On Thu, Nov 7, 2013 at 12:53 AM, William ML Leslie < [email protected]> wrote:
> On 3 November 2013 01:28, Jonathan S. Shapiro <[email protected]> wrote: > > Addendum about the M bit. > > > > To see why object headers want to live with their objects, consider what > > happens to the fast-path store barrier on the Pentium > > Could you explain this a bit more? What are we doing with the mark > bit of the referrer when we store into it? It's not the mark bit we're worried about, so much as the bit 'M' that indicates the object has been logged. That gets checked at every write barrier, so you don't want to have to do a computation to find it.
_______________________________________________ bitc-dev mailing list [email protected] http://www.coyotos.org/mailman/listinfo/bitc-dev
