peterr@utcsrgv.UUCP (Peter Rowley) (11/06/83)
From the information I have (PC Tech. Ref. Man., Intel 8086/88 Family Man., DOS 1.0, 1.1, and 2.0 manuals) it would appear that it is not the writer of MS-DOS that goofed with respect to the interrupt vectors (MS-DOS itself uses only the user, non-reserved, vectors), but rather the writer of the ROM BIOS routines that made the mistake. No specific authorship of those routines is given in the Tech. Ref. Man., but they clearly do violate the reserved-vector restrictions given in the Intel manual. A related point that I've never been clear on: the DOS manual mentions some- thing to the effect that hidden files called IBMBIO.COM and IBMDOS.COM are loaded into memory at boot time and, among other things, the code in IBMBIO sets some of the interrupt vectors, implying that it is a device driver that takes over from the ROM routines (page B-2, DOS 1.0 manual). Does anyone know what is happening here?