[PATCH] kerberos: Fix expiry time conversion.

[Dmitry Timoshkov]

This got broken during transition to unixlib interface.

Signed-off-by: Dmitry Timoshkov 
---
 dlls/kerberos/unixlib.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/dlls/kerberos/unixlib.c b/dlls/kerberos/unixlib.c
index a735a1c6a47..c66557efcca 100644
--- a/dlls/kerberos/unixlib.c
+++ b/dlls/kerberos/unixlib.c
@@ -477,6 +477,7 @@ static void expirytime_gss_to_sspi( OM_uint32 expirytime, TimeStamp *timestamp )

NtQuerySystemTime( &time );
     RtlSystemTimeToLocalTime( &time, &time );
+    time.QuadPart += expirytime;
     timestamp->LowPart  = time.QuadPart;
     timestamp->HighPart = time.QuadPart >> 32;
 }

[Alexandre Julliard]

Dmitry Timoshkov  writes:
> @@ -477,6 +477,7 @@ static void expirytime_gss_to_sspi( OM_uint32 expirytime, TimeStamp *timestamp )
>  NtQuerySystemTime( &time );
>  RtlSystemTimeToLocalTime( &time, &time );
> - time.QuadPart += expirytime;

Isn't expirytime supposed to be in seconds?

[Dmitry Timoshkov]

Alexandre Julliard  wrote:
> Dmitry Timoshkov  writes:
> > @@ -477,6 +477,7 @@ static void expirytime_gss_to_sspi( OM_uint32 expirytime, TimeStamp *timestamp )
> >  NtQuerySystemTime( &time );
> >  RtlSystemTimeToLocalTime( &time, &time );
> > - time.QuadPart += expirytime;
>
> Isn't expirytime supposed to be in seconds?

Good catch, thanks, that's correct, however that's how it was in old code where
I copied it from. Would you mind adding a multiplier before committing? Or should
I resend the patch?

[Alexandre Julliard]

Dmitry Timoshkov  writes:
> Alexandre Julliard  wrote:
> > Dmitry Timoshkov  writes:
> > > @@ -477,6 +477,7 @@ static void expirytime_gss_to_sspi( OM_uint32 expirytime, TimeStamp *timestamp )
> > >  NtQuerySystemTime( &time );
> > >  RtlSystemTimeToLocalTime( &time, &time );
> > > - time.QuadPart += expirytime;
> >
> > Isn't expirytime supposed to be in seconds?
>
> Good catch, thanks, that's correct, however that's how it was in old code where
> I copied it from. Would you mind adding a multiplier before committing? Or should
> I resend the patch?

I'm reworking that code anyway, so I'll fix it.