Patch to implement rotates by zero

Maciej W. Rozycki
Tue Aug 20 10:23:00 GMT 2002

On 20 Aug 2002, Eric Christopher wrote:

> >  Well, the behaviour is defined explicitly for "sll" and "sllv" by "MIPS
> > R4000 Microprocessor User's Manual" (pp. A-140, A-141), the original spec
> > for the original MIPS III processor, so the assumption should be any MIPS
> > III+ implementation must do them this way; otherwise it is non-compliant
> > (or simply broken).  There is a note on broken code generated by some
> > assemblers there, though.  The book is available from 
> That's a pretty big assumption though :) I think I like it the other
> way.

 Why should we "punish" good implementations?  If there are any bad ones,
they can be handled with CPU-specific workarounds. 

+  Maciej W. Rozycki, Technical University of Gdansk, Poland   +
+        e-mail:, PGP key available        +

More information about the Binutils mailing list