nvme-multipath: introduce adaptive I/O policy
Subsystems (2)
v3 cover nvme-multipath: introduce adaptive I/O policy
Hi,
This series introduces a new adaptive I/O policy for NVMe native multipath. Existing policies such as numa, round-robin, and queue-depth are static and do not adapt to real-time transport performance. The numa selects the path closest to the NUMA node of the current CPU, optimizing memory and path locality, but ignores actual path performance. The round-robin distributes I/O evenly across all paths, providing fairness but not performance awareness. The queue-depth reacts to instantaneous queue occupancy, avoiding heavily loaded paths, but does not account for actual latency, throughput, or link speed.
The new adaptive policy addresses these gaps selecting paths dynamically based on measured I/O latency for both PCIe and fabrics. Latency is derived by passively sampling I/O completions. Each path is assigned a weight proportional to its latency score, and I/Os are then forwarded accordingly. As condition changes (e.g. latency spikes, bandwidth differences), path weights are updated, automatically steering traffic toward better-performing paths.
Early results show reduced tail latency under mixed workloads and improved throughput by exploiting higher-speed links more effectively. For example, with NVMf/TCP using two paths (one throttled with ~30 ms delay), fio results with random read/write/rw workloads (direct I/O) showed:
numa round-robin queue-depth adaptive
----------- ----------- ----------- ---------
READ: 50.0 MiB/s 105 MiB/s 230 MiB/s 350 MiB/s
WRITE: 65.9 MiB/s 125 MiB/s 385 MiB/s 446 MiB/s
RW: R:30.6 MiB/s R:56.5 MiB/s R:122 MiB/s R:175 MiB/s
W:30.7 MiB/s W:56.5 MiB/s W:122 MiB/s W:175 MiB/sThis pathcset includes totla 6 patches:
[PATCH 1/6] block: expose blk_stat_{enable,disable}_accounting()
- Make blk_stat APIs available to block drivers.
- Needed for per-path latency measurement in adaptive policy.
[PATCH 2/6] nvme-multipath: add adaptive I/O policy
- Implement path scoring based on latency (EWMA).
- Distribute I/O proportionally to per-path weights.
[PATCH 3/6] nvme: add sysfs attribute adp_ewma_shift
- Adds a sysfs attribute to control ewma shift
[PATCH 4/6] nvme: add sysfs attribute adp_weight_timeout
- Adds a sysfs attribute to control path weight calculation timeout
[PATCH 5/6] nvme: add generic debugfs support
- Introduce generic debugfs support for NVMe module
[PATCH 6/6] nvme-multipath: add debugfs attribute for adaptive I/O policy stats
- Add “adaptive_stat” under per-path and head debugfs directories to expose adaptive policy state and statistics.
As ususal, feedback and suggestions are most welcome!
Thanks!
Changes from v2:
- Addede a new patch to allow user to configure EWMA shift through sysfs (Hannes Reinecke)
- Added a new patch to allow user to configure path weight calculation timeout (Hannes Reinecke)
- Distinguish between read/write and other commands (e.g. admin comamnd) and calculate path weight for other commands which is separate from read/write weight. (Hannes Reinecke)
- Normalize per-path weight in the range from 0-128 instead
of 0-100 (Hannes Reinecke)
- Restructure and optimize adaptive I/O forwarding code to use
one loop instead of two (Hannes Reinecke)
Link to v2: https://lore.kernel.org/all/[email protected]/
Changes from v1:
- Ensure that the completion of I/O occurs on the same CPU as the submitting I/O CPU (Hannes Reinecke)
- Remove adapter link speed from the path weight calculation
(Hannes Reinecke)
- Add adaptive I/O stat under debugfs instead of current sysfs (Hannes Reinecke)
- Move path weight calculation to a workqueue from IO completion code path
Link to v1: https://lore.kernel.org/all/[email protected]/
Nilay Shroff (6):
block: expose blk_stat_{enable,disable}_accounting() to drivers
nvme-multipath: add support for adaptive I/O policy
nvme: add sysfs attribute adp_ewma_shift
nvme: add sysfs attribute adp_weight_timeout
nvme: add generic debugfs support
nvme-multipath: add debugfs attribute for adaptive I/O policy stat
block/blk-stat.h | 4 - drivers/nvme/host/Makefile | 2 +- drivers/nvme/host/core.c | 31 ++- drivers/nvme/host/debugfs.c | 236 ++++++++++++++++ drivers/nvme/host/ioctl.c | 31 ++- drivers/nvme/host/multipath.c | 487 +++++++++++++++++++++++++++++++++- drivers/nvme/host/nvme.h | 86 +++++- drivers/nvme/host/pr.c | 6 +- drivers/nvme/host/sysfs.c | 4 +- include/linux/blk-mq.h | 4 + 10 files changed, 862 insertions(+), 29 deletions(-) create mode 100644 drivers/nvme/host/debugfs.c
Signature · 1 line
2.51.0
RFC PATCH v3 2/6 nvme-multipath: add support for adaptive I/O policy
This commit introduces a new I/O policy named "adaptive". Users can configure it by writing "adaptive" to "/sys/class/nvme-subsystem/nvme- subsystemX/iopolicy"
The adaptive policy dynamically distributes I/O based on measured completion latency. The main idea is to calculate latency for each path, derive a weight, and then proportionally forward I/O according to those weights.
To ensure scalability, path latency is measured per-CPU. Each CPU maintains its own statistics, and I/O forwarding uses these per-CPU values. Every ~15 seconds, a simple average latency of per-CPU batched samples are computed and fed into an Exponentially Weighted Moving Average (EWMA):
avg_latency = div_u64(batch, batch_count);new_ewma_latency = (prev_ewma_latency * (WEIGHT-1) + avg_latency)/WEIGHT
With WEIGHT = 8, this assigns 7/8 (~87.5%) weight to the previous latency value and 1/8 (~12.5%) to the most recent latency. This smoothing reduces jitter, adapts quickly to changing conditions, avoids storing historical samples, and works well for both low and high I/O rates. Path weights are then derived from the smoothed (EWMA) latency as follows (example with two paths A and B):
path_A_score = NSEC_PER_SEC / path_A_ewma_latency path_B_score = NSEC_PER_SEC / path_B_ewma_latency total_score = path_A_score + path_B_score
path_A_weight = (path_A_score * 100) / total_score path_B_weight = (path_B_score * 100) / total_score
where:
- path_X_ewma_latency is the smoothed latency of a path in nanoseconds
- NSEC_PER_SEC is used as a scaling factor since valid latencies are < 1 second
- weights are normalized to a 0–100 scale across all paths.
Path credits are refilled based on this weight, with one credit consumed per I/O. When all credits are consumed, the credits are refilled again based on the current weight. This ensures that I/O is distributed across paths proportionally to their calculated weight.
--- drivers/nvme/host/core.c | 12 +- drivers/nvme/host/ioctl.c | 31 ++- drivers/nvme/host/multipath.c | 421 ++++++++++++++++++++++++++++++++-- drivers/nvme/host/nvme.h | 71 +++++- drivers/nvme/host/pr.c | 6 +- drivers/nvme/host/sysfs.c | 2 +- 6 files changed, 520 insertions(+), 23 deletions(-)
drivers/nvme/host/core.c +10 -2
@@ -689,6 +689,7 @@ static void nvme_free_ns(struct kref *kref) { struct nvme_ns *ns = container_of(kref, struct nvme_ns, kref); +nvme_free_ns_stat (ns);put_disk (ns->disk);nvme_put_ns_head (ns->head);nvme_put_ctrl (ns->ctrl);
@@ -4137,6 +4138,9 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info)if (nvme_init_ns_head (ns, info)) goto out_cleanup_disk; +if (nvme_alloc_ns_stat (ns)) + goto out_unlink_ns; +/* * If multipathing is enabled, the device name for all disks and not * just those that represent shared namespaces needs to be based on the
@@ -4161,7 +4165,7 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info) }if (nvme_update_ns_info (ns, info)) - goto out_unlink_ns; + goto out_free_ns_stat;mutex_lock (&ctrl->namespaces_lock);/*
@@ -4170,7 +4174,7 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info)*/ if (test_bit (NVME_CTRL_FROZEN , &ctrl->flags)) {mutex_unlock (&ctrl->namespaces_lock); - goto out_unlink_ns; + goto out_free_ns_stat; }nvme_ns_add_to_ctrl_list (ns);mutex_unlock (&ctrl->namespaces_lock);
@@ -4201,6 +4205,8 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info)list_del_rcu (&ns->list);mutex_unlock (&ctrl->namespaces_lock);synchronize_srcu (&ctrl->srcu); +out_free_ns_stat: +nvme_free_ns_stat (ns); out_unlink_ns:mutex_lock (&ctrl->subsys->lock);list_del_rcu (&ns->siblings);
@@ -4244,6 +4250,8 @@ static void nvme_ns_remove(struct nvme_ns *ns)*/ synchronize_srcu (&ns->head->srcu); +nvme_mpath_cancel_current_adaptive_weight_work (ns); +/* wait for concurrent submissions */ if (nvme_mpath_clear_current_path (ns))synchronize_srcu (&ns->head->srcu);
drivers/nvme/host/ioctl.c +28 -3
@@ -700,18 +700,29 @@ static int nvme_ns_head_ctrl_ioctl(struct nvme_ns *ns, unsigned int cmd,int nvme_ns_head_ioctl (struct block_device *bdev,blk_mode_t mode,unsigned int cmd,unsigned long arg) { +u8 opcode;struct nvme_ns_head *head = bdev->bd_disk->private_data;bool open_for_write = mode &BLK_OPEN_WRITE ;void __user *argp = (void __user *)arg;struct nvme_ns *ns;int srcu_idx, ret = -EWOULDBLOCK ;unsigned int flags =0 ; +unsigned int op_type =NVME_STAT_OTHER ;if (bdev_is_partition (bdev)) flags |=NVME_IOCTL_PARTITION ; +if (cmd ==NVME_IOCTL_SUBMIT_IO ) { +if (get_user (opcode, (u8 *)argp)) +return -EFAULT ; +if (opcode == nvme_cmd_write) + op_type =NVME_STAT_WRITE ; +else if (opcode == nvme_cmd_read) + op_type =NVME_STAT_READ ; + } + srcu_idx =srcu_read_lock (&head->srcu); - ns =nvme_find_path (head); + ns =nvme_find_path (head, op_type);if (!ns) goto out_unlock;
@@ -733,6 +744,7 @@ int nvme_ns_head_ioctl(struct block_device *bdev, blk_mode_t mode,long nvme_ns_head_chr_ioctl (struct file *file,unsigned int cmd,unsigned long arg) { +u8 opcode;bool open_for_write = file->f_mode &FMODE_WRITE ;struct cdev *cdev =file_inode (file)->i_cdev; struct nvme_ns_head *head =
@@ -740,9 +752,19 @@ long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd,void __user *argp = (void __user *)arg;struct nvme_ns *ns;int srcu_idx, ret = -EWOULDBLOCK ; +unsigned int op_type =NVME_STAT_OTHER ; + +if (cmd ==NVME_IOCTL_SUBMIT_IO ) { +if (get_user (opcode, (u8 *)argp)) +return -EFAULT ; +if (opcode == nvme_cmd_write) + op_type =NVME_STAT_WRITE ; +else if (opcode == nvme_cmd_read) + op_type =NVME_STAT_READ ; + } srcu_idx =srcu_read_lock (&head->srcu); - ns =nvme_find_path (head); + ns =nvme_find_path (head, op_type);if (!ns) goto out_unlock;
@@ -762,7 +784,10 @@ int nvme_ns_head_chr_uring_cmd(struct io_uring_cmd *ioucmd,struct cdev *cdev =file_inode (ioucmd->file)->i_cdev; struct nvme_ns_head *head = container_of(cdev, struct nvme_ns_head, cdev);int srcu_idx =srcu_read_lock (&head->srcu); -struct nvme_ns *ns =nvme_find_path (head); +const struct nvme_uring_cmd *cmd =io_uring_sqe_cmd (ioucmd->sqe); +struct nvme_ns *ns =nvme_find_path (head, +READ_ONCE (cmd->opcode) &1 ? +NVME_STAT_WRITE :NVME_STAT_READ );int ret = -EINVAL ;if (ns)
drivers/nvme/host/multipath.c +408 -13
528 lines not on this page · raw message
drivers/nvme/host/nvme.h +69 -2
131 lines not on this page · raw message
drivers/nvme/host/pr.c +4 -2
@@ -53,10 +53,12 @@ static int nvme_send_ns_head_pr_command(struct block_device *bdev, struct nvme_command *c, void *data, unsigned int data_len) { struct nvme_ns_head *head = bdev->bd_disk->private_data; -int srcu_idx =srcu_read_lock (&head->srcu); -struct nvme_ns *ns =nvme_find_path (head); +int srcu_idx; +struct nvme_ns *ns;int ret = -EWOULDBLOCK ; + srcu_idx =srcu_read_lock (&head->srcu); + ns =nvme_find_path (head,NVME_STAT_OTHER );if (ns) { c->common.nsid =cpu_to_le32 (ns->head->ns_id); ret =nvme_submit_sync_cmd (ns->queue, c, data, data_len);
drivers/nvme/host/sysfs.c +1 -1
@@ -194,7 +194,7 @@ static int ns_head_update_nuse(struct nvme_ns_head *head)return 0 ; srcu_idx =srcu_read_lock (&head->srcu); - ns =nvme_find_path (head); + ns =nvme_find_path (head,NVME_STAT_OTHER );if (!ns) goto out_unlock;
Signature · 1 line
2.51.0
RFC PATCH v3 1/6 block: expose blk_stat_{enable,disable}_accounting() to drivers
In reply to Nilay ShroffThe functions blk_stat_enable_accounting() and blk_stat_disable_accounting() are currently exported, but their prototypes are only defined in a private header. Move these prototypes into a common header so that block drivers can directly use these APIs.
--- block/blk-stat.h | 4 ---- include/linux/blk-mq.h | 4 ++++ 2 files changed, 4 insertions(+), 4 deletions(-)
block/blk-stat.h +0 -4
@@ -67,10 +67,6 @@ void blk_free_queue_stats(struct blk_queue_stats *);void blk_stat_add (struct request *rq,u64 now); -/* record time/size info in request but not add a callback */ -void blk_stat_enable_accounting (struct request_queue *q); -void blk_stat_disable_accounting (struct request_queue *q); -/** * blk_stat_alloc_callback() - Allocate a block statistics callback. * @timer_fn: Timer callback function.
include/linux/blk-mq.h +4 -0
@@ -735,6 +735,10 @@ int blk_rq_poll(struct request *rq, struct io_comp_batch *iob,bool blk_mq_queue_inflight (struct request_queue *q); +/* record time/size info in request but not add a callback */ +void blk_stat_enable_accounting (struct request_queue *q); +void blk_stat_disable_accounting (struct request_queue *q); +enum {/* return when out of requests */ BLK_MQ_REQ_NOWAIT = (__force blk_mq_req_flags_t)(1 <<0 ),
Signature · 1 line
2.51.0
RFC PATCH v3 4/6 nvme: add sysfs attribute adp_weight_timeout
In reply to Nilay ShroffBy default, the adaptive I/O policy accumulates latency samples over a 15-second window. When this window expires, the driver computes the average latency and updates the smoothed (EWMA) latency value. The path weight is then recalculated based on this data.
A 15-second window provides a good balance for most workloads, as it helps smooth out transient latency spikes and produces a more stable path weight profile. However, some workloads may benefit from faster or slower adaptation to changing latency conditions.
This commit introduces a new sysfs attribute, adp_weight_timeout, which allows users to configure the path weight calculation interval based on their workload requirements.
--- drivers/nvme/host/core.c | 6 ++++++ drivers/nvme/host/multipath.c | 38 +++++++++++++++++++++++++++++++++-- drivers/nvme/host/nvme.h | 4 +++- drivers/nvme/host/sysfs.c | 1 + 4 files changed, 46 insertions(+), 3 deletions(-)
drivers/nvme/host/core.c +6 -0
@@ -3248,6 +3248,12 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)* used while adding latency sample for adaptive iopolicy. */ subsys->adp_ewma_shift =NVME_DEFAULT_ADP_EWMA_SHIFT ; +/* +* Path weight calculation timeout interval used for adaptive iopolicy. +* The default value of this paremeter is set to 15 seconds. However, it +* could be also changed through sysfs. +*/ + subsys->adp_weight_timeout =NVME_DEFAULT_ADP_WEIGHT_TIMEOUT ;#endif subsys->dev.class = &nvme_subsys_class; subsys->dev.release = nvme_release_subsystem;
drivers/nvme/host/multipath.c +36 -2
@@ -362,8 +362,11 @@ static void nvme_mpath_add_sample(struct request *rq, struct nvme_ns *ns) stat->batch_count++; stat->nr_samples++; -if (now > stat->last_weight_ts && - (now - stat->last_weight_ts) >=NVME_DEFAULT_ADP_WEIGHT_TIMEOUT ) { +if (now > stat->last_weight_ts) { +u64 timeout =READ_ONCE (head->subsys->adp_weight_timeout); + +if ((now - stat->last_weight_ts) < timeout) +return ; stat->last_weight_ts = now;
@@ -1495,6 +1498,37 @@ static ssize_t nvme_subsys_adp_ewma_shift_store(struct device *dev,SUBSYS_ATTR_RW (adp_ewma_shift,0644 , nvme_subsys_adp_ewma_shift_show, nvme_subsys_adp_ewma_shift_store); +static ssize_t nvme_subsys_adp_weight_timeout_show (struct device *dev, +struct device_attribute *attr,char *buf) +{ +struct nvme_subsystem *subsys = + container_of(dev, struct nvme_subsystem, dev); + +return sysfs_emit (buf,"%llu\n" , +div_u64 (READ_ONCE (subsys->adp_weight_timeout),NSEC_PER_SEC )); +} + +static ssize_t nvme_subsys_adp_weight_timeout_store (struct device *dev, +struct device_attribute *attr,const char *buf,size_t count) +{ +int timeout, err; +struct nvme_subsystem *subsys = + container_of(dev, struct nvme_subsystem, dev); + + err =kstrtoint (buf,0 , &timeout); +if (err) +return -EINVAL ; + +if (timeout <=0 ) +return -EINVAL ; + +WRITE_ONCE (subsys->adp_weight_timeout, timeout *NSEC_PER_SEC ); +return count; +} + +SUBSYS_ATTR_RW (adp_weight_timeout,0644 , nvme_subsys_adp_weight_timeout_show, + nvme_subsys_adp_weight_timeout_store); +static ssize_t ana_grpid_show (struct device *dev,struct device_attribute *attr,char *buf) {
drivers/nvme/host/nvme.h +3 -1
@@ -450,7 +450,8 @@ struct nvme_subsystem {struct ida ns_ida;#ifdef CONFIG_NVME_MULTIPATH enum nvme_iopolicy iopolicy; -int adp_ewma_shift;/* used for adaptive iopolicy */ +int adp_ewma_shift;/* used for adaptive iopolicy */ +u64 adp_weight_timeout;/* used for adaptive iopolicy */ #endif };
@@ -1045,6 +1046,7 @@ extern struct device_attribute dev_attr_numa_nodes;extern struct device_attribute dev_attr_delayed_removal_secs;extern struct device_attribute subsys_attr_iopolicy;extern struct device_attribute subsys_attr_adp_ewma_shift; +extern struct device_attribute subsys_attr_adp_weight_timeout;static inline bool nvme_disk_is_ns_head (struct gendisk *disk) {
drivers/nvme/host/sysfs.c +1 -0
@@ -918,6 +918,7 @@ static struct attribute *nvme_subsys_attrs[] = {#ifdef CONFIG_NVME_MULTIPATH &subsys_attr_iopolicy.attr, &subsys_attr_adp_ewma_shift.attr, + &subsys_attr_adp_weight_timeout.attr, #endif NULL, };
Signature · 1 line
2.51.0
RFC PATCH v3 3/6 nvme: add sysfs attribute adp_ewma_shift
In reply to Nilay ShroffBy default, the EWMA (Exponentially Weighted Moving Average) shift value, used for storing latency samples for adaptive iopolicy, is set to 3. The EWMA is calculated using the following formula:
ewma = (old * ((1 << ewma_shift) - 1) + new) >> ewma_shift;
The default value of 3 assigns ~87.5% weight to the existing EWMA value and ~12.5% weight to the new latency sample. This provides a stable average that smooths out short-term variations.
However, different workloads may require faster or slower adaptation to changing conditions. This commit introduces a new sysfs attribute, adp_ewma_shift, allowing users to tune the weighting factor.
For example:
- adp_ewma_shift = 1 => 50% old, 50% new
- adp_ewma_shift = 0 => 0% old, 100% new
--- drivers/nvme/host/core.c | 10 ++++++++- drivers/nvme/host/multipath.c | 38 +++++++++++++++++++++++++++++++---- drivers/nvme/host/nvme.h | 2 ++ drivers/nvme/host/sysfs.c | 1 + 4 files changed, 46 insertions(+), 5 deletions(-)
drivers/nvme/host/core.c +9 -1
@@ -3240,7 +3240,15 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)return -EINVAL ; }nvme_mpath_default_iopolicy (subsys); - +#ifdef CONFIG_NVME_MULTIPATH +/* +* Default value of emwa_shift is set to 3 so that we can assign ~87.5 % +* weight to the existing ewma and ~12.5% weight to the new latency +* sample. This default could be changed through sysfs. This value is +* used while adding latency sample for adaptive iopolicy. +*/ + subsys->adp_ewma_shift =NVME_DEFAULT_ADP_EWMA_SHIFT ; +#endif subsys->dev.class = &nvme_subsys_class; subsys->dev.release = nvme_release_subsystem; subsys->dev.groups = nvme_subsys_attrs_groups;
drivers/nvme/host/multipath.c +34 -4
@@ -294,10 +294,9 @@ static void nvme_mpath_weight_work(struct work_struct *weight_work)* For instance, with EWMA_SHIFT = 3, this assigns 7/8 (~87.5 %) weight to * the existing/old ewma and 1/8 (~12.5%) weight to the new sample. */ -static inline u64 ewma_update (u64 old,u64 new) +static inline u64 ewma_update (u64 old,u64 new,int ewma_shift) { -return (old * ((1 <<NVME_DEFAULT_ADP_EWMA_SHIFT ) -1 ) - + new) >>NVME_DEFAULT_ADP_EWMA_SHIFT ; +return (old * ((1 << ewma_shift) -1 ) + new) >> ewma_shift; }static void nvme_mpath_add_sample (struct request *rq,struct nvme_ns *ns)
@@ -389,7 +388,8 @@ static void nvme_mpath_add_sample(struct request *rq, struct nvme_ns *ns)if (unlikely (!stat->slat_ns))WRITE_ONCE (stat->slat_ns, avg_lat_ns);else { - slat_ns =ewma_update (stat->slat_ns, avg_lat_ns); + slat_ns =ewma_update (stat->slat_ns, avg_lat_ns, +READ_ONCE (head->subsys->adp_ewma_shift));WRITE_ONCE (stat->slat_ns, slat_ns); }
@@ -1465,6 +1465,36 @@ static ssize_t nvme_subsys_iopolicy_store(struct device *dev,SUBSYS_ATTR_RW (iopolicy,S_IRUGO |S_IWUSR , nvme_subsys_iopolicy_show, nvme_subsys_iopolicy_store); +static ssize_t nvme_subsys_adp_ewma_shift_show (struct device *dev, +struct device_attribute *attr,char *buf) +{ +struct nvme_subsystem *subsys = + container_of(dev, struct nvme_subsystem, dev); + +return sysfs_emit (buf,"%d\n" ,READ_ONCE (subsys->adp_ewma_shift)); +} + +static ssize_t nvme_subsys_adp_ewma_shift_store (struct device *dev, +struct device_attribute *attr,const char *buf,size_t count) +{ +int shift, err; +struct nvme_subsystem *subsys = + container_of(dev, struct nvme_subsystem, dev); + + err =kstrtoint (buf,0 , &shift); +if (err) +return -EINVAL ; + +if (shift <0 ) +return -EINVAL ; + +WRITE_ONCE (subsys->adp_ewma_shift, shift); +return count; +} + +SUBSYS_ATTR_RW (adp_ewma_shift,0644 , nvme_subsys_adp_ewma_shift_show, + nvme_subsys_adp_ewma_shift_store); +static ssize_t ana_grpid_show (struct device *dev,struct device_attribute *attr,char *buf) {
drivers/nvme/host/nvme.h +2 -0
@@ -450,6 +450,7 @@ struct nvme_subsystem {struct ida ns_ida;#ifdef CONFIG_NVME_MULTIPATH enum nvme_iopolicy iopolicy; +int adp_ewma_shift;/* used for adaptive iopolicy */ #endif };
@@ -1043,6 +1044,7 @@ extern struct device_attribute dev_attr_queue_depth;extern struct device_attribute dev_attr_numa_nodes;extern struct device_attribute dev_attr_delayed_removal_secs;extern struct device_attribute subsys_attr_iopolicy; +extern struct device_attribute subsys_attr_adp_ewma_shift;static inline bool nvme_disk_is_ns_head (struct gendisk *disk) {
drivers/nvme/host/sysfs.c +1 -0
@@ -917,6 +917,7 @@ static struct attribute *nvme_subsys_attrs[] = { &subsys_attr_subsystype.attr, #ifdef CONFIG_NVME_MULTIPATH &subsys_attr_iopolicy.attr, + &subsys_attr_adp_ewma_shift.attr, #endif NULL, };
Signature · 1 line
2.51.0
RFC PATCH v3 5/6 nvme: add generic debugfs support
In reply to Nilay ShroffAdd generic infrastructure for creating and managing debugfs files in the NVMe module. This introduces helper APIs that allow NVMe drivers to register and unregister debugfs entries, along with a reusable attribute structure for defining new debugfs files.
The implementation uses seq_file interfaces to safely expose per-NS and per-NS-head statistics, while supporting both simple show callbacks and full seq_operations.
--- drivers/nvme/host/Makefile | 2 +- drivers/nvme/host/debugfs.c | 122 ++++++++++++++++++++++++++++++++++++ drivers/nvme/host/nvme.h | 11 ++++ 3 files changed, 134 insertions(+), 1 deletion(-) create mode 100644 drivers/nvme/host/debugfs.c
drivers/nvme/host/Makefile +1 -1
@@ -10,7 +10,7 @@ obj-$(CONFIG_NVME_FC) += nvme-fc.o obj-$(CONFIG_NVME_TCP) += nvme-tcp.o obj-$(CONFIG_NVME_APPLE) += nvme-apple.o -nvme-core-y += core.o ioctl.o sysfs.o pr.o +nvme-core-y += core.o ioctl.o sysfs.o pr.o debugfs.o nvme-core-$(CONFIG_NVME_VERBOSE_ERRORS) += constants.o nvme-core-$(CONFIG_TRACING) += trace.o nvme-core-$(CONFIG_NVME_MULTIPATH) += multipath.o
drivers/nvme/host/debugfs.c new file +122 -0
122 lines not on this page · raw message
drivers/nvme/host/nvme.h +11 -0
@@ -988,6 +988,7 @@ extern const struct attribute_group *nvme_dev_attr_groups[];extern const struct block_device_operations nvme_bdev_ops;void nvme_delete_ctrl_sync (struct nvme_ctrl *ctrl); +struct nvme_ns *nvme_find_path (struct nvme_ns_head *head,unsigned int op_type);static inline int nvme_data_dir (const enum req_op op) {
@@ -999,6 +1000,16 @@ static inline int nvme_data_dir(const enum req_op op)return NVME_STAT_OTHER ; } +void nvme_debugfs_register (struct gendisk *disk); +static inline void nvme_debugfs_unregister (struct gendisk *disk) +{ +/* +* Nothing to do for now. When the request queue is unregistered, +* all files under q->debugfs_dir are recursively deleted. +* This is just a placeholder; the compiler will optimize it out. +*/ +} +#ifdef CONFIG_NVME_MULTIPATH static inline bool nvme_ctrl_use_ana (struct nvme_ctrl *ctrl) {
Signature · 1 line
2.51.0
RFC PATCH v3 6/6 nvme-multipath: add debugfs attribute for adaptive I/O policy stat
In reply to Nilay ShroffThis commit introduces a new debugfs attribute, "adaptive_stat", under both per-path and head debugfs directories (defined under /sys/kernel/ debug/block/). This attribute provides visibility into the internal state of the adaptive I/O policy to aid in debugging and performance analysis.
For per-path entries, "adaptive_stat" reports the corresponding path statistics such as I/O weight, selection count, processed samples, and ignored samples.
For head entries, it reports per-CPU statistics for each reachable path, including I/O weight, path score, smoothed (EWMA) latency, selection count, processed samples, and ignored samples.
These additions enhance observability of the adaptive I/O path selection behavior and help diagnose imbalance or instability in multipath performance.
--- drivers/nvme/host/core.c | 3 + drivers/nvme/host/debugfs.c | 114 ++++++++++++++++++++++++++++++++++ drivers/nvme/host/multipath.c | 2 + 3 files changed, 119 insertions(+)
drivers/nvme/host/core.c +3 -0
@@ -4198,6 +4198,8 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info)if (device_add_disk (ctrl->device, ns->disk, nvme_ns_attr_groups)) goto out_cleanup_ns_from_list; +nvme_debugfs_register (ns->disk); +if (!nvme_ns_head_multipath (ns->head))nvme_add_ns_cdev (ns);
@@ -4287,6 +4289,7 @@ static void nvme_ns_remove(struct nvme_ns *ns)nvme_mpath_remove_sysfs_link (ns); +nvme_debugfs_unregister (ns->disk);del_gendisk (ns->disk);mutex_lock (&ns->ctrl->namespaces_lock);
drivers/nvme/host/debugfs.c +114 -0
126 lines not on this page · raw message
drivers/nvme/host/multipath.c +2 -0
@@ -1088,6 +1088,7 @@ static void nvme_remove_head(struct nvme_ns_head *head)nvme_cdev_del (&head->cdev, &head->cdev_device);synchronize_srcu (&head->srcu); +nvme_debugfs_unregister (head->disk);del_gendisk (head->disk); }nvme_put_ns_head (head);
@@ -1191,6 +1192,7 @@ static void nvme_mpath_set_live(struct nvme_ns *ns) }nvme_add_ns_head_cdev (head);kblockd_schedule_work (&head->partition_scan_work); +nvme_debugfs_register (head->disk); }nvme_mpath_add_sysfs_link (ns->head);
Signature · 1 line
2.51.0
Re: [RFC PATCHv3 2/6] nvme-multipath: add support for adaptive I/O policy
In reply to Nilay ShroffQuoted 489 lines · Nilay Shroff
This commit introduces a new I/O policy named "adaptive". Users can configure it by writing "adaptive" to "
/sys/class/nvme-subsystem/nvme-subsystemX/iopolicy"The adaptive policy dynamically distributes I/O based on measured completion latency. The main idea is to calculate latency for each path, derive a weight, and then proportionally forward I/O according to those weights.
To ensure scalability, path latency is measured per-CPU. Each CPU maintains its own statistics, and I/O forwarding uses these per-CPU values. Every ~15 seconds, a simple average latency of per-CPU batched samples are computed and fed into an Exponentially Weighted Moving Average (EWMA):
avg_latency=div_u64(batch, batch_count);new_ewma_latency= (prev_ewma_latency* (WEIGHT-1) + avg_latency)/WEIGHTWith WEIGHT = 8, this assigns 7/8 (~87.5%) weight to the previous latency value and 1/8 (~12.5%) to the most recent latency. This smoothing reduces jitter, adapts quickly to changing conditions, avoids storing historical samples, and works well for both low and high I/O rates. Path weights are then derived from the smoothed (EWMA) latency as follows (example with two paths A and B):
path_A_score=NSEC_PER_SEC/path_A_ewma_latencypath_B_score=NSEC_PER_SEC/path_B_ewma_latencytotal_score=path_A_score+path_B_score
path_A_weight= (path_A_score* 100) /total_scorepath_B_weight= (path_B_score* 100) /total_scorewhere:
-path_X_ewma_latencyis the smoothed latency of a path in nanoseconds
-NSEC_PER_SECis used as a scaling factor since valid latencies are < 1 second
- weights are normalized to a 0–100 scale across all paths.Path credits are refilled based on this weight, with one credit consumed per I/O. When all credits are consumed, the credits are refilled again based on the current weight. This ensures that I/O is distributed across paths proportionally to their calculated weight.
Signed-off-by Nilay Shroff <[email protected]>--- drivers/nvme/host/core.c | 12 +- drivers/nvme/host/ioctl.c | 31 ++- drivers/nvme/host/multipath.c | 421 ++++++++++++++++++++++++++++++++-- drivers/nvme/host/nvme.h | 71 +++++- drivers/nvme/host/pr.c | 6 +- drivers/nvme/host/sysfs.c | 2 +- 6 files changed, 520 insertions(+), 23 deletions(-)drivers/nvme/host/core.c +10
-2@@ -689,6 +689,7 @@ static void nvme_free_ns(struct kref *kref) { struct nvme_ns *ns = container_of(kref, struct nvme_ns, kref); +nvme_free_ns_stat (ns);put_disk (ns->disk);nvme_put_ns_head (ns->head);nvme_put_ctrl (ns->ctrl);@@ -4137,6 +4138,9 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info)if (nvme_init_ns_head (ns, info)) goto out_cleanup_disk; +if (nvme_alloc_ns_stat (ns)) + goto out_unlink_ns; +/* * If multipathing is enabled, the device name for all disks and not * just those that represent shared namespaces needs to be based on the @@ -4161,7 +4165,7 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info) }if (nvme_update_ns_info (ns, info)) - goto out_unlink_ns; + goto out_free_ns_stat;mutex_lock (&ctrl->namespaces_lock);/* @@ -4170,7 +4174,7 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info)*/ if (test_bit (NVME_CTRL_FROZEN , &ctrl->flags)) {mutex_unlock (&ctrl->namespaces_lock); - goto out_unlink_ns; + goto out_free_ns_stat; }nvme_ns_add_to_ctrl_list (ns);mutex_unlock (&ctrl->namespaces_lock);@@ -4201,6 +4205,8 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info)list_del_rcu (&ns->list);mutex_unlock (&ctrl->namespaces_lock);synchronize_srcu (&ctrl->srcu); +out_free_ns_stat: +nvme_free_ns_stat (ns); out_unlink_ns:mutex_lock (&ctrl->subsys->lock);list_del_rcu (&ns->siblings);@@ -4244,6 +4250,8 @@ static void nvme_ns_remove(struct nvme_ns *ns)*/ synchronize_srcu (&ns->head->srcu); +nvme_mpath_cancel_current_adaptive_weight_work (ns); +/* wait for concurrent submissions */ if (nvme_mpath_clear_current_path (ns))synchronize_srcu (&ns->head->srcu);drivers/nvme/host/ioctl.c +28
-3@@ -700,18 +700,29 @@ static int nvme_ns_head_ctrl_ioctl(struct nvme_ns *ns, unsigned int cmd,int nvme_ns_head_ioctl (struct block_device *bdev,blk_mode_t mode,unsigned int cmd,unsigned long arg) { +u8 opcode;struct nvme_ns_head *head = bdev->bd_disk->private_data;bool open_for_write = mode &BLK_OPEN_WRITE ;void __user *argp = (void __user *)arg;struct nvme_ns *ns;int srcu_idx, ret = -EWOULDBLOCK ;unsigned int flags =0 ; +unsigned int op_type =NVME_STAT_OTHER ;if (bdev_is_partition (bdev)) flags |=NVME_IOCTL_PARTITION ; +if (cmd ==NVME_IOCTL_SUBMIT_IO ) { +if (get_user (opcode, (u8 *)argp)) +return -EFAULT ; +if (opcode == nvme_cmd_write) + op_type =NVME_STAT_WRITE ; +else if (opcode == nvme_cmd_read) + op_type =NVME_STAT_READ ; + } + srcu_idx =srcu_read_lock (&head->srcu); - ns =nvme_find_path (head); + ns =nvme_find_path (head, op_type);if (!ns) goto out_unlock;@@ -733,6 +744,7 @@ int nvme_ns_head_ioctl(struct block_device *bdev, blk_mode_t mode,long nvme_ns_head_chr_ioctl (struct file *file,unsigned int cmd,unsigned long arg) { +u8 opcode;bool open_for_write = file->f_mode &FMODE_WRITE ;struct cdev *cdev =file_inode (file)->i_cdev; struct nvme_ns_head *head =@@ -740,9 +752,19 @@ long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd,void __user *argp = (void __user *)arg;struct nvme_ns *ns;int srcu_idx, ret = -EWOULDBLOCK ; +unsigned int op_type =NVME_STAT_OTHER ; + +if (cmd ==NVME_IOCTL_SUBMIT_IO ) { +if (get_user (opcode, (u8 *)argp)) +return -EFAULT ; +if (opcode == nvme_cmd_write) + op_type =NVME_STAT_WRITE ; +else if (opcode == nvme_cmd_read) + op_type =NVME_STAT_READ ; + } srcu_idx =srcu_read_lock (&head->srcu); - ns =nvme_find_path (head); + ns =nvme_find_path (head, op_type);if (!ns) goto out_unlock;@@ -762,7 +784,10 @@ int nvme_ns_head_chr_uring_cmd(struct io_uring_cmd *ioucmd,struct cdev *cdev =file_inode (ioucmd->file)->i_cdev; struct nvme_ns_head *head = container_of(cdev, struct nvme_ns_head, cdev);int srcu_idx =srcu_read_lock (&head->srcu); -struct nvme_ns *ns =nvme_find_path (head); +const struct nvme_uring_cmd *cmd =io_uring_sqe_cmd (ioucmd->sqe); +struct nvme_ns *ns =nvme_find_path (head, +READ_ONCE (cmd->opcode) &1 ? +NVME_STAT_WRITE :NVME_STAT_READ );int ret = -EINVAL ;if (ns)drivers/nvme/host/multipath.c +256
-3@@ -6,6 +6,9 @@#include <linux/backing-dev.h> #include <linux/moduleparam.h> #include <linux/vmalloc.h> +#include <linux/blk-mq.h> +#include <linux/math64.h> +#include <linux/rculist.h> #include <trace/events/block.h> #include "nvme.h" @@ -66,9 +69,10 @@ MODULE_PARM_DESC(multipath_always_on, "create multipath node always except for private namespace with non-unique nsid; note that this also implicitly enables native multipath support");static const char *nvme_iopolicy_names[] = { - [NVME_IOPOLICY_NUMA ] ="numa" , - [NVME_IOPOLICY_RR ] ="round-robin" , - [NVME_IOPOLICY_QD ] ="queue-depth" , + [NVME_IOPOLICY_NUMA ] ="numa" , + [NVME_IOPOLICY_RR ] ="round-robin" , + [NVME_IOPOLICY_QD ] ="queue-depth" , + [NVME_IOPOLICY_ADAPTIVE ] ="adaptive" , };static int iopolicy =NVME_IOPOLICY_NUMA ;@@ -83,6 +87,8 @@ static int nvme_set_iopolicy(const char *val, const struct kernel_param *kp) iopolicy =NVME_IOPOLICY_RR ; else if (!strncmp(val, "queue-depth", 11)) iopolicy =NVME_IOPOLICY_QD ; + else if (!strncmp(val, "adaptive", 8)) + iopolicy =NVME_IOPOLICY_ADAPTIVE ; else return -EINVAL;@@ -198,6 +204,204 @@ void nvme_mpath_start_request(struct request *rq) }EXPORT_SYMBOL_GPL (nvme_mpath_start_request); +static void nvme_mpath_weight_work (struct work_struct *weight_work) +{ +int cpu, srcu_idx; +u32 weight; +struct nvme_ns *ns; +struct nvme_path_stat *stat; +struct nvme_path_work *work =container_of (weight_work, + struct nvme_path_work, weight_work); +struct nvme_ns_head *head = work->ns->head; +int op_type = work->op_type; +u64 total_score =0 ; + + cpu =get_cpu (); + + srcu_idx =srcu_read_lock (&head->srcu); +list_for_each_entry_srcu (ns, &head->list, siblings, +srcu_read_lock_held (&head->srcu)) { + + stat = &this_cpu_ptr (ns->info)[op_type].stat; +if (!READ_ONCE (stat->slat_ns)) { + stat->score =0 ; +continue ; + } +/* +* Compute the path score as the inverse of smoothed +* latency, scaled by NSEC_PER_SEC. Floating point +* math is unavailable in the kernel, so fixed-point +* scaling is used instead. NSEC_PER_SEC is chosen +* because valid latencies are always < 1 second; longer +* latencies are ignored. +*/ + stat->score =div_u64 (NSEC_PER_SEC ,READ_ONCE (stat->slat_ns)); + +/* Compute total score. */ + total_score += stat->score; + } + +if (!total_score) + goto out; + +/* +* After computing the total slatency, we derive per-path weight +* (normalized to the range 0–100). The weight represents the +* relative share of I/O the path should receive. +* +* - lower smoothed latency -> higher weight +* - higher smoothed slatency -> lower weight +* +* Next, while forwarding I/O, we assign "credits" to each path +* based on its weight (please also refer nvme_adaptive_path()): +* - Initially, credits = weight. +* - Each time an I/O is dispatched on a path, its credits are +* decremented proportionally. +* - When a path runs out of credits, it becomes temporarily +* ineligible until credit is refilled. +* +* I/O distribution is therefore governed by available credits, +* ensuring that over time the proportion of I/O sent to each +* path matches its weight (and thus its performance). +*/ +list_for_each_entry_srcu (ns, &head->list, siblings, +srcu_read_lock_held (&head->srcu)) { + + stat = &this_cpu_ptr (ns->info)[op_type].stat; + weight =div_u64 (stat->score *128 , total_score); + +/* +* Ensure the path weight never drops below 1. A weight +* of 0 is used only for newly added paths. During +* bootstrap, a few I/Os are sent to such paths to +* establish an initial weight. Enforcing a minimum +* weight of 1 guarantees that no path is forgotten and +* that each path is probed at least occasionally. +*/ +if (!weight) + weight =1 ; + +WRITE_ONCE (stat->weight, weight); + } +out: +srcu_read_unlock (&head->srcu, srcu_idx); +put_cpu (); +} + +/* +* Formula to calculate the EWMA (Exponentially Weighted Moving Average): +* ewma = (old_ewma * (EWMA_SHIFT - 1) + (EWMA_SHIFT)) / EWMA_SHIFT +* For instance, with EWMA_SHIFT = 3, this assigns 7/8 (~87.5 %) weight to +* the existing/old ewma and 1/8 (~12.5%) weight to the new sample. +*/ +static inline u64 ewma_update (u64 old,u64 new) +{ +return (old * ((1 <<NVME_DEFAULT_ADP_EWMA_SHIFT ) -1 ) + + new) >>NVME_DEFAULT_ADP_EWMA_SHIFT ; +} + +static void nvme_mpath_add_sample (struct request *rq,struct nvme_ns *ns) +{ +int cpu; +unsigned int op_type; +struct nvme_path_info *info; +struct nvme_path_stat *stat; +u64 now, latency, slat_ns, avg_lat_ns; +struct nvme_ns_head *head = ns->head; + +if (list_is_singular (&head->list)) +return ; + + now =ktime_get_ns (); + latency = now >= rq->io_start_time_ns ? now - rq->io_start_time_ns :0 ; +if (!latency) +return ; + +/* +* As completion code path is serialized(i.e. no same completion queue +* update code could run simultaneously on multiple cpu) we can safely +* access per cpu nvme path stat here from another cpu (in case the +* completion cpu is different from submission cpu). +* The only field which could be accessed simultaneously here is the +* path ->weight which may be accessed by this function as well as I/O +* submission path during path selection logic and we protect ->weight +* using READ_ONCE/WRITE_ONCE. Yes this may not be 100% accurate but +* we also don't need to be so accurate here as the path credit would +* be anyways refilled, based on path weight, once path consumes all +* its credits. And we limit path weight/credit max up to 100. Please +* also refer nvme_adaptive_path(). +*/ + cpu =blk_mq_rq_cpu (rq); + op_type =nvme_data_dir (req_op (rq)); + info = &per_cpu_ptr (ns->info, cpu)[op_type]; + stat = &info->stat; + +/* +* If latency > ~1s then ignore this sample to prevent EWMA from being +* skewed by pathological outliers (multi-second waits, controller +* timeouts etc.). This keeps path scores representative of normal +* performance and avoids instability from rare spikes. If such high +* latency is real, ANA state reporting or keep-alive error counters +* will mark the path unhealthy and remove it from the head node list, +* so we safely skip such sample here. +*/ +if (unlikely (latency >NSEC_PER_SEC )) { + stat->nr_ignored++; +dev_warn_ratelimited (ns->ctrl->device, +"ignoring sample with >1s latency (possible controller stall or timeout)\n" ); +return ; + } + +/* +* Accumulate latency samples and increment the batch count for each +* ~15 second interval. When the interval expires, compute the simple +* average latency over that window, then update the smoothed (EWMA) +* latency. The path weight is recalculated based on this smoothed +* latency. +*/ + stat->batch += latency; + stat->batch_count++; + stat->nr_samples++; + +if (now > stat->last_weight_ts && + (now - stat->last_weight_ts) >=NVME_DEFAULT_ADP_WEIGHT_TIMEOUT ) { + + stat->last_weight_ts = now; + +/* +* Find simple average latency for the last epoch (~15 sec +* interval). +*/ + avg_lat_ns =div_u64 (stat->batch, stat->batch_count); + +/* +* Calculate smooth/EWMA (Exponentially Weighted Moving Average) +* latency. EWMA is preferred over simple average latency +* because it smooths naturally, reduces jitter from sudden +* spikes, and adapts faster to changing conditions. It also +* avoids storing historical samples, and works well for both +* slow and fast I/O rates. +* Formula: +* slat_ns = (prev_slat_ns * (WEIGHT - 1) + (latency)) / WEIGHT +* With WEIGHT = 8, this assigns 7/8 (~87.5 %) weight to the +* existing latency and 1/8 (~12.5%) weight to the new latency. +*/ +if (unlikely (!stat->slat_ns)) +WRITE_ONCE (stat->slat_ns, avg_lat_ns); +else { + slat_ns =ewma_update (stat->slat_ns, avg_lat_ns); +WRITE_ONCE (stat->slat_ns, slat_ns); + } + + stat->batch = stat->batch_count =0 ; + +/* +* Defer calculation of the path weight in per-cpu workqueue. +*/ +schedule_work_on (cpu, &info->work.weight_work); + } +} +void nvme_mpath_end_request (struct request *rq) {struct nvme_ns *ns = rq->q->queuedata;@@ -205,6 +409,9 @@ void nvme_mpath_end_request(struct request *rq)if (nvme_req (rq)->flags &NVME_MPATH_CNT_ACTIVE )atomic_dec_if_positive (&ns->ctrl->nr_active); +if (test_bit (NVME_NS_PATH_STAT , &ns->flags)) +nvme_mpath_add_sample (rq, ns); +if (!(nvme_req (rq)->flags &NVME_MPATH_IO_STATS ))return ; bdev_end_io_acct(ns->head->disk->part0, req_op(rq),@@ -238,6 +445,62 @@ static const char *nvme_ana_state_names[] = { [NVME_ANA_CHANGE] = "change", }; +static void nvme_mpath_reset_current_stat (struct nvme_ns *ns) +{ +int i, cpu; +struct nvme_path_stat *stat; + +for_each_possible_cpu (cpu) { +for (i =0 ; i <NVME_NUM_STAT_GROUPS ; i++) { + stat = &per_cpu_ptr (ns->info, cpu)[i].stat; +memset (stat,0 ,sizeof (struct nvme_path_stat )); + } + } +} + +void nvme_mpath_cancel_current_adaptive_weight_work (struct nvme_ns *ns) +{ +int i, cpu; +struct nvme_path_info *info; + +if (!test_bit (NVME_NS_PATH_STAT , &ns->flags)) +return ; + +for_each_online_cpu (cpu) { +for (i =0 ; i <NVME_NUM_STAT_GROUPS ; i++) { + info = &per_cpu_ptr (ns->info, cpu)[i]; +cancel_work_sync (&info->work.weight_work); + } + } +} + +static bool nvme_mpath_set_current_adaptive_path (struct nvme_ns *ns) +{ +struct nvme_ns_head *head = ns->head; + +if (!head->disk || head->subsys->iopolicy !=NVME_IOPOLICY_ADAPTIVE ) +return false; + +if (test_and_set_bit (NVME_NS_PATH_STAT , &ns->flags)) +return false; + +blk_queue_flag_set (QUEUE_FLAG_SAME_FORCE , ns->queue); +blk_stat_enable_accounting (ns->queue); +return true; +} + +static bool nvme_mpath_clear_current_adaptive_path (struct nvme_ns *ns) +{
drivers/nvme/host/multipath.c static const char *nvme_ana_state_names[] = { · line 503
Quoted 10 lines
drivers/nvme/host/multipath.c +10
-0+ +if (!test_and_clear_bit (NVME_NS_PATH_STAT , &ns->flags)) +return false; + +blk_stat_disable_accounting (ns->queue); +blk_queue_flag_clear (QUEUE_FLAG_SAME_FORCE , ns->queue); +nvme_mpath_reset_current_stat (ns); +return true; +} +
Hmm. I think this is a misnomer, as for the adaptive path policy we don't really have a 'current_path', and as such we don't clear the 'current_path' setting. Rather we disable the policy altogether here.
Can we reflect that in the function name and name it eg nvme_mpath_enable_adaptive_path_policy()
or somesuch?
Quoted 140 lines
drivers/nvme/host/multipath.c +112
-0bool nvme_mpath_clear_current_path (struct nvme_ns *ns) {struct nvme_ns_head *head = ns->head;@@ -253,6 +516,8 @@ bool nvme_mpath_clear_current_path(struct nvme_ns *ns) changed = true; } } +if (nvme_mpath_clear_current_adaptive_path (ns)) + changed = true; out:return changed; }@@ -271,6 +536,45 @@ void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl)srcu_read_unlock (&ctrl->srcu, srcu_idx); } +int nvme_alloc_ns_stat (struct nvme_ns *ns) +{ +int i, cpu; +struct nvme_path_work *work; +gfp_t gfp =GFP_KERNEL | __GFP_ZERO; + +if (!ns->head->disk) +return 0 ; + + ns->info =__alloc_percpu_gfp (NVME_NUM_STAT_GROUPS * +sizeof (struct nvme_path_info ), + __alignof__(struct nvme_path_info ), gfp); +if (!ns->info) +return -ENOMEM ; + +for_each_possible_cpu (cpu) { +for (i =0 ; i <NVME_NUM_STAT_GROUPS ; i++) { + work = &per_cpu_ptr (ns->info, cpu)[i].work; + work->ns = ns; + work->op_type = i; +INIT_WORK (&work->weight_work, nvme_mpath_weight_work); + } + } + +return 0 ; +} + +static void nvme_mpath_set_ctrl_paths (struct nvme_ctrl *ctrl) +{ +struct nvme_ns *ns; +int srcu_idx; + + srcu_idx =srcu_read_lock (&ctrl->srcu); +list_for_each_entry_srcu (ns, &ctrl->namespaces, list, +srcu_read_lock_held (&ctrl->srcu)) +nvme_mpath_set_current_adaptive_path (ns); +srcu_read_unlock (&ctrl->srcu, srcu_idx); +} +void nvme_mpath_revalidate_paths (struct nvme_ns *ns) {struct nvme_ns_head *head = ns->head;@@ -283,6 +587,8 @@ void nvme_mpath_revalidate_paths(struct nvme_ns *ns) srcu_read_lock_held(&head->srcu)) {if (capacity !=get_capacity (ns->disk))clear_bit (NVME_NS_READY , &ns->flags); + +nvme_mpath_reset_current_stat (ns); }srcu_read_unlock (&head->srcu, srcu_idx);@@ -407,6 +713,91 @@ static struct nvme_ns *nvme_round_robin_path(struct nvme_ns_head *head)return found; } +static inline bool nvme_state_is_live (enum nvme_ana_state state) +{ +return state ==NVME_ANA_OPTIMIZED || state ==NVME_ANA_NONOPTIMIZED ; +} + +static struct nvme_ns *nvme_adaptive_path (struct nvme_ns_head *head, +unsigned int op_type) +{ +struct nvme_ns *ns, *start, *found =NULL ; +struct nvme_path_stat *stat; +u32 weight; + +get_cpu (); + ns =srcu_dereference (head->current_path[0 ], &head->srcu); +if (unlikely (!ns)) { + ns =list_first_or_null_rcu (&head->list, + struct nvme_ns, siblings); +if (unlikely (!ns)) + goto out; + } +found_ns: + start = ns; +while (nvme_path_is_disabled (ns) || + !nvme_state_is_live (ns->ana_state)) { + ns =list_next_entry_circular (ns, &head->list, siblings); + +/* +* If we iterate through all paths in the list but find each +* path in list is either disabled or dead then bail out. +*/ +if (ns == start) + goto out; + } + + stat = &this_cpu_ptr (ns->info)[op_type].stat; + +/* +* When the head path-list is singular we don't calculate the +* only path weight for optimization as we don't need to forward +* I/O to more than one path. The another possibility is whenthe +* path is newly added, we don't know its weight. So we go round +* -robin for each such path and forward I/O to it.Once we start +* getting response for such I/Os, the path weight calculation +* would kick in and then we start using path credit for +* forwarding I/O. +*/ + weight =READ_ONCE (stat->weight); +if (!weight) { + found = ns; + goto out; + } + +/* +* To keep path selection logic simple, we don't distinguish +* between ANA optimized and non-optimized states. The non- +* optimized path is expected to have a lower weight, and +* therefore fewer credits. As a result, only a small number of +* I/Os will be forwarded to paths in the non-optimized state. +*/ +if (stat->credit >0 ) { + --stat->credit; + found = ns; + goto out; + }else { +/* +* Refill credit from path weight and move to next path. The +* refilled credit of the current path will be used next when +* all remainng paths exhaust its credits. +*/
drivers/nvme/host/multipath.c static struct nvme_ns *nvme_round_robin_path(struct nvme_ns_head *head) · line 794
Quoted 10 lines
drivers/nvme/host/multipath.c +10
-0+ weight =READ_ONCE (stat->weight); + stat->credit = weight; + ns =list_next_entry_circular (ns, &head->list, siblings); +if (likely (ns)) + goto found_ns; + } +out: +if (found) { + stat->sel++; +rcu_assign_pointer (head->current_path[0 ], found);
See my discussion about the 'current_path' here.
Do we really need it for the adaptive policy?
Quoted 255 lines
drivers/nvme/host/multipath.c +20
-10+ } + +put_cpu (); +return found; +} +static struct nvme_ns *nvme_queue_depth_path (struct nvme_ns_head *head) {struct nvme_ns *best_opt =NULL , *best_nonopt =NULL , *ns;@@ -463,9 +854,12 @@ static struct nvme_ns *nvme_numa_path(struct nvme_ns_head *head)return ns; } -inline struct nvme_ns *nvme_find_path (struct nvme_ns_head *head) +inline struct nvme_ns *nvme_find_path (struct nvme_ns_head *head, +unsigned int op_type) {switch (READ_ONCE (head->subsys->iopolicy)) { +case NVME_IOPOLICY_ADAPTIVE : +return nvme_adaptive_path (head, op_type);case NVME_IOPOLICY_QD :return nvme_queue_depth_path (head);case NVME_IOPOLICY_RR :@@ -525,7 +919,7 @@ static void nvme_ns_head_submit_bio(struct bio *bio)return ; srcu_idx =srcu_read_lock (&head->srcu); - ns =nvme_find_path (head); + ns =nvme_find_path (head,nvme_data_dir (bio_op (bio)));if (likely (ns)) {bio_set_dev (bio, ns->disk->part0); bio->bi_opf |=REQ_NVME_MPATH ;@@ -567,7 +961,7 @@ static int nvme_ns_head_get_unique_id(struct gendisk *disk, u8 id[16],int srcu_idx, ret = -EWOULDBLOCK ; srcu_idx =srcu_read_lock (&head->srcu); - ns =nvme_find_path (head); + ns =nvme_find_path (head,NVME_STAT_OTHER );if (ns) ret =nvme_ns_get_unique_id (ns, id, type);srcu_read_unlock (&head->srcu, srcu_idx);@@ -583,7 +977,7 @@ static int nvme_ns_head_report_zones(struct gendisk *disk, sector_t sector,int srcu_idx, ret = -EWOULDBLOCK ; srcu_idx =srcu_read_lock (&head->srcu); - ns =nvme_find_path (head); + ns =nvme_find_path (head,NVME_STAT_OTHER );if (ns) ret =nvme_ns_report_zones (ns, sector, nr_zones, cb, data);srcu_read_unlock (&head->srcu, srcu_idx);@@ -809,6 +1203,10 @@ static void nvme_mpath_set_live(struct nvme_ns *ns) }mutex_unlock (&head->lock); +mutex_lock (&nvme_subsystems_lock); +nvme_mpath_set_current_adaptive_path (ns); +mutex_unlock (&nvme_subsystems_lock); +synchronize_srcu (&head->srcu);kblockd_schedule_work (&head->requeue_work); }@@ -857,11 +1255,6 @@ static int nvme_parse_ana_log(struct nvme_ctrl *ctrl, void *data,return 0 ; } -static inline bool nvme_state_is_live (enum nvme_ana_state state) -{ -return state ==NVME_ANA_OPTIMIZED || state ==NVME_ANA_NONOPTIMIZED ; -} -static void nvme_update_ns_ana_state (struct nvme_ana_group_desc *desc,struct nvme_ns *ns) {@@ -1039,10 +1432,12 @@ static void nvme_subsys_iopolicy_update(struct nvme_subsystem *subsys,WRITE_ONCE (subsys->iopolicy, iopolicy); -/* iopolicy changes clear the mpath by design */ +/* iopolicy changes clear/reset the mpath by design */ mutex_lock (&nvme_subsystems_lock);list_for_each_entry (ctrl, &subsys->ctrls, subsys_entry)nvme_mpath_clear_ctrl_paths (ctrl); +list_for_each_entry (ctrl, &subsys->ctrls, subsys_entry) +nvme_mpath_set_ctrl_paths (ctrl);mutex_unlock (&nvme_subsystems_lock); pr_notice("subsysnqn %s iopolicy changed from %s to %s\n",drivers/nvme/host/nvme.h +69
-2@@ -28,7 +28,10 @@ extern unsigned int nvme_io_timeout;extern unsigned int admin_timeout;#define NVME_ADMIN_TIMEOUT (admin_timeout * HZ) -#define NVME_DEFAULT_KATO 5 +#define NVME_DEFAULT_KATO 5 + +#define NVME_DEFAULT_ADP_EWMA_SHIFT 3 +#define NVME_DEFAULT_ADP_WEIGHT_TIMEOUT (15 * NSEC_PER_SEC)#ifdef CONFIG_ARCH_NO_SG_CHAIN #define NVME_INLINE_SG_CNT 0@@ -421,6 +424,7 @@ enum nvme_iopolicy {NVME_IOPOLICY_NUMA ,NVME_IOPOLICY_RR ,NVME_IOPOLICY_QD , + NVME_IOPOLICY_ADAPTIVE, }; struct nvme_subsystem {@@ -459,6 +463,37 @@ struct nvme_ns_ids {u8 csi; }; +enum nvme_stat_group { +NVME_STAT_READ , +NVME_STAT_WRITE , +NVME_STAT_OTHER , +NVME_NUM_STAT_GROUPS +}; + +struct nvme_path_stat { +u64 nr_samples;/* total num of samples processed */ +u64 nr_ignored;/* num. of samples ignored */ +u64 slat_ns;/* smoothed (ewma) latency in nanoseconds */ +u64 score;/* score used for weight calculation */ +u64 last_weight_ts;/* timestamp of the last weight calculation */ +u64 sel;/* num of times this path is selcted for I/O */ +u64 batch;/* accumulated latency sum for current window */ +u32 batch_count;/* num of samples accumulated in current window */ +u32 weight;/* path weight */ +u32 credit;/* path credit for I/O forwarding */ +}; + +struct nvme_path_work { +struct nvme_ns *ns;/* owning namespace */ +struct work_struct weight_work;/* deferred work for weight calculation */ +int op_type;/* op type : READ/WRITE/OTHER */ +}; + +struct nvme_path_info { +struct nvme_path_stat stat;/* path statistics */ +struct nvme_path_work work;/* background worker context */ +}; +/* * Anchor structure for namespaces. There is one for each namespace in a * NVMe subsystem that any of our controllers can see, and the namespace @@ -534,6 +569,7 @@ struct nvme_ns {#ifdef CONFIG_NVME_MULTIPATH enum nvme_ana_state ana_state;u32 ana_grpid; +struct nvme_path_info __percpu *info;#endif struct list_head siblings;struct kref kref;@@ -545,6 +581,7 @@ struct nvme_ns {#define NVME_NS_FORCE_RO 3#define NVME_NS_READY 4#define NVME_NS_SYSFS_ATTR_LINK 5 +#define NVME_NS_PATH_STAT 6struct cdev cdev;struct device cdev_device;@@ -949,7 +986,17 @@ extern const struct attribute_group *nvme_dev_attr_groups[];extern const struct block_device_operations nvme_bdev_ops;void nvme_delete_ctrl_sync (struct nvme_ctrl *ctrl); -struct nvme_ns *nvme_find_path (struct nvme_ns_head *head); +struct nvme_ns *nvme_find_path (struct nvme_ns_head *head,unsigned int op_type); +static inline int nvme_data_dir (const enum req_op op) +{ +if (op ==REQ_OP_READ ) +return NVME_STAT_READ ; +else if (op_is_write (op)) +return NVME_STAT_WRITE ; +else +return NVME_STAT_OTHER ; +} +#ifdef CONFIG_NVME_MULTIPATH static inline bool nvme_ctrl_use_ana (struct nvme_ctrl *ctrl) {@@ -972,12 +1019,14 @@ void nvme_mpath_init_ctrl(struct nvme_ctrl *ctrl);void nvme_mpath_update (struct nvme_ctrl *ctrl);void nvme_mpath_uninit (struct nvme_ctrl *ctrl);void nvme_mpath_stop (struct nvme_ctrl *ctrl); +void nvme_mpath_cancel_current_adaptive_weight_work (struct nvme_ns *ns);bool nvme_mpath_clear_current_path (struct nvme_ns *ns);void nvme_mpath_revalidate_paths (struct nvme_ns *ns);void nvme_mpath_clear_ctrl_paths (struct nvme_ctrl *ctrl);void nvme_mpath_remove_disk (struct nvme_ns_head *head);void nvme_mpath_start_request (struct request *rq);void nvme_mpath_end_request (struct request *rq); +int nvme_alloc_ns_stat (struct nvme_ns *ns);static inline void nvme_trace_bio_complete (struct request *req) {@@ -1005,6 +1054,13 @@ static inline bool nvme_mpath_queue_if_no_path(struct nvme_ns_head *head)return true; return false; } +static inline void nvme_free_ns_stat (struct nvme_ns *ns) +{ +if (!ns->head->disk) +return ; + +free_percpu (ns->info); +}#else #define multipath falsestatic inline bool nvme_ctrl_use_ana (struct nvme_ctrl *ctrl)@@ -1096,6 +1152,17 @@ static inline bool nvme_mpath_queue_if_no_path(struct nvme_ns_head *head) {return false; } +static inline void nvme_mpath_cancel_current_adaptive_weight_work ( +struct nvme_ns *ns) +{ +} +static inline int nvme_alloc_ns_stat (struct nvme_ns *ns) +{ +return 0 ; +} +static inline void nvme_free_ns_stat (struct nvme_ns *ns) +{ +}#endif /* CONFIG_NVME_MULTIPATH */ int nvme_ns_get_unique_id(struct nvme_ns *ns, u8 id[16],drivers/nvme/host/pr.c +4
-2@@ -53,10 +53,12 @@ static int nvme_send_ns_head_pr_command(struct block_device *bdev, struct nvme_command *c, void *data, unsigned int data_len) { struct nvme_ns_head *head = bdev->bd_disk->private_data; -int srcu_idx =srcu_read_lock (&head->srcu); -struct nvme_ns *ns =nvme_find_path (head); +int srcu_idx; +struct nvme_ns *ns;int ret = -EWOULDBLOCK ; + srcu_idx =srcu_read_lock (&head->srcu); + ns =nvme_find_path (head,NVME_STAT_OTHER );if (ns) { c->common.nsid =cpu_to_le32 (ns->head->ns_id); ret =nvme_submit_sync_cmd (ns->queue, c, data, data_len);drivers/nvme/host/sysfs.c +0
-0
drivers/nvme/host/sysfs.c static int ns_head_update_nuse(struct nvme_ns_head *head) · line 200
Quoted 8 lines
drivers/nvme/host/sysfs.c +1
-1@@ -194,7 +194,7 @@ static int ns_head_update_nuse(struct nvme_ns_head *head)return 0 ; srcu_idx =srcu_read_lock (&head->srcu); - ns =nvme_find_path (head); + ns =nvme_find_path (head,NVME_STAT_OTHER );if (!ns) goto out_unlock;
Otherwise looks good.
Cheers,
Hannes
Signature · 4 lines
Dr. Hannes Reinecke Kernel Storage Architect [email protected] +49 911 74053 688 SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich
Re: [RFC PATCHv3 3/6] nvme: add sysfs attribute adp_ewma_shift
In reply to Nilay ShroffQuoted 136 lines · Nilay Shroff
By default, the EWMA (Exponentially Weighted Moving Average) shift value, used for storing latency samples for adaptive iopolicy, is set to 3. The EWMA is calculated using the following formula:
ewma = (old * ((1 << ewma_shift) - 1) + new) >> ewma_shift;The default value of 3 assigns ~87.5% weight to the existing EWMA value and ~12.5% weight to the new latency sample. This provides a stable average that smooths out short-term variations.
However, different workloads may require faster or slower adaptation to changing conditions. This commit introduces a new sysfs attribute,
adp_ewma_shift, allowing users to tune the weighting factor.For example:
-adp_ewma_shift= 1 => 50% old, 50% new
-adp_ewma_shift= 0 => 0% old, 100% newSigned-off-by Nilay Shroff <[email protected]>--- drivers/nvme/host/core.c | 10 ++++++++- drivers/nvme/host/multipath.c | 38 +++++++++++++++++++++++++++++++---- drivers/nvme/host/nvme.h | 2 ++ drivers/nvme/host/sysfs.c | 1 + 4 files changed, 46 insertions(+), 5 deletions(-)drivers/nvme/host/core.c +9
-1@@ -3240,7 +3240,15 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)return -EINVAL ; }nvme_mpath_default_iopolicy (subsys); - +#ifdef CONFIG_NVME_MULTIPATH +/* +* Default value of emwa_shift is set to 3 so that we can assign ~87.5 % +* weight to the existing ewma and ~12.5% weight to the new latency +* sample. This default could be changed through sysfs. This value is +* used while adding latency sample for adaptive iopolicy. +*/ + subsys->adp_ewma_shift =NVME_DEFAULT_ADP_EWMA_SHIFT ; +#endif subsys->dev.class = &nvme_subsys_class; subsys->dev.release = nvme_release_subsystem; subsys->dev.groups = nvme_subsys_attrs_groups;drivers/nvme/host/multipath.c +34
-4@@ -294,10 +294,9 @@ static void nvme_mpath_weight_work(struct work_struct *weight_work)* For instance, with EWMA_SHIFT = 3, this assigns 7/8 (~87.5 %) weight to * the existing/old ewma and 1/8 (~12.5%) weight to the new sample. */ -static inline u64 ewma_update (u64 old,u64 new) +static inline u64 ewma_update (u64 old,u64 new,int ewma_shift) { -return (old * ((1 <<NVME_DEFAULT_ADP_EWMA_SHIFT ) -1 ) - + new) >>NVME_DEFAULT_ADP_EWMA_SHIFT ; +return (old * ((1 << ewma_shift) -1 ) + new) >> ewma_shift; }static void nvme_mpath_add_sample (struct request *rq,struct nvme_ns *ns)@@ -389,7 +388,8 @@ static void nvme_mpath_add_sample(struct request *rq, struct nvme_ns *ns)if (unlikely (!stat->slat_ns))WRITE_ONCE (stat->slat_ns, avg_lat_ns);else { - slat_ns =ewma_update (stat->slat_ns, avg_lat_ns); + slat_ns =ewma_update (stat->slat_ns, avg_lat_ns, +READ_ONCE (head->subsys->adp_ewma_shift));WRITE_ONCE (stat->slat_ns, slat_ns); }@@ -1465,6 +1465,36 @@ static ssize_t nvme_subsys_iopolicy_store(struct device *dev,SUBSYS_ATTR_RW (iopolicy,S_IRUGO |S_IWUSR , nvme_subsys_iopolicy_show, nvme_subsys_iopolicy_store); +static ssize_t nvme_subsys_adp_ewma_shift_show (struct device *dev, +struct device_attribute *attr,char *buf) +{ +struct nvme_subsystem *subsys = + container_of(dev, struct nvme_subsystem, dev); + +return sysfs_emit (buf,"%d\n" ,READ_ONCE (subsys->adp_ewma_shift)); +} + +static ssize_t nvme_subsys_adp_ewma_shift_store (struct device *dev, +struct device_attribute *attr,const char *buf,size_t count) +{ +int shift, err; +struct nvme_subsystem *subsys = + container_of(dev, struct nvme_subsystem, dev); + + err =kstrtoint (buf,0 , &shift); +if (err) +return -EINVAL ; + +if (shift <0 ) +return -EINVAL ; + +WRITE_ONCE (subsys->adp_ewma_shift, shift); +return count; +} + +SUBSYS_ATTR_RW (adp_ewma_shift,0644 , nvme_subsys_adp_ewma_shift_show, + nvme_subsys_adp_ewma_shift_store); +static ssize_t ana_grpid_show (struct device *dev,struct device_attribute *attr,char *buf) {drivers/nvme/host/nvme.h +2
-0@@ -450,6 +450,7 @@ struct nvme_subsystem {struct ida ns_ida;#ifdef CONFIG_NVME_MULTIPATH enum nvme_iopolicy iopolicy; +int adp_ewma_shift;/* used for adaptive iopolicy */ #endif };@@ -1043,6 +1044,7 @@ extern struct device_attribute dev_attr_queue_depth;extern struct device_attribute dev_attr_numa_nodes;extern struct device_attribute dev_attr_delayed_removal_secs;extern struct device_attribute subsys_attr_iopolicy; +extern struct device_attribute subsys_attr_adp_ewma_shift;static inline bool nvme_disk_is_ns_head (struct gendisk *disk) {drivers/nvme/host/sysfs.c +0
-0
drivers/nvme/host/sysfs.c static struct attribute *nvme_subsys_attrs[] = { · line 923
Quoted 7 lines
drivers/nvme/host/sysfs.c +1
-0@@ -917,6 +917,7 @@ static struct attribute *nvme_subsys_attrs[] = { &subsys_attr_subsystype.attr, #ifdef CONFIG_NVME_MULTIPATH &subsys_attr_iopolicy.attr, + &subsys_attr_adp_ewma_shift.attr, #endif NULL, };
Hmm. Can we please move that to debugfs?
I'd rather not litter sysfs with module-specific attributes which come and go depending on the configuration ...
Cheers,
Hannes
Signature · 4 lines
Dr. Hannes Reinecke Kernel Storage Architect [email protected] +49 911 74053 688 SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich
Re: [RFC PATCHv3 4/6] nvme: add sysfs attribute adp_weight_timeout
In reply to Nilay ShroffQuoted 122 lines · Nilay Shroff
By default, the adaptive I/O policy accumulates latency samples over a 15-second window. When this window expires, the driver computes the average latency and updates the smoothed (EWMA) latency value. The path weight is then recalculated based on this data.
A 15-second window provides a good balance for most workloads, as it helps smooth out transient latency spikes and produces a more stable path weight profile. However, some workloads may benefit from faster or slower adaptation to changing latency conditions.
This commit introduces a new sysfs attribute,
adp_weight_timeout, which allows users to configure the path weight calculation interval based on their workload requirements.Signed-off-by Nilay Shroff <[email protected]>--- drivers/nvme/host/core.c | 6 ++++++ drivers/nvme/host/multipath.c | 38 +++++++++++++++++++++++++++++++++-- drivers/nvme/host/nvme.h | 4 +++- drivers/nvme/host/sysfs.c | 1 + 4 files changed, 46 insertions(+), 3 deletions(-)drivers/nvme/host/core.c +6
-0@@ -3248,6 +3248,12 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)* used while adding latency sample for adaptive iopolicy. */ subsys->adp_ewma_shift =NVME_DEFAULT_ADP_EWMA_SHIFT ; +/* +* Path weight calculation timeout interval used for adaptive iopolicy. +* The default value of this paremeter is set to 15 seconds. However, it +* could be also changed through sysfs. +*/ + subsys->adp_weight_timeout =NVME_DEFAULT_ADP_WEIGHT_TIMEOUT ;#endif subsys->dev.class = &nvme_subsys_class; subsys->dev.release = nvme_release_subsystem;drivers/nvme/host/multipath.c +36
-2@@ -362,8 +362,11 @@ static void nvme_mpath_add_sample(struct request *rq, struct nvme_ns *ns) stat->batch_count++; stat->nr_samples++; -if (now > stat->last_weight_ts && - (now - stat->last_weight_ts) >=NVME_DEFAULT_ADP_WEIGHT_TIMEOUT ) { +if (now > stat->last_weight_ts) { +u64 timeout =READ_ONCE (head->subsys->adp_weight_timeout); + +if ((now - stat->last_weight_ts) < timeout) +return ; stat->last_weight_ts = now;@@ -1495,6 +1498,37 @@ static ssize_t nvme_subsys_adp_ewma_shift_store(struct device *dev,SUBSYS_ATTR_RW (adp_ewma_shift,0644 , nvme_subsys_adp_ewma_shift_show, nvme_subsys_adp_ewma_shift_store); +static ssize_t nvme_subsys_adp_weight_timeout_show (struct device *dev, +struct device_attribute *attr,char *buf) +{ +struct nvme_subsystem *subsys = + container_of(dev, struct nvme_subsystem, dev); + +return sysfs_emit (buf,"%llu\n" , +div_u64 (READ_ONCE (subsys->adp_weight_timeout),NSEC_PER_SEC )); +} + +static ssize_t nvme_subsys_adp_weight_timeout_store (struct device *dev, +struct device_attribute *attr,const char *buf,size_t count) +{ +int timeout, err; +struct nvme_subsystem *subsys = + container_of(dev, struct nvme_subsystem, dev); + + err =kstrtoint (buf,0 , &timeout); +if (err) +return -EINVAL ; + +if (timeout <=0 ) +return -EINVAL ; + +WRITE_ONCE (subsys->adp_weight_timeout, timeout *NSEC_PER_SEC ); +return count; +} + +SUBSYS_ATTR_RW (adp_weight_timeout,0644 , nvme_subsys_adp_weight_timeout_show, + nvme_subsys_adp_weight_timeout_store); +static ssize_t ana_grpid_show (struct device *dev,struct device_attribute *attr,char *buf) {drivers/nvme/host/nvme.h +3
-1@@ -450,7 +450,8 @@ struct nvme_subsystem {struct ida ns_ida;#ifdef CONFIG_NVME_MULTIPATH enum nvme_iopolicy iopolicy; -int adp_ewma_shift;/* used for adaptive iopolicy */ +int adp_ewma_shift;/* used for adaptive iopolicy */ +u64 adp_weight_timeout;/* used for adaptive iopolicy */ #endif };@@ -1045,6 +1046,7 @@ extern struct device_attribute dev_attr_numa_nodes;extern struct device_attribute dev_attr_delayed_removal_secs;extern struct device_attribute subsys_attr_iopolicy;extern struct device_attribute subsys_attr_adp_ewma_shift; +extern struct device_attribute subsys_attr_adp_weight_timeout;static inline bool nvme_disk_is_ns_head (struct gendisk *disk) {drivers/nvme/host/sysfs.c +0
-0
drivers/nvme/host/sysfs.c static struct attribute *nvme_subsys_attrs[] = { · line 924
Quoted 7 lines
drivers/nvme/host/sysfs.c +1
-0@@ -918,6 +918,7 @@ static struct attribute *nvme_subsys_attrs[] = {#ifdef CONFIG_NVME_MULTIPATH &subsys_attr_iopolicy.attr, &subsys_attr_adp_ewma_shift.attr, + &subsys_attr_adp_weight_timeout.attr, #endif NULL, };
Similar comment: can we please move it to debugfs?
Cheers,
Hannes
Signature · 4 lines
Dr. Hannes Reinecke Kernel Storage Architect [email protected] +49 911 74053 688 SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich
Re: [RFC PATCHv3 5/6] nvme: add generic debugfs support
In reply to Nilay ShroffQuoted 178 lines · Nilay Shroff
Add generic infrastructure for creating and managing debugfs files in the NVMe module. This introduces helper APIs that allow NVMe drivers to register and unregister debugfs entries, along with a reusable attribute structure for defining new debugfs files.
The implementation uses
seq_fileinterfaces to safely expose per-NS and per-NS-head statistics, while supporting both simple show callbacks and fullseq_operations.Reviewed-by Hannes Reinecke <[email protected]>Signed-off-by Nilay Shroff <[email protected]>--- drivers/nvme/host/Makefile | 2 +- drivers/nvme/host/debugfs.c | 122 ++++++++++++++++++++++++++++++++++++ drivers/nvme/host/nvme.h | 11 ++++ 3 files changed, 134 insertions(+), 1 deletion(-)create mode 100644
drivers/nvme/host/debugfs.cdrivers/nvme/host/Makefile +1
-1@@ -10,7 +10,7 @@ obj-$(CONFIG_NVME_FC) += nvme-fc.o obj-$(CONFIG_NVME_TCP) += nvme-tcp.o obj-$(CONFIG_NVME_APPLE) += nvme-apple.o -nvme-core-y += core.o ioctl.o sysfs.o pr.o +nvme-core-y += core.o ioctl.o sysfs.o pr.o debugfs.o nvme-core-$(CONFIG_NVME_VERBOSE_ERRORS) += constants.o nvme-core-$(CONFIG_TRACING) += trace.o nvme-core-$(CONFIG_NVME_MULTIPATH) += multipath.odrivers/nvme/host/debugfs.c new file +122
-0@@ -0,0 +1,122 @@ +// SPDX-License-Identifier: GPL-2.0 +/* +* Copyright (c) 2025 IBM Corporation +* Nilay Shroff <[email protected]> +*/ + +#include <linux/debugfs.h> +#include <linux/seq_file.h> + +#include "nvme.h" + +struct nvme_debugfs_attr { +const char *name; +umode_t mode; +int (*show)(void *data,struct seq_file *m); +const struct seq_operations *seq_ops; +}; + +struct nvme_debugfs_ctx { +void *data; +struct nvme_debugfs_attr *attr; +int srcu_idx; +}; + +static int nvme_debugfs_show (struct seq_file *m,void *v) +{ +struct nvme_debugfs_ctx *ctx = m->private; +void *data = ctx->data; +struct nvme_debugfs_attr *attr = ctx->attr; + +return attr->show (data, m); +} + +static int nvme_debugfs_open (struct inode *inode,struct file *file) +{ +void *data = inode->i_private; +struct nvme_debugfs_attr *attr =debugfs_get_aux (file); +struct nvme_debugfs_ctx *ctx; +struct seq_file *m; +int ret; + + ctx =kzalloc (sizeof (*ctx),GFP_KERNEL ); +if (WARN_ON_ONCE (!ctx)) +return -ENOMEM ; + + ctx->data = data; + ctx->attr = attr; + +if (attr->seq_ops) { + ret =seq_open (file, attr->seq_ops); +if (ret) { +kfree (ctx); +return ret; + } + m = file->private_data; + m->private = ctx; +return ret; + } + +if (WARN_ON_ONCE (!attr->show)) { +kfree (ctx); +return -EPERM ; + } + +return single_open (file, nvme_debugfs_show, ctx); +} + +static int nvme_debugfs_release (struct inode *inode,struct file *file) +{ +struct seq_file *m = file->private_data; +struct nvme_debugfs_ctx *ctx = m->private; +struct nvme_debugfs_attr *attr = ctx->attr; +int ret; + +if (attr->seq_ops) + ret =seq_release (inode, file); +else + ret =single_release (inode, file); + +kfree (ctx); +return ret; +} + +static const struct file_operations nvme_debugfs_fops = { + .owner =THIS_MODULE , + .open = nvme_debugfs_open, + .read = seq_read, + .llseek = seq_lseek, + .release = nvme_debugfs_release, +}; + + +static const struct nvme_debugfs_attr nvme_mpath_debugfs_attrs[] = { + {}, +}; + +static const struct nvme_debugfs_attr nvme_ns_debugfs_attrs[] = { + {}, +}; + +static void nvme_debugfs_create_files (struct request_queue *q, +const struct nvme_debugfs_attr *attr,void *data) +{ +if (WARN_ON_ONCE (!q->debugfs_dir)) +return ; + +for (; attr->name; attr++) +debugfs_create_file_aux (attr->name, attr->mode, q->debugfs_dir, + data, (void *)attr, &nvme_debugfs_fops); +} + +void nvme_debugfs_register (struct gendisk *disk) +{ +const struct nvme_debugfs_attr *attr; + +if (nvme_disk_is_ns_head (disk)) + attr = nvme_mpath_debugfs_attrs; +else + attr = nvme_ns_debugfs_attrs; + +nvme_debugfs_create_files (disk->queue, attr, disk->private_data); +}drivers/nvme/host/nvme.h +4
-0@@ -988,6 +988,7 @@ extern const struct attribute_group *nvme_dev_attr_groups[];extern const struct block_device_operations nvme_bdev_ops;void nvme_delete_ctrl_sync (struct nvme_ctrl *ctrl); +struct nvme_ns *nvme_find_path (struct nvme_ns_head *head,unsigned int op_type);static inline int nvme_data_dir (const enum req_op op) {@@ -999,6 +1000,16 @@ static inline int nvme_data_dir(const enum req_op op)return NVME_STAT_OTHER ; } +void nvme_debugfs_register (struct gendisk *disk); +static inline void nvme_debugfs_unregister (struct gendisk *disk) +{
drivers/nvme/host/nvme.h static inline int nvme_data_dir(const enum req_op op) · line 1015
Quoted 10 lines
drivers/nvme/host/nvme.h +7
-0+/* +* Nothing to do for now. When the request queue is unregistered, +* all files under q->debugfs_dir are recursively deleted. +* This is just a placeholder; the compiler will optimize it out. +*/ +} +#ifdef CONFIG_NVME_MULTIPATH static inline bool nvme_ctrl_use_ana (struct nvme_ctrl *ctrl) {
Especially as you already add a debugfs entry, so moving the values from the previous two patches should be easy.
Cheers,
Hannes
Signature · 4 lines
Dr. Hannes Reinecke Kernel Storage Architect [email protected] +49 911 74053 688 SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich
Re: [RFC PATCHv3 5/6] nvme: add generic debugfs support
Quoted 181 lines · Hannes Reinecke
Quoted 180 lines · Nilay Shroff
Add generic infrastructure for creating and managing debugfs files in the NVMe module. This introduces helper APIs that allow NVMe drivers to register and unregister debugfs entries, along with a reusable attribute structure for defining new debugfs files.
The implementation uses
seq_fileinterfaces to safely expose per-NS and per-NS-head statistics, while supporting both simple show callbacks and fullseq_operations.Reviewed-by Hannes Reinecke <[email protected]>Signed-off-by Nilay Shroff <[email protected]>---
drivers/nvme/host/Makefile | 2 +-drivers/nvme/host/debugfs.c| 122 ++++++++++++++++++++++++++++++++++++ drivers/nvme/host/nvme.h | 11 ++++3 files changed, 134 insertions(+), 1 deletion(-)create mode 100644
drivers/nvme/host/debugfs.cdrivers/nvme/host/Makefile +0
-0@@ -10,7 +10,7 @@ obj-$(CONFIG_NVME_FC) += nvme-fc.oobj-$(CONFIG_NVME_TCP) += nvme-tcp.o obj-$(CONFIG_NVME_APPLE) += nvme-apple.o -nvme-core-y += core.o ioctl.o sysfs.o pr.o +nvme-core-y += core.o ioctl.o sysfs.o pr.o debugfs.o nvme-core-$(CONFIG_NVME_VERBOSE_ERRORS) += constants.o nvme-core-$(CONFIG_TRACING) += trace.o nvme-core-$(CONFIG_NVME_MULTIPATH) += multipath.o
drivers/nvme/host/debugfs.c new file +122
-0@@ -0,0 +1,122 @@ +// SPDX-License-Identifier: GPL-2.0 +/* +* Copyright (c) 2025 IBM Corporation +* Nilay Shroff <[email protected]> +*/ + +#include <linux/debugfs.h> +#include <linux/seq_file.h> + +#include "nvme.h" + +struct nvme_debugfs_attr { + const char *name; + umode_t mode; + int (*show)(void *data, struct seq_file *m); + const struct seq_operations *seq_ops; +}; + +struct nvme_debugfs_ctx { + void *data; + struct nvme_debugfs_attr *attr; + int srcu_idx; +}; + +static int nvme_debugfs_show (struct seq_file *m,void *v) +{ + struct nvme_debugfs_ctx *ctx = m->private; + void *data = ctx->data; + struct nvme_debugfs_attr *attr = ctx->attr; + + return attr->show(data, m); +} + +static int nvme_debugfs_open (struct inode *inode,struct file *file) +{ + void *data = inode->i_private; + struct nvme_debugfs_attr *attr = debugfs_get_aux(file); + struct nvme_debugfs_ctx *ctx; + struct seq_file *m; + int ret; + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (WARN_ON_ONCE(!ctx)) + return -ENOMEM; + + ctx->data = data; + ctx->attr = attr; + + if (attr->seq_ops) { + ret = seq_open(file, attr->seq_ops); + if (ret) { + kfree(ctx); + return ret; + } + m = file->private_data; + m->private = ctx; + return ret; + } + + if (WARN_ON_ONCE(!attr->show)) { + kfree(ctx); + return -EPERM; + } + + return single_open(file, nvme_debugfs_show, ctx); +} + +static int nvme_debugfs_release (struct inode *inode,struct file *file) +{ + struct seq_file *m = file->private_data; + struct nvme_debugfs_ctx *ctx = m->private; + struct nvme_debugfs_attr *attr = ctx->attr; + int ret; + + if (attr->seq_ops) + ret = seq_release(inode, file); + else + ret = single_release(inode, file); + + kfree(ctx); + return ret; +} + +static const struct file_operations nvme_debugfs_fops = { + .owner = THIS_MODULE, + .open = nvme_debugfs_open, + .read = seq_read, + .llseek = seq_lseek, + .release = nvme_debugfs_release, +}; + + +static const struct nvme_debugfs_attr nvme_mpath_debugfs_attrs[] = { + {}, +}; + +static const struct nvme_debugfs_attr nvme_ns_debugfs_attrs[] = { + {}, +}; + +static void nvme_debugfs_create_files (struct request_queue *q, + const struct nvme_debugfs_attr *attr, void *data) +{ + if (WARN_ON_ONCE(!q->debugfs_dir)) + return; + + for (; attr->name; attr++) + debugfs_create_file_aux(attr->name, attr->mode, q->debugfs_dir, + data, (void *)attr, &nvme_debugfs_fops); +} + +void nvme_debugfs_register (struct gendisk *disk) +{ + const struct nvme_debugfs_attr *attr; + + if (nvme_disk_is_ns_head(disk)) + attr = nvme_mpath_debugfs_attrs; + else + attr = nvme_ns_debugfs_attrs; + + nvme_debugfs_create_files(disk->queue, attr, disk->private_data); +}drivers/nvme/host/nvme.h +8
-0@@ -988,6 +988,7 @@ extern const struct attribute_group *nvme_dev_attr_groups[]; extern const struct block_device_operations nvme_bdev_ops; void nvme_delete_ctrl_sync(struct nvme_ctrl *ctrl); + struct nvme_ns *nvme_find_path(struct nvme_ns_head *head, unsigned int op_type); static inline int nvme_data_dir(const enum req_op op) {@@ -999,6 +1000,16 @@ static inline int nvme_data_dir(const enum req_op op) return NVME_STAT_OTHER; } +void nvme_debugfs_register(struct gendisk *disk); +static inline void nvme_debugfs_unregister (struct gendisk *disk) +{ + /* +* Nothing to do for now. When the request queue is unregistered, +* all files under q->debugfs_dir are recursively deleted. +* This is just a placeholder; the compiler will optimize it out. +*/
Quoted 9 lines
drivers/nvme/host/nvme.h static inline int nvme_data_dir(const enum req_op op) · line 1015
Quoted 5 lines
drivers/nvme/host/nvme.h +2
-0+} + #ifdef CONFIG_NVME_MULTIPATH static inline bool nvme_ctrl_use_ana(struct nvme_ctrl *ctrl) {Especially as you already add a debugfs entry, so moving the values from the previous two patches should be easy.
Okay will move those two entries under debugfs.
Thanks,
--Nilay
Re: [RFC PATCHv3 2/6] nvme-multipath: add support for adaptive I/O policy
In reply to Hannes ReineckeQuoted 28 lines · Hannes Reinecke
Quoted 17 lines
drivers/nvme/host/multipath.c +17
-0+static bool nvme_mpath_set_current_adaptive_path (struct nvme_ns *ns) +{ + struct nvme_ns_head *head = ns->head; + + if (!head->disk || head->subsys->iopolicy != NVME_IOPOLICY_ADAPTIVE) + return false; + + if (test_and_set_bit(NVME_NS_PATH_STAT, &ns->flags)) + return false; + + blk_queue_flag_set(QUEUE_FLAG_SAME_FORCE, ns->queue); + blk_stat_enable_accounting(ns->queue); + return true; +} + +static bool nvme_mpath_clear_current_adaptive_path (struct nvme_ns *ns) +{drivers/nvme/host/multipath.c static const char *nvme_ana_state_names[] = { · line 503
Quoted 10 lines
drivers/nvme/host/multipath.c +10
-0+ + if (!test_and_clear_bit(NVME_NS_PATH_STAT, &ns->flags)) + return false; + + blk_stat_disable_accounting(ns->queue); + blk_queue_flag_clear(QUEUE_FLAG_SAME_FORCE, ns->queue); + nvme_mpath_reset_current_stat(ns); + return true; +} +
Quoted 7 lines
Hmm. I think this is a misnomer, as for the adaptive path policy we don't really have a '
current_path', and as such we don't clear the 'current_path' setting. Rather we disable the policy altogether here.
Can we reflect that in the function name and name it egnvme_mpath_enable_adaptive_path_policy()
or somesuch?
Yeah you are right, I'll update function name as suggested.
Thanks,
--Nilay
Christoph Hellwig
maintainerRe: [RFC PATCHv3 1/6] block: expose blk_stat_{enable,disable}_accounting() to drivers
In reply to Nilay ShroffQuoted 4 lines · Nilay Shroff
The functions
blk_stat_enable_accounting()andblk_stat_disable_accounting()are currently exported, but their prototypes are only defined in a private header. Move these prototypes into a common header so that block drivers can directly use these APIs.
That's because they are only used by I/O scheduler, which are the only valid users. I don't think drivers should touch this, but I'll stay tuned..
Christoph Hellwig
maintainerRe: [RFC PATCHv3 2/6] nvme-multipath: add support for adaptive I/O policy
In reply to Nilay ShroffQuoted 8 lines · Nilay Shroff
This commit introduces a new I/O policy named "adaptive". Users can configure it by writing "adaptive" to "
/sys/class/nvme-subsystem/nvme-subsystemX/iopolicy"The adaptive policy dynamically distributes I/O based on measured completion latency. The main idea is to calculate latency for each path, derive a weight, and then proportionally forward I/O according to those weights.
This really sounds like a lot of overhead, and really smells of all that horrible old-school FC SAN thinking we've been carefully trying to avoid in nvme.
What's the point here?
Re: [RFC PATCHv3 2/6] nvme-multipath: add support for adaptive I/O policy
Hi Christoph,
Quoted 10 lines · Christoph Hellwig
Quoted 8 lines · Nilay Shroff
This commit introduces a new I/O policy named "adaptive". Users can configure it by writing "adaptive" to "
/sys/class/nvme-subsystem/nvme-subsystemX/iopolicy"The adaptive policy dynamically distributes I/O based on measured completion latency. The main idea is to calculate latency for each path, derive a weight, and then proportionally forward I/O according to those weights.
Quoted 5 lines
This really sounds like a lot of overhead, and really smells of all that horrible old-school FC SAN thinking we've been carefully trying to avoid in nvme.
What's the point here?
Thanks for the feedback!
The primary goal of the adaptive I/O policy is to improve I/O distribution across NVMe paths in cases where the backend latency varies significantly — for example, with asymmetric paths in multipath environments (different fabrics, load imbalance, or transient congestion). So it's mainly focused on the NVMeOF deployment.
The mechanism itself is lightweight. The policy samples I/O completion latency periodically (with configurable sample window and EWMA smoothing) and derives proportional weights per path. The intent is to avoid excessive control overhead while providing better path utilization than simple round-robin or queue-depth based policies in heterogeneous environments.
That said, I’m happy to share some profiling data and latency distributions from test runs to demonstrate that the overhead remains negligible (<1% CPU impact in current measurements) and that throughput improves when path
latencies diverge.
For example, with NVMf/TCP using two paths (one throttled with ~30 ms delay), fio results with random read/write/rw workloads (direct I/O) showed:
numa round-robin queue-depth adaptive
----------- ----------- ----------- ---------
READ: 50.0 MiB/s 105 MiB/s 230 MiB/s 350 MiB/s
WRITE: 65.9 MiB/s 125 MiB/s 385 MiB/s 446 MiB/s
RW: R:30.6 MiB/s R:56.5 MiB/s R:122 MiB/s R:175 MiB/s
W:30.7 MiB/s W:56.5 MiB/s W:122 MiB/s W:175 MiB/sWould that context and data help clarify the motivation?
Furthermore, the goal isn’t to micromanage path scheduling in real time, but to let the host bias I/O slightly toward faster or healthier paths without violating NVMe’s design principles of simplicity and scalability. If this still feels too policy-heavy for the block layer, I’m happy to consider ways to simplify or confine it so that it stays aligned with NVMe’s overall design philosophy.
Thanks,
--Nilay
Re: [RFC PATCHv3 1/6] block: expose blk_stat_{enable,disable}_accounting() to drivers
In reply to Christoph HellwigQuoted 10 lines · Christoph Hellwig
Quoted 4 lines · Nilay Shroff
The functions
blk_stat_enable_accounting()andblk_stat_disable_accounting()are currently exported, but their prototypes are only defined in a private header. Move these prototypes into a common header so that block drivers can directly use these APIs.That's because they are only used by I/O scheduler, which are the only valid users. I don't think drivers should touch this, but I'll stay tuned..
Well, these stat APIs are already exported by the block layer, so I assumed the intention was to allow drivers to make use of them. If that’s not the case, then perhaps they were exported unintentionally.
On a related note, I also looked into using nvme_req(rq)->start_time, which is initialized using jiffies when the request is submitted to the device. However, for my use case, I need timestamp data with nanosecond precision, which is why I opted to use the blk-stat APIs instead.
Thanks,
--Nilay