Fundamental defect of the concept of shared libraries

[Mike My Watch Has Windows Meyer]

In article <197 at titccy.cc.titech.ac.jp> mohta at necom830.cc.titech.ac.jp (Masataka Ohta) writes:
   >There may actually not be any "right" implementations extant at the
   >moment (this is debatable), but that's not the point.

   Without any fact, your claim is nothing.

Neither is yours.

   If we take 1), the hardware architecture must support PC relative jump,
   of course. Moreover, to access library private data, it must also
   address data PC relative. Aside from effeciency, not all architechture
   support this.

   Even worse, with some architechture, it is impossible to map several virtual
   addresses to a physical address. Virtually tagged cache and inverted
   page tables are notable examples.

So some architechtures can't support shared libraries? Well, don't put
shared libraries on them. Some architechtures can't support demand
paged memory, or virtual address spaces, or preemptive scheduling.
Does that mean we have to live without them on machines that can
support them?  No; it doesn't.

   I hope you can now understand how complex the shared library is.

No, I understand that you aren't qualified to do systems design work.
Using your logic, I can show that you can't do any of the things I
mentioned above "correctly".  They are still usefull in lots of
places.  The solution is not to "just not do them;" the solution is to
understand them and the various implementations, and to know the
tradeoffs involved in using those implementatins, and use them where
it's appropriate.

	<mike
--
Kiss me with your mouth.				Mike Meyer
Your love is better than wine.				mwm at pa.dec.com
But wine is all I have.					decwrl!mwm
Will your love ever be mine?