 4)
static void siw_unmap_pages(struct page **pages, int hdr_len, int num_maps)
static void siw_unmap_pages(struct page **pp, unsigned long kmap_mask)
{
if (hdr_len) {
++pages;
--num_maps;
}
while (num_maps-- > 0) {
kunmap(*pages);
pages++;
while (kmap_mask) {
if (kmap_mask & BIT(0))
kunmap(*pp);
pp++;
kmap_mask >>= 1;
}
}
@ -437,6 +435,7 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
unsigned int data_len = c_tx->bytes_unsent, hdr_len = 0, trl_len = 0,
sge_off = c_tx->sge_off, sge_idx = c_tx->sge_idx,
pbl_idx = c_tx->pbl_idx;
unsigned long kmap_mask = 0L;
if (c_tx->state == SIW_SEND_HDR) {
if (c_tx->use_sendpage) {
@ -463,8 +462,7 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
if (!(tx_flags(wqe) & SIW_WQE_INLINE)) {
mem = wqe->mem[sge_idx];
if (!mem->mem_obj)
is_kva = 1;
is_kva = mem->mem_obj == NULL ? 1 : 0;
} else {
is_kva = 1;
}
@ -473,7 +471,8 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
* tx from kernel virtual address: either inline data
* or memory region with assigned kernel buffer
*/
iov[seg].iov_base = (void *)(sge->laddr + sge_off);
iov[seg].iov_base =
(void *)(uintptr_t)(sge->laddr + sge_off);
iov[seg].iov_len = sge_len;
if (do_crc)
@ -500,12 +499,7 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
p = siw_get_upage(mem->umem,
sge->laddr + sge_off);
if (unlikely(!p)) {
if (hdr_len)
seg--;
if (!c_tx->use_sendpage && seg) {
siw_unmap_pages(page_array,
hdr_len, seg);
}
siw_unmap_pages(page_array, kmap_mask);
wqe->processed -= c_tx->bytes_unsent;
rv = -EFAULT;
goto done_crc;
@ -515,6 +509,10 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
if (!c_tx->use_sendpage) {
iov[seg].iov_base = kmap(p) + fp_off;
iov[seg].iov_len = plen;
/* Remember for later kunmap() */
kmap_mask |= BIT(seg);
if (do_crc)
crypto_shash_update(
c_tx->mpa_crc_hd,
@ -526,13 +524,13 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
page_address(p) + fp_off,
plen);
} else {
u64 pa = ((sge->laddr + sge_off) & PAGE_MASK);
u64 va = sge->laddr + sge_off;
page_array[seg] = virt_to_page(pa);
page_array[seg] = virt_to_page(va & PAGE_MASK);
if (do_crc)
crypto_shash_update(
c_tx->mpa_crc_hd,
(void *)(sge->laddr + sge_off),
(void *)(uintptr_t)va,
plen);
}
@ -543,10 +541,7 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
if (++seg > (int)MAX_ARRAY) {
siw_dbg_qp(tx_qp(c_tx), "to many fragments\n");
if (!is_kva && !c_tx->use_sendpage) {
siw_unmap_pages(page_array, hdr_len,
seg - 1);
}
siw_unmap_pages(page_array, kmap_mask);
wqe->processed -= c_tx->bytes_unsent;
rv = -EMSGSIZE;
goto done_crc;
@ -597,8 +592,7 @@ sge_done:
} else {
rv = kernel_sendmsg(s, &msg, iov, seg + 1,
hdr_len + data_len + trl_len);
if (!is_kva)
siw_unmap_pages(page_array, hdr_len, seg);
siw_unmap_pages(page_array, kmap_mask);
}
if (rv < (int)hdr_len) {
/* Not even complete hdr pushed or negative rv */
@ -829,7 +823,8 @@ static int siw_qp_sq_proc_tx(struct siw_qp *qp, struct siw_wqe *wqe)
rv = -EINVAL;
goto tx_error;
}
wqe->sqe.sge[0].laddr = (u64)&wqe->sqe.sge[1];
wqe->sqe.sge[0].laddr =
(u64)(uintptr_t)&wqe->sqe.sge[1];
}
}
wqe->wr_status = SIW_WR_INPROGRESS;
@ -924,7 +919,7 @@ tx_error:
static int siw_fastreg_mr(struct ib_pd *pd, struct siw_sqe *sqe)
{
struct ib_mr *base_mr = (struct ib_mr *)sqe->base_mr;
struct ib_mr *base_mr = (struct ib_mr *)(uintptr_t)sqe->base_mr;
struct siw_device *sdev = to_siw_dev(pd->device);
struct siw_mem *mem = siw_mem_id2obj(sdev, sqe->rkey >> 8);
int rv = 0;
@ -954,8 +949,7 @@ static int siw_fastreg_mr(struct ib_pd *pd, struct siw_sqe *sqe)
mem->stag = sqe->rkey;
mem->perms = sqe->access;
siw_dbg_mem(mem, "STag now valid, MR va: 0x%016llx -> 0x%016llx\n",
mem->va, base_mr->iova);
siw_dbg_mem(mem, "STag 0x%08x now valid\n", sqe->rkey);
mem->va = base_mr->iova;
mem->stag_valid = 1;
out:

40
drivers/infiniband/sw/siw/siw_verbs.c
View File

@ -424,8 +424,7 @@ struct ib_qp *siw_create_qp(struct ib_pd *pd,
*/
qp->srq = to_siw_srq(attrs->srq);
qp->attrs.rq_size = 0;
siw_dbg(base_dev, "QP [%u]: [SRQ 0x%p] attached\n",
qp->qp_num, qp->srq);
siw_dbg(base_dev, "QP [%u]: SRQ attached\n", qp->qp_num);
} else if (num_rqe) {
if (qp->kernel_verbs)
qp->recvq = vzalloc(num_rqe * sizeof(struct siw_rqe));
@ -610,7 +609,7 @@ int siw_destroy_qp(struct ib_qp *base_qp, struct ib_udata *udata)
base_ucontext);
struct siw_qp_attrs qp_attrs;
siw_dbg_qp(qp, "state %d, cep 0x%p\n", qp->attrs.state, qp->cep);
siw_dbg_qp(qp, "state %d\n", qp->attrs.state);
/*
* Mark QP as in process of destruction to prevent from
@ -662,7 +661,7 @@ static int siw_copy_inline_sgl(const struct ib_send_wr *core_wr,
void *kbuf = &sqe->sge[1];
int num_sge = core_wr->num_sge, bytes = 0;
sqe->sge[0].laddr = (u64)kbuf;
sqe->sge[0].laddr = (uintptr_t)kbuf;
sqe->sge[0].lkey = 0;
while (num_sge--) {
@ -825,7 +824,7 @@ int siw_post_send(struct ib_qp *base_qp, const struct ib_send_wr *wr,
break;
case IB_WR_REG_MR:
sqe->base_mr = (uint64_t)reg_wr(wr)->mr;
sqe->base_mr = (uintptr_t)reg_wr(wr)->mr;
sqe->rkey = reg_wr(wr)->key;
sqe->access = reg_wr(wr)->access & IWARP_ACCESS_MASK;
sqe->opcode = SIW_OP_REG_MR;
@ -842,8 +841,9 @@ int siw_post_send(struct ib_qp *base_qp, const struct ib_send_wr *wr,
rv = -EINVAL;
break;
}
siw_dbg_qp(qp, "opcode %d, flags 0x%x, wr_id 0x%p\n",
sqe->opcode, sqe->flags, (void *)sqe->id);
siw_dbg_qp(qp, "opcode %d, flags 0x%x, wr_id 0x%pK\n",
sqe->opcode, sqe->flags,
(void *)(uintptr_t)sqe->id);
if (unlikely(rv < 0))
break;
@ -1205,8 +1205,8 @@ struct ib_mr *siw_reg_user_mr(struct ib_pd *pd, u64 start, u64 len,
unsigned long mem_limit = rlimit(RLIMIT_MEMLOCK);
int rv;
siw_dbg_pd(pd, "start: 0x%016llx, va: 0x%016llx, len: %llu\n",
(unsigned long long)start, (unsigned long long)rnic_va,
siw_dbg_pd(pd, "start: 0x%pK, va: 0x%pK, len: %llu\n",
(void *)(uintptr_t)start, (void *)(uintptr_t)rnic_va,
(unsigned long long)len);
if (atomic_inc_return(&sdev->num_mr) > SIW_MAX_MR) {
@ -1363,7 +1363,7 @@ int siw_map_mr_sg(struct ib_mr *base_mr, struct scatterlist *sl, int num_sle,
struct siw_mem *mem = mr->mem;
struct siw_pbl *pbl = mem->pbl;
struct siw_pble *pble;
u64 pbl_size;
unsigned long pbl_size;
int i, rv;
if (!pbl) {
@ -1402,16 +1402,18 @@ int siw_map_mr_sg(struct ib_mr *base_mr, struct scatterlist *sl, int num_sle,
pbl_size += sg_dma_len(slp);
}
siw_dbg_mem(mem,
"sge[%d], size %llu, addr 0x%016llx, total %llu\n",
i, pble->size, pble->addr, pbl_size);
"sge[%d], size %u, addr 0x%p, total %lu\n",
i, pble->size, (void *)(uintptr_t)pble->addr,
pbl_size);
}
rv = ib_sg_to_pages(base_mr, sl, num_sle, sg_off, siw_set_pbl_page);
if (rv > 0) {
mem->len = base_mr->length;
mem->va = base_mr->iova;
siw_dbg_mem(mem,
"%llu bytes, start 0x%016llx, %u SLE to %u entries\n",
mem->len, mem->va, num_sle, pbl->num_buf);
"%llu bytes, start 0x%pK, %u SLE to %u entries\n",
mem->len, (void *)(uintptr_t)mem->va, num_sle,
pbl->num_buf);
}
return rv;
}
@ -1529,7 +1531,7 @@ int siw_create_srq(struct ib_srq *base_srq,
}
spin_lock_init(&srq->lock);
siw_dbg_pd(base_srq->pd, "[SRQ 0x%p]: success\n", srq);
siw_dbg_pd(base_srq->pd, "[SRQ]: success\n");
return 0;
@ -1650,8 +1652,7 @@ int siw_post_srq_recv(struct ib_srq *base_srq, const struct ib_recv_wr *wr,
if (unlikely(!srq->kernel_verbs)) {
siw_dbg_pd(base_srq->pd,
"[SRQ 0x%p]: no kernel post_recv for mapped srq\n",
srq);
"[SRQ]: no kernel post_recv for mapped srq\n");
rv = -EINVAL;
goto out;
}
@ -1673,8 +1674,7 @@ int siw_post_srq_recv(struct ib_srq *base_srq, const struct ib_recv_wr *wr,
}
if (unlikely(wr->num_sge > srq->max_sge)) {
siw_dbg_pd(base_srq->pd,
"[SRQ 0x%p]: too many sge's: %d\n", srq,
wr->num_sge);
"[SRQ]: too many sge's: %d\n", wr->num_sge);
rv = -EINVAL;
break;
}
@ -1693,7 +1693,7 @@ int siw_post_srq_recv(struct ib_srq *base_srq, const struct ib_recv_wr *wr,
spin_unlock_irqrestore(&srq->lock, flags);
out:
if (unlikely(rv < 0)) {
siw_dbg_pd(base_srq->pd, "[SRQ 0x%p]: error %d\n", srq, rv);
siw_dbg_pd(base_srq->pd, "[SRQ]: error %d\n", rv);
*bad_wr = wr;
}
return rv;

3
include/rdma/restrack.h
View File

@ -105,8 +105,7 @@ struct rdma_restrack_entry {
};
int rdma_restrack_count(struct ib_device *dev,
enum rdma_restrack_type type,
struct pid_namespace *ns);
enum rdma_restrack_type type);
void rdma_restrack_kadd(struct rdma_restrack_entry *res);
void rdma_restrack_uadd(struct rdma_restrack_entry *res);