# Contents of /trunk/blas/ddot.f

Revision 1 - (show annotations) (download)
Fri Oct 29 20:54:12 2004 UTC (17 years, 9 months ago) by aw0a
File size: 1388 byte(s)
```Setting up web subdirectory in repository
```
 1 C ddot.f 2 C is freely available from netlib. It is not subject to any GNU License 3 C set by the authors of the ASCEND math programming system. 4 C \$Date: 1996/04/30 18:14:07 \$ \$Revision: 1.1.1.1 \$ 5 C 6 double precision function ddot(n,dx,incx,dy,incy) 7 c 8 c forms the dot product of two vectors. 9 c uses unrolled loops for increments equal to one. 10 c jack dongarra, linpack, 3/11/78. 11 c 12 double precision dx(*),dy(*),dtemp 13 integer i,incx,incy,ix,iy,m,mp1,n 14 c 15 ddot = 0.0d0 16 dtemp = 0.0d0 17 if(n.le.0)return 18 if(incx.eq.1.and.incy.eq.1)go to 20 19 c 20 c code for unequal increments or equal increments 21 c not equal to 1 22 c 23 ix = 1 24 iy = 1 25 if(incx.lt.0)ix = (-n+1)*incx + 1 26 if(incy.lt.0)iy = (-n+1)*incy + 1 27 do 10 i = 1,n 28 dtemp = dtemp + dx(ix)*dy(iy) 29 ix = ix + incx 30 iy = iy + incy 31 10 continue 32 ddot = dtemp 33 return 34 c 35 c code for both increments equal to 1 36 c 37 c 38 c clean-up loop 39 c 40 20 m = mod(n,5) 41 if( m .eq. 0 ) go to 40 42 do 30 i = 1,m 43 dtemp = dtemp + dx(i)*dy(i) 44 30 continue 45 if( n .lt. 5 ) go to 60 46 40 mp1 = m + 1 47 do 50 i = mp1,n,5 48 dtemp = dtemp + dx(i)*dy(i) + dx(i + 1)*dy(i + 1) + 49 * dx(i + 2)*dy(i + 2) + dx(i + 3)*dy(i + 3) + dx(i + 4)*dy(i + 4) 50 50 continue 51 60 ddot = dtemp 52 return 53 end 54 55

 john.pye@anu.edu.au ViewVC Help Powered by ViewVC 1.1.22