Hi,
ron minnich rminnich@gmail.com wrote:
I read both, all parts, from 2004 to now.
Muting this discussion. Peter, your comments are not helpful. Sorry I bothered you, folks. Have a nice day.
I'm probably doing something wrong by continuing that discussion, but that kind of things seem really interesting.
Sadly the kind of fixes mentioned before in this thread seems to be a kind of black magic: I guess that only the vendors know why they're really needed and what they really do.
_start in src/cpu/x86/16bit/entry16.inc has also some interesting comments( including one on the cache invalidation).
I came accross all that with my first attempts of my newer fallback mecanism work (the idea is to make "what's after the ramstage" declare that the computer booted successfuly).
Denis.