On Wed, Jan 24, 2024, at 19:02, Elizabeth Figura wrote:

On Wednesday, 24 January 2024 01:56:52 CST Arnd Bergmann wrote:

quoted

On Wed, Jan 24, 2024, at 01:40, Elizabeth Figura wrote:

quoted

+	if (args->timeout) {
+		struct timespec64 to;
+
+		if (get_timespec64(&to, u64_to_user_ptr(args->timeout)))
+			return -EFAULT;
+		if (!timespec64_valid(&to))
+			return -EINVAL;
+
+		timeout = timespec64_to_ns(&to);
+	}

Have you considered just passing the nanosecond value here?
Since you do not appear to write it back, that would avoid
the complexities of dealing with timespec layout differences
and indirection.

That'd be nicer in general. I think there was some documentation that advised
using timespec64 for new ioctl interfaces but it may have been outdated or
misread.

It's probably something I wrote. It depends a bit on
whether you have an absolute or relative timeout. If
the timeout is relative to the current time as I understand
it is here, a 64-bit number seems more logical to me.

For absolute times, I would usually use a __kernel_timespec,
especially if it's CLOCK_REALTIME. In this case you would
also need to specify the time domain.

      Arnd