| Juha Nieminen <nospam@thanks.invalid>: May 27 11:20AM > Given that there will not be many threads each individual thread must > run at the highest possible speed. Which requires caches, pipelines, and > all that gubbins, which is where the side channel attacks come in. You seem to be implying that without C we would nowadays have really slow-ass CPUs that might be able to run thousands of simultaneous threads, but run each one at a snail's pace (ie. we would still be in the days when a single register assignment would take at least 3 or 4 clock cycles, with more complex opcodes taking a dozen). If that's really the argument, then perhaps C did some good to CPU design after all. (Not that I believe any of this to be the case, but assuming if.) |
| Vir Campestris <vir.campestris@invalid.invalid>: May 27 09:24PM +0100 On 25/05/2018 22:30, Scott Lurndal wrote: > Vir Campestris <vir.campestris@invalid.invalid> writes: <snip> >> that side channel stuff. > flushing the cache is a side effect of the exception, thus a covert > channel - useful to determine valid kernel addresses, for example. You get an exception when you access a page that you're not allowed to. System flushes the cache. There is no way to determine whether this is an illegal address, or a forbidden address - unless of course the system chooses to tell you. What will you do with the list of valid kernel addresses anyway? It's their contents that are interesting. Andy |
| Vir Campestris <vir.campestris@invalid.invalid>: May 27 09:26PM +0100 On 27/05/2018 12:20, Juha Nieminen wrote: > threads, but run each one at a snail's pace (ie. we would still > be in the days when a single register assignment would take at > least 3 or 4 clock cycles, with more complex opcodes taking a dozen). Without C, yes. Or Fortran, Cobol, Algol... maybe APL would like a properly parallel machine, but I can't think of another language of that vintage that isn't essentially single threaded. > If that's really the argument, then perhaps C did some good to CPU > design after all. (Not that I believe any of this to be the case, > but assuming if.) Andy |
| You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page. To unsubscribe from this group and stop receiving emails from it send an email to comp.lang.c+++unsubscribe@googlegroups.com. |
No comments:
Post a Comment