[Devel] [PATCH RHEL7 COMMIT] ve: Add interface for @start_timespec and @real_start_timespec adjustmen

[Konstantin Khorenko]

The commit is pushed to "branch-rh7-3.10.0-327.3.1-vz7.10.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git
after rh7-3.10.0-327.3.1.vz7.10.10
------>
commit 15b4ab52bd8700f5ab02613f91aeb102f511c20e
Author: Cyrill Gorcunov <gorcunov at virtuozzo.com>
Date:   Thu Feb 11 12:59:24 2016 +0400

    ve: Add interface for @start_timespec and @real_start_timespec adjustmen
    
    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>
---
 kernel/ve/ve.c | 77 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 75 insertions(+), 2 deletions(-)

diff --git a/kernel/ve/ve.c b/kernel/ve/ve.c
index 2408c3e..3b151d5 100644
--- a/kernel/ve/ve.c
+++ b/kernel/ve/ve.c
@@ -439,6 +439,8 @@ static void ve_drop_context(struct ve_struct *ve)
 	ve->init_cred = NULL;
 }
 
+static const struct timespec zero_time = { };
+
 /* under ve->op_sem write-lock */
 int ve_start_container(struct ve_struct *ve)
 {
@@ -454,8 +456,16 @@ int ve_start_container(struct ve_struct *ve)
 	if (tsk->task_ve != ve || !is_child_reaper(task_pid(tsk)))
 		return -ECHILD;
 
-	ve->start_timespec = tsk->start_time;
-	ve->real_start_timespec = tsk->real_start_time;
+	/*
+	 * Setup uptime for new containers only, if restored
+	 * the velue won't be zero here already but setup from
+	 * cgroup write while resuming the container.
+	 */
+	if (timespec_equal(&ve->start_timespec, &zero_time)) {
+		ve->start_timespec = tsk->start_time;
+		ve->real_start_timespec = tsk->real_start_time;
+	}
+
 	/* The value is wrong, but it is never compared to process
 	 * start times */
 	ve->start_jiffies = get_jiffies_64();
@@ -1141,8 +1151,57 @@ enum {
 	VE_CF_FEATURES,
 	VE_CF_IPTABLES_MASK,
 	VE_CF_PSEUDOSUPER,
+	VE_CF_CLOCK_MONOTONIC,
+	VE_CF_CLOCK_BOOTBASED
 };
 
+static int ve_ts_read(struct cgroup *cg, struct cftype *cft, struct seq_file *m)
+{
+	struct ve_struct *ve = cgroup_ve(cg);
+	struct timespec ts, *delta;
+
+	do_posix_clock_monotonic_gettime(&ts);
+	if (cft->private == VE_CF_CLOCK_MONOTONIC) {
+		delta = &ve->start_timespec;
+	} else if (cft->private == VE_CF_CLOCK_BOOTBASED) {
+		delta = &ve->real_start_timespec;
+		monotonic_to_bootbased(&ts);
+	} else {
+		delta = &ts;
+		memset(&ts, 0, sizeof(ts));
+		WARN_ON_ONCE(1);
+	}
+
+	set_normalized_timespec(&ts, ts.tv_sec - delta->tv_sec,
+				ts.tv_nsec - delta->tv_nsec);
+	seq_printf(m, "%ld %ld", ts.tv_sec, ts.tv_nsec);
+	return 0;
+}
+
+static int ve_ts_write(struct cgroup *cg, struct cftype *cft, const char *buffer)
+{
+	struct ve_struct *ve = cgroup_ve(cg);
+	struct timespec ts, delta, *target;
+
+	if (sscanf(buffer, "%ld %ld", &delta.tv_sec, &delta.tv_nsec) != 2)
+		return -EINVAL;
+
+	do_posix_clock_monotonic_gettime(&ts);
+	if (cft->private == VE_CF_CLOCK_MONOTONIC) {
+		target = &ve->start_timespec;
+	} else if (cft->private == VE_CF_CLOCK_BOOTBASED) {
+		target = &ve->real_start_timespec;
+		monotonic_to_bootbased(&ts);
+	} else {
+		WARN_ON_ONCE(1);
+		return -EINVAL;
+	}
+
+	set_normalized_timespec(target, ts.tv_sec - delta.tv_sec,
+				ts.tv_nsec - delta.tv_nsec);
+	return 0;
+}
+
 static u64 ve_read_u64(struct cgroup *cg, struct cftype *cft)
 {
 	if (cft->private == VE_CF_FEATURES)
@@ -1258,6 +1317,20 @@ static struct cftype ve_cftypes[] = {
 		.write_u64		= ve_write_pseudosuper,
 		.private		= VE_CF_PSEUDOSUPER,
 	},
+	{
+		.name			= "clock_monotonic",
+		.flags			= CFTYPE_NOT_ON_ROOT,
+		.read_seq_string	= ve_ts_read,
+		.write_string		= ve_ts_write,
+		.private		= VE_CF_CLOCK_MONOTONIC,
+	},
+	{
+		.name			= "clock_bootbased",
+		.flags			= CFTYPE_NOT_ON_ROOT,
+		.read_seq_string	= ve_ts_read,
+		.write_string		= ve_ts_write,
+		.private		= VE_CF_CLOCK_BOOTBASED,
+	},
 	{ }
 };