linux/drivers/infiniband/hw/vmw_pvrdma/pvrdma_mr.c
Adit Ranadive 29c8d9eba5 IB: Add vmw_pvrdma driver
This patch series adds a driver for a paravirtual RDMA device. The
device is developed for VMware's Virtual Machines and allows existing RDMA
applications to continue to use existing Verbs API when deployed in VMs
on ESXi. We recently did a presentation in the OFA Workshop [1] regarding
this device.

Description and RDMA Support
============================
The virtual device is exposed as a dual function PCIe device. One part
is a virtual network device (VMXNet3) which provides networking properties
like MAC, IP addresses to the RDMA part of the device. The networking
properties are used to register GIDs required by RDMA applications to
communicate.

These patches add support and the all required infrastructure for
letting applications use such a device. We support the mandatory Verbs API as
well as the base memory management extensions (Local Inv, Send with Inv and
Fast Register Work Requests). We currently support both Reliable Connected
and Unreliable Datagram QPs but do not support Shared Receive Queues
(SRQs).

Also, we support the following types of Work Requests:
 o Send/Receive (with or without Immediate Data)
 o RDMA Write (with or without Immediate Data)
 o RDMA Read
 o Local Invalidate
 o Send with Invalidate
 o Fast Register Work Requests

This version only adds support for version 1 of RoCE. We will add RoCEv2
support in a future patch. We do support registration of both MAC-based
and IP-based GIDs. I have also created a git tree for our user-level driver
[2].

Testing
=======
We have tested this internally for various types of Guest OS - Red Hat,
Centos, Ubuntu 12.04/14.04/16.04, Oracle Enterprise Linux, SLES 12
using backported versions of this driver. The tests included several
runs of the performance tests (included with OFED), Intel MPI PingPong
benchmark on OpenMPI, krping for FRWRs. Mellanox has been kind enough
to test the backported version of the driver internally on their hardware
using a VMware provided ESX build. I have also applied and tested this
with Doug's k.o/for-4.9 branch (commit 5603910b). Note, that this patch
series should be applied all together. I split out the commits so that
it may be easier to review.

PVRDMA Resources
================
[1] OFA Workshop Presentation -
https://openfabrics.org/images/eventpresos/2016presentations/102parardma.pdf

[2] Libpvrdma User-level library -
http://git.openfabrics.org/?p=~aditr/libpvrdma.git;a=summary

Reviewed-by: Jorgen Hansen <jhansen@vmware.com>
Reviewed-by: George Zhang <georgezhang@vmware.com>
Reviewed-by: Aditya Sarwade <asarwade@vmware.com>
Reviewed-by: Bryan Tan <bryantan@vmware.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Adit Ranadive <aditr@vmware.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-14 14:55:10 -05:00

335 lines
9.0 KiB
C

/*
* Copyright (c) 2012-2016 VMware, Inc. All rights reserved.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of EITHER the GNU General Public License
* version 2 as published by the Free Software Foundation or the BSD
* 2-Clause License. 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 version 2 for more details at
* http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html.
*
* You should have received a copy of the GNU General Public License
* along with this program available in the file COPYING in the main
* directory of this source tree.
*
* The BSD 2-Clause 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.
*
* 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 HOLDER 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.
*/
#include <linux/list.h>
#include <linux/slab.h>
#include "pvrdma.h"
/**
* pvrdma_get_dma_mr - get a DMA memory region
* @pd: protection domain
* @acc: access flags
*
* @return: ib_mr pointer on success, otherwise returns an errno.
*/
struct ib_mr *pvrdma_get_dma_mr(struct ib_pd *pd, int acc)
{
struct pvrdma_dev *dev = to_vdev(pd->device);
struct pvrdma_user_mr *mr;
union pvrdma_cmd_req req;
union pvrdma_cmd_resp rsp;
struct pvrdma_cmd_create_mr *cmd = &req.create_mr;
struct pvrdma_cmd_create_mr_resp *resp = &rsp.create_mr_resp;
int ret;
/* Support only LOCAL_WRITE flag for DMA MRs */
if (acc & ~IB_ACCESS_LOCAL_WRITE) {
dev_warn(&dev->pdev->dev,
"unsupported dma mr access flags %#x\n", acc);
return ERR_PTR(-EOPNOTSUPP);
}
mr = kzalloc(sizeof(*mr), GFP_KERNEL);
if (!mr)
return ERR_PTR(-ENOMEM);
memset(cmd, 0, sizeof(*cmd));
cmd->hdr.cmd = PVRDMA_CMD_CREATE_MR;
cmd->pd_handle = to_vpd(pd)->pd_handle;
cmd->access_flags = acc;
cmd->flags = PVRDMA_MR_FLAG_DMA;
ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_CREATE_MR_RESP);
if (ret < 0) {
dev_warn(&dev->pdev->dev,
"could not get DMA mem region, error: %d\n", ret);
kfree(mr);
return ERR_PTR(ret);
}
mr->mmr.mr_handle = resp->mr_handle;
mr->ibmr.lkey = resp->lkey;
mr->ibmr.rkey = resp->rkey;
return &mr->ibmr;
}
/**
* pvrdma_reg_user_mr - register a userspace memory region
* @pd: protection domain
* @start: starting address
* @length: length of region
* @virt_addr: I/O virtual address
* @access_flags: access flags for memory region
* @udata: user data
*
* @return: ib_mr pointer on success, otherwise returns an errno.
*/
struct ib_mr *pvrdma_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
u64 virt_addr, int access_flags,
struct ib_udata *udata)
{
struct pvrdma_dev *dev = to_vdev(pd->device);
struct pvrdma_user_mr *mr = NULL;
struct ib_umem *umem;
union pvrdma_cmd_req req;
union pvrdma_cmd_resp rsp;
struct pvrdma_cmd_create_mr *cmd = &req.create_mr;
struct pvrdma_cmd_create_mr_resp *resp = &rsp.create_mr_resp;
int nchunks;
int ret;
int entry;
struct scatterlist *sg;
if (length == 0 || length > dev->dsr->caps.max_mr_size) {
dev_warn(&dev->pdev->dev, "invalid mem region length\n");
return ERR_PTR(-EINVAL);
}
umem = ib_umem_get(pd->uobject->context, start,
length, access_flags, 0);
if (IS_ERR(umem)) {
dev_warn(&dev->pdev->dev,
"could not get umem for mem region\n");
return ERR_CAST(umem);
}
nchunks = 0;
for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry)
nchunks += sg_dma_len(sg) >> PAGE_SHIFT;
if (nchunks < 0 || nchunks > PVRDMA_PAGE_DIR_MAX_PAGES) {
dev_warn(&dev->pdev->dev, "overflow %d pages in mem region\n",
nchunks);
ret = -EINVAL;
goto err_umem;
}
mr = kzalloc(sizeof(*mr), GFP_KERNEL);
if (!mr) {
ret = -ENOMEM;
goto err_umem;
}
mr->mmr.iova = virt_addr;
mr->mmr.size = length;
mr->umem = umem;
ret = pvrdma_page_dir_init(dev, &mr->pdir, nchunks, false);
if (ret) {
dev_warn(&dev->pdev->dev,
"could not allocate page directory\n");
goto err_umem;
}
ret = pvrdma_page_dir_insert_umem(&mr->pdir, mr->umem, 0);
if (ret)
goto err_pdir;
memset(cmd, 0, sizeof(*cmd));
cmd->hdr.cmd = PVRDMA_CMD_CREATE_MR;
cmd->start = start;
cmd->length = length;
cmd->pd_handle = to_vpd(pd)->pd_handle;
cmd->access_flags = access_flags;
cmd->nchunks = nchunks;
cmd->pdir_dma = mr->pdir.dir_dma;
ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_CREATE_MR_RESP);
if (ret < 0) {
dev_warn(&dev->pdev->dev,
"could not register mem region, error: %d\n", ret);
goto err_pdir;
}
mr->mmr.mr_handle = resp->mr_handle;
mr->ibmr.lkey = resp->lkey;
mr->ibmr.rkey = resp->rkey;
return &mr->ibmr;
err_pdir:
pvrdma_page_dir_cleanup(dev, &mr->pdir);
err_umem:
ib_umem_release(umem);
kfree(mr);
return ERR_PTR(ret);
}
/**
* pvrdma_alloc_mr - allocate a memory region
* @pd: protection domain
* @mr_type: type of memory region
* @max_num_sg: maximum number of pages
*
* @return: ib_mr pointer on success, otherwise returns an errno.
*/
struct ib_mr *pvrdma_alloc_mr(struct ib_pd *pd, enum ib_mr_type mr_type,
u32 max_num_sg)
{
struct pvrdma_dev *dev = to_vdev(pd->device);
struct pvrdma_user_mr *mr;
union pvrdma_cmd_req req;
union pvrdma_cmd_resp rsp;
struct pvrdma_cmd_create_mr *cmd = &req.create_mr;
struct pvrdma_cmd_create_mr_resp *resp = &rsp.create_mr_resp;
int size = max_num_sg * sizeof(u64);
int ret;
if (mr_type != IB_MR_TYPE_MEM_REG ||
max_num_sg > PVRDMA_MAX_FAST_REG_PAGES)
return ERR_PTR(-EINVAL);
mr = kzalloc(sizeof(*mr), GFP_KERNEL);
if (!mr)
return ERR_PTR(-ENOMEM);
mr->pages = kzalloc(size, GFP_KERNEL);
if (!mr->pages) {
ret = -ENOMEM;
goto freemr;
}
ret = pvrdma_page_dir_init(dev, &mr->pdir, max_num_sg, false);
if (ret) {
dev_warn(&dev->pdev->dev,
"failed to allocate page dir for mr\n");
ret = -ENOMEM;
goto freepages;
}
memset(cmd, 0, sizeof(*cmd));
cmd->hdr.cmd = PVRDMA_CMD_CREATE_MR;
cmd->pd_handle = to_vpd(pd)->pd_handle;
cmd->access_flags = 0;
cmd->flags = PVRDMA_MR_FLAG_FRMR;
cmd->nchunks = max_num_sg;
ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_CREATE_MR_RESP);
if (ret < 0) {
dev_warn(&dev->pdev->dev,
"could not create FR mem region, error: %d\n", ret);
goto freepdir;
}
mr->max_pages = max_num_sg;
mr->mmr.mr_handle = resp->mr_handle;
mr->ibmr.lkey = resp->lkey;
mr->ibmr.rkey = resp->rkey;
mr->page_shift = PAGE_SHIFT;
mr->umem = NULL;
return &mr->ibmr;
freepdir:
pvrdma_page_dir_cleanup(dev, &mr->pdir);
freepages:
kfree(mr->pages);
freemr:
kfree(mr);
return ERR_PTR(ret);
}
/**
* pvrdma_dereg_mr - deregister a memory region
* @ibmr: memory region
*
* @return: 0 on success.
*/
int pvrdma_dereg_mr(struct ib_mr *ibmr)
{
struct pvrdma_user_mr *mr = to_vmr(ibmr);
struct pvrdma_dev *dev = to_vdev(ibmr->device);
union pvrdma_cmd_req req;
struct pvrdma_cmd_destroy_mr *cmd = &req.destroy_mr;
int ret;
memset(cmd, 0, sizeof(*cmd));
cmd->hdr.cmd = PVRDMA_CMD_DESTROY_MR;
cmd->mr_handle = mr->mmr.mr_handle;
ret = pvrdma_cmd_post(dev, &req, NULL, 0);
if (ret < 0)
dev_warn(&dev->pdev->dev,
"could not deregister mem region, error: %d\n", ret);
pvrdma_page_dir_cleanup(dev, &mr->pdir);
if (mr->umem)
ib_umem_release(mr->umem);
kfree(mr->pages);
kfree(mr);
return 0;
}
static int pvrdma_set_page(struct ib_mr *ibmr, u64 addr)
{
struct pvrdma_user_mr *mr = to_vmr(ibmr);
if (mr->npages == mr->max_pages)
return -ENOMEM;
mr->pages[mr->npages++] = addr;
return 0;
}
int pvrdma_map_mr_sg(struct ib_mr *ibmr, struct scatterlist *sg, int sg_nents,
unsigned int *sg_offset)
{
struct pvrdma_user_mr *mr = to_vmr(ibmr);
struct pvrdma_dev *dev = to_vdev(ibmr->device);
int ret;
mr->npages = 0;
ret = ib_sg_to_pages(ibmr, sg, sg_nents, sg_offset, pvrdma_set_page);
if (ret < 0)
dev_warn(&dev->pdev->dev, "could not map sg to pages\n");
return ret;
}