Use SSE intrinsics instead of asm() for vertical SSE, finally offloading the register...

[qscale] / README

diff --git a/README b/README
index 98b88c3f406aba47113484f73520ebdf7cd11d41..be6feb972d34f9ede69615b1eb5059712ebeb6b0 100644 (file)
--- a/README
+++ b/README
@@ -75,9 +75,6 @@ The following optimizations are possible but not done (yet?):
    filter kernel is large (ie. large amounts of scaling).
  - qscale does not use multiple processors or cores (although different cores 
    can of course work on different images at the same time).
- - qscale does not make very good use of the extra eight SSE registers found
-   on 64-bit x86-compatible (usually called amd64 or x86-64) machines. In
-   fact, out of the box it might not even compile on such machines.
 
 Finally, a word of advice: Even the fastest scaler can never be as fast as not
 scaling. If you can do less work, or move it to a time where it is less user