On Thu, Mar 2, 2017 at 11:50 AM, Jonathan Scott <[email protected] > wrote:
> Ref: Your note of 2 March 2017, 11:17:08 -0600 > > I think the clue is in the corresponding restriction for both > forms of Prefetch Data when the code is 6 or 7. STCMH with a > mask of zero clearly doesn't have any functional use, but I > presume it was used as an interim form of Prefetch Data which > allowed new machines to use it while remaining a no-operation on > machines which did not. This is supported by the fact that Google > finds "STCMH" mentioned in a Share presentation "What's new in > z/OS performance" from 2011. If it uses the same conventions as > Prefetch Data, then codes 6 and 7 would cause a cache line to be > explicitly released from exclusive control, which sounds very > likely to be incompatible with transactional execution. > Ah, that makes sense. Thank you for the enlightenment. For some reason I've become just a tad curious about Transaction Execution. Conceptually, I understand it. It is basically a way to do a series of instructions "atomically". I guess the hardware people got a bit tired of extending the PLO instruction. > > Jonathan Scott > IBM Hursley, UK > > -- "Irrigation of the land with seawater desalinated by fusion power is ancient. It's called 'rain'." -- Michael McClary, in alt.fusion Maranatha! <>< John McKown
