On Fri, 18 Feb 2005, Jeff Garzik wrote:
> 
> Speaking of which, on the code generation side, I was wondering if 
> generating PIC code would require any changes in sparse.

Hmm. Yes and no.

I don't think there is anything really fundamentally wrong with just
considering PIC to be a pure code generation function, and in that case
no, none of the current code should care (since none of the current code
does any code generation nor should be making any real assumptions about
it either, except for the example thing).

On the other hand, maybe you would want CSE to be aware of base registers
etc (I don't know if that is actually true, though), in which case you'd
want to teach the linearizer to output PIC accesses as explicitly based on
a new pseudo.

I have absolutely zero idea what real compilers do.

                        Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-sparse" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to