Nobody has commented on the potential performance hit of the proposed extended storage key lookup. As described, it looks to me like it would add a few cycles to storage accesses which are currently (I assume) 1 cycle.
Also... On Thu, Sep 29, 2005 at 09:28:03AM -0600, Anne & Lynn Wheeler ([EMAIL PROTECTED]) wrote: [snip] > Infrastructures that grew up with multiple virtual address spaces were > heavily message passing based ... rather than pointer passing paradigm > (which implied access in the same address space) [snip] Maybe that's why those other architectures require so much more compute power to accomplish the same workload? /Leonard ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
