Given that we're going down the path of having typed operands, it would seem that new operands (like MOVDB) make sense.
Keith On Mon, 2010-01-04 at 05:18 -0800, Igor Oliveira wrote: > Right, > > i was thinking tgsi supporting DOUBLE types. The first solution that i > was considering is use all operations like double value. Other way is > create Double instructions like movdb ou something like that. what do > you think? > > Igor > > On Mon, Jan 4, 2010 at 8:35 AM, michal <mic...@vmware.com> wrote: > > Igor Oliveira wrote on 2010-01-04 12:49: > >> > >> Hi > >> > >> i was seeing the changes done by gallium-integer-opcodes the mov > >> operation using dst like float type in embedded systems can not be > >> slow? > >> > >> > > > > Igor, > > > > The MOV instruction always interprets the source operand as FLOAT, thus > > source operand modifiers (negation and absolute) work as before. This > > implies you cannot use those modifiers on integer data and instead must use > > INEG and IMAX opcodes. > > ------------------------------------------------------------------------------ This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev _______________________________________________ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
