Page 1 of 1
IO Port Access Times (was: not unique keyboard scancodes)
Posted: Mon Nov 19, 2012 1:30 pm
by bluemoon
SparrowOS wrote:Your time-out delay is hoeky, but works for reasons you probably don't understand. IN/OUT instructions move at ISA bus speed which is roughly 1uS.
My machine has no ISA support and the IO is handled by south-bridge which operates at PCI bus speed.
Furthermore, newer CPU like Sandy Bridge has the south-bridge directly connected to the CPU with quickpath, and I suppose it run at even faster speed.
Anyway, IN/OUT instruction has
no defined speed for modern machine, there are hundred of things can affect the timing.
Stop telling the un-true.
Re: not unique keyboard scancodes
Posted: Mon Nov 19, 2012 1:46 pm
by SparrowOS
bluemoon wrote:SparrowOS wrote:Your time-out delay is hoeky, but works for reasons you probably don't understand. IN/OUT instructions move at ISA bus speed which is roughly 1uS.
My machine has no ISA support and the IO is handled by south-bridge which operates at PCI bus speed.
Furthermore, newer CPU like Sandy Bridge has the south-bridge directly connected to the CPU with quickpath, and I suppose it run at even faster speed.
Anyway, IN/OUT instruction has
no defined speed for modern machine, there are hundred of things can affect the timing.
Stop telling the un-true.
I would love to get a machine where my port I/O for harddrive/CDROM ATA/ATAPI was faster than 1uS.