Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.arch    |    Apparently more than just beeps & boops    |    131,241 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 130,358 of 131,241    |
|    Robert Finch to All    |
|    Re: Tonights Tradeoff - NaN boxed precis    |
|    25 Nov 25 21:08:45    |
      From: robfi680@gmail.com              > In this case, put the cause in a container the instruction drags down       > the pipe, and retrieve it when you do have address access to where it       > needs to go.              I may change things to pass the address around in the float package.       Putting the address into the NaN later may cause issues with timing. It       adds a mux into things. May be better to use the original NaN mux in the       float modules. May call it a NaN identity field instead of an address.              Modified NaN support in the float package to store to the HOBs.              Survey says:              The Qulps PUSH and POP instructions have room for six register fields.       Should one of the fields be used to identify the stack pointer register       allowing five registers to be pushed or popped? Or should the stack       pointer register be assumed so that six registers may be pushed or popped?              I think the SP should be identified as PUSH / POP would be the only       instructions assuming the SP register. Otherwise any register could be       chosen by the compiler.              --- SoupGate-Win32 v1.05        * Origin: you cannot sedate... all the things you hate (1:229/2)    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca