Revision 6e2df0581f569038719cf2bc2b3baa3fcc83cab4 authored by Peter Zijlstra on 08 November 2019, 10:11:52 UTC, committed by Peter Zijlstra on 08 November 2019, 21:34:14 UTC
Commit 67692435c411 ("sched: Rework pick_next_task() slow-path") inadvertly introduced a race because it changed a previously unexplored dependency between dropping the rq->lock and sched_class::put_prev_task(). The comments about dropping rq->lock, in for example newidle_balance(), only mentions the task being current and ->on_cpu being set. But when we look at the 'change' pattern (in for example sched_setnuma()): queued = task_on_rq_queued(p); /* p->on_rq == TASK_ON_RQ_QUEUED */ running = task_current(rq, p); /* rq->curr == p */ if (queued) dequeue_task(...); if (running) put_prev_task(...); /* change task properties */ if (queued) enqueue_task(...); if (running) set_next_task(...); It becomes obvious that if we do this after put_prev_task() has already been called on @p, things go sideways. This is exactly what the commit in question allows to happen when it does: prev->sched_class->put_prev_task(rq, prev, rf); if (!rq->nr_running) newidle_balance(rq, rf); The newidle_balance() call will drop rq->lock after we've called put_prev_task() and that allows the above 'change' pattern to interleave and mess up the state. Furthermore, it turns out we lost the RT-pull when we put the last DL task. Fix both problems by extracting the balancing from put_prev_task() and doing a multi-class balance() pass before put_prev_task(). Fixes: 67692435c411 ("sched: Rework pick_next_task() slow-path") Reported-by: Quentin Perret <qperret@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Tested-by: Quentin Perret <qperret@google.com> Tested-by: Valentin Schneider <valentin.schneider@arm.com>
1 parent e3b8b6a
quota.c
// SPDX-License-Identifier: GPL-2.0
/*
* quota.c - CephFS quota
*
* Copyright (C) 2017-2018 SUSE
*/
#include <linux/statfs.h>
#include "super.h"
#include "mds_client.h"
void ceph_adjust_quota_realms_count(struct inode *inode, bool inc)
{
struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
if (inc)
atomic64_inc(&mdsc->quotarealms_count);
else
atomic64_dec(&mdsc->quotarealms_count);
}
static inline bool ceph_has_realms_with_quotas(struct inode *inode)
{
struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
struct super_block *sb = mdsc->fsc->sb;
if (atomic64_read(&mdsc->quotarealms_count) > 0)
return true;
/* if root is the real CephFS root, we don't have quota realms */
if (sb->s_root->d_inode &&
(sb->s_root->d_inode->i_ino == CEPH_INO_ROOT))
return false;
/* otherwise, we can't know for sure */
return true;
}
void ceph_handle_quota(struct ceph_mds_client *mdsc,
struct ceph_mds_session *session,
struct ceph_msg *msg)
{
struct super_block *sb = mdsc->fsc->sb;
struct ceph_mds_quota *h = msg->front.iov_base;
struct ceph_vino vino;
struct inode *inode;
struct ceph_inode_info *ci;
if (msg->front.iov_len < sizeof(*h)) {
pr_err("%s corrupt message mds%d len %d\n", __func__,
session->s_mds, (int)msg->front.iov_len);
ceph_msg_dump(msg);
return;
}
/* increment msg sequence number */
mutex_lock(&session->s_mutex);
session->s_seq++;
mutex_unlock(&session->s_mutex);
/* lookup inode */
vino.ino = le64_to_cpu(h->ino);
vino.snap = CEPH_NOSNAP;
inode = ceph_find_inode(sb, vino);
if (!inode) {
pr_warn("Failed to find inode %llu\n", vino.ino);
return;
}
ci = ceph_inode(inode);
spin_lock(&ci->i_ceph_lock);
ci->i_rbytes = le64_to_cpu(h->rbytes);
ci->i_rfiles = le64_to_cpu(h->rfiles);
ci->i_rsubdirs = le64_to_cpu(h->rsubdirs);
__ceph_update_quota(ci, le64_to_cpu(h->max_bytes),
le64_to_cpu(h->max_files));
spin_unlock(&ci->i_ceph_lock);
/* avoid calling iput_final() in dispatch thread */
ceph_async_iput(inode);
}
static struct ceph_quotarealm_inode *
find_quotarealm_inode(struct ceph_mds_client *mdsc, u64 ino)
{
struct ceph_quotarealm_inode *qri = NULL;
struct rb_node **node, *parent = NULL;
mutex_lock(&mdsc->quotarealms_inodes_mutex);
node = &(mdsc->quotarealms_inodes.rb_node);
while (*node) {
parent = *node;
qri = container_of(*node, struct ceph_quotarealm_inode, node);
if (ino < qri->ino)
node = &((*node)->rb_left);
else if (ino > qri->ino)
node = &((*node)->rb_right);
else
break;
}
if (!qri || (qri->ino != ino)) {
/* Not found, create a new one and insert it */
qri = kmalloc(sizeof(*qri), GFP_KERNEL);
if (qri) {
qri->ino = ino;
qri->inode = NULL;
qri->timeout = 0;
mutex_init(&qri->mutex);
rb_link_node(&qri->node, parent, node);
rb_insert_color(&qri->node, &mdsc->quotarealms_inodes);
} else
pr_warn("Failed to alloc quotarealms_inode\n");
}
mutex_unlock(&mdsc->quotarealms_inodes_mutex);
return qri;
}
/*
* This function will try to lookup a realm inode which isn't visible in the
* filesystem mountpoint. A list of these kind of inodes (not visible) is
* maintained in the mdsc and freed only when the filesystem is umounted.
*
* Note that these inodes are kept in this list even if the lookup fails, which
* allows to prevent useless lookup requests.
*/
static struct inode *lookup_quotarealm_inode(struct ceph_mds_client *mdsc,
struct super_block *sb,
struct ceph_snap_realm *realm)
{
struct ceph_quotarealm_inode *qri;
struct inode *in;
qri = find_quotarealm_inode(mdsc, realm->ino);
if (!qri)
return NULL;
mutex_lock(&qri->mutex);
if (qri->inode && ceph_is_any_caps(qri->inode)) {
/* A request has already returned the inode */
mutex_unlock(&qri->mutex);
return qri->inode;
}
/* Check if this inode lookup has failed recently */
if (qri->timeout &&
time_before_eq(jiffies, qri->timeout)) {
mutex_unlock(&qri->mutex);
return NULL;
}
if (qri->inode) {
/* get caps */
int ret = __ceph_do_getattr(qri->inode, NULL,
CEPH_STAT_CAP_INODE, true);
if (ret >= 0)
in = qri->inode;
else
in = ERR_PTR(ret);
} else {
in = ceph_lookup_inode(sb, realm->ino);
}
if (IS_ERR(in)) {
pr_warn("Can't lookup inode %llx (err: %ld)\n",
realm->ino, PTR_ERR(in));
qri->timeout = jiffies + msecs_to_jiffies(60 * 1000); /* XXX */
} else {
qri->timeout = 0;
qri->inode = in;
}
mutex_unlock(&qri->mutex);
return in;
}
void ceph_cleanup_quotarealms_inodes(struct ceph_mds_client *mdsc)
{
struct ceph_quotarealm_inode *qri;
struct rb_node *node;
/*
* It should now be safe to clean quotarealms_inode tree without holding
* mdsc->quotarealms_inodes_mutex...
*/
mutex_lock(&mdsc->quotarealms_inodes_mutex);
while (!RB_EMPTY_ROOT(&mdsc->quotarealms_inodes)) {
node = rb_first(&mdsc->quotarealms_inodes);
qri = rb_entry(node, struct ceph_quotarealm_inode, node);
rb_erase(node, &mdsc->quotarealms_inodes);
iput(qri->inode);
kfree(qri);
}
mutex_unlock(&mdsc->quotarealms_inodes_mutex);
}
/*
* This function walks through the snaprealm for an inode and returns the
* ceph_snap_realm for the first snaprealm that has quotas set (either max_files
* or max_bytes). If the root is reached, return the root ceph_snap_realm
* instead.
*
* Note that the caller is responsible for calling ceph_put_snap_realm() on the
* returned realm.
*
* Callers of this function need to hold mdsc->snap_rwsem. However, if there's
* a need to do an inode lookup, this rwsem will be temporarily dropped. Hence
* the 'retry' argument: if rwsem needs to be dropped and 'retry' is 'false'
* this function will return -EAGAIN; otherwise, the snaprealms walk-through
* will be restarted.
*/
static struct ceph_snap_realm *get_quota_realm(struct ceph_mds_client *mdsc,
struct inode *inode, bool retry)
{
struct ceph_inode_info *ci = NULL;
struct ceph_snap_realm *realm, *next;
struct inode *in;
bool has_quota;
if (ceph_snap(inode) != CEPH_NOSNAP)
return NULL;
restart:
realm = ceph_inode(inode)->i_snap_realm;
if (realm)
ceph_get_snap_realm(mdsc, realm);
else
pr_err_ratelimited("get_quota_realm: ino (%llx.%llx) "
"null i_snap_realm\n", ceph_vinop(inode));
while (realm) {
bool has_inode;
spin_lock(&realm->inodes_with_caps_lock);
has_inode = realm->inode;
in = has_inode ? igrab(realm->inode) : NULL;
spin_unlock(&realm->inodes_with_caps_lock);
if (has_inode && !in)
break;
if (!in) {
up_read(&mdsc->snap_rwsem);
in = lookup_quotarealm_inode(mdsc, inode->i_sb, realm);
down_read(&mdsc->snap_rwsem);
if (IS_ERR_OR_NULL(in))
break;
ceph_put_snap_realm(mdsc, realm);
if (!retry)
return ERR_PTR(-EAGAIN);
goto restart;
}
ci = ceph_inode(in);
has_quota = __ceph_has_any_quota(ci);
/* avoid calling iput_final() while holding mdsc->snap_rwsem */
ceph_async_iput(in);
next = realm->parent;
if (has_quota || !next)
return realm;
ceph_get_snap_realm(mdsc, next);
ceph_put_snap_realm(mdsc, realm);
realm = next;
}
if (realm)
ceph_put_snap_realm(mdsc, realm);
return NULL;
}
bool ceph_quota_is_same_realm(struct inode *old, struct inode *new)
{
struct ceph_mds_client *mdsc = ceph_inode_to_client(old)->mdsc;
struct ceph_snap_realm *old_realm, *new_realm;
bool is_same;
restart:
/*
* We need to lookup 2 quota realms atomically, i.e. with snap_rwsem.
* However, get_quota_realm may drop it temporarily. By setting the
* 'retry' parameter to 'false', we'll get -EAGAIN if the rwsem was
* dropped and we can then restart the whole operation.
*/
down_read(&mdsc->snap_rwsem);
old_realm = get_quota_realm(mdsc, old, true);
new_realm = get_quota_realm(mdsc, new, false);
if (PTR_ERR(new_realm) == -EAGAIN) {
up_read(&mdsc->snap_rwsem);
if (old_realm)
ceph_put_snap_realm(mdsc, old_realm);
goto restart;
}
is_same = (old_realm == new_realm);
up_read(&mdsc->snap_rwsem);
if (old_realm)
ceph_put_snap_realm(mdsc, old_realm);
if (new_realm)
ceph_put_snap_realm(mdsc, new_realm);
return is_same;
}
enum quota_check_op {
QUOTA_CHECK_MAX_FILES_OP, /* check quota max_files limit */
QUOTA_CHECK_MAX_BYTES_OP, /* check quota max_files limit */
QUOTA_CHECK_MAX_BYTES_APPROACHING_OP /* check if quota max_files
limit is approaching */
};
/*
* check_quota_exceeded() will walk up the snaprealm hierarchy and, for each
* realm, it will execute quota check operation defined by the 'op' parameter.
* The snaprealm walk is interrupted if the quota check detects that the quota
* is exceeded or if the root inode is reached.
*/
static bool check_quota_exceeded(struct inode *inode, enum quota_check_op op,
loff_t delta)
{
struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
struct ceph_inode_info *ci;
struct ceph_snap_realm *realm, *next;
struct inode *in;
u64 max, rvalue;
bool exceeded = false;
if (ceph_snap(inode) != CEPH_NOSNAP)
return false;
down_read(&mdsc->snap_rwsem);
restart:
realm = ceph_inode(inode)->i_snap_realm;
if (realm)
ceph_get_snap_realm(mdsc, realm);
else
pr_err_ratelimited("check_quota_exceeded: ino (%llx.%llx) "
"null i_snap_realm\n", ceph_vinop(inode));
while (realm) {
bool has_inode;
spin_lock(&realm->inodes_with_caps_lock);
has_inode = realm->inode;
in = has_inode ? igrab(realm->inode) : NULL;
spin_unlock(&realm->inodes_with_caps_lock);
if (has_inode && !in)
break;
if (!in) {
up_read(&mdsc->snap_rwsem);
in = lookup_quotarealm_inode(mdsc, inode->i_sb, realm);
down_read(&mdsc->snap_rwsem);
if (IS_ERR_OR_NULL(in))
break;
ceph_put_snap_realm(mdsc, realm);
goto restart;
}
ci = ceph_inode(in);
spin_lock(&ci->i_ceph_lock);
if (op == QUOTA_CHECK_MAX_FILES_OP) {
max = ci->i_max_files;
rvalue = ci->i_rfiles + ci->i_rsubdirs;
} else {
max = ci->i_max_bytes;
rvalue = ci->i_rbytes;
}
spin_unlock(&ci->i_ceph_lock);
switch (op) {
case QUOTA_CHECK_MAX_FILES_OP:
exceeded = (max && (rvalue >= max));
break;
case QUOTA_CHECK_MAX_BYTES_OP:
exceeded = (max && (rvalue + delta > max));
break;
case QUOTA_CHECK_MAX_BYTES_APPROACHING_OP:
if (max) {
if (rvalue >= max)
exceeded = true;
else {
/*
* when we're writing more that 1/16th
* of the available space
*/
exceeded =
(((max - rvalue) >> 4) < delta);
}
}
break;
default:
/* Shouldn't happen */
pr_warn("Invalid quota check op (%d)\n", op);
exceeded = true; /* Just break the loop */
}
/* avoid calling iput_final() while holding mdsc->snap_rwsem */
ceph_async_iput(in);
next = realm->parent;
if (exceeded || !next)
break;
ceph_get_snap_realm(mdsc, next);
ceph_put_snap_realm(mdsc, realm);
realm = next;
}
if (realm)
ceph_put_snap_realm(mdsc, realm);
up_read(&mdsc->snap_rwsem);
return exceeded;
}
/*
* ceph_quota_is_max_files_exceeded - check if we can create a new file
* @inode: directory where a new file is being created
*
* This functions returns true is max_files quota allows a new file to be
* created. It is necessary to walk through the snaprealm hierarchy (until the
* FS root) to check all realms with quotas set.
*/
bool ceph_quota_is_max_files_exceeded(struct inode *inode)
{
if (!ceph_has_realms_with_quotas(inode))
return false;
WARN_ON(!S_ISDIR(inode->i_mode));
return check_quota_exceeded(inode, QUOTA_CHECK_MAX_FILES_OP, 0);
}
/*
* ceph_quota_is_max_bytes_exceeded - check if we can write to a file
* @inode: inode being written
* @newsize: new size if write succeeds
*
* This functions returns true is max_bytes quota allows a file size to reach
* @newsize; it returns false otherwise.
*/
bool ceph_quota_is_max_bytes_exceeded(struct inode *inode, loff_t newsize)
{
loff_t size = i_size_read(inode);
if (!ceph_has_realms_with_quotas(inode))
return false;
/* return immediately if we're decreasing file size */
if (newsize <= size)
return false;
return check_quota_exceeded(inode, QUOTA_CHECK_MAX_BYTES_OP, (newsize - size));
}
/*
* ceph_quota_is_max_bytes_approaching - check if we're reaching max_bytes
* @inode: inode being written
* @newsize: new size if write succeeds
*
* This function returns true if the new file size @newsize will be consuming
* more than 1/16th of the available quota space; it returns false otherwise.
*/
bool ceph_quota_is_max_bytes_approaching(struct inode *inode, loff_t newsize)
{
loff_t size = ceph_inode(inode)->i_reported_size;
if (!ceph_has_realms_with_quotas(inode))
return false;
/* return immediately if we're decreasing file size */
if (newsize <= size)
return false;
return check_quota_exceeded(inode, QUOTA_CHECK_MAX_BYTES_APPROACHING_OP,
(newsize - size));
}
/*
* ceph_quota_update_statfs - if root has quota update statfs with quota status
* @fsc: filesystem client instance
* @buf: statfs to update
*
* If the mounted filesystem root has max_bytes quota set, update the filesystem
* statistics with the quota status.
*
* This function returns true if the stats have been updated, false otherwise.
*/
bool ceph_quota_update_statfs(struct ceph_fs_client *fsc, struct kstatfs *buf)
{
struct ceph_mds_client *mdsc = fsc->mdsc;
struct ceph_inode_info *ci;
struct ceph_snap_realm *realm;
struct inode *in;
u64 total = 0, used, free;
bool is_updated = false;
down_read(&mdsc->snap_rwsem);
realm = get_quota_realm(mdsc, d_inode(fsc->sb->s_root), true);
up_read(&mdsc->snap_rwsem);
if (!realm)
return false;
spin_lock(&realm->inodes_with_caps_lock);
in = realm->inode ? igrab(realm->inode) : NULL;
spin_unlock(&realm->inodes_with_caps_lock);
if (in) {
ci = ceph_inode(in);
spin_lock(&ci->i_ceph_lock);
if (ci->i_max_bytes) {
total = ci->i_max_bytes >> CEPH_BLOCK_SHIFT;
used = ci->i_rbytes >> CEPH_BLOCK_SHIFT;
/* It is possible for a quota to be exceeded.
* Report 'zero' in that case
*/
free = total > used ? total - used : 0;
}
spin_unlock(&ci->i_ceph_lock);
if (total) {
buf->f_blocks = total;
buf->f_bfree = free;
buf->f_bavail = free;
is_updated = true;
}
iput(in);
}
ceph_put_snap_realm(mdsc, realm);
return is_updated;
}
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...