kevin/linux
1
Fork 0
Code
e46a5e3ff0
linux/drivers/infiniband/hw/cxgb3/iwch_qp.c
Steve Wise f8b0dfd152 RDMA/cxgb3: Support peer-2-peer connection setup 
Open MPI, Intel MPI and other applications don't respect the iWARP
requirement that the client (active) side of the connection send the
first RDMA message.  This class of application connection setup is
called peer-to-peer.  Typically once the connection is setup, _both_
sides want to send data.

This patch enables supporting peer-to-peer over the chelsio RNIC by
enforcing this iWARP requirement in the driver itself as part of RDMA
connection setup.

Connection setup is extended, when the peer2peer module option is 1,
such that the MPA initiator will send a 0B Read (the RTR) just after
connection setup.  The MPA responder will suspend SQ processing until
the RTR message is received and reply-to.

In the longer term, this will be handled in a standardized way by
enhancing the MPA negotiation so peers can indicate whether they
want/need the RTR and what type of RTR (0B read, 0B write, or 0B send)
should be sent.  This will be done by standardizing a few bits of the
private data in order to negotiate all this.  However this patch
enables peer-to-peer applications now and allows most of the required
firmware and driver changes to be done and tested now.

Design:

 - Add a module option, peer2peer, to enable this mode.

 - New firmware support for peer-to-peer mode:

	- a new bit in the rdma_init WR to tell it to do peer-2-peer
	  and what form of RTR message to send or expect.

	- process _all_ preposted recvs before moving the connection
	  into rdma mode.

	- passive side: defer completing the rdma_init WR until all
	  pre-posted recvs are processed.  Suspend SQ processing until
	  the RTR is received.

	- active side: expect and process the 0B read WR on offload TX
	  queue. Defer completing the rdma_init WR until all
	  pre-posted recvs are processed.  Suspend SQ processing until
	  the 0B read WR is processed from the offload TX queue.

 - If peer2peer is set, driver posts 0B read request on offload TX
   queue just after posting the rdma_init WR to the offload TX queue.

 - Add CQ poll logic to ignore unsolicitied read responses.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-29 13:46:52 -07:00

1051 lines
28 KiB
C
Raw Blame History

/*
* Copyright (c) 2006 Chelsio, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#include "iwch_provider.h"
#include "iwch.h"
#include "iwch_cm.h"
#include "cxio_hal.h"
#define NO_SUPPORT -1
static int iwch_build_rdma_send(union t3_wr *wqe, struct ib_send_wr *wr,
u8 * flit_cnt)
{
int i;
u32 plen;
switch (wr->opcode) {
case IB_WR_SEND:
case IB_WR_SEND_WITH_IMM:
if (wr->send_flags & IB_SEND_SOLICITED)
wqe->send.rdmaop = T3_SEND_WITH_SE;
else
wqe->send.rdmaop = T3_SEND;
wqe->send.rem_stag = 0;
break;
#if 0 /* Not currently supported */
case TYPE_SEND_INVALIDATE:
case TYPE_SEND_INVALIDATE_IMMEDIATE:
wqe->send.rdmaop = T3_SEND_WITH_INV;
wqe->send.rem_stag = cpu_to_be32(wr->wr.rdma.rkey);
break;
case TYPE_SEND_SE_INVALIDATE:
wqe->send.rdmaop = T3_SEND_WITH_SE_INV;
wqe->send.rem_stag = cpu_to_be32(wr->wr.rdma.rkey);
break;
#endif
default:
break;
}
if (wr->num_sge > T3_MAX_SGE)
return -EINVAL;
wqe->send.reserved[0] = 0;
wqe->send.reserved[1] = 0;
wqe->send.reserved[2] = 0;
if (wr->opcode == IB_WR_SEND_WITH_IMM) {
plen = 4;
wqe->send.sgl[0].stag = wr->ex.imm_data;
wqe->send.sgl[0].len = __constant_cpu_to_be32(0);
wqe->send.num_sgle = __constant_cpu_to_be32(0);
*flit_cnt = 5;
} else {
plen = 0;
for (i = 0; i < wr->num_sge; i++) {
if ((plen + wr->sg_list[i].length) < plen) {
return -EMSGSIZE;
}
plen += wr->sg_list[i].length;
wqe->send.sgl[i].stag =
cpu_to_be32(wr->sg_list[i].lkey);
wqe->send.sgl[i].len =
cpu_to_be32(wr->sg_list[i].length);
wqe->send.sgl[i].to = cpu_to_be64(wr->sg_list[i].addr);
}
wqe->send.num_sgle = cpu_to_be32(wr->num_sge);
*flit_cnt = 4 + ((wr->num_sge) << 1);
}
wqe->send.plen = cpu_to_be32(plen);
return 0;
}
static int iwch_build_rdma_write(union t3_wr *wqe, struct ib_send_wr *wr,
u8 *flit_cnt)
{
int i;
u32 plen;
if (wr->num_sge > T3_MAX_SGE)
return -EINVAL;
wqe->write.rdmaop = T3_RDMA_WRITE;
wqe->write.reserved[0] = 0;
wqe->write.reserved[1] = 0;
wqe->write.reserved[2] = 0;
wqe->write.stag_sink = cpu_to_be32(wr->wr.rdma.rkey);
wqe->write.to_sink = cpu_to_be64(wr->wr.rdma.remote_addr);
if (wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM) {
plen = 4;
wqe->write.sgl[0].stag = wr->ex.imm_data;
wqe->write.sgl[0].len = __constant_cpu_to_be32(0);
wqe->write.num_sgle = __constant_cpu_to_be32(0);
*flit_cnt = 6;
} else {
plen = 0;
for (i = 0; i < wr->num_sge; i++) {
if ((plen + wr->sg_list[i].length) < plen) {
return -EMSGSIZE;
}
plen += wr->sg_list[i].length;
wqe->write.sgl[i].stag =
cpu_to_be32(wr->sg_list[i].lkey);
wqe->write.sgl[i].len =
cpu_to_be32(wr->sg_list[i].length);
wqe->write.sgl[i].to =
cpu_to_be64(wr->sg_list[i].addr);
}
wqe->write.num_sgle = cpu_to_be32(wr->num_sge);
*flit_cnt = 5 + ((wr->num_sge) << 1);
}
wqe->write.plen = cpu_to_be32(plen);
return 0;
}
static int iwch_build_rdma_read(union t3_wr *wqe, struct ib_send_wr *wr,
u8 *flit_cnt)
{
if (wr->num_sge > 1)
return -EINVAL;
wqe->read.rdmaop = T3_READ_REQ;
wqe->read.reserved[0] = 0;
wqe->read.reserved[1] = 0;
wqe->read.reserved[2] = 0;
wqe->read.rem_stag = cpu_to_be32(wr->wr.rdma.rkey);
wqe->read.rem_to = cpu_to_be64(wr->wr.rdma.remote_addr);
wqe->read.local_stag = cpu_to_be32(wr->sg_list[0].lkey);
wqe->read.local_len = cpu_to_be32(wr->sg_list[0].length);
wqe->read.local_to = cpu_to_be64(wr->sg_list[0].addr);
*flit_cnt = sizeof(struct t3_rdma_read_wr) >> 3;
return 0;
}
/*
* TBD: this is going to be moved to firmware. Missing pdid/qpid check for now.
*/
static int iwch_sgl2pbl_map(struct iwch_dev *rhp, struct ib_sge *sg_list,
u32 num_sgle, u32 * pbl_addr, u8 * page_size)
{
int i;
struct iwch_mr *mhp;
u32 offset;
for (i = 0; i < num_sgle; i++) {
mhp = get_mhp(rhp, (sg_list[i].lkey) >> 8);
if (!mhp) {
PDBG("%s %d\n", __func__, __LINE__);
return -EIO;
}
if (!mhp->attr.state) {
PDBG("%s %d\n", __func__, __LINE__);
return -EIO;
}
if (mhp->attr.zbva) {
PDBG("%s %d\n", __func__, __LINE__);
return -EIO;
}
if (sg_list[i].addr < mhp->attr.va_fbo) {
PDBG("%s %d\n", __func__, __LINE__);
return -EINVAL;
}
if (sg_list[i].addr + ((u64) sg_list[i].length) <
sg_list[i].addr) {
PDBG("%s %d\n", __func__, __LINE__);
return -EINVAL;
}
if (sg_list[i].addr + ((u64) sg_list[i].length) >
mhp->attr.va_fbo + ((u64) mhp->attr.len)) {
PDBG("%s %d\n", __func__, __LINE__);
return -EINVAL;
}
offset = sg_list[i].addr - mhp->attr.va_fbo;
offset += ((u32) mhp->attr.va_fbo) %
(1UL << (12 + mhp->attr.page_size));
pbl_addr[i] = ((mhp->attr.pbl_addr -
rhp->rdev.rnic_info.pbl_base) >> 3) +
(offset >> (12 + mhp->attr.page_size));
page_size[i] = mhp->attr.page_size;
}
return 0;
}
static int iwch_build_rdma_recv(struct iwch_dev *rhp, union t3_wr *wqe,
struct ib_recv_wr *wr)
{
int i;
if (wr->num_sge > T3_MAX_SGE)
return -EINVAL;
wqe->recv.num_sgle = cpu_to_be32(wr->num_sge);
for (i = 0; i < wr->num_sge; i++) {
wqe->recv.sgl[i].stag = cpu_to_be32(wr->sg_list[i].lkey);
wqe->recv.sgl[i].len = cpu_to_be32(wr->sg_list[i].length);
wqe->recv.sgl[i].to = cpu_to_be64(wr->sg_list[i].addr);
}
for (; i < T3_MAX_SGE; i++) {
wqe->recv.sgl[i].stag = 0;
wqe->recv.sgl[i].len = 0;
wqe->recv.sgl[i].to = 0;
}
return 0;
}
int iwch_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
struct ib_send_wr **bad_wr)
{
int err = 0;
u8 t3_wr_flit_cnt;
enum t3_wr_opcode t3_wr_opcode = 0;
enum t3_wr_flags t3_wr_flags;
struct iwch_qp *qhp;
u32 idx;
union t3_wr *wqe;
u32 num_wrs;
unsigned long flag;
struct t3_swsq *sqp;
qhp = to_iwch_qp(ibqp);
spin_lock_irqsave(&qhp->lock, flag);
if (qhp->attr.state > IWCH_QP_STATE_RTS) {
spin_unlock_irqrestore(&qhp->lock, flag);
return -EINVAL;
}
num_wrs = Q_FREECNT(qhp->wq.sq_rptr, qhp->wq.sq_wptr,
qhp->wq.sq_size_log2);
if (num_wrs <= 0) {
spin_unlock_irqrestore(&qhp->lock, flag);
return -ENOMEM;
}
while (wr) {
if (num_wrs == 0) {
err = -ENOMEM;
*bad_wr = wr;
break;
}
idx = Q_PTR2IDX(qhp->wq.wptr, qhp->wq.size_log2);
wqe = (union t3_wr *) (qhp->wq.queue + idx);
t3_wr_flags = 0;
if (wr->send_flags & IB_SEND_SOLICITED)
t3_wr_flags |= T3_SOLICITED_EVENT_FLAG;
if (wr->send_flags & IB_SEND_FENCE)
t3_wr_flags |= T3_READ_FENCE_FLAG;
if (wr->send_flags & IB_SEND_SIGNALED)
t3_wr_flags |= T3_COMPLETION_FLAG;
sqp = qhp->wq.sq +
Q_PTR2IDX(qhp->wq.sq_wptr, qhp->wq.sq_size_log2);
switch (wr->opcode) {
case IB_WR_SEND:
case IB_WR_SEND_WITH_IMM:
t3_wr_opcode = T3_WR_SEND;
err = iwch_build_rdma_send(wqe, wr, &t3_wr_flit_cnt);
break;
case IB_WR_RDMA_WRITE:
case IB_WR_RDMA_WRITE_WITH_IMM:
t3_wr_opcode = T3_WR_WRITE;
err = iwch_build_rdma_write(wqe, wr, &t3_wr_flit_cnt);
break;
case IB_WR_RDMA_READ:
t3_wr_opcode = T3_WR_READ;
t3_wr_flags = 0; /* T3 reads are always signaled */
err = iwch_build_rdma_read(wqe, wr, &t3_wr_flit_cnt);
if (err)
break;
sqp->read_len = wqe->read.local_len;
if (!qhp->wq.oldest_read)
qhp->wq.oldest_read = sqp;
break;
default:
PDBG("%s post of type=%d TBD!\n", __func__,
wr->opcode);
err = -EINVAL;
}
if (err) {
*bad_wr = wr;
break;
}
wqe->send.wrid.id0.hi = qhp->wq.sq_wptr;
sqp->wr_id = wr->wr_id;
sqp->opcode = wr2opcode(t3_wr_opcode);
sqp->sq_wptr = qhp->wq.sq_wptr;
sqp->complete = 0;
sqp->signaled = (wr->send_flags & IB_SEND_SIGNALED);
build_fw_riwrh((void *) wqe, t3_wr_opcode, t3_wr_flags,
Q_GENBIT(qhp->wq.wptr, qhp->wq.size_log2),
0, t3_wr_flit_cnt);
PDBG("%s cookie 0x%llx wq idx 0x%x swsq idx %ld opcode %d\n",
__func__, (unsigned long long) wr->wr_id, idx,
Q_PTR2IDX(qhp->wq.sq_wptr, qhp->wq.sq_size_log2),
sqp->opcode);
wr = wr->next;
num_wrs--;
++(qhp->wq.wptr);
++(qhp->wq.sq_wptr);
}
spin_unlock_irqrestore(&qhp->lock, flag);
ring_doorbell(qhp->wq.doorbell, qhp->wq.qpid);
return err;
}
int iwch_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr,
struct ib_recv_wr **bad_wr)
{
int err = 0;
struct iwch_qp *qhp;
u32 idx;
union t3_wr *wqe;
u32 num_wrs;
unsigned long flag;
qhp = to_iwch_qp(ibqp);
spin_lock_irqsave(&qhp->lock, flag);
if (qhp->attr.state > IWCH_QP_STATE_RTS) {
spin_unlock_irqrestore(&qhp->lock, flag);
return -EINVAL;
}
num_wrs = Q_FREECNT(qhp->wq.rq_rptr, qhp->wq.rq_wptr,
qhp->wq.rq_size_log2) - 1;
if (!wr) {
spin_unlock_irqrestore(&qhp->lock, flag);
return -EINVAL;
}
while (wr) {
idx = Q_PTR2IDX(qhp->wq.wptr, qhp->wq.size_log2);
wqe = (union t3_wr *) (qhp->wq.queue + idx);
if (num_wrs)
err = iwch_build_rdma_recv(qhp->rhp, wqe, wr);
else
err = -ENOMEM;
if (err) {
*bad_wr = wr;
break;
}
qhp->wq.rq[Q_PTR2IDX(qhp->wq.rq_wptr, qhp->wq.rq_size_log2)] =
wr->wr_id;
build_fw_riwrh((void *) wqe, T3_WR_RCV, T3_COMPLETION_FLAG,
Q_GENBIT(qhp->wq.wptr, qhp->wq.size_log2),
0, sizeof(struct t3_receive_wr) >> 3);
PDBG("%s cookie 0x%llx idx 0x%x rq_wptr 0x%x rw_rptr 0x%x "
"wqe %p \n", __func__, (unsigned long long) wr->wr_id,
idx, qhp->wq.rq_wptr, qhp->wq.rq_rptr, wqe);
++(qhp->wq.rq_wptr);
++(qhp->wq.wptr);
wr = wr->next;
num_wrs--;
}
spin_unlock_irqrestore(&qhp->lock, flag);
ring_doorbell(qhp->wq.doorbell, qhp->wq.qpid);
return err;
}
int iwch_bind_mw(struct ib_qp *qp,
struct ib_mw *mw,
struct ib_mw_bind *mw_bind)
{
struct iwch_dev *rhp;
struct iwch_mw *mhp;
struct iwch_qp *qhp;
union t3_wr *wqe;
u32 pbl_addr;
u8 page_size;
u32 num_wrs;
unsigned long flag;
struct ib_sge sgl;
int err=0;
enum t3_wr_flags t3_wr_flags;
u32 idx;
struct t3_swsq *sqp;
qhp = to_iwch_qp(qp);
mhp = to_iwch_mw(mw);
rhp = qhp->rhp;
spin_lock_irqsave(&qhp->lock, flag);
if (qhp->attr.state > IWCH_QP_STATE_RTS) {
spin_unlock_irqrestore(&qhp->lock, flag);
return -EINVAL;
}
num_wrs = Q_FREECNT(qhp->wq.sq_rptr, qhp->wq.sq_wptr,
qhp->wq.sq_size_log2);
if ((num_wrs) <= 0) {
spin_unlock_irqrestore(&qhp->lock, flag);
return -ENOMEM;
}
idx = Q_PTR2IDX(qhp->wq.wptr, qhp->wq.size_log2);
PDBG("%s: idx 0x%0x, mw 0x%p, mw_bind 0x%p\n", __func__, idx,
mw, mw_bind);
wqe = (union t3_wr *) (qhp->wq.queue + idx);
t3_wr_flags = 0;
if (mw_bind->send_flags & IB_SEND_SIGNALED)
t3_wr_flags = T3_COMPLETION_FLAG;
sgl.addr = mw_bind->addr;
sgl.lkey = mw_bind->mr->lkey;
sgl.length = mw_bind->length;
wqe->bind.reserved = 0;
wqe->bind.type = T3_VA_BASED_TO;
/* TBD: check perms */
wqe->bind.perms = iwch_ib_to_mwbind_access(mw_bind->mw_access_flags);
wqe->bind.mr_stag = cpu_to_be32(mw_bind->mr->lkey);
wqe->bind.mw_stag = cpu_to_be32(mw->rkey);
wqe->bind.mw_len = cpu_to_be32(mw_bind->length);
wqe->bind.mw_va = cpu_to_be64(mw_bind->addr);
err = iwch_sgl2pbl_map(rhp, &sgl, 1, &pbl_addr, &page_size);
if (err) {
spin_unlock_irqrestore(&qhp->lock, flag);
return err;
}
wqe->send.wrid.id0.hi = qhp->wq.sq_wptr;
sqp = qhp->wq.sq + Q_PTR2IDX(qhp->wq.sq_wptr, qhp->wq.sq_size_log2);
sqp->wr_id = mw_bind->wr_id;
sqp->opcode = T3_BIND_MW;
sqp->sq_wptr = qhp->wq.sq_wptr;
sqp->complete = 0;
sqp->signaled = (mw_bind->send_flags & IB_SEND_SIGNALED);
wqe->bind.mr_pbl_addr = cpu_to_be32(pbl_addr);
wqe->bind.mr_pagesz = page_size;
wqe->flit[T3_SQ_COOKIE_FLIT] = mw_bind->wr_id;
build_fw_riwrh((void *)wqe, T3_WR_BIND, t3_wr_flags,
Q_GENBIT(qhp->wq.wptr, qhp->wq.size_log2), 0,
sizeof(struct t3_bind_mw_wr) >> 3);
++(qhp->wq.wptr);
++(qhp->wq.sq_wptr);
spin_unlock_irqrestore(&qhp->lock, flag);
ring_doorbell(qhp->wq.doorbell, qhp->wq.qpid);
return err;
}
static inline void build_term_codes(struct respQ_msg_t *rsp_msg,
u8 *layer_type, u8 *ecode)
{
int status = TPT_ERR_INTERNAL_ERR;
int tagged = 0;
int opcode = -1;
int rqtype = 0;
int send_inv = 0;
if (rsp_msg) {
status = CQE_STATUS(rsp_msg->cqe);
opcode = CQE_OPCODE(rsp_msg->cqe);
rqtype = RQ_TYPE(rsp_msg->cqe);
send_inv = (opcode == T3_SEND_WITH_INV) ||
(opcode == T3_SEND_WITH_SE_INV);
tagged = (opcode == T3_RDMA_WRITE) ||
(rqtype && (opcode == T3_READ_RESP));
}
switch (status) {
case TPT_ERR_STAG:
if (send_inv) {
*layer_type = LAYER_RDMAP|RDMAP_REMOTE_OP;
*ecode = RDMAP_CANT_INV_STAG;
} else {
*layer_type = LAYER_RDMAP|RDMAP_REMOTE_PROT;
*ecode = RDMAP_INV_STAG;
}
break;
case TPT_ERR_PDID:
*layer_type = LAYER_RDMAP|RDMAP_REMOTE_PROT;
if ((opcode == T3_SEND_WITH_INV) ||
(opcode == T3_SEND_WITH_SE_INV))
*ecode = RDMAP_CANT_INV_STAG;
else
*ecode = RDMAP_STAG_NOT_ASSOC;
break;
case TPT_ERR_QPID:
*layer_type = LAYER_RDMAP|RDMAP_REMOTE_PROT;
*ecode = RDMAP_STAG_NOT_ASSOC;
break;
case TPT_ERR_ACCESS:
*layer_type = LAYER_RDMAP|RDMAP_REMOTE_PROT;
*ecode = RDMAP_ACC_VIOL;
break;
case TPT_ERR_WRAP:
*layer_type = LAYER_RDMAP|RDMAP_REMOTE_PROT;
*ecode = RDMAP_TO_WRAP;
break;
case TPT_ERR_BOUND:
if (tagged) {
*layer_type = LAYER_DDP|DDP_TAGGED_ERR;
*ecode = DDPT_BASE_BOUNDS;
} else {
*layer_type = LAYER_RDMAP|RDMAP_REMOTE_PROT;
*ecode = RDMAP_BASE_BOUNDS;
}
break;
case TPT_ERR_INVALIDATE_SHARED_MR:
case TPT_ERR_INVALIDATE_MR_WITH_MW_BOUND:
*layer_type = LAYER_RDMAP|RDMAP_REMOTE_OP;
*ecode = RDMAP_CANT_INV_STAG;
break;
case TPT_ERR_ECC:
case TPT_ERR_ECC_PSTAG:
case TPT_ERR_INTERNAL_ERR:
*layer_type = LAYER_RDMAP|RDMAP_LOCAL_CATA;
*ecode = 0;
break;
case TPT_ERR_OUT_OF_RQE:
*layer_type = LAYER_DDP|DDP_UNTAGGED_ERR;
*ecode = DDPU_INV_MSN_NOBUF;
break;
case TPT_ERR_PBL_ADDR_BOUND:
*layer_type = LAYER_DDP|DDP_TAGGED_ERR;
*ecode = DDPT_BASE_BOUNDS;
break;
case TPT_ERR_CRC:
*layer_type = LAYER_MPA|DDP_LLP;
*ecode = MPA_CRC_ERR;
break;
case TPT_ERR_MARKER:
*layer_type = LAYER_MPA|DDP_LLP;
*ecode = MPA_MARKER_ERR;
break;
case TPT_ERR_PDU_LEN_ERR:
*layer_type = LAYER_DDP|DDP_UNTAGGED_ERR;
*ecode = DDPU_MSG_TOOBIG;
break;
case TPT_ERR_DDP_VERSION:
if (tagged) {
*layer_type = LAYER_DDP|DDP_TAGGED_ERR;
*ecode = DDPT_INV_VERS;
} else {
*layer_type = LAYER_DDP|DDP_UNTAGGED_ERR;
*ecode = DDPU_INV_VERS;
}
break;
case TPT_ERR_RDMA_VERSION:
*layer_type = LAYER_RDMAP|RDMAP_REMOTE_OP;
*ecode = RDMAP_INV_VERS;
break;
case TPT_ERR_OPCODE:
*layer_type = LAYER_RDMAP|RDMAP_REMOTE_OP;
*ecode = RDMAP_INV_OPCODE;
break;
case TPT_ERR_DDP_QUEUE_NUM:
*layer_type = LAYER_DDP|DDP_UNTAGGED_ERR;
*ecode = DDPU_INV_QN;
break;
case TPT_ERR_MSN:
case TPT_ERR_MSN_GAP:
case TPT_ERR_MSN_RANGE:
case TPT_ERR_IRD_OVERFLOW:
*layer_type = LAYER_DDP|DDP_UNTAGGED_ERR;
*ecode = DDPU_INV_MSN_RANGE;
break;
case TPT_ERR_TBIT:
*layer_type = LAYER_DDP|DDP_LOCAL_CATA;
*ecode = 0;
break;
case TPT_ERR_MO:
*layer_type = LAYER_DDP|DDP_UNTAGGED_ERR;
*ecode = DDPU_INV_MO;
break;
default:
*layer_type = LAYER_RDMAP|DDP_LOCAL_CATA;
*ecode = 0;
break;
}
}
int iwch_post_zb_read(struct iwch_qp *qhp)
{
union t3_wr *wqe;
struct sk_buff *skb;
u8 flit_cnt = sizeof(struct t3_rdma_read_wr) >> 3;
PDBG("%s enter\n", __func__);
skb = alloc_skb(40, GFP_KERNEL);
if (!skb) {
printk(KERN_ERR "%s cannot send zb_read!!\n", __func__);
return -ENOMEM;
}
wqe = (union t3_wr *)skb_put(skb, sizeof(struct t3_rdma_read_wr));
memset(wqe, 0, sizeof(struct t3_rdma_read_wr));
wqe->read.rdmaop = T3_READ_REQ;
wqe->read.reserved[0] = 0;
wqe->read.reserved[1] = 0;
wqe->read.reserved[2] = 0;
wqe->read.rem_stag = cpu_to_be32(1);
wqe->read.rem_to = cpu_to_be64(1);
wqe->read.local_stag = cpu_to_be32(1);
wqe->read.local_len = cpu_to_be32(0);
wqe->read.local_to = cpu_to_be64(1);
wqe->send.wrh.op_seop_flags = cpu_to_be32(V_FW_RIWR_OP(T3_WR_READ));
wqe->send.wrh.gen_tid_len = cpu_to_be32(V_FW_RIWR_TID(qhp->ep->hwtid)|
V_FW_RIWR_LEN(flit_cnt));
skb->priority = CPL_PRIORITY_DATA;
return cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb);
}
/*
* This posts a TERMINATE with layer=RDMA, type=catastrophic.
*/
int iwch_post_terminate(struct iwch_qp *qhp, struct respQ_msg_t *rsp_msg)
{
union t3_wr *wqe;
struct terminate_message *term;
struct sk_buff *skb;
PDBG("%s %d\n", __func__, __LINE__);
skb = alloc_skb(40, GFP_ATOMIC);
if (!skb) {
printk(KERN_ERR "%s cannot send TERMINATE!\n", __func__);
return -ENOMEM;
}
wqe = (union t3_wr *)skb_put(skb, 40);
memset(wqe, 0, 40);
wqe->send.rdmaop = T3_TERMINATE;
/* immediate data length */
wqe->send.plen = htonl(4);
/* immediate data starts here. */
term = (struct terminate_message *)wqe->send.sgl;
build_term_codes(rsp_msg, &term->layer_etype, &term->ecode);
wqe->send.wrh.op_seop_flags = cpu_to_be32(V_FW_RIWR_OP(T3_WR_SEND) |
V_FW_RIWR_FLAGS(T3_COMPLETION_FLAG | T3_NOTIFY_FLAG));
wqe->send.wrh.gen_tid_len = cpu_to_be32(V_FW_RIWR_TID(qhp->ep->hwtid));
skb->priority = CPL_PRIORITY_DATA;
return cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb);
}
/*
* Assumes qhp lock is held.
*/
static void __flush_qp(struct iwch_qp *qhp, unsigned long *flag)
{
struct iwch_cq *rchp, *schp;
int count;
rchp = get_chp(qhp->rhp, qhp->attr.rcq);
schp = get_chp(qhp->rhp, qhp->attr.scq);
PDBG("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp);
/* take a ref on the qhp since we must release the lock */
atomic_inc(&qhp->refcnt);
spin_unlock_irqrestore(&qhp->lock, *flag);
/* locking heirarchy: cq lock first, then qp lock. */
spin_lock_irqsave(&rchp->lock, *flag);
spin_lock(&qhp->lock);
cxio_flush_hw_cq(&rchp->cq);
cxio_count_rcqes(&rchp->cq, &qhp->wq, &count);
cxio_flush_rq(&qhp->wq, &rchp->cq, count);
spin_unlock(&qhp->lock);
spin_unlock_irqrestore(&rchp->lock, *flag);
(*rchp->ibcq.comp_handler)(&rchp->ibcq, rchp->ibcq.cq_context);
/* locking heirarchy: cq lock first, then qp lock. */
spin_lock_irqsave(&schp->lock, *flag);
spin_lock(&qhp->lock);
cxio_flush_hw_cq(&schp->cq);
cxio_count_scqes(&schp->cq, &qhp->wq, &count);
cxio_flush_sq(&qhp->wq, &schp->cq, count);
spin_unlock(&qhp->lock);
spin_unlock_irqrestore(&schp->lock, *flag);
(*schp->ibcq.comp_handler)(&schp->ibcq, schp->ibcq.cq_context);
/* deref */
if (atomic_dec_and_test(&qhp->refcnt))
wake_up(&qhp->wait);
spin_lock_irqsave(&qhp->lock, *flag);
}
static void flush_qp(struct iwch_qp *qhp, unsigned long *flag)
{
if (qhp->ibqp.uobject)
cxio_set_wq_in_error(&qhp->wq);
else
__flush_qp(qhp, flag);
}
/*
* Return count of RECV WRs posted
*/
u16 iwch_rqes_posted(struct iwch_qp *qhp)
{
union t3_wr *wqe = qhp->wq.queue;
u16 count = 0;
while ((count+1) != 0 && fw_riwrh_opcode((struct fw_riwrh *)wqe) == T3_WR_RCV) {
count++;
wqe++;
}
PDBG("%s qhp %p count %u\n", __func__, qhp, count);
return count;
}
static int rdma_init(struct iwch_dev *rhp, struct iwch_qp *qhp,
enum iwch_qp_attr_mask mask,
struct iwch_qp_attributes *attrs)
{
struct t3_rdma_init_attr init_attr;
int ret;
init_attr.tid = qhp->ep->hwtid;
init_attr.qpid = qhp->wq.qpid;
init_attr.pdid = qhp->attr.pd;
init_attr.scqid = qhp->attr.scq;
init_attr.rcqid = qhp->attr.rcq;
init_attr.rq_addr = qhp->wq.rq_addr;
init_attr.rq_size = 1 << qhp->wq.rq_size_log2;
init_attr.mpaattrs = uP_RI_MPA_IETF_ENABLE |
qhp->attr.mpa_attr.recv_marker_enabled |
(qhp->attr.mpa_attr.xmit_marker_enabled << 1) |
(qhp->attr.mpa_attr.crc_enabled << 2);
/*
* XXX - The IWCM doesn't quite handle getting these
* attrs set before going into RTS. For now, just turn
* them on always...
*/
#if 0
init_attr.qpcaps = qhp->attr.enableRdmaRead |
(qhp->attr.enableRdmaWrite << 1) |
(qhp->attr.enableBind << 2) |
(qhp->attr.enable_stag0_fastreg << 3) |
(qhp->attr.enable_stag0_fastreg << 4);
#else
init_attr.qpcaps = 0x1f;
#endif
init_attr.tcp_emss = qhp->ep->emss;
init_attr.ord = qhp->attr.max_ord;
init_attr.ird = qhp->attr.max_ird;
init_attr.qp_dma_addr = qhp->wq.dma_addr;
init_attr.qp_dma_size = (1UL << qhp->wq.size_log2);
init_attr.rqe_count = iwch_rqes_posted(qhp);
init_attr.flags = qhp->attr.mpa_attr.initiator ? MPA_INITIATOR : 0;
init_attr.flags |= capable(CAP_NET_BIND_SERVICE) ? PRIV_QP : 0;
if (peer2peer) {
init_attr.rtr_type = RTR_READ;
if (init_attr.ord == 0 && qhp->attr.mpa_attr.initiator)
init_attr.ord = 1;
if (init_attr.ird == 0 && !qhp->attr.mpa_attr.initiator)
init_attr.ird = 1;
} else
init_attr.rtr_type = 0;
init_attr.irs = qhp->ep->rcv_seq;
PDBG("%s init_attr.rq_addr 0x%x init_attr.rq_size = %d "
"flags 0x%x qpcaps 0x%x\n", __func__,
init_attr.rq_addr, init_attr.rq_size,
init_attr.flags, init_attr.qpcaps);
ret = cxio_rdma_init(&rhp->rdev, &init_attr);
PDBG("%s ret %d\n", __func__, ret);
return ret;
}
int iwch_modify_qp(struct iwch_dev *rhp, struct iwch_qp *qhp,
enum iwch_qp_attr_mask mask,
struct iwch_qp_attributes *attrs,
int internal)
{
int ret = 0;
struct iwch_qp_attributes newattr = qhp->attr;
unsigned long flag;
int disconnect = 0;
int terminate = 0;
int abort = 0;
int free = 0;
struct iwch_ep *ep = NULL;
PDBG("%s qhp %p qpid 0x%x ep %p state %d -> %d\n", __func__,
qhp, qhp->wq.qpid, qhp->ep, qhp->attr.state,
(mask & IWCH_QP_ATTR_NEXT_STATE) ? attrs->next_state : -1);
spin_lock_irqsave(&qhp->lock, flag);
/* Process attr changes if in IDLE */
if (mask & IWCH_QP_ATTR_VALID_MODIFY) {
if (qhp->attr.state != IWCH_QP_STATE_IDLE) {
ret = -EIO;
goto out;
}
if (mask & IWCH_QP_ATTR_ENABLE_RDMA_READ)
newattr.enable_rdma_read = attrs->enable_rdma_read;
if (mask & IWCH_QP_ATTR_ENABLE_RDMA_WRITE)
newattr.enable_rdma_write = attrs->enable_rdma_write;
if (mask & IWCH_QP_ATTR_ENABLE_RDMA_BIND)
newattr.enable_bind = attrs->enable_bind;
if (mask & IWCH_QP_ATTR_MAX_ORD) {
if (attrs->max_ord >
rhp->attr.max_rdma_read_qp_depth) {
ret = -EINVAL;
goto out;
}
newattr.max_ord = attrs->max_ord;
}
if (mask & IWCH_QP_ATTR_MAX_IRD) {
if (attrs->max_ird >
rhp->attr.max_rdma_reads_per_qp) {
ret = -EINVAL;
goto out;
}
newattr.max_ird = attrs->max_ird;
}
qhp->attr = newattr;
}
if (!(mask & IWCH_QP_ATTR_NEXT_STATE))
goto out;
if (qhp->attr.state == attrs->next_state)
goto out;
switch (qhp->attr.state) {
case IWCH_QP_STATE_IDLE:
switch (attrs->next_state) {
case IWCH_QP_STATE_RTS:
if (!(mask & IWCH_QP_ATTR_LLP_STREAM_HANDLE)) {
ret = -EINVAL;
goto out;
}
if (!(mask & IWCH_QP_ATTR_MPA_ATTR)) {
ret = -EINVAL;
goto out;
}
qhp->attr.mpa_attr = attrs->mpa_attr;
qhp->attr.llp_stream_handle = attrs->llp_stream_handle;
qhp->ep = qhp->attr.llp_stream_handle;
qhp->attr.state = IWCH_QP_STATE_RTS;
/*
* Ref the endpoint here and deref when we
* disassociate the endpoint from the QP. This
* happens in CLOSING->IDLE transition or *->ERROR
* transition.
*/
get_ep(&qhp->ep->com);
spin_unlock_irqrestore(&qhp->lock, flag);
ret = rdma_init(rhp, qhp, mask, attrs);
spin_lock_irqsave(&qhp->lock, flag);
if (ret)
goto err;
break;
case IWCH_QP_STATE_ERROR:
qhp->attr.state = IWCH_QP_STATE_ERROR;
flush_qp(qhp, &flag);
break;
default:
ret = -EINVAL;
goto out;
}
break;
case IWCH_QP_STATE_RTS:
switch (attrs->next_state) {
case IWCH_QP_STATE_CLOSING:
BUG_ON(atomic_read(&qhp->ep->com.kref.refcount) < 2);
qhp->attr.state = IWCH_QP_STATE_CLOSING;
if (!internal) {
abort=0;
disconnect = 1;
ep = qhp->ep;
get_ep(&ep->com);
}
flush_qp(qhp, &flag);
break;
case IWCH_QP_STATE_TERMINATE:
qhp->attr.state = IWCH_QP_STATE_TERMINATE;
if (qhp->ibqp.uobject)
cxio_set_wq_in_error(&qhp->wq);
if (!internal)
terminate = 1;
break;
case IWCH_QP_STATE_ERROR:
qhp->attr.state = IWCH_QP_STATE_ERROR;
if (!internal) {
abort=1;
disconnect = 1;
ep = qhp->ep;
get_ep(&ep->com);
}
goto err;
break;
default:
ret = -EINVAL;
goto out;
}
break;
case IWCH_QP_STATE_CLOSING:
if (!internal) {
ret = -EINVAL;
goto out;
}
switch (attrs->next_state) {
case IWCH_QP_STATE_IDLE:
qhp->attr.state = IWCH_QP_STATE_IDLE;
qhp->attr.llp_stream_handle = NULL;
put_ep(&qhp->ep->com);
qhp->ep = NULL;
wake_up(&qhp->wait);
break;
case IWCH_QP_STATE_ERROR:
goto err;
default:
ret = -EINVAL;
goto err;
}
break;
case IWCH_QP_STATE_ERROR:
if (attrs->next_state != IWCH_QP_STATE_IDLE) {
ret = -EINVAL;
goto out;
}
if (!Q_EMPTY(qhp->wq.sq_rptr, qhp->wq.sq_wptr) ||
!Q_EMPTY(qhp->wq.rq_rptr, qhp->wq.rq_wptr)) {
ret = -EINVAL;
goto out;
}
qhp->attr.state = IWCH_QP_STATE_IDLE;
memset(&qhp->attr, 0, sizeof(qhp->attr));
break;
case IWCH_QP_STATE_TERMINATE:
if (!internal) {
ret = -EINVAL;
goto out;
}
goto err;
break;
default:
printk(KERN_ERR "%s in a bad state %d\n",
__func__, qhp->attr.state);
ret = -EINVAL;
goto err;
break;
}
goto out;
err:
PDBG("%s disassociating ep %p qpid 0x%x\n", __func__, qhp->ep,
qhp->wq.qpid);
/* disassociate the LLP connection */
qhp->attr.llp_stream_handle = NULL;
ep = qhp->ep;
qhp->ep = NULL;
qhp->attr.state = IWCH_QP_STATE_ERROR;
free=1;
wake_up(&qhp->wait);
BUG_ON(!ep);
flush_qp(qhp, &flag);
out:
spin_unlock_irqrestore(&qhp->lock, flag);
if (terminate)
iwch_post_terminate(qhp, NULL);
/*
* If disconnect is 1, then we need to initiate a disconnect
* on the EP. This can be a normal close (RTS->CLOSING) or
* an abnormal close (RTS/CLOSING->ERROR).
*/
if (disconnect) {
iwch_ep_disconnect(ep, abort, GFP_KERNEL);
put_ep(&ep->com);
}
/*
* If free is 1, then we've disassociated the EP from the QP
* and we need to dereference the EP.
*/
if (free)
put_ep(&ep->com);
PDBG("%s exit state %d\n", __func__, qhp->attr.state);
return ret;
}
static int quiesce_qp(struct iwch_qp *qhp)
{
spin_lock_irq(&qhp->lock);
iwch_quiesce_tid(qhp->ep);
qhp->flags |= QP_QUIESCED;
spin_unlock_irq(&qhp->lock);
return 0;
}
static int resume_qp(struct iwch_qp *qhp)
{
spin_lock_irq(&qhp->lock);
iwch_resume_tid(qhp->ep);
qhp->flags &= ~QP_QUIESCED;
spin_unlock_irq(&qhp->lock);
return 0;
}
int iwch_quiesce_qps(struct iwch_cq *chp)
{
int i;
struct iwch_qp *qhp;
for (i=0; i < T3_MAX_NUM_QP; i++) {
qhp = get_qhp(chp->rhp, i);
if (!qhp)
continue;
if ((qhp->attr.rcq == chp->cq.cqid) && !qp_quiesced(qhp)) {
quiesce_qp(qhp);
continue;
}
if ((qhp->attr.scq == chp->cq.cqid) && !qp_quiesced(qhp))
quiesce_qp(qhp);
}
return 0;
}
int iwch_resume_qps(struct iwch_cq *chp)
{
int i;
struct iwch_qp *qhp;
for (i=0; i < T3_MAX_NUM_QP; i++) {
qhp = get_qhp(chp->rhp, i);
if (!qhp)
continue;
if ((qhp->attr.rcq == chp->cq.cqid) && qp_quiesced(qhp)) {
resume_qp(qhp);
continue;
}
if ((qhp->attr.scq == chp->cq.cqid) && qp_quiesced(qhp))
resume_qp(qhp);
}
return 0;
}
View Git Blame Copy Permalink