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 129,872 of 131,241    |
|    John Savard to All    |
|    Re: Linus Torvalds on bad architectural     |
|    09 Oct 25 21:41:03    |
      From: quadibloc@invalid.invalid              On Fri, 03 Oct 2025 08:58:32 +0000, Anton Ertl quoted:       > |If somebody really wants to create bad hardware in this day and age,       > |please do make it big-endian, and also add the following very       > |traditional features for sh*t-for-brains hardware:              I think that for a computer to be big-endian is a good thing.              It makes it easier to understand core dumps, as numbers are stored just as       they are written.              But more importantly, it means that binary integers are ordered the same       way as packed decimal integers, which are ordered the same way as integers       in character text form.              As for the _rest_ of the items, though, all of them are indeed bad things.              But some are worse than others.              > | - only do aligned memory accesses              Nearly all memory access are, or could be, aligned. Performance is       improved if they are. As long as there's some provision to handle       unaligned data, such as a move characters instruction, data structures can       be dealt with for things like communications formats.       I'm not saying it isn't bad, just that it was excusable before we had as       many transistors available as we do now.              > | - expose your pipeline details in the ISA              The original MIPS did this. This is bad indeed, as whatever you do in this       direction won't be applicable to later iterations of the ISA as technology       advances.              Failing to support the entire IEEE 754 floating-point standard just needs       to be documented. Expecting software to fake it being implemented is not       reasonable: as long as denormals instead produce zero as the result, one       just has an inferior floating-point format, not a computer that doesn't       work.       Once again, bad, but not all that terrible.              But anything that means that programs could randomly fail because       interrupts don't properly save or restore the entire machine state...       *that* is catastrophically bad, and hardly compares to his other examples.              John Savard              --- 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