home bbs files messages ]

Forums before death by AOL, social media and spammers... "We can't have nice things"

   comp.lang.asm.x86      Ahh, the lost art of x86 assembly      4,675 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 3,764 of 4,675   
   Terje Mathisen to All   
   Re: Detect physical memory addressing li   
   07 Jan 19 08:52:36   
   
   From: terje.mathisen@nospicedham.tmsw.no   
      
   JJ wrote:   
   > On Sun, 06 Jan 2019 10:44:31 -0600, Robert Wessel wrote:   
   >>   
   >> I've never used it, but CPUID with EAX=80000008H looks like it   
   >> would work.   
   >>   
   >> I'm not sure if this is universally supported by x86-64 CPUs, but   
   >> the documentation goes back to AMD's and Inte'ls original,   
   >> pre-release, documentation for that mode.   
   >   
   > Thank you. It looks promising.   
      
   CPUID isn't just promising, it is probably the only way you would be   
   able to do this.   
      
   OTOH, if you have actual, physical access to a machine with more than 64   
   TB of RAM, then lucky you!   
      
   If you do then you could just have a tiny piece of test code which tries   
   to allocated (memmap?) a 65TB (or larger) array, with flags to specify   
   that it has to be physical memory that is locked in places. If this   
   fails then you obviously cannot use this much RAM even if the machine   
   contains it.   
      
   >   
   > I might need to find a database of CPUIDs on the net, and check it   
   > manually using a CPU specific emulator.   
      
   I would expect CPUID nooks and crannies to be locations where even   
   cpu-specific emulators have lots of bugs!   
      
   Terje   
      
   --   
   -    
   "almost all programming can be viewed as an exercise in caching"   
      
   --- 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