Re: [Qemu-devel] [PATCH] qemu-ga-win: Fail loudly on bare 'set-time'

[Michael Roth <mdroth@linux.vnet.ibm.com>]

Quoting Michal Privoznik (2015-01-21 05:09:50)
> The command is not implemented correctly yet. The documentation allows
> to not pass any value to set, in which case the time is re-read from
> RTC. However, reading CMOS on Windows is not trivial to implement. So
> instead of pretending we've set the correct time, fail explicitly.
> 
> Signed-off-by: Michal Privoznik <mprivozn@redhat.com>

Thanks, applied to QGA tree:

https://github.com/mdroth/qemu/commits/qga

> ---
>  qga/commands-win32.c | 44 ++++++++++++++++++++++----------------------
>  qga/qapi-schema.json |  5 ++++-
>  2 files changed, 26 insertions(+), 23 deletions(-)
> 
> diff --git a/qga/commands-win32.c b/qga/commands-win32.c
> index 3bcbeae..4fe45f0 100644
> --- a/qga/commands-win32.c
> +++ b/qga/commands-win32.c
> @@ -395,31 +395,31 @@ void qmp_guest_set_time(bool has_time, int64_t time_ns, Error **errp)
>      FILETIME tf;
>      LONGLONG time;
> 
> -    if (has_time) {
> -        /* Okay, user passed a time to set. Validate it. */
> -        if (time_ns < 0 || time_ns / 100 > INT64_MAX - W32_FT_OFFSET) {
> -            error_setg(errp, "Time %" PRId64 "is invalid", time_ns);
> -            return;
> -        }
> +    if (!has_time) {
> +        /* Unfortunately, Windows libraries don't provide an easy way to access
> +         * RTC yet:
> +         *
> +         * https://msdn.microsoft.com/en-us/library/aa908981.aspx
> +         */
> +        error_setg(errp, "Time argument is required on this platform");
> +        return;
> +    }
> 
> -        time = time_ns / 100 + W32_FT_OFFSET;
> +    /* Validate time passed by user. */
> +    if (time_ns < 0 || time_ns / 100 > INT64_MAX - W32_FT_OFFSET) {
> +        error_setg(errp, "Time %" PRId64 "is invalid", time_ns);
> +        return;
> +    }
> 
> -        tf.dwLowDateTime = (DWORD) time;
> -        tf.dwHighDateTime = (DWORD) (time >> 32);
> +    time = time_ns / 100 + W32_FT_OFFSET;
> 
> -        if (!FileTimeToSystemTime(&tf, &ts)) {
> -            error_setg(errp, "Failed to convert system time %d",
> -                       (int)GetLastError());
> -            return;
> -        }
> -    } else {
> -        /* Otherwise read the time from RTC which contains the correct value.
> -         * Hopefully. */
> -        GetSystemTime(&ts);
> -        if (ts.wYear < 1601 || ts.wYear > 30827) {
> -            error_setg(errp, "Failed to get time");
> -            return;
> -        }
> +    tf.dwLowDateTime = (DWORD) time;
> +    tf.dwHighDateTime = (DWORD) (time >> 32);
> +
> +    if (!FileTimeToSystemTime(&tf, &ts)) {
> +        error_setg(errp, "Failed to convert system time %d",
> +                   (int)GetLastError());
> +        return;
>      }
> 
>      acquire_privilege(SE_SYSTEMTIME_NAME, &local_err);
> diff --git a/qga/qapi-schema.json b/qga/qapi-schema.json
> index 376e79f..ce73dd3 100644
> --- a/qga/qapi-schema.json
> +++ b/qga/qapi-schema.json
> @@ -121,7 +121,10 @@
>  # given value, then sets the Hardware Clock (RTC) to the
>  # current System Time. This will make it easier for a guest
>  # to resynchronize without waiting for NTP. If no @time is
> -# specified, then the time to set is read from RTC.
> +# specified, then the time to set is read from RTC. However,
> +# this may not be supported on all platforms (i.e. Windows).
> +# If that's the case users are advised to always pass a
> +# value.
>  #
>  # @time: #optional time of nanoseconds, relative to the Epoch
>  #        of 1970-01-01 in UTC.
> -- 
> 2.0.5