time: Always make sure wall_to_monotonic isn't positive

PATCH In review Harsh Shandilya 2 participants 2 messages Nov 26, 2017 – Nov 27, 2017

Subsystems (2)

Stable branch Timekeeping, clocksource core, NTP, alarmtimer

lore

Share & tools

lore

Harsh Shandilya

author

2017-11-26 07:25 UTC #0 lore raw

PATCH 1/1 time: Always make sure wall_to_monotonic isn't positive

From: Wang YanQing <[email protected]>

commit e1d7ba8735551ed79c7a0463a042353574b96da3 upstream.

Two issues were found on an IMX6 development board without an enabled RTC device(resulting in the boot time and monotonic time being initialized to 0).

Issue 1:exportfs -a generate:
"exportfs: /opt/nfs/arm does not support NFS export"
Issue 2:cat /proc/stat:

       "btime 4294967236"

The same issues can be reproduced on x86 after running the following code:

	int main(void)
	{
	    struct timeval val;
	    int ret;

	    val.tv_sec = 0;
	    val.tv_usec = 0;
	    ret = settimeofday(&val, NULL);
	    return 0;
	}

Two issues are different symptoms of same problem:
The reason is a positive wall_to_monotonic pushes boot time back to the time before Epoch, and getboottime will return negative value.

In symptom 1:
negative boot time cause get_expiry() to overflow time_t when input expire time is 2147483647, then cache_flush() always clears entries just added in ip_map_parse.
In symptom 2:
show_stat() uses "unsigned long" to print negative btime value returned by getboottime.

This patch fix the problem by prohibiting time from being set to a value which would cause a negative boot time. As a result one can't set the CLOCK_REALTIME time prior to (1970 + system uptime).

Change-Id I19acf5df5cc34dd388de0dc633723fe73adc077e

Cc Prarit Bhargava <[email protected]>

Cc Richard Cochran <[email protected]>

Cc Ingo Molnar <[email protected]>

Cc Thomas Gleixner <[email protected]>

Signed-off-by Wang YanQing <[email protected]>

[jstultz: reworded commit message]

[msfjarvis: Backport to 3.18 as we are missing the do_settimeofday64 function the upstream commit patches, so we apply the changes to do_settimeofday]

Signed-off-by John Stultz <[email protected]>

Signed-off-by Harsh Shandilya <[email protected]>

---
 kernel/time/timekeeping.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

kernel/time/timekeeping.c +10 -3

index 07b1c31898bd..caece7c8e44c 100644
@@ -713,6 +713,7 @@ int do_settimeofday(const struct timespec *tv)
 	struct timekeeper *tk = &tk_core.timekeeper;
 	struct timespec64 ts_delta, xt, tmp;
 	unsigned long flags;
+	int ret = 0;
 
 	if (!timespec_valid_strict(tv))
 		return -EINVAL;

@@ -726,11 +727,16 @@ int do_settimeofday(const struct timespec *tv)
 	ts_delta.tv_sec = tv->tv_sec - xt.tv_sec;
 	ts_delta.tv_nsec = tv->tv_nsec - xt.tv_nsec;
 
+	if (timespec64_compare(&tk->wall_to_monotonic, &ts_delta) > 0) {
+		ret = -EINVAL;
+		goto out;
+	}
+
 	tk_set_wall_to_mono(tk, timespec64_sub(tk->wall_to_monotonic, ts_delta));
 
 	tmp = timespec_to_timespec64(*tv);
 	tk_set_xtime(tk, &tmp);
-
+out:
 	timekeeping_update(tk, TK_CLEAR_NTP | TK_MIRROR | TK_CLOCK_WAS_SET);
 
 	write_seqcount_end(&tk_core.seq);

@@ -739,7 +745,7 @@ int do_settimeofday(const struct timespec *tv)
 	/* signal hrtimers about time change */
 	clock_was_set();
 
-	return 0;
+	return ret;
 }
 EXPORT_SYMBOL(do_settimeofday);

@@ -768,7 +774,8 @@ int timekeeping_inject_offset(struct timespec *ts)
 
 	/* Make sure the proposed value is valid */
 	tmp = timespec64_add(tk_xtime(tk),  ts64);
-	if (!timespec64_valid_strict(&tmp)) {
+	if (timespec64_compare(&tk->wall_to_monotonic, &ts64) > 0 ||
+		!timespec64_valid_strict(&tmp)) {
 		ret = -EINVAL;
 		goto error;
 	}

Signature · 1 line

2.15.0.631.g7ddcec0

Greg KH

2017-11-27 16:41 UTC #1 lore raw

Quoted 50 lines · Harsh Shandilya

From: Wang YanQing <[email protected]>
commit e1d7ba8735551ed79c7a0463a042353574b96da3 upstream.
Two issues were found on an IMX6 development board without an enabled RTC device(resulting in the boot time and monotonic time being initialized to 0).
Issue 1:exportfs -a generate:
"exportfs: /opt/nfs/arm does not support NFS export"
Issue 2:cat /proc/stat:
       "btime 4294967236"
The same issues can be reproduced on x86 after running the following code:
	int main(void)
	{
	    struct timeval val;
	    int ret;

	    val.tv_sec = 0;
	    val.tv_usec = 0;
	    ret = settimeofday(&val, NULL);
	    return 0;
	}
Two issues are different symptoms of same problem:
The reason is a positive wall_to_monotonic pushes boot time back to the time before Epoch, and getboottime will return negative value.
In symptom 1:
negative boot time cause get_expiry() to overflow time_t when input expire time is 2147483647, then cache_flush() always clears entries just added in ip_map_parse.
In symptom 2:
show_stat() uses "unsigned long" to print negative btime value returned by getboottime.
This patch fix the problem by prohibiting time from being set to a value which would cause a negative boot time. As a result one can't set the CLOCK_REALTIME time prior to (1970 + system uptime).
Change-Id I19acf5df5cc34dd388de0dc633723fe73adc077e
Cc Prarit Bhargava <[email protected]>
Cc Richard Cochran <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
Signed-off-by Wang YanQing <[email protected]>
[jstultz: reworded commit message]

Quoted 8 lines

[msfjarvis: Backport to 3.18 as we are missing the do_settimeofday64 function the upstream commit patches, so we apply the changes to do_settimeofday]
Signed-off-by John Stultz <[email protected]>
Signed-off-by Harsh Shandilya <[email protected]>
---
 kernel/time/timekeeping.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

Thanks for this, now queued up.

greg k-h