freebsd-nq/sys/dev/nvme/nvme_sysctl.c

298 lines
8.5 KiB
C
Raw Normal View History

This is the first of several commits which will add NVM Express (NVMe) support to FreeBSD. A full description of the overall functionality being added is below. nvmexpress.org defines NVM Express as "an optimized register interface, command set and feature set fo PCI Express (PCIe)-based Solid-State Drives (SSDs)." This commit adds nvme(4) and nvd(4) driver source code and Makefiles to the tree. Full NVMe functionality description: Add nvme(4) and nvd(4) drivers and nvmecontrol(8) for NVM Express (NVMe) device support. There will continue to be ongoing work on NVM Express support, but there is more than enough to allow for evaluation of pre-production NVM Express devices as well as soliciting feedback. Questions and feedback are welcome. nvme(4) implements NVMe hardware abstraction and is a provider of NVMe namespaces. The closest equivalent of an NVMe namespace is a SCSI LUN. nvd(4) is an NVMe consumer, surfacing NVMe namespaces as GEOM disks. nvmecontrol(8) is used for NVMe configuration and management. The following are currently supported: nvme(4) - full mandatory NVM command set support - per-CPU IO queues (enabled by default but configurable) - per-queue sysctls for statistics and full command/completion queue dumps for debugging - registration API for NVMe namespace consumers - I/O error handling (except for timeoutsee below) - compilation switches for support back to stable-7 nvd(4) - BIO_DELETE and BIO_FLUSH (if supported by controller) - proper BIO_ORDERED handling nvmecontrol(8) - devlist: list NVMe controllers and their namespaces - identify: display controller or namespace identify data in human-readable or hex format - perftest: quick and dirty performance test to measure raw performance of NVMe device without userspace/physio/GEOM overhead The following are still work in progress and will be completed over the next 3-6 months in rough priority order: - complete man pages - firmware download and activation - asynchronous error requests - command timeout error handling - controller resets - nvmecontrol(8) log page retrieval This has been primarily tested on amd64, with light testing on i386. I would be happy to provide assistance to anyone interested in porting this to other architectures, but am not currently planning to do this work myself. Big-endian and dmamap sync for command/completion queues are the main areas that would need to be addressed. The nvme(4) driver currently has references to Chatham, which is an Intel-developed prototype board which is not fully spec compliant. These references will all be removed over time. Sponsored by: Intel Contributions from: Joe Golio/EMC <joseph dot golio at emc dot com>
2012-09-17 19:23:01 +00:00
/*-
* Copyright (C) 2012-2013 Intel Corporation
This is the first of several commits which will add NVM Express (NVMe) support to FreeBSD. A full description of the overall functionality being added is below. nvmexpress.org defines NVM Express as "an optimized register interface, command set and feature set fo PCI Express (PCIe)-based Solid-State Drives (SSDs)." This commit adds nvme(4) and nvd(4) driver source code and Makefiles to the tree. Full NVMe functionality description: Add nvme(4) and nvd(4) drivers and nvmecontrol(8) for NVM Express (NVMe) device support. There will continue to be ongoing work on NVM Express support, but there is more than enough to allow for evaluation of pre-production NVM Express devices as well as soliciting feedback. Questions and feedback are welcome. nvme(4) implements NVMe hardware abstraction and is a provider of NVMe namespaces. The closest equivalent of an NVMe namespace is a SCSI LUN. nvd(4) is an NVMe consumer, surfacing NVMe namespaces as GEOM disks. nvmecontrol(8) is used for NVMe configuration and management. The following are currently supported: nvme(4) - full mandatory NVM command set support - per-CPU IO queues (enabled by default but configurable) - per-queue sysctls for statistics and full command/completion queue dumps for debugging - registration API for NVMe namespace consumers - I/O error handling (except for timeoutsee below) - compilation switches for support back to stable-7 nvd(4) - BIO_DELETE and BIO_FLUSH (if supported by controller) - proper BIO_ORDERED handling nvmecontrol(8) - devlist: list NVMe controllers and their namespaces - identify: display controller or namespace identify data in human-readable or hex format - perftest: quick and dirty performance test to measure raw performance of NVMe device without userspace/physio/GEOM overhead The following are still work in progress and will be completed over the next 3-6 months in rough priority order: - complete man pages - firmware download and activation - asynchronous error requests - command timeout error handling - controller resets - nvmecontrol(8) log page retrieval This has been primarily tested on amd64, with light testing on i386. I would be happy to provide assistance to anyone interested in porting this to other architectures, but am not currently planning to do this work myself. Big-endian and dmamap sync for command/completion queues are the main areas that would need to be addressed. The nvme(4) driver currently has references to Chatham, which is an Intel-developed prototype board which is not fully spec compliant. These references will all be removed over time. Sponsored by: Intel Contributions from: Joe Golio/EMC <joseph dot golio at emc dot com>
2012-09-17 19:23:01 +00:00
* All rights reserved.
*
* 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.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/bus.h>
#include <sys/sysctl.h>
#include "nvme_private.h"
/*
* CTLTYPE_S64 and sysctl_handle_64 were added in r217616. Define these
* explicitly here for older kernels that don't include the r217616
* changeset.
*/
#ifndef CTLTYPE_S64
#define CTLTYPE_S64 CTLTYPE_QUAD
#define sysctl_handle_64 sysctl_handle_quad
#endif
This is the first of several commits which will add NVM Express (NVMe) support to FreeBSD. A full description of the overall functionality being added is below. nvmexpress.org defines NVM Express as "an optimized register interface, command set and feature set fo PCI Express (PCIe)-based Solid-State Drives (SSDs)." This commit adds nvme(4) and nvd(4) driver source code and Makefiles to the tree. Full NVMe functionality description: Add nvme(4) and nvd(4) drivers and nvmecontrol(8) for NVM Express (NVMe) device support. There will continue to be ongoing work on NVM Express support, but there is more than enough to allow for evaluation of pre-production NVM Express devices as well as soliciting feedback. Questions and feedback are welcome. nvme(4) implements NVMe hardware abstraction and is a provider of NVMe namespaces. The closest equivalent of an NVMe namespace is a SCSI LUN. nvd(4) is an NVMe consumer, surfacing NVMe namespaces as GEOM disks. nvmecontrol(8) is used for NVMe configuration and management. The following are currently supported: nvme(4) - full mandatory NVM command set support - per-CPU IO queues (enabled by default but configurable) - per-queue sysctls for statistics and full command/completion queue dumps for debugging - registration API for NVMe namespace consumers - I/O error handling (except for timeoutsee below) - compilation switches for support back to stable-7 nvd(4) - BIO_DELETE and BIO_FLUSH (if supported by controller) - proper BIO_ORDERED handling nvmecontrol(8) - devlist: list NVMe controllers and their namespaces - identify: display controller or namespace identify data in human-readable or hex format - perftest: quick and dirty performance test to measure raw performance of NVMe device without userspace/physio/GEOM overhead The following are still work in progress and will be completed over the next 3-6 months in rough priority order: - complete man pages - firmware download and activation - asynchronous error requests - command timeout error handling - controller resets - nvmecontrol(8) log page retrieval This has been primarily tested on amd64, with light testing on i386. I would be happy to provide assistance to anyone interested in porting this to other architectures, but am not currently planning to do this work myself. Big-endian and dmamap sync for command/completion queues are the main areas that would need to be addressed. The nvme(4) driver currently has references to Chatham, which is an Intel-developed prototype board which is not fully spec compliant. These references will all be removed over time. Sponsored by: Intel Contributions from: Joe Golio/EMC <joseph dot golio at emc dot com>
2012-09-17 19:23:01 +00:00
static void
nvme_dump_queue(struct nvme_qpair *qpair)
{
struct nvme_completion *cpl;
struct nvme_command *cmd;
int i;
printf("id:%04Xh phase:%d\n", qpair->id, qpair->phase);
printf("Completion queue:\n");
for (i = 0; i < qpair->num_entries; i++) {
cpl = &qpair->cpl[i];
printf("%05d: ", i);
nvme_dump_completion(cpl);
}
printf("Submission queue:\n");
for (i = 0; i < qpair->num_entries; i++) {
cmd = &qpair->cmd[i];
printf("%05d: ", i);
nvme_dump_command(cmd);
}
}
static int
nvme_sysctl_dump_debug(SYSCTL_HANDLER_ARGS)
{
struct nvme_qpair *qpair = arg1;
uint32_t val = 0;
int error = sysctl_handle_int(oidp, &val, 0, req);
if (error)
return (error);
if (val != 0)
nvme_dump_queue(qpair);
return (0);
}
static int
nvme_sysctl_int_coal_time(SYSCTL_HANDLER_ARGS)
{
struct nvme_controller *ctrlr = arg1;
uint32_t oldval = ctrlr->int_coal_time;
int error = sysctl_handle_int(oidp, &ctrlr->int_coal_time, 0,
req);
if (error)
return (error);
if (oldval != ctrlr->int_coal_time)
nvme_ctrlr_cmd_set_interrupt_coalescing(ctrlr,
ctrlr->int_coal_time, ctrlr->int_coal_threshold, NULL,
NULL);
return (0);
}
static int
nvme_sysctl_int_coal_threshold(SYSCTL_HANDLER_ARGS)
{
struct nvme_controller *ctrlr = arg1;
uint32_t oldval = ctrlr->int_coal_threshold;
int error = sysctl_handle_int(oidp, &ctrlr->int_coal_threshold, 0,
req);
if (error)
return (error);
if (oldval != ctrlr->int_coal_threshold)
nvme_ctrlr_cmd_set_interrupt_coalescing(ctrlr,
ctrlr->int_coal_time, ctrlr->int_coal_threshold, NULL,
NULL);
return (0);
}
static int
nvme_sysctl_timeout_period(SYSCTL_HANDLER_ARGS)
{
struct nvme_controller *ctrlr = arg1;
uint32_t oldval = ctrlr->timeout_period;
int error = sysctl_handle_int(oidp, &ctrlr->timeout_period, 0, req);
if (error)
return (error);
if (ctrlr->timeout_period > NVME_MAX_TIMEOUT_PERIOD ||
ctrlr->timeout_period < NVME_MIN_TIMEOUT_PERIOD) {
ctrlr->timeout_period = oldval;
return (EINVAL);
}
return (0);
}
static void
nvme_qpair_reset_stats(struct nvme_qpair *qpair)
{
qpair->num_cmds = 0;
qpair->num_intr_handler_calls = 0;
}
static int
nvme_sysctl_num_cmds(SYSCTL_HANDLER_ARGS)
{
struct nvme_controller *ctrlr = arg1;
int64_t num_cmds = 0;
int i;
num_cmds = ctrlr->adminq.num_cmds;
for (i = 0; i < ctrlr->num_io_queues; i++)
num_cmds += ctrlr->ioq[i].num_cmds;
return (sysctl_handle_64(oidp, &num_cmds, 0, req));
}
static int
nvme_sysctl_num_intr_handler_calls(SYSCTL_HANDLER_ARGS)
{
struct nvme_controller *ctrlr = arg1;
int64_t num_intr_handler_calls = 0;
int i;
num_intr_handler_calls = ctrlr->adminq.num_intr_handler_calls;
for (i = 0; i < ctrlr->num_io_queues; i++)
num_intr_handler_calls += ctrlr->ioq[i].num_intr_handler_calls;
return (sysctl_handle_64(oidp, &num_intr_handler_calls, 0, req));
}
static int
nvme_sysctl_reset_stats(SYSCTL_HANDLER_ARGS)
{
struct nvme_controller *ctrlr = arg1;
uint32_t i, val = 0;
int error = sysctl_handle_int(oidp, &val, 0, req);
if (error)
return (error);
if (val != 0) {
nvme_qpair_reset_stats(&ctrlr->adminq);
for (i = 0; i < ctrlr->num_io_queues; i++)
nvme_qpair_reset_stats(&ctrlr->ioq[i]);
}
return (0);
}
This is the first of several commits which will add NVM Express (NVMe) support to FreeBSD. A full description of the overall functionality being added is below. nvmexpress.org defines NVM Express as "an optimized register interface, command set and feature set fo PCI Express (PCIe)-based Solid-State Drives (SSDs)." This commit adds nvme(4) and nvd(4) driver source code and Makefiles to the tree. Full NVMe functionality description: Add nvme(4) and nvd(4) drivers and nvmecontrol(8) for NVM Express (NVMe) device support. There will continue to be ongoing work on NVM Express support, but there is more than enough to allow for evaluation of pre-production NVM Express devices as well as soliciting feedback. Questions and feedback are welcome. nvme(4) implements NVMe hardware abstraction and is a provider of NVMe namespaces. The closest equivalent of an NVMe namespace is a SCSI LUN. nvd(4) is an NVMe consumer, surfacing NVMe namespaces as GEOM disks. nvmecontrol(8) is used for NVMe configuration and management. The following are currently supported: nvme(4) - full mandatory NVM command set support - per-CPU IO queues (enabled by default but configurable) - per-queue sysctls for statistics and full command/completion queue dumps for debugging - registration API for NVMe namespace consumers - I/O error handling (except for timeoutsee below) - compilation switches for support back to stable-7 nvd(4) - BIO_DELETE and BIO_FLUSH (if supported by controller) - proper BIO_ORDERED handling nvmecontrol(8) - devlist: list NVMe controllers and their namespaces - identify: display controller or namespace identify data in human-readable or hex format - perftest: quick and dirty performance test to measure raw performance of NVMe device without userspace/physio/GEOM overhead The following are still work in progress and will be completed over the next 3-6 months in rough priority order: - complete man pages - firmware download and activation - asynchronous error requests - command timeout error handling - controller resets - nvmecontrol(8) log page retrieval This has been primarily tested on amd64, with light testing on i386. I would be happy to provide assistance to anyone interested in porting this to other architectures, but am not currently planning to do this work myself. Big-endian and dmamap sync for command/completion queues are the main areas that would need to be addressed. The nvme(4) driver currently has references to Chatham, which is an Intel-developed prototype board which is not fully spec compliant. These references will all be removed over time. Sponsored by: Intel Contributions from: Joe Golio/EMC <joseph dot golio at emc dot com>
2012-09-17 19:23:01 +00:00
static void
nvme_sysctl_initialize_queue(struct nvme_qpair *qpair,
struct sysctl_ctx_list *ctrlr_ctx, struct sysctl_oid *que_tree)
{
struct sysctl_oid_list *que_list = SYSCTL_CHILDREN(que_tree);
SYSCTL_ADD_UINT(ctrlr_ctx, que_list, OID_AUTO, "num_entries",
CTLFLAG_RD, &qpair->num_entries, 0,
"Number of entries in hardware queue");
SYSCTL_ADD_UINT(ctrlr_ctx, que_list, OID_AUTO, "num_trackers",
CTLFLAG_RD, &qpair->num_trackers, 0,
"Number of trackers pre-allocated for this queue pair");
This is the first of several commits which will add NVM Express (NVMe) support to FreeBSD. A full description of the overall functionality being added is below. nvmexpress.org defines NVM Express as "an optimized register interface, command set and feature set fo PCI Express (PCIe)-based Solid-State Drives (SSDs)." This commit adds nvme(4) and nvd(4) driver source code and Makefiles to the tree. Full NVMe functionality description: Add nvme(4) and nvd(4) drivers and nvmecontrol(8) for NVM Express (NVMe) device support. There will continue to be ongoing work on NVM Express support, but there is more than enough to allow for evaluation of pre-production NVM Express devices as well as soliciting feedback. Questions and feedback are welcome. nvme(4) implements NVMe hardware abstraction and is a provider of NVMe namespaces. The closest equivalent of an NVMe namespace is a SCSI LUN. nvd(4) is an NVMe consumer, surfacing NVMe namespaces as GEOM disks. nvmecontrol(8) is used for NVMe configuration and management. The following are currently supported: nvme(4) - full mandatory NVM command set support - per-CPU IO queues (enabled by default but configurable) - per-queue sysctls for statistics and full command/completion queue dumps for debugging - registration API for NVMe namespace consumers - I/O error handling (except for timeoutsee below) - compilation switches for support back to stable-7 nvd(4) - BIO_DELETE and BIO_FLUSH (if supported by controller) - proper BIO_ORDERED handling nvmecontrol(8) - devlist: list NVMe controllers and their namespaces - identify: display controller or namespace identify data in human-readable or hex format - perftest: quick and dirty performance test to measure raw performance of NVMe device without userspace/physio/GEOM overhead The following are still work in progress and will be completed over the next 3-6 months in rough priority order: - complete man pages - firmware download and activation - asynchronous error requests - command timeout error handling - controller resets - nvmecontrol(8) log page retrieval This has been primarily tested on amd64, with light testing on i386. I would be happy to provide assistance to anyone interested in porting this to other architectures, but am not currently planning to do this work myself. Big-endian and dmamap sync for command/completion queues are the main areas that would need to be addressed. The nvme(4) driver currently has references to Chatham, which is an Intel-developed prototype board which is not fully spec compliant. These references will all be removed over time. Sponsored by: Intel Contributions from: Joe Golio/EMC <joseph dot golio at emc dot com>
2012-09-17 19:23:01 +00:00
SYSCTL_ADD_UINT(ctrlr_ctx, que_list, OID_AUTO, "sq_head",
CTLFLAG_RD, &qpair->sq_head, 0,
"Current head of submission queue (as observed by driver)");
SYSCTL_ADD_UINT(ctrlr_ctx, que_list, OID_AUTO, "sq_tail",
CTLFLAG_RD, &qpair->sq_tail, 0,
"Current tail of submission queue (as observed by driver)");
SYSCTL_ADD_UINT(ctrlr_ctx, que_list, OID_AUTO, "cq_head",
CTLFLAG_RD, &qpair->cq_head, 0,
"Current head of completion queue (as observed by driver)");
SYSCTL_ADD_QUAD(ctrlr_ctx, que_list, OID_AUTO, "num_cmds",
CTLFLAG_RD, &qpair->num_cmds, "Number of commands submitted");
SYSCTL_ADD_QUAD(ctrlr_ctx, que_list, OID_AUTO, "num_intr_handler_calls",
CTLFLAG_RD, &qpair->num_intr_handler_calls,
"Number of times interrupt handler was invoked (will typically be "
"less than number of actual interrupts generated due to "
"coalescing)");
This is the first of several commits which will add NVM Express (NVMe) support to FreeBSD. A full description of the overall functionality being added is below. nvmexpress.org defines NVM Express as "an optimized register interface, command set and feature set fo PCI Express (PCIe)-based Solid-State Drives (SSDs)." This commit adds nvme(4) and nvd(4) driver source code and Makefiles to the tree. Full NVMe functionality description: Add nvme(4) and nvd(4) drivers and nvmecontrol(8) for NVM Express (NVMe) device support. There will continue to be ongoing work on NVM Express support, but there is more than enough to allow for evaluation of pre-production NVM Express devices as well as soliciting feedback. Questions and feedback are welcome. nvme(4) implements NVMe hardware abstraction and is a provider of NVMe namespaces. The closest equivalent of an NVMe namespace is a SCSI LUN. nvd(4) is an NVMe consumer, surfacing NVMe namespaces as GEOM disks. nvmecontrol(8) is used for NVMe configuration and management. The following are currently supported: nvme(4) - full mandatory NVM command set support - per-CPU IO queues (enabled by default but configurable) - per-queue sysctls for statistics and full command/completion queue dumps for debugging - registration API for NVMe namespace consumers - I/O error handling (except for timeoutsee below) - compilation switches for support back to stable-7 nvd(4) - BIO_DELETE and BIO_FLUSH (if supported by controller) - proper BIO_ORDERED handling nvmecontrol(8) - devlist: list NVMe controllers and their namespaces - identify: display controller or namespace identify data in human-readable or hex format - perftest: quick and dirty performance test to measure raw performance of NVMe device without userspace/physio/GEOM overhead The following are still work in progress and will be completed over the next 3-6 months in rough priority order: - complete man pages - firmware download and activation - asynchronous error requests - command timeout error handling - controller resets - nvmecontrol(8) log page retrieval This has been primarily tested on amd64, with light testing on i386. I would be happy to provide assistance to anyone interested in porting this to other architectures, but am not currently planning to do this work myself. Big-endian and dmamap sync for command/completion queues are the main areas that would need to be addressed. The nvme(4) driver currently has references to Chatham, which is an Intel-developed prototype board which is not fully spec compliant. These references will all be removed over time. Sponsored by: Intel Contributions from: Joe Golio/EMC <joseph dot golio at emc dot com>
2012-09-17 19:23:01 +00:00
SYSCTL_ADD_PROC(ctrlr_ctx, que_list, OID_AUTO,
"dump_debug", CTLTYPE_UINT | CTLFLAG_RW, qpair, 0,
nvme_sysctl_dump_debug, "IU", "Dump debug data");
}
void
nvme_sysctl_initialize_ctrlr(struct nvme_controller *ctrlr)
{
struct sysctl_ctx_list *ctrlr_ctx;
struct sysctl_oid *ctrlr_tree, *que_tree;
struct sysctl_oid_list *ctrlr_list;
#define QUEUE_NAME_LENGTH 16
char queue_name[QUEUE_NAME_LENGTH];
int i;
ctrlr_ctx = device_get_sysctl_ctx(ctrlr->dev);
ctrlr_tree = device_get_sysctl_tree(ctrlr->dev);
ctrlr_list = SYSCTL_CHILDREN(ctrlr_tree);
SYSCTL_ADD_PROC(ctrlr_ctx, ctrlr_list, OID_AUTO,
"int_coal_time", CTLTYPE_UINT | CTLFLAG_RW, ctrlr, 0,
nvme_sysctl_int_coal_time, "IU",
"Interrupt coalescing timeout (in microseconds)");
SYSCTL_ADD_PROC(ctrlr_ctx, ctrlr_list, OID_AUTO,
"int_coal_threshold", CTLTYPE_UINT | CTLFLAG_RW, ctrlr, 0,
nvme_sysctl_int_coal_threshold, "IU",
"Interrupt coalescing threshold");
SYSCTL_ADD_PROC(ctrlr_ctx, ctrlr_list, OID_AUTO,
"timeout_period", CTLTYPE_UINT | CTLFLAG_RW, ctrlr, 0,
nvme_sysctl_timeout_period, "IU",
"Timeout period (in seconds)");
SYSCTL_ADD_PROC(ctrlr_ctx, ctrlr_list, OID_AUTO,
"num_cmds", CTLTYPE_S64 | CTLFLAG_RD,
ctrlr, 0, nvme_sysctl_num_cmds, "IU",
"Number of commands submitted");
SYSCTL_ADD_PROC(ctrlr_ctx, ctrlr_list, OID_AUTO,
"num_intr_handler_calls", CTLTYPE_S64 | CTLFLAG_RD,
ctrlr, 0, nvme_sysctl_num_intr_handler_calls, "IU",
"Number of times interrupt handler was invoked (will "
"typically be less than number of actual interrupts "
"generated due to coalescing)");
SYSCTL_ADD_PROC(ctrlr_ctx, ctrlr_list, OID_AUTO,
"reset_stats", CTLTYPE_UINT | CTLFLAG_RW, ctrlr, 0,
nvme_sysctl_reset_stats, "IU", "Reset statistics to zero");
This is the first of several commits which will add NVM Express (NVMe) support to FreeBSD. A full description of the overall functionality being added is below. nvmexpress.org defines NVM Express as "an optimized register interface, command set and feature set fo PCI Express (PCIe)-based Solid-State Drives (SSDs)." This commit adds nvme(4) and nvd(4) driver source code and Makefiles to the tree. Full NVMe functionality description: Add nvme(4) and nvd(4) drivers and nvmecontrol(8) for NVM Express (NVMe) device support. There will continue to be ongoing work on NVM Express support, but there is more than enough to allow for evaluation of pre-production NVM Express devices as well as soliciting feedback. Questions and feedback are welcome. nvme(4) implements NVMe hardware abstraction and is a provider of NVMe namespaces. The closest equivalent of an NVMe namespace is a SCSI LUN. nvd(4) is an NVMe consumer, surfacing NVMe namespaces as GEOM disks. nvmecontrol(8) is used for NVMe configuration and management. The following are currently supported: nvme(4) - full mandatory NVM command set support - per-CPU IO queues (enabled by default but configurable) - per-queue sysctls for statistics and full command/completion queue dumps for debugging - registration API for NVMe namespace consumers - I/O error handling (except for timeoutsee below) - compilation switches for support back to stable-7 nvd(4) - BIO_DELETE and BIO_FLUSH (if supported by controller) - proper BIO_ORDERED handling nvmecontrol(8) - devlist: list NVMe controllers and their namespaces - identify: display controller or namespace identify data in human-readable or hex format - perftest: quick and dirty performance test to measure raw performance of NVMe device without userspace/physio/GEOM overhead The following are still work in progress and will be completed over the next 3-6 months in rough priority order: - complete man pages - firmware download and activation - asynchronous error requests - command timeout error handling - controller resets - nvmecontrol(8) log page retrieval This has been primarily tested on amd64, with light testing on i386. I would be happy to provide assistance to anyone interested in porting this to other architectures, but am not currently planning to do this work myself. Big-endian and dmamap sync for command/completion queues are the main areas that would need to be addressed. The nvme(4) driver currently has references to Chatham, which is an Intel-developed prototype board which is not fully spec compliant. These references will all be removed over time. Sponsored by: Intel Contributions from: Joe Golio/EMC <joseph dot golio at emc dot com>
2012-09-17 19:23:01 +00:00
que_tree = SYSCTL_ADD_NODE(ctrlr_ctx, ctrlr_list, OID_AUTO, "adminq",
CTLFLAG_RD, NULL, "Admin Queue");
nvme_sysctl_initialize_queue(&ctrlr->adminq, ctrlr_ctx, que_tree);
for (i = 0; i < ctrlr->num_io_queues; i++) {
snprintf(queue_name, QUEUE_NAME_LENGTH, "ioq%d", i);
que_tree = SYSCTL_ADD_NODE(ctrlr_ctx, ctrlr_list, OID_AUTO,
queue_name, CTLFLAG_RD, NULL, "IO Queue");
nvme_sysctl_initialize_queue(&ctrlr->ioq[i], ctrlr_ctx,
que_tree);
}
}