Skip to contentSkip to conversation

ptp: Limit time setting of PTP clocks

PATCH Superseded Miroslav Lichvar 2 participants 3 messages Sep 9, 2024 – Sep 10, 2024

Subsystems (2)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 ptp: Limit time setting of PTP clocks

Networking drivers implementing PTP clocks and kernel socket code handling hardware timestamps use the 64-bit signed ktime_t type counting nanoseconds. When a PTP clock reaches the maximum value in year 2262, the timestamps returned to applications will overflow into year 1667. The same thing happens when injecting a large offset with clock_adjtime(ADJ_SETOFFSET).

The commit 7a8e61f84786 ("timekeeping: Force upper bound for setting CLOCK_REALTIME") limited the maximum accepted value setting the system clock to 30 years before the maximum representable value (i.e. year 2232) to avoid the overflow, assuming the system will not run for more than 30 years.

Enforce the same limit for PTP clocks. Don't allow negative values and values closer than 30 years to the maximum value. Drivers may implement an even lower limit if the hardware registers cannot represent the whole interval between years 1970 and 2262 in the required resolution.

Cc Richard Cochran <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc John Stultz <[email protected]>
Cc Arnd Bergmann <[email protected]>
---
 drivers/ptp/ptp_clock.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)
drivers/ptp/ptp_clock.c +12 -2
index c56cd0f63909..bd7c5f534be6 100644
@@ -100,6 +100,9 @@ static int ptp_clock_settime(struct posix_clock *pc, const struct timespec64 *tp
 		return -EBUSY;
 	}
 
+	if (!timespec64_valid_settod(tp))
+		return -EINVAL;
+
 	return  ptp->info->settime64(ptp->info, tp);
 }
 
@@ -129,7 +132,7 @@ static int ptp_clock_adjtime(struct posix_clock *pc, struct __kernel_timex *tx)
 	ops = ptp->info;
 
 	if (tx->modes & ADJ_SETOFFSET) {
-		struct timespec64 ts;
+		struct timespec64 ts, ts2;
 		ktime_t kt;
 		s64 delta;
 
@@ -139,7 +142,14 @@ static int ptp_clock_adjtime(struct posix_clock *pc, struct __kernel_timex *tx)
 		if (!(tx->modes & ADJ_NANO))
 			ts.tv_nsec *= 1000;
 
-		if ((unsigned long) ts.tv_nsec >= NSEC_PER_SEC)
+		/* Make sure the offset is valid */
+		err = ptp_clock_gettime(pc, &ts2);
+		if (err)
+			return err;
+		ts2 = timespec64_add(ts2, ts);
+
+		if ((unsigned long) ts.tv_nsec >= NSEC_PER_SEC ||
+		    !timespec64_valid_settod(&ts2))
 			return -EINVAL;
 
 		kt = timespec64_to_ktime(ts);
Signature · 1 line
2.46.0
maintainer
#1 lore raw

drivers/ptp/ptp_clock.c static int ptp_clock_settime(struct posix_clock *pc, const struct timespec64 *tp · line 104

Quoted 11 lines
drivers/ptp/ptp_clock.c +2 -0
index c56cd0f63909..bd7c5f534be6 100644
@@ -100,6 +100,9 @@ static int ptp_clock_settime(struct posix_clock *pc, const struct timespec64 *tp
 		return -EBUSY;
 	}
 
+	if (!timespec64_valid_settod(tp))
+		return -EINVAL;

Why not perform the test earlier, in SYSCALL_DEFINE2(clock_settime, ...) ?

Thanks,
Richard

#2 lore raw
Quoted 8 lines · Richard Cochran
Quoted 8 lines
drivers/ptp/ptp_clock.c +0 -0
index c56cd0f63909..bd7c5f534be6 100644
@@ -100,6 +100,9 @@ static int ptp_clock_settime(struct posix_clock *pc, const struct timespec64 *tp
 		return -EBUSY;
 	}
 
Quoted 4 lines

drivers/ptp/ptp_clock.c static int ptp_clock_settime(struct posix_clock *pc, const struct timespec64 *tp · line 104

Quoted 2 lines
drivers/ptp/ptp_clock.c +2 -0
+	if (!timespec64_valid_settod(tp))
+		return -EINVAL;

Why not perform the test earlier, in SYSCALL_DEFINE2(clock_settime, ...) ?

It would be more code as there are also the 32-bit compat variants of the functions. The adjtime function would probably need to handle missing gettime.

The limit would apply to all clocks, not just PTP clocks. If nothing else, I suspect it would change returned errors for other clocks trying to set the an unacceptable timespec value.

Signature · 1 line
Miroslav Lichvar