linux-sg2042/block/blk-mq-cpu.c

59 lines
1.4 KiB
C
Raw Normal View History

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 16:20:05 +08:00
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/blkdev.h>
#include <linux/list.h>
#include <linux/llist.h>
#include <linux/smp.h>
#include <linux/cpu.h>
#include <linux/blk-mq.h>
#include "blk-mq.h"
static LIST_HEAD(blk_mq_cpu_notify_list);
rt,blk,mq: Make blk_mq_cpu_notify_lock a raw spinlock [ 365.164040] BUG: sleeping function called from invalid context at kernel/rtmutex.c:674 [ 365.164041] in_atomic(): 1, irqs_disabled(): 1, pid: 26, name: migration/1 [ 365.164043] no locks held by migration/1/26. [ 365.164044] irq event stamp: 6648 [ 365.164056] hardirqs last enabled at (6647): [<ffffffff8153d377>] restore_args+0x0/0x30 [ 365.164062] hardirqs last disabled at (6648): [<ffffffff810ed98d>] multi_cpu_stop+0x9d/0x120 [ 365.164070] softirqs last enabled at (0): [<ffffffff810543bc>] copy_process.part.28+0x6fc/0x1920 [ 365.164072] softirqs last disabled at (0): [< (null)>] (null) [ 365.164076] CPU: 1 PID: 26 Comm: migration/1 Tainted: GF N 3.12.12-rt19-0.gcb6c4a2-rt #3 [ 365.164078] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011 [ 365.164091] 0000000000000001 ffff880a42ea7c30 ffffffff815367e6 ffffffff81a086c0 [ 365.164099] ffff880a42ea7c40 ffffffff8108919c ffff880a42ea7c60 ffffffff8153c24f [ 365.164107] ffff880a42ea91f0 00000000ffffffe1 ffff880a42ea7c88 ffffffff81297ec0 [ 365.164108] Call Trace: [ 365.164119] [<ffffffff810060b1>] try_stack_unwind+0x191/0x1a0 [ 365.164127] [<ffffffff81004872>] dump_trace+0x92/0x360 [ 365.164133] [<ffffffff81006108>] show_trace_log_lvl+0x48/0x60 [ 365.164138] [<ffffffff81004c18>] show_stack_log_lvl+0xd8/0x1d0 [ 365.164143] [<ffffffff81006160>] show_stack+0x20/0x50 [ 365.164153] [<ffffffff815367e6>] dump_stack+0x54/0x9a [ 365.164163] [<ffffffff8108919c>] __might_sleep+0xfc/0x140 [ 365.164173] [<ffffffff8153c24f>] rt_spin_lock+0x1f/0x70 [ 365.164182] [<ffffffff81297ec0>] blk_mq_main_cpu_notify+0x20/0x70 [ 365.164191] [<ffffffff81540a1c>] notifier_call_chain+0x4c/0x70 [ 365.164201] [<ffffffff81083499>] __raw_notifier_call_chain+0x9/0x10 [ 365.164207] [<ffffffff810567be>] cpu_notify+0x1e/0x40 [ 365.164217] [<ffffffff81525da2>] take_cpu_down+0x22/0x40 [ 365.164223] [<ffffffff810ed9c6>] multi_cpu_stop+0xd6/0x120 [ 365.164229] [<ffffffff810edd97>] cpu_stopper_thread+0xd7/0x1e0 [ 365.164235] [<ffffffff810863a3>] smpboot_thread_fn+0x203/0x380 [ 365.164241] [<ffffffff8107cbf8>] kthread+0xc8/0xd0 [ 365.164250] [<ffffffff8154440c>] ret_from_fork+0x7c/0xb0 [ 365.164429] smpboot: CPU 1 is now offline Signed-off-by: Mike Galbraith <bitbucket@online.de> Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-03 12:57:26 +08:00
static DEFINE_RAW_SPINLOCK(blk_mq_cpu_notify_lock);
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 16:20:05 +08:00
static int blk_mq_main_cpu_notify(struct notifier_block *self,
unsigned long action, void *hcpu)
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 16:20:05 +08:00
{
unsigned int cpu = (unsigned long) hcpu;
struct blk_mq_cpu_notifier *notify;
rt,blk,mq: Make blk_mq_cpu_notify_lock a raw spinlock [ 365.164040] BUG: sleeping function called from invalid context at kernel/rtmutex.c:674 [ 365.164041] in_atomic(): 1, irqs_disabled(): 1, pid: 26, name: migration/1 [ 365.164043] no locks held by migration/1/26. [ 365.164044] irq event stamp: 6648 [ 365.164056] hardirqs last enabled at (6647): [<ffffffff8153d377>] restore_args+0x0/0x30 [ 365.164062] hardirqs last disabled at (6648): [<ffffffff810ed98d>] multi_cpu_stop+0x9d/0x120 [ 365.164070] softirqs last enabled at (0): [<ffffffff810543bc>] copy_process.part.28+0x6fc/0x1920 [ 365.164072] softirqs last disabled at (0): [< (null)>] (null) [ 365.164076] CPU: 1 PID: 26 Comm: migration/1 Tainted: GF N 3.12.12-rt19-0.gcb6c4a2-rt #3 [ 365.164078] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011 [ 365.164091] 0000000000000001 ffff880a42ea7c30 ffffffff815367e6 ffffffff81a086c0 [ 365.164099] ffff880a42ea7c40 ffffffff8108919c ffff880a42ea7c60 ffffffff8153c24f [ 365.164107] ffff880a42ea91f0 00000000ffffffe1 ffff880a42ea7c88 ffffffff81297ec0 [ 365.164108] Call Trace: [ 365.164119] [<ffffffff810060b1>] try_stack_unwind+0x191/0x1a0 [ 365.164127] [<ffffffff81004872>] dump_trace+0x92/0x360 [ 365.164133] [<ffffffff81006108>] show_trace_log_lvl+0x48/0x60 [ 365.164138] [<ffffffff81004c18>] show_stack_log_lvl+0xd8/0x1d0 [ 365.164143] [<ffffffff81006160>] show_stack+0x20/0x50 [ 365.164153] [<ffffffff815367e6>] dump_stack+0x54/0x9a [ 365.164163] [<ffffffff8108919c>] __might_sleep+0xfc/0x140 [ 365.164173] [<ffffffff8153c24f>] rt_spin_lock+0x1f/0x70 [ 365.164182] [<ffffffff81297ec0>] blk_mq_main_cpu_notify+0x20/0x70 [ 365.164191] [<ffffffff81540a1c>] notifier_call_chain+0x4c/0x70 [ 365.164201] [<ffffffff81083499>] __raw_notifier_call_chain+0x9/0x10 [ 365.164207] [<ffffffff810567be>] cpu_notify+0x1e/0x40 [ 365.164217] [<ffffffff81525da2>] take_cpu_down+0x22/0x40 [ 365.164223] [<ffffffff810ed9c6>] multi_cpu_stop+0xd6/0x120 [ 365.164229] [<ffffffff810edd97>] cpu_stopper_thread+0xd7/0x1e0 [ 365.164235] [<ffffffff810863a3>] smpboot_thread_fn+0x203/0x380 [ 365.164241] [<ffffffff8107cbf8>] kthread+0xc8/0xd0 [ 365.164250] [<ffffffff8154440c>] ret_from_fork+0x7c/0xb0 [ 365.164429] smpboot: CPU 1 is now offline Signed-off-by: Mike Galbraith <bitbucket@online.de> Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-03 12:57:26 +08:00
raw_spin_lock(&blk_mq_cpu_notify_lock);
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 16:20:05 +08:00
list_for_each_entry(notify, &blk_mq_cpu_notify_list, list)
notify->notify(notify->data, action, cpu);
rt,blk,mq: Make blk_mq_cpu_notify_lock a raw spinlock [ 365.164040] BUG: sleeping function called from invalid context at kernel/rtmutex.c:674 [ 365.164041] in_atomic(): 1, irqs_disabled(): 1, pid: 26, name: migration/1 [ 365.164043] no locks held by migration/1/26. [ 365.164044] irq event stamp: 6648 [ 365.164056] hardirqs last enabled at (6647): [<ffffffff8153d377>] restore_args+0x0/0x30 [ 365.164062] hardirqs last disabled at (6648): [<ffffffff810ed98d>] multi_cpu_stop+0x9d/0x120 [ 365.164070] softirqs last enabled at (0): [<ffffffff810543bc>] copy_process.part.28+0x6fc/0x1920 [ 365.164072] softirqs last disabled at (0): [< (null)>] (null) [ 365.164076] CPU: 1 PID: 26 Comm: migration/1 Tainted: GF N 3.12.12-rt19-0.gcb6c4a2-rt #3 [ 365.164078] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011 [ 365.164091] 0000000000000001 ffff880a42ea7c30 ffffffff815367e6 ffffffff81a086c0 [ 365.164099] ffff880a42ea7c40 ffffffff8108919c ffff880a42ea7c60 ffffffff8153c24f [ 365.164107] ffff880a42ea91f0 00000000ffffffe1 ffff880a42ea7c88 ffffffff81297ec0 [ 365.164108] Call Trace: [ 365.164119] [<ffffffff810060b1>] try_stack_unwind+0x191/0x1a0 [ 365.164127] [<ffffffff81004872>] dump_trace+0x92/0x360 [ 365.164133] [<ffffffff81006108>] show_trace_log_lvl+0x48/0x60 [ 365.164138] [<ffffffff81004c18>] show_stack_log_lvl+0xd8/0x1d0 [ 365.164143] [<ffffffff81006160>] show_stack+0x20/0x50 [ 365.164153] [<ffffffff815367e6>] dump_stack+0x54/0x9a [ 365.164163] [<ffffffff8108919c>] __might_sleep+0xfc/0x140 [ 365.164173] [<ffffffff8153c24f>] rt_spin_lock+0x1f/0x70 [ 365.164182] [<ffffffff81297ec0>] blk_mq_main_cpu_notify+0x20/0x70 [ 365.164191] [<ffffffff81540a1c>] notifier_call_chain+0x4c/0x70 [ 365.164201] [<ffffffff81083499>] __raw_notifier_call_chain+0x9/0x10 [ 365.164207] [<ffffffff810567be>] cpu_notify+0x1e/0x40 [ 365.164217] [<ffffffff81525da2>] take_cpu_down+0x22/0x40 [ 365.164223] [<ffffffff810ed9c6>] multi_cpu_stop+0xd6/0x120 [ 365.164229] [<ffffffff810edd97>] cpu_stopper_thread+0xd7/0x1e0 [ 365.164235] [<ffffffff810863a3>] smpboot_thread_fn+0x203/0x380 [ 365.164241] [<ffffffff8107cbf8>] kthread+0xc8/0xd0 [ 365.164250] [<ffffffff8154440c>] ret_from_fork+0x7c/0xb0 [ 365.164429] smpboot: CPU 1 is now offline Signed-off-by: Mike Galbraith <bitbucket@online.de> Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-03 12:57:26 +08:00
raw_spin_unlock(&blk_mq_cpu_notify_lock);
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 16:20:05 +08:00
return NOTIFY_OK;
}
void blk_mq_register_cpu_notifier(struct blk_mq_cpu_notifier *notifier)
{
BUG_ON(!notifier->notify);
rt,blk,mq: Make blk_mq_cpu_notify_lock a raw spinlock [ 365.164040] BUG: sleeping function called from invalid context at kernel/rtmutex.c:674 [ 365.164041] in_atomic(): 1, irqs_disabled(): 1, pid: 26, name: migration/1 [ 365.164043] no locks held by migration/1/26. [ 365.164044] irq event stamp: 6648 [ 365.164056] hardirqs last enabled at (6647): [<ffffffff8153d377>] restore_args+0x0/0x30 [ 365.164062] hardirqs last disabled at (6648): [<ffffffff810ed98d>] multi_cpu_stop+0x9d/0x120 [ 365.164070] softirqs last enabled at (0): [<ffffffff810543bc>] copy_process.part.28+0x6fc/0x1920 [ 365.164072] softirqs last disabled at (0): [< (null)>] (null) [ 365.164076] CPU: 1 PID: 26 Comm: migration/1 Tainted: GF N 3.12.12-rt19-0.gcb6c4a2-rt #3 [ 365.164078] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011 [ 365.164091] 0000000000000001 ffff880a42ea7c30 ffffffff815367e6 ffffffff81a086c0 [ 365.164099] ffff880a42ea7c40 ffffffff8108919c ffff880a42ea7c60 ffffffff8153c24f [ 365.164107] ffff880a42ea91f0 00000000ffffffe1 ffff880a42ea7c88 ffffffff81297ec0 [ 365.164108] Call Trace: [ 365.164119] [<ffffffff810060b1>] try_stack_unwind+0x191/0x1a0 [ 365.164127] [<ffffffff81004872>] dump_trace+0x92/0x360 [ 365.164133] [<ffffffff81006108>] show_trace_log_lvl+0x48/0x60 [ 365.164138] [<ffffffff81004c18>] show_stack_log_lvl+0xd8/0x1d0 [ 365.164143] [<ffffffff81006160>] show_stack+0x20/0x50 [ 365.164153] [<ffffffff815367e6>] dump_stack+0x54/0x9a [ 365.164163] [<ffffffff8108919c>] __might_sleep+0xfc/0x140 [ 365.164173] [<ffffffff8153c24f>] rt_spin_lock+0x1f/0x70 [ 365.164182] [<ffffffff81297ec0>] blk_mq_main_cpu_notify+0x20/0x70 [ 365.164191] [<ffffffff81540a1c>] notifier_call_chain+0x4c/0x70 [ 365.164201] [<ffffffff81083499>] __raw_notifier_call_chain+0x9/0x10 [ 365.164207] [<ffffffff810567be>] cpu_notify+0x1e/0x40 [ 365.164217] [<ffffffff81525da2>] take_cpu_down+0x22/0x40 [ 365.164223] [<ffffffff810ed9c6>] multi_cpu_stop+0xd6/0x120 [ 365.164229] [<ffffffff810edd97>] cpu_stopper_thread+0xd7/0x1e0 [ 365.164235] [<ffffffff810863a3>] smpboot_thread_fn+0x203/0x380 [ 365.164241] [<ffffffff8107cbf8>] kthread+0xc8/0xd0 [ 365.164250] [<ffffffff8154440c>] ret_from_fork+0x7c/0xb0 [ 365.164429] smpboot: CPU 1 is now offline Signed-off-by: Mike Galbraith <bitbucket@online.de> Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-03 12:57:26 +08:00
raw_spin_lock(&blk_mq_cpu_notify_lock);
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 16:20:05 +08:00
list_add_tail(&notifier->list, &blk_mq_cpu_notify_list);
rt,blk,mq: Make blk_mq_cpu_notify_lock a raw spinlock [ 365.164040] BUG: sleeping function called from invalid context at kernel/rtmutex.c:674 [ 365.164041] in_atomic(): 1, irqs_disabled(): 1, pid: 26, name: migration/1 [ 365.164043] no locks held by migration/1/26. [ 365.164044] irq event stamp: 6648 [ 365.164056] hardirqs last enabled at (6647): [<ffffffff8153d377>] restore_args+0x0/0x30 [ 365.164062] hardirqs last disabled at (6648): [<ffffffff810ed98d>] multi_cpu_stop+0x9d/0x120 [ 365.164070] softirqs last enabled at (0): [<ffffffff810543bc>] copy_process.part.28+0x6fc/0x1920 [ 365.164072] softirqs last disabled at (0): [< (null)>] (null) [ 365.164076] CPU: 1 PID: 26 Comm: migration/1 Tainted: GF N 3.12.12-rt19-0.gcb6c4a2-rt #3 [ 365.164078] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011 [ 365.164091] 0000000000000001 ffff880a42ea7c30 ffffffff815367e6 ffffffff81a086c0 [ 365.164099] ffff880a42ea7c40 ffffffff8108919c ffff880a42ea7c60 ffffffff8153c24f [ 365.164107] ffff880a42ea91f0 00000000ffffffe1 ffff880a42ea7c88 ffffffff81297ec0 [ 365.164108] Call Trace: [ 365.164119] [<ffffffff810060b1>] try_stack_unwind+0x191/0x1a0 [ 365.164127] [<ffffffff81004872>] dump_trace+0x92/0x360 [ 365.164133] [<ffffffff81006108>] show_trace_log_lvl+0x48/0x60 [ 365.164138] [<ffffffff81004c18>] show_stack_log_lvl+0xd8/0x1d0 [ 365.164143] [<ffffffff81006160>] show_stack+0x20/0x50 [ 365.164153] [<ffffffff815367e6>] dump_stack+0x54/0x9a [ 365.164163] [<ffffffff8108919c>] __might_sleep+0xfc/0x140 [ 365.164173] [<ffffffff8153c24f>] rt_spin_lock+0x1f/0x70 [ 365.164182] [<ffffffff81297ec0>] blk_mq_main_cpu_notify+0x20/0x70 [ 365.164191] [<ffffffff81540a1c>] notifier_call_chain+0x4c/0x70 [ 365.164201] [<ffffffff81083499>] __raw_notifier_call_chain+0x9/0x10 [ 365.164207] [<ffffffff810567be>] cpu_notify+0x1e/0x40 [ 365.164217] [<ffffffff81525da2>] take_cpu_down+0x22/0x40 [ 365.164223] [<ffffffff810ed9c6>] multi_cpu_stop+0xd6/0x120 [ 365.164229] [<ffffffff810edd97>] cpu_stopper_thread+0xd7/0x1e0 [ 365.164235] [<ffffffff810863a3>] smpboot_thread_fn+0x203/0x380 [ 365.164241] [<ffffffff8107cbf8>] kthread+0xc8/0xd0 [ 365.164250] [<ffffffff8154440c>] ret_from_fork+0x7c/0xb0 [ 365.164429] smpboot: CPU 1 is now offline Signed-off-by: Mike Galbraith <bitbucket@online.de> Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-03 12:57:26 +08:00
raw_spin_unlock(&blk_mq_cpu_notify_lock);
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 16:20:05 +08:00
}
void blk_mq_unregister_cpu_notifier(struct blk_mq_cpu_notifier *notifier)
{
rt,blk,mq: Make blk_mq_cpu_notify_lock a raw spinlock [ 365.164040] BUG: sleeping function called from invalid context at kernel/rtmutex.c:674 [ 365.164041] in_atomic(): 1, irqs_disabled(): 1, pid: 26, name: migration/1 [ 365.164043] no locks held by migration/1/26. [ 365.164044] irq event stamp: 6648 [ 365.164056] hardirqs last enabled at (6647): [<ffffffff8153d377>] restore_args+0x0/0x30 [ 365.164062] hardirqs last disabled at (6648): [<ffffffff810ed98d>] multi_cpu_stop+0x9d/0x120 [ 365.164070] softirqs last enabled at (0): [<ffffffff810543bc>] copy_process.part.28+0x6fc/0x1920 [ 365.164072] softirqs last disabled at (0): [< (null)>] (null) [ 365.164076] CPU: 1 PID: 26 Comm: migration/1 Tainted: GF N 3.12.12-rt19-0.gcb6c4a2-rt #3 [ 365.164078] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011 [ 365.164091] 0000000000000001 ffff880a42ea7c30 ffffffff815367e6 ffffffff81a086c0 [ 365.164099] ffff880a42ea7c40 ffffffff8108919c ffff880a42ea7c60 ffffffff8153c24f [ 365.164107] ffff880a42ea91f0 00000000ffffffe1 ffff880a42ea7c88 ffffffff81297ec0 [ 365.164108] Call Trace: [ 365.164119] [<ffffffff810060b1>] try_stack_unwind+0x191/0x1a0 [ 365.164127] [<ffffffff81004872>] dump_trace+0x92/0x360 [ 365.164133] [<ffffffff81006108>] show_trace_log_lvl+0x48/0x60 [ 365.164138] [<ffffffff81004c18>] show_stack_log_lvl+0xd8/0x1d0 [ 365.164143] [<ffffffff81006160>] show_stack+0x20/0x50 [ 365.164153] [<ffffffff815367e6>] dump_stack+0x54/0x9a [ 365.164163] [<ffffffff8108919c>] __might_sleep+0xfc/0x140 [ 365.164173] [<ffffffff8153c24f>] rt_spin_lock+0x1f/0x70 [ 365.164182] [<ffffffff81297ec0>] blk_mq_main_cpu_notify+0x20/0x70 [ 365.164191] [<ffffffff81540a1c>] notifier_call_chain+0x4c/0x70 [ 365.164201] [<ffffffff81083499>] __raw_notifier_call_chain+0x9/0x10 [ 365.164207] [<ffffffff810567be>] cpu_notify+0x1e/0x40 [ 365.164217] [<ffffffff81525da2>] take_cpu_down+0x22/0x40 [ 365.164223] [<ffffffff810ed9c6>] multi_cpu_stop+0xd6/0x120 [ 365.164229] [<ffffffff810edd97>] cpu_stopper_thread+0xd7/0x1e0 [ 365.164235] [<ffffffff810863a3>] smpboot_thread_fn+0x203/0x380 [ 365.164241] [<ffffffff8107cbf8>] kthread+0xc8/0xd0 [ 365.164250] [<ffffffff8154440c>] ret_from_fork+0x7c/0xb0 [ 365.164429] smpboot: CPU 1 is now offline Signed-off-by: Mike Galbraith <bitbucket@online.de> Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-03 12:57:26 +08:00
raw_spin_lock(&blk_mq_cpu_notify_lock);
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 16:20:05 +08:00
list_del(&notifier->list);
rt,blk,mq: Make blk_mq_cpu_notify_lock a raw spinlock [ 365.164040] BUG: sleeping function called from invalid context at kernel/rtmutex.c:674 [ 365.164041] in_atomic(): 1, irqs_disabled(): 1, pid: 26, name: migration/1 [ 365.164043] no locks held by migration/1/26. [ 365.164044] irq event stamp: 6648 [ 365.164056] hardirqs last enabled at (6647): [<ffffffff8153d377>] restore_args+0x0/0x30 [ 365.164062] hardirqs last disabled at (6648): [<ffffffff810ed98d>] multi_cpu_stop+0x9d/0x120 [ 365.164070] softirqs last enabled at (0): [<ffffffff810543bc>] copy_process.part.28+0x6fc/0x1920 [ 365.164072] softirqs last disabled at (0): [< (null)>] (null) [ 365.164076] CPU: 1 PID: 26 Comm: migration/1 Tainted: GF N 3.12.12-rt19-0.gcb6c4a2-rt #3 [ 365.164078] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011 [ 365.164091] 0000000000000001 ffff880a42ea7c30 ffffffff815367e6 ffffffff81a086c0 [ 365.164099] ffff880a42ea7c40 ffffffff8108919c ffff880a42ea7c60 ffffffff8153c24f [ 365.164107] ffff880a42ea91f0 00000000ffffffe1 ffff880a42ea7c88 ffffffff81297ec0 [ 365.164108] Call Trace: [ 365.164119] [<ffffffff810060b1>] try_stack_unwind+0x191/0x1a0 [ 365.164127] [<ffffffff81004872>] dump_trace+0x92/0x360 [ 365.164133] [<ffffffff81006108>] show_trace_log_lvl+0x48/0x60 [ 365.164138] [<ffffffff81004c18>] show_stack_log_lvl+0xd8/0x1d0 [ 365.164143] [<ffffffff81006160>] show_stack+0x20/0x50 [ 365.164153] [<ffffffff815367e6>] dump_stack+0x54/0x9a [ 365.164163] [<ffffffff8108919c>] __might_sleep+0xfc/0x140 [ 365.164173] [<ffffffff8153c24f>] rt_spin_lock+0x1f/0x70 [ 365.164182] [<ffffffff81297ec0>] blk_mq_main_cpu_notify+0x20/0x70 [ 365.164191] [<ffffffff81540a1c>] notifier_call_chain+0x4c/0x70 [ 365.164201] [<ffffffff81083499>] __raw_notifier_call_chain+0x9/0x10 [ 365.164207] [<ffffffff810567be>] cpu_notify+0x1e/0x40 [ 365.164217] [<ffffffff81525da2>] take_cpu_down+0x22/0x40 [ 365.164223] [<ffffffff810ed9c6>] multi_cpu_stop+0xd6/0x120 [ 365.164229] [<ffffffff810edd97>] cpu_stopper_thread+0xd7/0x1e0 [ 365.164235] [<ffffffff810863a3>] smpboot_thread_fn+0x203/0x380 [ 365.164241] [<ffffffff8107cbf8>] kthread+0xc8/0xd0 [ 365.164250] [<ffffffff8154440c>] ret_from_fork+0x7c/0xb0 [ 365.164429] smpboot: CPU 1 is now offline Signed-off-by: Mike Galbraith <bitbucket@online.de> Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-03 12:57:26 +08:00
raw_spin_unlock(&blk_mq_cpu_notify_lock);
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 16:20:05 +08:00
}
void blk_mq_init_cpu_notifier(struct blk_mq_cpu_notifier *notifier,
void (*fn)(void *, unsigned long, unsigned int),
void *data)
{
notifier->notify = fn;
notifier->data = data;
}
void __init blk_mq_cpu_init(void)
{
hotcpu_notifier(blk_mq_main_cpu_notify, 0);
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 16:20:05 +08:00
}