[Bug middle-end/19391] [4.0 Regression] missed optimization with size of 8 vectors

2005-01-11 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-01-12 06:34 --- Note this worked with 20041124. -- What|Removed |Added Target Milestone|---

[Bug middle-end/19391] [4.0 Regression] missed optimization with size of 8 vectors

2005-01-12 Thread steven at gcc dot gnu dot org
--- Additional Comments From steven at gcc dot gnu dot org 2005-01-12 09:23 --- Is this a regression that may be caused by RTH's rewrite? -- What|Removed |Added

[Bug middle-end/19391] [4.0 Regression] missed optimization with size of 8 vectors

2005-01-12 Thread rth at gcc dot gnu dot org
--- Additional Comments From rth at gcc dot gnu dot org 2005-01-12 09:54 --- Try again with functions. It is absolutely ESSENTIAL that we do NOT emit mmx vector operations UNLESS the routines are used. Doing so without the compiler also being able to insert emms instructions is wron

[Bug middle-end/19391] [4.0 Regression] missed optimization with size of 8 vectors

2005-01-12 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-01-12 15:31 --- (In reply to comment #3) > Try again with functions. > > It is absolutely ESSENTIAL that we do NOT emit mmx vector operations UNLESS > the routines are used. Doing so without the compiler also > being a