mirror/eigen
2
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2024-12-27 07:29:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

Get rid of redundant pabs instruction in complex square root.

Browse Source
This commit is contained in:
Rasmus Munk Larsen 2021-06-29 23:26:15 +00:00
parent 3a087ccb99
commit 5aebbe9098
1 changed files with 1 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
Eigen/src/Core/arch/Default/GenericPacketMathFunctions.h
View File

@ -802,9 +802,8 @@ Packet psqrt_complex(const Packet& a) {
// l0 = (min0 == 0 ? max0 : max0 * sqrt(1 + (min0/max0)**2)), // l0 = (min0 == 0 ? max0 : max0 * sqrt(1 + (min0/max0)**2)),
// where max0 = max(|x0|, |y0|), min0 = min(|x0|, |y0|), and similarly for l1. // where max0 = max(|x0|, |y0|), min0 = min(|x0|, |y0|), and similarly for l1.
Packet a_flip = pcplxflip(a);
RealPacket a_abs = pabs(a.v); // [|x0|, |y0|, |x1|, |y1|] RealPacket a_abs = pabs(a.v); // [|x0|, |y0|, |x1|, |y1|]
RealPacket a_abs_flip = pabs(a_flip.v); // [|y0|, |x0|, |y1|, |x1|] RealPacket a_abs_flip = pcplxflip(Packet(a_abs)).v; // [|y0|, |x0|, |y1|, |x1|]
RealPacket a_max = pmax(a_abs, a_abs_flip); RealPacket a_max = pmax(a_abs, a_abs_flip);
RealPacket a_min = pmin(a_abs, a_abs_flip); RealPacket a_min = pmin(a_abs, a_abs_flip);
RealPacket a_min_zero_mask = pcmp_eq(a_min, pzero(a_min)); RealPacket a_min_zero_mask = pcmp_eq(a_min, pzero(a_min));