Hi!
28-Янв-2004 18:06 [EMAIL PROTECTED] (tom ehlert) wrote to "Arkady V.Belousov"
<[EMAIL PROTECTED]>:
AVB>> As Pat note in email, this is unimportant if A20 will be waited
AVB>> infinitely or waiting code simply returns after some retries - in both cases
AVB>> we get crash when returns in HMA code
te> how about the kernel's
te> DisableA20()
te> and often saying 'A20 can't be disabled - nothing to worry about'
te> would be a problem if it would wait infinitively
I suggest, infinite waiting/diagnosing should be only part of
EnableA20(). Disabling isn't (so) critical.
te> enable/disable requires some OUT's/IN's.
te> each in requires ~1 mikrosecond, regardless on maschine speed.
This depends on bus speed, I think.
AVB>> Again: if there interrupts are enabled (STI),
te> a big IF, and it's NOT a good idea to enable/disable interrupts.
With long (non-critical) sections of code, you _should_ enable
interrupts, or you lost some of them.
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Freedos-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/freedos-devel
