Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.compilers    |    Compiler construction, theory, etc. (Mod    |    2,753 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 1,642 of 2,753    |
|    Ken Rose to cyril.cressent@gmail.com    |
|    Re: Guidelines for instruction set desig    |
|    01 May 09 17:30:55    |
      From: rose@acm.org              cyril.cressent@gmail.com wrote:              > I was wondering if there are some general guidelines one should       > observe when designing an instruction set so that a C compiler can       > easily be ported to that CPU.              > [Interesting question. C should be pretty straightforward on anything       > with flat byte addressing and enough registers to handle stack frames.       > What makes this architecture hard? -John]              I don't know anything about the poster's instruction set, but from my       own experience, it's a pain to have to have a scratch register to spill,       which makes a register+offset or register+register addressing mode very       handy.              Be careful that the instruction set supports all the branches that C       calls for. For instance, one machine I've worked on doesn't have an       overflow flag, which makes signed comparisons awkward.               - ken              --- 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