On Friday 02 June 2006 23:34, Jeff Dike wrote:
> On Fri, Jun 02, 2006 at 08:28:37PM +0200, Blaisorblade wrote:
> > Ok, since I now I'll never finish it:
> > $ ll old-patch-scripts/patches/uml-fix-timers.patch
> > -rw-r--r-- 1 paolo paolo 6763 2005-07-24 06:41
> > old-patch-scripts/patches/uml-fix-t
On Fri, Jun 02, 2006 at 08:28:37PM +0200, Blaisorblade wrote:
> Ok, since I now I'll never finish it:
> $ ll old-patch-scripts/patches/uml-fix-timers.patch
> -rw-r--r-- 1 paolo paolo 6763 2005-07-24 06:41
> old-patch-scripts/patches/uml-fix-timers.patch
>
> I'm attaching this incomplete patch. It
On Friday 02 June 2006 17:19, Jeff Dike wrote:
> On Fri, Jun 02, 2006 at 08:54:22AM +0200, Thomas Gleixner wrote:
> > NAK. ktime_t is defined that ist must be normalized the same way as
> > timespecs. The nsec part must be >= 0 and < NSEC_PER_SEC. Fix the part
> > which is feeding non normalized va
On Fri, Jun 02, 2006 at 08:54:22AM +0200, Thomas Gleixner wrote:
> NAK. ktime_t is defined that ist must be normalized the same way as
> timespecs. The nsec part must be >= 0 and < NSEC_PER_SEC. Fix the part
> which is feeding non normalized values.
Aha, that would be me, initializing wall_to_mono
The use of 64-bit additions and subtractions on something which is
nominally a struct containing 32-bit second and nanosecond field is
broken when a negative time is involved. When the structure is
treated as a 64-bit integer, the increment of the upper 32 bits that's
part of two's-complement subt
