Hello! In article <slrn9d6n13.vlg.kc5tja at garnet.armored.net>, Samuel A. Falvo II <kc5tja at armored.net> wrote: >On 10 Apr 2001 16:08:24 +0200, Hannah Schroeter wrote: >[...] >>Segmentation is coarse grained and cannot be used to alleviate fine-grained >>double indirections. >Why not? Simply telling me that "it's not possible" isn't enough. I, >however, have an overwhelming amount of evidence that it *can* be used for >this very purpose, especially since Intel segments can be byte-sized >granular for objects less than 1MB in size. In addition, the PC/GEOS >operating system used segmentation to a capacity very close to this. Should >I send them an e-mail telling them to please stop, as it's clearly not >possible? You *can* use segments of small size, however not too many, IIRC. As, IIRC, the x86 segment selector registers are of a size of 16 bits, of which 3 bits are reserved (2 bits mark the privilege level, 1 bit marks whether the segment is in the LDT or the GDT). Usually the OS grabs the GDT for itself, so you have just the use of 2^13 LDT segments. If you want to be able to move around single objects without adjusting pointers, you are limited to max. 2^13 living objects. Kind regards, Hannah.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4