commit | 682393c4b1995c209e2cf71780b0fb9023150213 | [log] [tgz] |
---|---|---|
author | Roland Levillain <rpl@google.com> | Tue Apr 14 15:57:52 2015 +0100 |
committer | Roland Levillain <rpl@google.com> | Wed Jun 03 12:37:13 2015 +0100 |
tree | 48ec13295adcabdd8de203ce6081034e19e12ee2 | |
parent | 6a0d5e7fe6dc0c9d3dd941ab991203f2d5d1c354 [diff] |
Improve the performance of long-to-double conversions on ARM. Use a VMLA instruction instead of VADD & VMUL instructions in long-to-double conversions on ARM. This change reduces code size and improves execution times (but does not alter precision). It trades one temporary FPU register for two temporary core registers. Change-Id: I1dc35bef6c12be8f305e5b46da98c2421686b60d