Re: TCP/IP Checksumming

Alan Cox (
Tue, 26 Nov 1996 21:27:54 +0000 (GMT)

> The performance hit will be the same for all procedures tested. The
> purpose of the test-bench is to perform relative testing. The ratio
> changes less than 1 percent although the machine-cycles are less.

Your test bench is invalid, your assumptions for the pipelines being
the same are broken.

> Looking at my data book for the i486, I see that it MAY be possible
> to get back to real mode from protected mode without a processor reset.

Its trivial. The code is even in the current kernel source tree.