So that is wikipedias "opinion" which any one could edit. I should edit that to reflect where the performance came from. Always check sources. ( The reference comes from intel386.
com)
The cited website went offline this morning, and it does not discuss the execution rate of the 286, NOR does it discuss the reserved bytes in the GDT, and LDTs.
" For those descriptors that are common to both the 80286 and the 80386, the presence of zeros in the final word causes the 80386 to interpret these descriptors exactly as 80286 does; for example: "
This is patently and documented as untrue. The final word is reserved for use as a 365 GDT, and breaks 286 code, specifically Xenix 286, on a 386.
