Commit 87d9ad02 authored by Jens Axboe's avatar Jens Axboe

Merge tag 'nvme-5.13-2021-04-22' of git://git.infradead.org/nvme into for-5.13/drivers

Pull NVMe updates from Christoph:

"- add support for a per-namespace character device (Minwoo Im)
 - various KATO fixes and cleanups (Hou Pu, Hannes Reinecke)
 - APST fix and cleanup"

* tag 'nvme-5.13-2021-04-22' of git://git.infradead.org/nvme:
  nvme: introduce generic per-namespace chardev
  nvme: cleanup nvme_configure_apst
  nvme: do not try to reconfigure APST when the controller is not live
  nvme: add 'kato' sysfs attribute
  nvme: sanitize KATO setting
  nvmet: avoid queuing keep-alive timer if it is disabled
parents f4be591f 2637baed
......@@ -89,6 +89,10 @@ static dev_t nvme_ctrl_base_chr_devt;
static struct class *nvme_class;
static struct class *nvme_subsys_class;
static DEFINE_IDA(nvme_ns_chr_minor_ida);
static dev_t nvme_ns_chr_devt;
static struct class *nvme_ns_chr_class;
static void nvme_put_subsystem(struct nvme_subsystem *subsys);
static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl,
unsigned nsid);
......@@ -1109,6 +1113,17 @@ void nvme_execute_passthru_rq(struct request *rq)
}
EXPORT_SYMBOL_NS_GPL(nvme_execute_passthru_rq, NVME_TARGET_PASSTHRU);
/*
* Recommended frequency for KATO commands per NVMe 1.4 section 7.12.1:
*
* The host should send Keep Alive commands at half of the Keep Alive Timeout
* accounting for transport roundtrip times [..].
*/
static void nvme_queue_keep_alive_work(struct nvme_ctrl *ctrl)
{
queue_delayed_work(nvme_wq, &ctrl->ka_work, ctrl->kato * HZ / 2);
}
static void nvme_keep_alive_end_io(struct request *rq, blk_status_t status)
{
struct nvme_ctrl *ctrl = rq->end_io_data;
......@@ -1131,7 +1146,7 @@ static void nvme_keep_alive_end_io(struct request *rq, blk_status_t status)
startka = true;
spin_unlock_irqrestore(&ctrl->lock, flags);
if (startka)
queue_delayed_work(nvme_wq, &ctrl->ka_work, ctrl->kato * HZ);
nvme_queue_keep_alive_work(ctrl);
}
static int nvme_keep_alive(struct nvme_ctrl *ctrl)
......@@ -1161,7 +1176,7 @@ static void nvme_keep_alive_work(struct work_struct *work)
dev_dbg(ctrl->device,
"reschedule traffic based keep-alive timer\n");
ctrl->comp_seen = false;
queue_delayed_work(nvme_wq, &ctrl->ka_work, ctrl->kato * HZ);
nvme_queue_keep_alive_work(ctrl);
return;
}
......@@ -1178,7 +1193,7 @@ static void nvme_start_keep_alive(struct nvme_ctrl *ctrl)
if (unlikely(ctrl->kato == 0))
return;
queue_delayed_work(nvme_wq, &ctrl->ka_work, ctrl->kato * HZ);
nvme_queue_keep_alive_work(ctrl);
}
void nvme_stop_keep_alive(struct nvme_ctrl *ctrl)
......@@ -2170,28 +2185,28 @@ static int nvme_configure_acre(struct nvme_ctrl *ctrl)
return ret;
}
static int nvme_configure_apst(struct nvme_ctrl *ctrl)
{
/*
* APST (Autonomous Power State Transition) lets us program a
* table of power state transitions that the controller will
* perform automatically. We configure it with a simple
* heuristic: we are willing to spend at most 2% of the time
* transitioning between power states. Therefore, when running
* in any given state, we will enter the next lower-power
* non-operational state after waiting 50 * (enlat + exlat)
* microseconds, as long as that state's exit latency is under
* the requested maximum latency.
/*
* APST (Autonomous Power State Transition) lets us program a table of power
* state transitions that the controller will perform automatically.
* We configure it with a simple heuristic: we are willing to spend at most 2%
* of the time transitioning between power states. Therefore, when running in
* any given state, we will enter the next lower-power non-operational state
* after waiting 50 * (enlat + exlat) microseconds, as long as that state's exit
* latency is under the requested maximum latency.
*
* We will not autonomously enter any non-operational state for
* which the total latency exceeds ps_max_latency_us. Users
* can set ps_max_latency_us to zero to turn off APST.
* We will not autonomously enter any non-operational state for which the total
* latency exceeds ps_max_latency_us.
*
* Users can set ps_max_latency_us to zero to turn off APST.
*/
unsigned apste;
static int nvme_configure_apst(struct nvme_ctrl *ctrl)
{
struct nvme_feat_auto_pst *table;
unsigned apste = 0;
u64 max_lat_us = 0;
__le64 target = 0;
int max_ps = -1;
int state;
int ret;
/*
......@@ -2212,17 +2227,15 @@ static int nvme_configure_apst(struct nvme_ctrl *ctrl)
if (!ctrl->apst_enabled || ctrl->ps_max_latency_us == 0) {
/* Turn off APST. */
apste = 0;
dev_dbg(ctrl->device, "APST disabled\n");
} else {
__le64 target = cpu_to_le64(0);
int state;
goto done;
}
/*
* Walk through all states from lowest- to highest-power.
* According to the spec, lower-numbered states use more
* power. NPSS, despite the name, is the index of the
* lowest-power state, not the number of states.
* According to the spec, lower-numbered states use more power. NPSS,
* despite the name, is the index of the lowest-power state, not the
* number of states.
*/
for (state = (int)ctrl->npss; state >= 0; state--) {
u64 total_latency_us, exit_latency_us, transition_ms;
......@@ -2231,64 +2244,55 @@ static int nvme_configure_apst(struct nvme_ctrl *ctrl)
table->entries[state] = target;
/*
* Don't allow transitions to the deepest state
* if it's quirked off.
* Don't allow transitions to the deepest state if it's quirked
* off.
*/
if (state == ctrl->npss &&
(ctrl->quirks & NVME_QUIRK_NO_DEEPEST_PS))
continue;
/*
* Is this state a useful non-operational state for
* higher-power states to autonomously transition to?
* Is this state a useful non-operational state for higher-power
* states to autonomously transition to?
*/
if (!(ctrl->psd[state].flags &
NVME_PS_FLAGS_NON_OP_STATE))
if (!(ctrl->psd[state].flags & NVME_PS_FLAGS_NON_OP_STATE))
continue;
exit_latency_us =
(u64)le32_to_cpu(ctrl->psd[state].exit_lat);
exit_latency_us = (u64)le32_to_cpu(ctrl->psd[state].exit_lat);
if (exit_latency_us > ctrl->ps_max_latency_us)
continue;
total_latency_us =
exit_latency_us +
total_latency_us = exit_latency_us +
le32_to_cpu(ctrl->psd[state].entry_lat);
/*
* This state is good. Use it as the APST idle
* target for higher power states.
* This state is good. Use it as the APST idle target for
* higher power states.
*/
transition_ms = total_latency_us + 19;
do_div(transition_ms, 20);
if (transition_ms > (1 << 24) - 1)
transition_ms = (1 << 24) - 1;
target = cpu_to_le64((state << 3) |
(transition_ms << 8));
target = cpu_to_le64((state << 3) | (transition_ms << 8));
if (max_ps == -1)
max_ps = state;
if (total_latency_us > max_lat_us)
max_lat_us = total_latency_us;
}
apste = 1;
if (max_ps == -1) {
if (max_ps == -1)
dev_dbg(ctrl->device, "APST enabled but no non-operational states are available\n");
} else {
else
dev_dbg(ctrl->device, "APST enabled: max PS = %d, max round-trip latency = %lluus, table = %*phN\n",
max_ps, max_lat_us, (int)sizeof(*table), table);
}
}
apste = 1;
done:
ret = nvme_set_features(ctrl, NVME_FEAT_AUTO_PST, apste,
table, sizeof(*table), NULL);
if (ret)
dev_err(ctrl->device, "failed to set APST feature (%d)\n", ret);
kfree(table);
return ret;
}
......@@ -2310,6 +2314,7 @@ static void nvme_set_latency_tolerance(struct device *dev, s32 val)
if (ctrl->ps_max_latency_us != latency) {
ctrl->ps_max_latency_us = latency;
if (ctrl->state == NVME_CTRL_LIVE)
nvme_configure_apst(ctrl);
}
}
......@@ -3161,6 +3166,7 @@ nvme_show_int_function(cntlid);
nvme_show_int_function(numa_node);
nvme_show_int_function(queue_count);
nvme_show_int_function(sqsize);
nvme_show_int_function(kato);
static ssize_t nvme_sysfs_delete(struct device *dev,
struct device_attribute *attr, const char *buf,
......@@ -3358,6 +3364,7 @@ static struct attribute *nvme_dev_attrs[] = {
&dev_attr_ctrl_loss_tmo.attr,
&dev_attr_reconnect_delay.attr,
&dev_attr_fast_io_fail_tmo.attr,
&dev_attr_kato.attr,
NULL
};
......@@ -3426,6 +3433,66 @@ static int __nvme_check_ids(struct nvme_subsystem *subsys,
return 0;
}
void nvme_cdev_del(struct cdev *cdev, struct device *cdev_device)
{
cdev_device_del(cdev, cdev_device);
ida_simple_remove(&nvme_ns_chr_minor_ida, MINOR(cdev_device->devt));
}
int nvme_cdev_add(struct cdev *cdev, struct device *cdev_device,
const struct file_operations *fops, struct module *owner)
{
int minor, ret;
minor = ida_simple_get(&nvme_ns_chr_minor_ida, 0, 0, GFP_KERNEL);
if (minor < 0)
return minor;
cdev_device->devt = MKDEV(MAJOR(nvme_ns_chr_devt), minor);
cdev_device->class = nvme_ns_chr_class;
device_initialize(cdev_device);
cdev_init(cdev, fops);
cdev->owner = owner;
ret = cdev_device_add(cdev, cdev_device);
if (ret)
ida_simple_remove(&nvme_ns_chr_minor_ida, minor);
return ret;
}
static int nvme_ns_chr_open(struct inode *inode, struct file *file)
{
return nvme_ns_open(container_of(inode->i_cdev, struct nvme_ns, cdev));
}
static int nvme_ns_chr_release(struct inode *inode, struct file *file)
{
nvme_ns_release(container_of(inode->i_cdev, struct nvme_ns, cdev));
return 0;
}
static const struct file_operations nvme_ns_chr_fops = {
.owner = THIS_MODULE,
.open = nvme_ns_chr_open,
.release = nvme_ns_chr_release,
.unlocked_ioctl = nvme_ns_chr_ioctl,
.compat_ioctl = compat_ptr_ioctl,
};
static int nvme_add_ns_cdev(struct nvme_ns *ns)
{
int ret;
ns->cdev_device.parent = ns->ctrl->device;
ret = dev_set_name(&ns->cdev_device, "ng%dn%d",
ns->ctrl->instance, ns->head->instance);
if (ret)
return ret;
ret = nvme_cdev_add(&ns->cdev, &ns->cdev_device, &nvme_ns_chr_fops,
ns->ctrl->ops->module);
if (ret)
kfree_const(ns->cdev_device.kobj.name);
return ret;
}
static struct nvme_ns_head *nvme_alloc_ns_head(struct nvme_ctrl *ctrl,
unsigned nsid, struct nvme_ns_ids *ids)
{
......@@ -3627,6 +3694,8 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid,
nvme_get_ctrl(ctrl);
device_add_disk(ctrl->device, ns->disk, nvme_ns_id_attr_groups);
if (!nvme_ns_head_multipath(ns->head))
nvme_add_ns_cdev(ns);
nvme_mpath_add_disk(ns, id);
nvme_fault_inject_init(&ns->fault_inject, ns->disk->disk_name);
......@@ -3671,6 +3740,8 @@ static void nvme_ns_remove(struct nvme_ns *ns)
synchronize_srcu(&ns->head->srcu); /* wait for concurrent submissions */
if (ns->disk->flags & GENHD_FL_UP) {
if (!nvme_ns_head_multipath(ns->head))
nvme_cdev_del(&ns->cdev, &ns->cdev_device);
del_gendisk(ns->disk);
blk_cleanup_queue(ns->queue);
if (blk_get_integrity(ns->disk))
......@@ -4461,8 +4532,24 @@ static int __init nvme_core_init(void)
result = PTR_ERR(nvme_subsys_class);
goto destroy_class;
}
result = alloc_chrdev_region(&nvme_ns_chr_devt, 0, NVME_MINORS,
"nvme-generic");
if (result < 0)
goto destroy_subsys_class;
nvme_ns_chr_class = class_create(THIS_MODULE, "nvme-generic");
if (IS_ERR(nvme_ns_chr_class)) {
result = PTR_ERR(nvme_ns_chr_class);
goto unregister_generic_ns;
}
return 0;
unregister_generic_ns:
unregister_chrdev_region(nvme_ns_chr_devt, NVME_MINORS);
destroy_subsys_class:
class_destroy(nvme_subsys_class);
destroy_class:
class_destroy(nvme_class);
unregister_chrdev:
......@@ -4479,12 +4566,15 @@ static int __init nvme_core_init(void)
static void __exit nvme_core_exit(void)
{
class_destroy(nvme_ns_chr_class);
class_destroy(nvme_subsys_class);
class_destroy(nvme_class);
unregister_chrdev_region(nvme_ns_chr_devt, NVME_MINORS);
unregister_chrdev_region(nvme_ctrl_base_chr_devt, NVME_MINORS);
destroy_workqueue(nvme_delete_wq);
destroy_workqueue(nvme_reset_wq);
destroy_workqueue(nvme_wq);
ida_destroy(&nvme_ns_chr_minor_ida);
ida_destroy(&nvme_instance_ida);
}
......
......@@ -379,10 +379,8 @@ int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl)
/*
* Set keep-alive timeout in seconds granularity (ms * 1000)
* and add a grace period for controller kato enforcement
*/
cmd.connect.kato = ctrl->kato ?
cpu_to_le32((ctrl->kato + NVME_KATO_GRACE) * 1000) : 0;
cmd.connect.kato = cpu_to_le32(ctrl->kato * 1000);
if (ctrl->opts->disable_sqflow)
cmd.connect.cattr |= NVME_CONNECT_DISABLE_SQFLOW;
......
......@@ -346,15 +346,27 @@ static int nvme_ns_ioctl(struct nvme_ns *ns, unsigned int cmd,
}
}
static int __nvme_ioctl(struct nvme_ns *ns, unsigned int cmd, void __user *arg)
{
if (is_ctrl_ioctl(cmd))
return nvme_ctrl_ioctl(ns->ctrl, cmd, arg);
return nvme_ns_ioctl(ns, cmd, arg);
}
int nvme_ioctl(struct block_device *bdev, fmode_t mode,
unsigned int cmd, unsigned long arg)
{
struct nvme_ns *ns = bdev->bd_disk->private_data;
void __user *argp = (void __user *)arg;
if (is_ctrl_ioctl(cmd))
return nvme_ctrl_ioctl(ns->ctrl, cmd, argp);
return nvme_ns_ioctl(ns, cmd, argp);
return __nvme_ioctl(ns, cmd, (void __user *)arg);
}
long nvme_ns_chr_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
struct nvme_ns *ns =
container_of(file_inode(file)->i_cdev, struct nvme_ns, cdev);
return __nvme_ioctl(ns, cmd, (void __user *)arg);
}
#ifdef CONFIG_NVME_MULTIPATH
......@@ -388,10 +400,24 @@ int nvme_ns_head_ioctl(struct block_device *bdev, fmode_t mode,
unsigned int cmd, unsigned long arg)
{
struct nvme_ns_head *head = bdev->bd_disk->private_data;
void __user *argp = (void __user *)arg;
if (is_ctrl_ioctl(cmd))
return nvme_ns_head_ctrl_ioctl(head, cmd, argp);
return nvme_ns_head_ns_ioctl(head, cmd, argp);
}
long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd,
unsigned long arg)
{
struct cdev *cdev = file_inode(file)->i_cdev;
struct nvme_ns_head *head =
container_of(cdev, struct nvme_ns_head, cdev);
void __user *argp = (void __user *)arg;
if (is_ctrl_ioctl(cmd))
return nvme_ns_head_ctrl_ioctl(head, cmd, (void __user *)arg);
return nvme_ns_head_ns_ioctl(head, cmd, (void __user *)arg);
return nvme_ns_head_ctrl_ioctl(head, cmd, argp);
return nvme_ns_head_ns_ioctl(head, cmd, argp);
}
#endif /* CONFIG_NVME_MULTIPATH */
......
......@@ -357,6 +357,48 @@ const struct block_device_operations nvme_ns_head_ops = {
.pr_ops = &nvme_pr_ops,
};
static inline struct nvme_ns_head *cdev_to_ns_head(struct cdev *cdev)
{
return container_of(cdev, struct nvme_ns_head, cdev);
}
static int nvme_ns_head_chr_open(struct inode *inode, struct file *file)
{
if (!nvme_tryget_ns_head(cdev_to_ns_head(inode->i_cdev)))
return -ENXIO;
return 0;
}
static int nvme_ns_head_chr_release(struct inode *inode, struct file *file)
{
nvme_put_ns_head(cdev_to_ns_head(inode->i_cdev));
return 0;
}
static const struct file_operations nvme_ns_head_chr_fops = {
.owner = THIS_MODULE,
.open = nvme_ns_head_chr_open,
.release = nvme_ns_head_chr_release,
.unlocked_ioctl = nvme_ns_head_chr_ioctl,
.compat_ioctl = compat_ptr_ioctl,
};
static int nvme_add_ns_head_cdev(struct nvme_ns_head *head)
{
int ret;
head->cdev_device.parent = &head->subsys->dev;
ret = dev_set_name(&head->cdev_device, "ng%dn%d",
head->subsys->instance, head->instance);
if (ret)
return ret;
ret = nvme_cdev_add(&head->cdev, &head->cdev_device,
&nvme_ns_head_chr_fops, THIS_MODULE);
if (ret)
kfree_const(head->cdev_device.kobj.name);
return ret;
}
static void nvme_requeue_work(struct work_struct *work)
{
struct nvme_ns_head *head =
......@@ -435,9 +477,11 @@ static void nvme_mpath_set_live(struct nvme_ns *ns)
if (!head->disk)
return;
if (!test_and_set_bit(NVME_NSHEAD_DISK_LIVE, &head->flags))
if (!test_and_set_bit(NVME_NSHEAD_DISK_LIVE, &head->flags)) {
device_add_disk(&head->subsys->dev, head->disk,
nvme_ns_id_attr_groups);
nvme_add_ns_head_cdev(head);
}
mutex_lock(&head->lock);
if (nvme_path_is_optimized(ns)) {
......@@ -714,8 +758,10 @@ void nvme_mpath_remove_disk(struct nvme_ns_head *head)
{
if (!head->disk)
return;
if (head->disk->flags & GENHD_FL_UP)
if (head->disk->flags & GENHD_FL_UP) {
nvme_cdev_del(&head->cdev, &head->cdev_device);
del_gendisk(head->disk);
}
blk_set_queue_dying(head->disk->queue);
/* make sure all pending bios are cleaned up */
kblockd_schedule_work(&head->requeue_work);
......@@ -785,4 +831,3 @@ void nvme_mpath_uninit(struct nvme_ctrl *ctrl)
kfree(ctrl->ana_log_buf);
ctrl->ana_log_buf = NULL;
}
......@@ -27,7 +27,6 @@ extern unsigned int admin_timeout;
#define NVME_ADMIN_TIMEOUT (admin_timeout * HZ)
#define NVME_DEFAULT_KATO 5
#define NVME_KATO_GRACE 10
#ifdef CONFIG_ARCH_NO_SG_CHAIN
#define NVME_INLINE_SG_CNT 0
......@@ -413,6 +412,10 @@ struct nvme_ns_head {
bool shared;
int instance;
struct nvme_effects_log *effects;
struct cdev cdev;
struct device cdev_device;
struct gendisk *disk;
#ifdef CONFIG_NVME_MULTIPATH
struct bio_list requeue_list;
......@@ -465,6 +468,9 @@ struct nvme_ns {
#define NVME_NS_ANA_PENDING 2
#define NVME_NS_FORCE_RO 3
struct cdev cdev;
struct device cdev_device;
struct nvme_fault_inject fault_inject;
};
......@@ -659,10 +665,16 @@ void nvme_put_ns_from_disk(struct nvme_ns_head *head, int idx);
bool nvme_tryget_ns_head(struct nvme_ns_head *head);
void nvme_put_ns_head(struct nvme_ns_head *head);
struct nvme_ctrl *nvme_find_get_live_ctrl(struct nvme_subsystem *subsys);
int nvme_cdev_add(struct cdev *cdev, struct device *cdev_device,
const struct file_operations *fops, struct module *owner);
void nvme_cdev_del(struct cdev *cdev, struct device *cdev_device);
int nvme_ioctl(struct block_device *bdev, fmode_t mode,
unsigned int cmd, unsigned long arg);
long nvme_ns_chr_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
int nvme_ns_head_ioctl(struct block_device *bdev, fmode_t mode,
unsigned int cmd, unsigned long arg);
long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd,
unsigned long arg);
long nvme_dev_ioctl(struct file *file, unsigned int cmd,
unsigned long arg);
int nvme_getgeo(struct block_device *bdev, struct hd_geometry *geo);
......
......@@ -919,15 +919,21 @@ void nvmet_execute_async_event(struct nvmet_req *req)
void nvmet_execute_keep_alive(struct nvmet_req *req)
{
struct nvmet_ctrl *ctrl = req->sq->ctrl;
u16 status = 0;
if (!nvmet_check_transfer_len(req, 0))
return;
if (!ctrl->kato) {
status = NVME_SC_KA_TIMEOUT_INVALID;
goto out;
}
pr_debug("ctrl %d update keep-alive timer for %d secs\n",
ctrl->cntlid, ctrl->kato);
mod_delayed_work(system_wq, &ctrl->ka_work, ctrl->kato * HZ);
nvmet_req_complete(req, 0);
out:
nvmet_req_complete(req, status);
}
u16 nvmet_parse_admin_cmd(struct nvmet_req *req)
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment