Hardware:
News ToolsReg Shops |
Comments on ‘Sun slots transactional memory into Rock’Prays for code revolutionPublished Tuesday 21st August 2007 21:21 GMT
Minor bugBy Osvaldo
Posted Wednesday 22nd August 2007 04:29 GMT
"you see an all or nothing scenarios where either all the loads and stores go through or none do." Obviously, that should be just "...either all the stores go through or none do.", since you cannot rollback reads ;-) re: Minor bugBy Anonymous Coward
Posted Wednesday 22nd August 2007 05:03 GMT
One needn't make only the stores transactional. One could conceive of a chip design that allows the whole register file to be checkpointed, making loads and any other instruction transactional. Rolling back the soupBy Chris Croughton
Posted Wednesday 22nd August 2007 08:11 GMT
Thank you for that image... re: re: Minor bugBy Robert Hill
Posted Wednesday 22nd August 2007 10:42 GMT
"One needn't make only the stores transactional. One could conceive of a chip design that allows the whole register file to be checkpointed, making loads and any other instruction transactional." Huh? If by register file you mean the processor's registers, then these are already 1:1 associated with the processor core and thus the thread, and therefore shouldn't have any parallel access issues - so I fail to see the use for making them transactional. Can you explain further? RollbackBy Daniel Ballado-Torres
Posted Wednesday 22nd August 2007 14:53 GMT
If you don't roll back the registers, you could have weird stuff happening with the rolled back code, as it may be dependent on the values such registers had when execution started. If you've played with assembly, you know what I'm talking about ;) That chef analogy got me laughing. I just imagined: "Waiter, the chef rolled back my soup. It's cold!" The period for commenting on this story has finished |
|
Top 20 stories • All The Week’s Headlines • Archive • Search