Re: PATCH: `__norestrict' type qualifier

[patl at cag dot lcs dot mit dot edu (Patrick J. LoPresti)]

 torvalds> In all fairness it also gathered a lot of people who didn't
 torvalds> like it, but none on tecnical grounds (several on the
 torvalds> grounds of "don't add extensions to the language").

(I also just finished reading most of the comp.std.c discussion)

To be completely fair, there was at least one reasonable technical
objection to the keyword as an extension to the standard: It would
seem to *require* that implementations allow access to the bits in an
object of one type via a pointer of another type.  This is counter to
the spirit of the standard, which attempts to include architectures
with unusual representations (padding bits and who-knows what else).

The hypothetical standard would probably allow `norestrict' to be
applied only to integral types, and even then with some amount of
implementation-defined behavior (but with clearly specified intent).

Does this have anything to do with the proposed patch?  You tell
me...

 - Pat