freebsd-dev/sys/dev/iscsi/icl_conn_if.m

139 lines
3.1 KiB
Mathematica
Raw Normal View History

#-
# SPDX-License-Identifier: BSD-2-Clause-FreeBSD
#
# Copyright (c) 2014 The FreeBSD Foundation
#
# This software was developed by Edward Tomasz Napierala under sponsorship
# from the FreeBSD Foundation.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions
# are met:
# 1. Redistributions of source code must retain the above copyright
# notice, this list of conditions and the following disclaimer.
# 2. 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.
#
# THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
#
# $FreeBSD$
#
#include <sys/socket.h>
#include <dev/iscsi/icl.h>
INTERFACE icl_conn;
Implement zero-copy iSCSI target transmission/read. Add ICL_NOCOPY flag to icl_pdu_append_data(), specifying that the method can just reference the data buffer instead of immediately copying it. Extend the offload KPI with optional PDU queue method, allowing to specify completion callback, called when all the data referenced by above has been transferred and won't be accessed any more (the buffers can be freed). Implement the above functionality in software iSCSI driver using mbufs with external storage and reference counter. Note that some NICs (ixl(4)) may keep the mbuf in TX queue for a long time, so CTL has to be ready. Add optional method to struct ctl_scsiio for buffer reference counting. Implement it for CTL block backend, allowing to delay free of the struct ctl_be_block_io and memory it references as needed. In first reincarnation of the patch I tried to delay whole I/O as it is done for FibreChannel, that was cleaner, but due to the above callback delays I had to rewrite it this way to not leave LUN referenced potentially for hours or more. All together on sequential read from ZFS ARC this saves about 30% of CPU time and memory bandwidth by avoiding one of 3 memory copies (the other two are from ZFS ARC to DMU cache and then from DMU cache to CTL buffers). On tests with 2x Xeon Silver 4114 this allows to reach full line rate of 100GigE NIC. Tests with Gold CPUs and two 100GigE NICs are stil TBD, but expectations to saturate them are pretty high. ;) Discussed with: Chelsio Sponsored by: iXsystems, Inc.
2020-06-08 20:53:57 +00:00
CODE {
static void null_pdu_queue_cb(struct icl_conn *ic,
struct icl_pdu *ip, icl_pdu_cb cb)
{
ICL_CONN_PDU_QUEUE(ic, ip);
if (cb)
cb(ip, 0);
}
};
METHOD size_t pdu_data_segment_length {
struct icl_conn *_ic;
const struct icl_pdu *_ip;
};
METHOD int pdu_append_data {
struct icl_conn *_ic;
struct icl_pdu *_ip;
const void *_addr;
size_t _len;
int _flags;
};
METHOD void pdu_get_data {
struct icl_conn *_ic;
struct icl_pdu *_ip;
size_t _off;
void *_addr;
size_t _len;
};
METHOD void pdu_queue {
struct icl_conn *_ic;
struct icl_pdu *_ip;
};
Implement zero-copy iSCSI target transmission/read. Add ICL_NOCOPY flag to icl_pdu_append_data(), specifying that the method can just reference the data buffer instead of immediately copying it. Extend the offload KPI with optional PDU queue method, allowing to specify completion callback, called when all the data referenced by above has been transferred and won't be accessed any more (the buffers can be freed). Implement the above functionality in software iSCSI driver using mbufs with external storage and reference counter. Note that some NICs (ixl(4)) may keep the mbuf in TX queue for a long time, so CTL has to be ready. Add optional method to struct ctl_scsiio for buffer reference counting. Implement it for CTL block backend, allowing to delay free of the struct ctl_be_block_io and memory it references as needed. In first reincarnation of the patch I tried to delay whole I/O as it is done for FibreChannel, that was cleaner, but due to the above callback delays I had to rewrite it this way to not leave LUN referenced potentially for hours or more. All together on sequential read from ZFS ARC this saves about 30% of CPU time and memory bandwidth by avoiding one of 3 memory copies (the other two are from ZFS ARC to DMU cache and then from DMU cache to CTL buffers). On tests with 2x Xeon Silver 4114 this allows to reach full line rate of 100GigE NIC. Tests with Gold CPUs and two 100GigE NICs are stil TBD, but expectations to saturate them are pretty high. ;) Discussed with: Chelsio Sponsored by: iXsystems, Inc.
2020-06-08 20:53:57 +00:00
METHOD void pdu_queue_cb {
struct icl_conn *_ic;
struct icl_pdu *_ip;
icl_pdu_cb cb;
} DEFAULT null_pdu_queue_cb;
METHOD void pdu_free {
struct icl_conn *_ic;
struct icl_pdu *_ip;
};
METHOD struct icl_pdu * new_pdu {
struct icl_conn *_ic;
int _flags;
};
METHOD void free {
struct icl_conn *_ic;
};
METHOD int handoff {
struct icl_conn *_ic;
int _fd;
};
METHOD void close {
struct icl_conn *_ic;
};
METHOD int task_setup {
struct icl_conn *_ic;
struct icl_pdu *_ip;
struct ccb_scsiio *_csio;
uint32_t *_task_tag;
void **_prvp;
};
METHOD void task_done {
struct icl_conn *_ic;
void *_prv;
};
METHOD int transfer_setup {
struct icl_conn *_ic;
union ctl_io *_io;
uint32_t *_transfer_tag;
void **_prvp;
};
METHOD void transfer_done {
struct icl_conn *_ic;
void *_prv;
};
#
# The function below is only used with ICL_KERNEL_PROXY.
#
METHOD int connect {
struct icl_conn *_ic;
int _domain;
int _socktype;
int _protocol;
struct sockaddr *_from_sa;
struct sockaddr *_to_sa;
};