Re: POST(Power On Self Test)

"Paul G. Tobey [eMVP]" <p space tobey no spam AT no instrument no
spam DOT com> wrote in news:eJ5WaMBJJHA.468@xxxxxxxxxxxxxxxxxxxx:

It's pretty hardware-dependent, as I said, generally. You might
need to know how much RAM there is or actually change RAM control
registers to find out. Whipping up something to step through a
given size range of addresses and write a pattern to all the
locations, then read back through and make sure they're all
correct shouldn't take more than five minutes to write. You could
do that right after entry to your bootloader.

unsigned char *p = 0x00000000;
unsigned long i;
for ( i = 0; i < 0x10000000; i++ )
*(p+i) = <pattern>;
if ( *(p+i) != <pattern> )
// You have a RAM failure. Indicate somehow (flash an LED

It's better to test with two patterns (0xAA,0x55 for example) to check
that each bit can change its state.
And maybe it's also better to do all the writes and after that the
reads (ensuring also that you disabled the cache).

Valter Minute
Training, support and development for Windows CE
(the reply address of this message is invalid)

Relevant Pages

  • Re: POST(Power On Self Test)
    ... You could fill in anything for <pattern>, including shifting a 1 bit through ... need to know how much RAM there is or actually change RAM control ... unsigned char *p = 0x00000000; ...
  • Re: 32-bit programs on Windows x64
    ... processes running that if you really ... need 4GB of physical memory available to hold your ... My system requires 4 GB to store its DFA pixel pattern ... I think that if enough RAM is available you are supposed to ...
  • Re: Tech - Eight Ball display weirdness
    ... gibberish, especially if it seems to follow a pattern, sounds more ... from the RAM. ... If I start a game, the gibberish kind of cures itself for awhile, then ...
  • Re: Ideas for a pulse programmer needed
    ... < system, implemented on spartan 3 starter kit board, a pulse programmer ... < digital lines for a given time and then changes the pattern according ... < block load's the contents of the first ram into a counter and latches ... the AC in incremented and the next time and pattern words are ...
  • Re: Emulator RAM init differences - encountered another problem withanold Apple program
    ... emulators would want to consider it. ... The initial RAM pattern is a deficiency in just about all Apple II ...