[Devel] [PATCH RHEL COMMIT] ve: Add interface for ve::clock_[monotonic|bootbased] adjustment

[Konstantin Khorenko]

The commit is pushed to "branch-rh9-5.14.vz9.1.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git
after ark-5.14
------>
commit e38514e2dacba61d6736325b26786cf49fe31eff
Author: Cyrill Gorcunov <gorcunov at virtuozzo.com>
Date:   Fri Oct 1 19:38:40 2021 +0300

    ve: Add interface for ve::clock_[monotonic|bootbased] adjustment
    
    This two members represent monotonic and bootbased clocks for
    container's uptime. When container is in suspended state (or
    moving to another node) we trest monotonic and bootbased
    clocks as being stopped so we need to account delta time
    on restore and adjust the members in subject.
    
    Moreover this timestamps are involved into posix-timers
    setup so once application tries to setup monotonic clocks
    after the restore (with absolute time specification) we
    adjust the values as well.
    
    The application which migrate a container must fetch
    the current settings from /sys/fs/cgroup/ve/$VE/ve.real_start_timespec
    and /sys/fs/cgroup/ve/$VE/ve.start_timespec, then write them
    back on the restore.
    
    https://jira.sw.ru/browse/PSBM-41311
    https://jira.sw.ru/browse/PSBM-41406
    
    v2:
     - use clock_[monotonic|bootbased] for cgroup entry names instead
    
    Original-by: Andrew Vagin <avagin at openvz.org>
    Signed-off-by: Cyrill Gorcunov <gorcunov at virtuozzo.com>
    
    Reviewed-by: Vladimir Davydov <vdavydov at virtuozzo.com>
    
    (cherry picked from vz7 commit 43f4b0c752abd84aa1b346373d152941123d2446
    ("ve: Add interface for @start_timespec and @real_start_timespec
    adjustmen"))
    
    Signed-off-by: Konstantin Khorenko <khorenko at virtuozzo.com>
    
    +++
    ve/time: Limit values to write in ve::clock_[monotonic|bootbased]
    
    What do we mean when write a valie XXX into, say, ve::ve.clock_bootbased?
    We mean that "up to now the CT worked for XXX secs/usecs already".
    And we store the delta between Node "now" and XXX into ve->start_time_real.
    
    If the CT worked less than the current Node, ve->start_time_real will
    contain positive value and we'll substitute it from Node's "now" each
    time when we need to get the time since the CT start.
    
    If the CT worked longer than the current CT (say, CT has been migrated
    from another HN), the stored delta will be negative and thus we'll "add"
    more time for Node's "now".
    
    So then what do we want to limit?
    1. Negative values written to ve::clock_[monotonic|bootbased].
       Indeed we can hardly imagine that the CT has been started, but the
       time since it's start is negative.
    
    2. A big positive value, so some time later when we read from
       ve::clock_[monotonic|bootbased] we get an overflowed value.
    
    Both these checks are performed by timespec_valid_strict().
    
    mFixes: 25cab3041305 ("ve: Add interface for
    ve::clock_[monotonic|bootbased] adjustment")
    
    Signed-off-by: Konstantin Khorenko <khorenko at virtuozzo.com>
    
    Reviewed-by: Kirill Tkhai <ktkhai at virtuozzo.com>
    
    Cherry-picked from vz8 commit ad5d9cc5fd62 ("ve: Add interface for
    ve::clock_[monotonic|bootbased] adjustment")).
    
    Ported to timespec64.
    Followed ve->real_start_time -> ve->start_boottime rename.
    Followed ktime_get_boot_ns() -> ktime_get_boottime_ns() rename.
    
    Signed-off-by: Nikita Yushchenko <nikita.yushchenko at virtuozzo.com>
---
 kernel/ve/ve.c | 76 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 76 insertions(+)

diff --git a/kernel/ve/ve.c b/kernel/ve/ve.c
index e3a07d4c9fe4..f3df12f8638b 100644
--- a/kernel/ve/ve.c
+++ b/kernel/ve/ve.c
@@ -955,6 +955,68 @@ static ssize_t ve_os_release_write(struct kernfs_open_file *of, char *buf,
 	return ret ? ret : nbytes;
 }
 
+enum {
+	VE_CF_CLOCK_MONOTONIC,
+	VE_CF_CLOCK_BOOTBASED,
+};
+
+static int ve_ts_read(struct seq_file *sf, void *v)
+{
+	struct ve_struct *ve = css_to_ve(seq_css(sf));
+	struct timespec64 ts;
+	u64 now, delta;
+
+	switch (seq_cft(sf)->private) {
+		case VE_CF_CLOCK_MONOTONIC:
+			now = ktime_get_ns();
+			delta = ve->start_time;
+			break;
+		case VE_CF_CLOCK_BOOTBASED:
+			now = ktime_get_boottime_ns();
+			delta = ve->start_boottime;
+			break;
+		default:
+			now = delta = 0;
+			WARN_ON_ONCE(1);
+			break;
+	}
+
+	ts = ns_to_timespec64(now - delta);
+	seq_printf(sf, "%lld %ld", ts.tv_sec, ts.tv_nsec);
+	return 0;
+}
+
+static ssize_t ve_ts_write(struct kernfs_open_file *of, char *buf,
+			   size_t nbytes, loff_t off)
+{
+	struct ve_struct *ve = css_to_ve(of_css(of));
+	struct timespec64 delta;
+	u64 delta_ns, now, *target;
+
+	if (sscanf(buf, "%lld %ld", &delta.tv_sec, &delta.tv_nsec) != 2)
+		return -EINVAL;
+	if (!timespec64_valid_strict(&delta))
+		return -EINVAL;
+	delta_ns = timespec64_to_ns(&delta);
+
+	switch (of_cft(of)->private) {
+		case VE_CF_CLOCK_MONOTONIC:
+			now = ktime_get_ns();
+			target = &ve->start_time;
+			break;
+		case VE_CF_CLOCK_BOOTBASED:
+			now = ktime_get_boottime_ns();
+			target = &ve->start_boottime;
+			break;
+		default:
+			WARN_ON_ONCE(1);
+			return -EINVAL;
+	}
+
+	*target = now - delta_ns;
+	return nbytes;
+}
+
 static struct cftype ve_cftypes[] = {
 
 	{
@@ -981,6 +1043,20 @@ static struct cftype ve_cftypes[] = {
 		.read_u64		= ve_reatures_read,
 		.write_u64		= ve_reatures_write,
 	},
+	{
+		.name			= "clock_monotonic",
+		.flags			= CFTYPE_NOT_ON_ROOT,
+		.seq_show		= ve_ts_read,
+		.write			= ve_ts_write,
+		.private		= VE_CF_CLOCK_MONOTONIC,
+	},
+	{
+		.name			= "clock_bootbased",
+		.flags			= CFTYPE_NOT_ON_ROOT,
+		.seq_show		= ve_ts_read,
+		.write			= ve_ts_write,
+		.private		= VE_CF_CLOCK_BOOTBASED,
+	},
 	{
 		.name			= "netns_max_nr",
 		.flags			= CFTYPE_NOT_ON_ROOT,