2019-04-30 18:42:43 +00:00
|
|
|
// SPDX-License-Identifier: GPL-2.0
|
2014-05-28 16:15:41 +00:00
|
|
|
/*
|
|
|
|
|
* CPU <-> hardware queue mapping helpers
|
|
|
|
|
*
|
|
|
|
|
* Copyright (C) 2013-2014 Jens Axboe
|
|
|
|
|
*/
|
blk-mq: new multi-queue block IO queueing mechanism
Linux currently has two models for block devices:
- The classic request_fn based approach, where drivers use struct
request units for IO. The block layer provides various helper
functionalities to let drivers share code, things like tag
management, timeout handling, queueing, etc.
- The "stacked" approach, where a driver squeezes in between the
block layer and IO submitter. Since this bypasses the IO stack,
driver generally have to manage everything themselves.
With drivers being written for new high IOPS devices, the classic
request_fn based driver doesn't work well enough. The design dates
back to when both SMP and high IOPS was rare. It has problems with
scaling to bigger machines, and runs into scaling issues even on
smaller machines when you have IOPS in the hundreds of thousands
per device.
The stacked approach is then most often selected as the model
for the driver. But this means that everybody has to re-invent
everything, and along with that we get all the problems again
that the shared approach solved.
This commit introduces blk-mq, block multi queue support. The
design is centered around per-cpu queues for queueing IO, which
then funnel down into x number of hardware submission queues.
We might have a 1:1 mapping between the two, or it might be
an N:M mapping. That all depends on what the hardware supports.
blk-mq provides various helper functions, which include:
- Scalable support for request tagging. Most devices need to
be able to uniquely identify a request both in the driver and
to the hardware. The tagging uses per-cpu caches for freed
tags, to enable cache hot reuse.
- Timeout handling without tracking request on a per-device
basis. Basically the driver should be able to get a notification,
if a request happens to fail.
- Optional support for non 1:1 mappings between issue and
submission queues. blk-mq can redirect IO completions to the
desired location.
- Support for per-request payloads. Drivers almost always need
to associate a request structure with some driver private
command structure. Drivers can tell blk-mq this at init time,
and then any request handed to the driver will have the
required size of memory associated with it.
- Support for merging of IO, and plugging. The stacked model
gets neither of these. Even for high IOPS devices, merging
sequential IO reduces per-command overhead and thus
increases bandwidth.
For now, this is provided as a potential 3rd queueing model, with
the hope being that, as it matures, it can replace both the classic
and stacked model. That would get us back to having just 1 real
model for block devices, leaving the stacked approach to dm/md
devices (as it was originally intended).
Contributions in this patch from the following people:
Shaohua Li <shli@fusionio.com>
Alexander Gordeev <agordeev@redhat.com>
Christoph Hellwig <hch@infradead.org>
Mike Christie <michaelc@cs.wisc.edu>
Matias Bjorling <m@bjorling.me>
Jeff Moyer <jmoyer@redhat.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
|
|
|
#include <linux/kernel.h>
|
|
|
|
|
#include <linux/threads.h>
|
|
|
|
|
#include <linux/module.h>
|
|
|
|
|
#include <linux/mm.h>
|
|
|
|
|
#include <linux/smp.h>
|
|
|
|
|
#include <linux/cpu.h>
|
blk-mq: Build default queue map via group_cpus_evenly()
The default queue mapping builder of blk_mq_map_queues doesn't take NUMA
topo into account, so the built mapping is pretty bad, since CPUs
belonging to different NUMA node are assigned to same queue. It is
observed that IOPS drops by ~30% when running two jobs on same hctx
of null_blk from two CPUs belonging to two NUMA nodes compared with
from same NUMA node.
Address the issue by reusing group_cpus_evenly() for building queue mapping
since group_cpus_evenly() does group cpus according to CPU/NUMA locality.
Also performance data becomes more stable with this given correct queue
mapping is applied wrt. numa locality viewpoint, for example, on one two
nodes arm64 machine with 160 cpus, node 0(cpu 0~79), node 1(cpu 80~159):
1) modprobe null_blk nr_devices=1 submit_queues=2
2) run 'fio(t/io_uring -p 0 -n 4 -r 20 /dev/nullb0)', and observe that
IOPS becomes much stable on multiple tests:
- unpatched: IOPS is 2.5M ~ 4.5M
- patched: IOPS is 4.3M ~ 5.0M
Lots of drivers may benefit from the change, such as nvme pci poll,
nvme tcp, ...
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: John Garry <john.g.garry@oracle.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/20221227022905.352674-7-ming.lei@redhat.com
2022-12-27 02:29:05 +00:00
|
|
|
#include <linux/group_cpus.h>
|
2024-12-02 14:00:12 +00:00
|
|
|
#include <linux/device/bus.h>
|
blk-mq: new multi-queue block IO queueing mechanism
Linux currently has two models for block devices:
- The classic request_fn based approach, where drivers use struct
request units for IO. The block layer provides various helper
functionalities to let drivers share code, things like tag
management, timeout handling, queueing, etc.
- The "stacked" approach, where a driver squeezes in between the
block layer and IO submitter. Since this bypasses the IO stack,
driver generally have to manage everything themselves.
With drivers being written for new high IOPS devices, the classic
request_fn based driver doesn't work well enough. The design dates
back to when both SMP and high IOPS was rare. It has problems with
scaling to bigger machines, and runs into scaling issues even on
smaller machines when you have IOPS in the hundreds of thousands
per device.
The stacked approach is then most often selected as the model
for the driver. But this means that everybody has to re-invent
everything, and along with that we get all the problems again
that the shared approach solved.
This commit introduces blk-mq, block multi queue support. The
design is centered around per-cpu queues for queueing IO, which
then funnel down into x number of hardware submission queues.
We might have a 1:1 mapping between the two, or it might be
an N:M mapping. That all depends on what the hardware supports.
blk-mq provides various helper functions, which include:
- Scalable support for request tagging. Most devices need to
be able to uniquely identify a request both in the driver and
to the hardware. The tagging uses per-cpu caches for freed
tags, to enable cache hot reuse.
- Timeout handling without tracking request on a per-device
basis. Basically the driver should be able to get a notification,
if a request happens to fail.
- Optional support for non 1:1 mappings between issue and
submission queues. blk-mq can redirect IO completions to the
desired location.
- Support for per-request payloads. Drivers almost always need
to associate a request structure with some driver private
command structure. Drivers can tell blk-mq this at init time,
and then any request handed to the driver will have the
required size of memory associated with it.
- Support for merging of IO, and plugging. The stacked model
gets neither of these. Even for high IOPS devices, merging
sequential IO reduces per-command overhead and thus
increases bandwidth.
For now, this is provided as a potential 3rd queueing model, with
the hope being that, as it matures, it can replace both the classic
and stacked model. That would get us back to having just 1 real
model for block devices, leaving the stacked approach to dm/md
devices (as it was originally intended).
Contributions in this patch from the following people:
Shaohua Li <shli@fusionio.com>
Alexander Gordeev <agordeev@redhat.com>
Christoph Hellwig <hch@infradead.org>
Mike Christie <michaelc@cs.wisc.edu>
Matias Bjorling <m@bjorling.me>
Jeff Moyer <jmoyer@redhat.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
|
|
|
|
|
|
|
|
#include "blk.h"
|
|
|
|
|
#include "blk-mq.h"
|
|
|
|
|
|
2022-08-15 17:00:43 +00:00
|
|
|
void blk_mq_map_queues(struct blk_mq_queue_map *qmap)
|
blk-mq: new multi-queue block IO queueing mechanism
Linux currently has two models for block devices:
- The classic request_fn based approach, where drivers use struct
request units for IO. The block layer provides various helper
functionalities to let drivers share code, things like tag
management, timeout handling, queueing, etc.
- The "stacked" approach, where a driver squeezes in between the
block layer and IO submitter. Since this bypasses the IO stack,
driver generally have to manage everything themselves.
With drivers being written for new high IOPS devices, the classic
request_fn based driver doesn't work well enough. The design dates
back to when both SMP and high IOPS was rare. It has problems with
scaling to bigger machines, and runs into scaling issues even on
smaller machines when you have IOPS in the hundreds of thousands
per device.
The stacked approach is then most often selected as the model
for the driver. But this means that everybody has to re-invent
everything, and along with that we get all the problems again
that the shared approach solved.
This commit introduces blk-mq, block multi queue support. The
design is centered around per-cpu queues for queueing IO, which
then funnel down into x number of hardware submission queues.
We might have a 1:1 mapping between the two, or it might be
an N:M mapping. That all depends on what the hardware supports.
blk-mq provides various helper functions, which include:
- Scalable support for request tagging. Most devices need to
be able to uniquely identify a request both in the driver and
to the hardware. The tagging uses per-cpu caches for freed
tags, to enable cache hot reuse.
- Timeout handling without tracking request on a per-device
basis. Basically the driver should be able to get a notification,
if a request happens to fail.
- Optional support for non 1:1 mappings between issue and
submission queues. blk-mq can redirect IO completions to the
desired location.
- Support for per-request payloads. Drivers almost always need
to associate a request structure with some driver private
command structure. Drivers can tell blk-mq this at init time,
and then any request handed to the driver will have the
required size of memory associated with it.
- Support for merging of IO, and plugging. The stacked model
gets neither of these. Even for high IOPS devices, merging
sequential IO reduces per-command overhead and thus
increases bandwidth.
For now, this is provided as a potential 3rd queueing model, with
the hope being that, as it matures, it can replace both the classic
and stacked model. That would get us back to having just 1 real
model for block devices, leaving the stacked approach to dm/md
devices (as it was originally intended).
Contributions in this patch from the following people:
Shaohua Li <shli@fusionio.com>
Alexander Gordeev <agordeev@redhat.com>
Christoph Hellwig <hch@infradead.org>
Mike Christie <michaelc@cs.wisc.edu>
Matias Bjorling <m@bjorling.me>
Jeff Moyer <jmoyer@redhat.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
|
|
|
{
|
blk-mq: Build default queue map via group_cpus_evenly()
The default queue mapping builder of blk_mq_map_queues doesn't take NUMA
topo into account, so the built mapping is pretty bad, since CPUs
belonging to different NUMA node are assigned to same queue. It is
observed that IOPS drops by ~30% when running two jobs on same hctx
of null_blk from two CPUs belonging to two NUMA nodes compared with
from same NUMA node.
Address the issue by reusing group_cpus_evenly() for building queue mapping
since group_cpus_evenly() does group cpus according to CPU/NUMA locality.
Also performance data becomes more stable with this given correct queue
mapping is applied wrt. numa locality viewpoint, for example, on one two
nodes arm64 machine with 160 cpus, node 0(cpu 0~79), node 1(cpu 80~159):
1) modprobe null_blk nr_devices=1 submit_queues=2
2) run 'fio(t/io_uring -p 0 -n 4 -r 20 /dev/nullb0)', and observe that
IOPS becomes much stable on multiple tests:
- unpatched: IOPS is 2.5M ~ 4.5M
- patched: IOPS is 4.3M ~ 5.0M
Lots of drivers may benefit from the change, such as nvme pci poll,
nvme tcp, ...
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: John Garry <john.g.garry@oracle.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/20221227022905.352674-7-ming.lei@redhat.com
2022-12-27 02:29:05 +00:00
|
|
|
const struct cpumask *masks;
|
|
|
|
|
unsigned int queue, cpu;
|
|
|
|
|
|
|
|
|
|
masks = group_cpus_evenly(qmap->nr_queues);
|
|
|
|
|
if (!masks) {
|
|
|
|
|
for_each_possible_cpu(cpu)
|
|
|
|
|
qmap->mq_map[cpu] = qmap->queue_offset;
|
|
|
|
|
return;
|
2019-07-25 09:41:46 +00:00
|
|
|
}
|
blk-mq: new multi-queue block IO queueing mechanism
Linux currently has two models for block devices:
- The classic request_fn based approach, where drivers use struct
request units for IO. The block layer provides various helper
functionalities to let drivers share code, things like tag
management, timeout handling, queueing, etc.
- The "stacked" approach, where a driver squeezes in between the
block layer and IO submitter. Since this bypasses the IO stack,
driver generally have to manage everything themselves.
With drivers being written for new high IOPS devices, the classic
request_fn based driver doesn't work well enough. The design dates
back to when both SMP and high IOPS was rare. It has problems with
scaling to bigger machines, and runs into scaling issues even on
smaller machines when you have IOPS in the hundreds of thousands
per device.
The stacked approach is then most often selected as the model
for the driver. But this means that everybody has to re-invent
everything, and along with that we get all the problems again
that the shared approach solved.
This commit introduces blk-mq, block multi queue support. The
design is centered around per-cpu queues for queueing IO, which
then funnel down into x number of hardware submission queues.
We might have a 1:1 mapping between the two, or it might be
an N:M mapping. That all depends on what the hardware supports.
blk-mq provides various helper functions, which include:
- Scalable support for request tagging. Most devices need to
be able to uniquely identify a request both in the driver and
to the hardware. The tagging uses per-cpu caches for freed
tags, to enable cache hot reuse.
- Timeout handling without tracking request on a per-device
basis. Basically the driver should be able to get a notification,
if a request happens to fail.
- Optional support for non 1:1 mappings between issue and
submission queues. blk-mq can redirect IO completions to the
desired location.
- Support for per-request payloads. Drivers almost always need
to associate a request structure with some driver private
command structure. Drivers can tell blk-mq this at init time,
and then any request handed to the driver will have the
required size of memory associated with it.
- Support for merging of IO, and plugging. The stacked model
gets neither of these. Even for high IOPS devices, merging
sequential IO reduces per-command overhead and thus
increases bandwidth.
For now, this is provided as a potential 3rd queueing model, with
the hope being that, as it matures, it can replace both the classic
and stacked model. That would get us back to having just 1 real
model for block devices, leaving the stacked approach to dm/md
devices (as it was originally intended).
Contributions in this patch from the following people:
Shaohua Li <shli@fusionio.com>
Alexander Gordeev <agordeev@redhat.com>
Christoph Hellwig <hch@infradead.org>
Mike Christie <michaelc@cs.wisc.edu>
Matias Bjorling <m@bjorling.me>
Jeff Moyer <jmoyer@redhat.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
|
|
|
|
blk-mq: Build default queue map via group_cpus_evenly()
The default queue mapping builder of blk_mq_map_queues doesn't take NUMA
topo into account, so the built mapping is pretty bad, since CPUs
belonging to different NUMA node are assigned to same queue. It is
observed that IOPS drops by ~30% when running two jobs on same hctx
of null_blk from two CPUs belonging to two NUMA nodes compared with
from same NUMA node.
Address the issue by reusing group_cpus_evenly() for building queue mapping
since group_cpus_evenly() does group cpus according to CPU/NUMA locality.
Also performance data becomes more stable with this given correct queue
mapping is applied wrt. numa locality viewpoint, for example, on one two
nodes arm64 machine with 160 cpus, node 0(cpu 0~79), node 1(cpu 80~159):
1) modprobe null_blk nr_devices=1 submit_queues=2
2) run 'fio(t/io_uring -p 0 -n 4 -r 20 /dev/nullb0)', and observe that
IOPS becomes much stable on multiple tests:
- unpatched: IOPS is 2.5M ~ 4.5M
- patched: IOPS is 4.3M ~ 5.0M
Lots of drivers may benefit from the change, such as nvme pci poll,
nvme tcp, ...
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: John Garry <john.g.garry@oracle.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/20221227022905.352674-7-ming.lei@redhat.com
2022-12-27 02:29:05 +00:00
|
|
|
for (queue = 0; queue < qmap->nr_queues; queue++) {
|
|
|
|
|
for_each_cpu(cpu, &masks[queue])
|
|
|
|
|
qmap->mq_map[cpu] = qmap->queue_offset + queue;
|
blk-mq: new multi-queue block IO queueing mechanism
Linux currently has two models for block devices:
- The classic request_fn based approach, where drivers use struct
request units for IO. The block layer provides various helper
functionalities to let drivers share code, things like tag
management, timeout handling, queueing, etc.
- The "stacked" approach, where a driver squeezes in between the
block layer and IO submitter. Since this bypasses the IO stack,
driver generally have to manage everything themselves.
With drivers being written for new high IOPS devices, the classic
request_fn based driver doesn't work well enough. The design dates
back to when both SMP and high IOPS was rare. It has problems with
scaling to bigger machines, and runs into scaling issues even on
smaller machines when you have IOPS in the hundreds of thousands
per device.
The stacked approach is then most often selected as the model
for the driver. But this means that everybody has to re-invent
everything, and along with that we get all the problems again
that the shared approach solved.
This commit introduces blk-mq, block multi queue support. The
design is centered around per-cpu queues for queueing IO, which
then funnel down into x number of hardware submission queues.
We might have a 1:1 mapping between the two, or it might be
an N:M mapping. That all depends on what the hardware supports.
blk-mq provides various helper functions, which include:
- Scalable support for request tagging. Most devices need to
be able to uniquely identify a request both in the driver and
to the hardware. The tagging uses per-cpu caches for freed
tags, to enable cache hot reuse.
- Timeout handling without tracking request on a per-device
basis. Basically the driver should be able to get a notification,
if a request happens to fail.
- Optional support for non 1:1 mappings between issue and
submission queues. blk-mq can redirect IO completions to the
desired location.
- Support for per-request payloads. Drivers almost always need
to associate a request structure with some driver private
command structure. Drivers can tell blk-mq this at init time,
and then any request handed to the driver will have the
required size of memory associated with it.
- Support for merging of IO, and plugging. The stacked model
gets neither of these. Even for high IOPS devices, merging
sequential IO reduces per-command overhead and thus
increases bandwidth.
For now, this is provided as a potential 3rd queueing model, with
the hope being that, as it matures, it can replace both the classic
and stacked model. That would get us back to having just 1 real
model for block devices, leaving the stacked approach to dm/md
devices (as it was originally intended).
Contributions in this patch from the following people:
Shaohua Li <shli@fusionio.com>
Alexander Gordeev <agordeev@redhat.com>
Christoph Hellwig <hch@infradead.org>
Mike Christie <michaelc@cs.wisc.edu>
Matias Bjorling <m@bjorling.me>
Jeff Moyer <jmoyer@redhat.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
|
|
|
}
|
blk-mq: Build default queue map via group_cpus_evenly()
The default queue mapping builder of blk_mq_map_queues doesn't take NUMA
topo into account, so the built mapping is pretty bad, since CPUs
belonging to different NUMA node are assigned to same queue. It is
observed that IOPS drops by ~30% when running two jobs on same hctx
of null_blk from two CPUs belonging to two NUMA nodes compared with
from same NUMA node.
Address the issue by reusing group_cpus_evenly() for building queue mapping
since group_cpus_evenly() does group cpus according to CPU/NUMA locality.
Also performance data becomes more stable with this given correct queue
mapping is applied wrt. numa locality viewpoint, for example, on one two
nodes arm64 machine with 160 cpus, node 0(cpu 0~79), node 1(cpu 80~159):
1) modprobe null_blk nr_devices=1 submit_queues=2
2) run 'fio(t/io_uring -p 0 -n 4 -r 20 /dev/nullb0)', and observe that
IOPS becomes much stable on multiple tests:
- unpatched: IOPS is 2.5M ~ 4.5M
- patched: IOPS is 4.3M ~ 5.0M
Lots of drivers may benefit from the change, such as nvme pci poll,
nvme tcp, ...
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: John Garry <john.g.garry@oracle.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/20221227022905.352674-7-ming.lei@redhat.com
2022-12-27 02:29:05 +00:00
|
|
|
kfree(masks);
|
blk-mq: new multi-queue block IO queueing mechanism
Linux currently has two models for block devices:
- The classic request_fn based approach, where drivers use struct
request units for IO. The block layer provides various helper
functionalities to let drivers share code, things like tag
management, timeout handling, queueing, etc.
- The "stacked" approach, where a driver squeezes in between the
block layer and IO submitter. Since this bypasses the IO stack,
driver generally have to manage everything themselves.
With drivers being written for new high IOPS devices, the classic
request_fn based driver doesn't work well enough. The design dates
back to when both SMP and high IOPS was rare. It has problems with
scaling to bigger machines, and runs into scaling issues even on
smaller machines when you have IOPS in the hundreds of thousands
per device.
The stacked approach is then most often selected as the model
for the driver. But this means that everybody has to re-invent
everything, and along with that we get all the problems again
that the shared approach solved.
This commit introduces blk-mq, block multi queue support. The
design is centered around per-cpu queues for queueing IO, which
then funnel down into x number of hardware submission queues.
We might have a 1:1 mapping between the two, or it might be
an N:M mapping. That all depends on what the hardware supports.
blk-mq provides various helper functions, which include:
- Scalable support for request tagging. Most devices need to
be able to uniquely identify a request both in the driver and
to the hardware. The tagging uses per-cpu caches for freed
tags, to enable cache hot reuse.
- Timeout handling without tracking request on a per-device
basis. Basically the driver should be able to get a notification,
if a request happens to fail.
- Optional support for non 1:1 mappings between issue and
submission queues. blk-mq can redirect IO completions to the
desired location.
- Support for per-request payloads. Drivers almost always need
to associate a request structure with some driver private
command structure. Drivers can tell blk-mq this at init time,
and then any request handed to the driver will have the
required size of memory associated with it.
- Support for merging of IO, and plugging. The stacked model
gets neither of these. Even for high IOPS devices, merging
sequential IO reduces per-command overhead and thus
increases bandwidth.
For now, this is provided as a potential 3rd queueing model, with
the hope being that, as it matures, it can replace both the classic
and stacked model. That would get us back to having just 1 real
model for block devices, leaving the stacked approach to dm/md
devices (as it was originally intended).
Contributions in this patch from the following people:
Shaohua Li <shli@fusionio.com>
Alexander Gordeev <agordeev@redhat.com>
Christoph Hellwig <hch@infradead.org>
Mike Christie <michaelc@cs.wisc.edu>
Matias Bjorling <m@bjorling.me>
Jeff Moyer <jmoyer@redhat.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
|
|
|
}
|
2016-11-01 14:12:47 +00:00
|
|
|
EXPORT_SYMBOL_GPL(blk_mq_map_queues);
|
blk-mq: new multi-queue block IO queueing mechanism
Linux currently has two models for block devices:
- The classic request_fn based approach, where drivers use struct
request units for IO. The block layer provides various helper
functionalities to let drivers share code, things like tag
management, timeout handling, queueing, etc.
- The "stacked" approach, where a driver squeezes in between the
block layer and IO submitter. Since this bypasses the IO stack,
driver generally have to manage everything themselves.
With drivers being written for new high IOPS devices, the classic
request_fn based driver doesn't work well enough. The design dates
back to when both SMP and high IOPS was rare. It has problems with
scaling to bigger machines, and runs into scaling issues even on
smaller machines when you have IOPS in the hundreds of thousands
per device.
The stacked approach is then most often selected as the model
for the driver. But this means that everybody has to re-invent
everything, and along with that we get all the problems again
that the shared approach solved.
This commit introduces blk-mq, block multi queue support. The
design is centered around per-cpu queues for queueing IO, which
then funnel down into x number of hardware submission queues.
We might have a 1:1 mapping between the two, or it might be
an N:M mapping. That all depends on what the hardware supports.
blk-mq provides various helper functions, which include:
- Scalable support for request tagging. Most devices need to
be able to uniquely identify a request both in the driver and
to the hardware. The tagging uses per-cpu caches for freed
tags, to enable cache hot reuse.
- Timeout handling without tracking request on a per-device
basis. Basically the driver should be able to get a notification,
if a request happens to fail.
- Optional support for non 1:1 mappings between issue and
submission queues. blk-mq can redirect IO completions to the
desired location.
- Support for per-request payloads. Drivers almost always need
to associate a request structure with some driver private
command structure. Drivers can tell blk-mq this at init time,
and then any request handed to the driver will have the
required size of memory associated with it.
- Support for merging of IO, and plugging. The stacked model
gets neither of these. Even for high IOPS devices, merging
sequential IO reduces per-command overhead and thus
increases bandwidth.
For now, this is provided as a potential 3rd queueing model, with
the hope being that, as it matures, it can replace both the classic
and stacked model. That would get us back to having just 1 real
model for block devices, leaving the stacked approach to dm/md
devices (as it was originally intended).
Contributions in this patch from the following people:
Shaohua Li <shli@fusionio.com>
Alexander Gordeev <agordeev@redhat.com>
Christoph Hellwig <hch@infradead.org>
Mike Christie <michaelc@cs.wisc.edu>
Matias Bjorling <m@bjorling.me>
Jeff Moyer <jmoyer@redhat.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
|
|
|
|
2019-05-31 00:00:53 +00:00
|
|
|
/**
|
|
|
|
|
* blk_mq_hw_queue_to_node - Look up the memory node for a hardware queue index
|
|
|
|
|
* @qmap: CPU to hardware queue map.
|
|
|
|
|
* @index: hardware queue index.
|
|
|
|
|
*
|
2014-05-27 18:06:53 +00:00
|
|
|
* We have no quick way of doing reverse lookups. This is only used at
|
|
|
|
|
* queue init time, so runtime isn't important.
|
|
|
|
|
*/
|
2018-10-29 19:06:14 +00:00
|
|
|
int blk_mq_hw_queue_to_node(struct blk_mq_queue_map *qmap, unsigned int index)
|
2014-05-27 18:06:53 +00:00
|
|
|
{
|
|
|
|
|
int i;
|
|
|
|
|
|
|
|
|
|
for_each_possible_cpu(i) {
|
2018-10-29 19:06:14 +00:00
|
|
|
if (index == qmap->mq_map[i])
|
2020-10-19 08:20:47 +00:00
|
|
|
return cpu_to_node(i);
|
2014-05-27 18:06:53 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return NUMA_NO_NODE;
|
|
|
|
|
}
|
2024-12-02 14:00:12 +00:00
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* blk_mq_map_hw_queues - Create CPU to hardware queue mapping
|
|
|
|
|
* @qmap: CPU to hardware queue map
|
|
|
|
|
* @dev: The device to map queues
|
|
|
|
|
* @offset: Queue offset to use for the device
|
|
|
|
|
*
|
|
|
|
|
* Create a CPU to hardware queue mapping in @qmap. The struct bus_type
|
|
|
|
|
* irq_get_affinity callback will be used to retrieve the affinity.
|
|
|
|
|
*/
|
|
|
|
|
void blk_mq_map_hw_queues(struct blk_mq_queue_map *qmap,
|
|
|
|
|
struct device *dev, unsigned int offset)
|
|
|
|
|
|
|
|
|
|
{
|
|
|
|
|
const struct cpumask *mask;
|
|
|
|
|
unsigned int queue, cpu;
|
|
|
|
|
|
|
|
|
|
if (!dev->bus->irq_get_affinity)
|
|
|
|
|
goto fallback;
|
|
|
|
|
|
|
|
|
|
for (queue = 0; queue < qmap->nr_queues; queue++) {
|
|
|
|
|
mask = dev->bus->irq_get_affinity(dev, queue + offset);
|
|
|
|
|
if (!mask)
|
|
|
|
|
goto fallback;
|
|
|
|
|
|
|
|
|
|
for_each_cpu(cpu, mask)
|
|
|
|
|
qmap->mq_map[cpu] = qmap->queue_offset + queue;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return;
|
|
|
|
|
|
|
|
|
|
fallback:
|
|
|
|
|
WARN_ON_ONCE(qmap->nr_queues > 1);
|
|
|
|
|
blk_mq_clear_mq_map(qmap);
|
|
|
|
|
}
|
|
|
|
|
EXPORT_SYMBOL_GPL(blk_mq_map_hw_queues);
|