linux/drivers/infiniband/hw/hfi1/verbs_txreq.h

119 lines
3.8 KiB
C
Raw Normal View History

/*
* Copyright(c) 2016 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
*
* GPL LICENSE SUMMARY
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* BSD LICENSE
*
* 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.
* - Neither the name of Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/
#ifndef HFI1_VERBS_TXREQ_H
#define HFI1_VERBS_TXREQ_H
#include <linux/types.h>
#include <linux/slab.h>
#include "verbs.h"
#include "sdma_txreq.h"
#include "iowait.h"
struct verbs_txreq {
struct hfi1_sdma_header phdr;
struct sdma_txreq txreq;
struct rvt_qp *qp;
struct rvt_swqe *wqe;
struct rvt_mregion *mr;
struct rvt_sge_state *ss;
struct sdma_engine *sde;
struct send_context *psc;
u16 hdr_dwords;
u16 s_cur_size;
};
struct hfi1_ibdev;
struct verbs_txreq *__get_txreq(struct hfi1_ibdev *dev,
struct rvt_qp *qp);
static inline struct verbs_txreq *get_txreq(struct hfi1_ibdev *dev,
struct rvt_qp *qp)
__must_hold(&qp->slock)
{
struct verbs_txreq *tx;
struct hfi1_qp_priv *priv = qp->priv;
tx = kmem_cache_alloc(dev->verbs_txreq_cache, GFP_ATOMIC);
if (unlikely(!tx)) {
/* call slow path to get the lock */
tx = __get_txreq(dev, qp);
if (IS_ERR(tx))
return tx;
}
tx->qp = qp;
tx->mr = NULL;
tx->sde = priv->s_sde;
tx->psc = priv->s_sendcontext;
/* so that we can test if the sdma decriptors are there */
tx->txreq.num_desc = 0;
return tx;
}
staging/rdma/hfi1: Adaptive PIO for short messages The change requires a new pio_busy field in the iowait structure to track the number of outstanding pios. The new counter together with the sdma counter serve as the basis for a packet by packet decision as to which egress mechanism to use. Since packets given to different egress mechanisms are not ordered, this scheme will preserve the order. The iowait drain/wait mechanisms are extended for a pio case. An additional qp wait flag is added for the PIO drain wait case. Currently the only pio wait is for buffers, so the no_bufs_available() routine name is changed to pio_wait() and a third argument is passed with one of the two pio wait flags to generalize the routine. A module parameter is added to hold a configurable threshold. For now, the module parameter is zero. A heuristic routine is added to return the func pointer of the proper egress routine to use. The heuristic is as follows: - SMI always uses pio - GSI,UD qps <= threshold use pio - UD qps > threadhold use sdma o No coordination with sdma is required because order is not required and this qp pio count is not maintained for UD - RC/UC ONLY packets <= threshold chose as follows: o If sdmas pending, use SDMA o Otherwise use pio and enable the pio tracking count at the time the pio buffer is allocated - RC/UC ONLY packets > threshold use SDMA o If pio's are pending the pio_wait with the new wait flag is called to delay for pios to drain The threshold is potentially reduced by the QP's mtu. The sc_buffer_alloc() has two additional args (a callback, a void *) which are exploited by the RC/UC cases to pass a new complete routine and a qp *. When the shadow ring completes the credit associated with a packet, the new complete routine is called. The verbs_pio_complete() will then decrement the busy count and trigger any drain waiters in qp destroy or reset. Reviewed-by: Jubin John <jubin.john@intel.com> Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-02-14 20:45:36 +00:00
static inline struct sdma_txreq *get_sdma_txreq(struct verbs_txreq *tx)
{
return &tx->txreq;
}
static inline struct verbs_txreq *get_waiting_verbs_txreq(struct rvt_qp *qp)
{
struct sdma_txreq *stx;
struct hfi1_qp_priv *priv = qp->priv;
stx = iowait_get_txhead(&priv->s_iowait);
if (stx)
return container_of(stx, struct verbs_txreq, txreq);
return NULL;
}
void hfi1_put_txreq(struct verbs_txreq *tx);
int verbs_txreq_init(struct hfi1_ibdev *dev);
void verbs_txreq_exit(struct hfi1_ibdev *dev);
#endif /* HFI1_VERBS_TXREQ_H */