source: NEMO/trunk/src/OCE/do_loop_substitute.h90

Last change on this file was 13296, checked in by acc, 3 months ago

Update comments in new do_loop_substitute.h90 on the trunk

File size: 2.9 KB
Line 
1#if defined show_comments
2! These comments are not intended to be retained during preprocessing; i.e. do not define "show_comments"
3!!----------------------------------------------------------------------
4!! NEMO/OCE 4.0 , NEMO Consortium (2018)
5!! Software governed by the CeCILL license (see ./LICENSE)
6!!----------------------------------------------------------------------
7! This header file contains preprocessor definitions and macros used in the do-loop substitutions introduced
8! between version 4.0 and 4.2. The primary aim of these macros is to assist in future applications of tiling
9! to improve performance. This is expected to be achieved by alternative versions of these macros in selected
10! locations. The initial introduction of these macros simply replaced all identifiable nested 2D- and 3D-loops
11! with single line statements (and adjusts indenting accordingly). Do loops were identifiable if they comformed
12! to either:
13!                                       DO jk = ....
14!   DO jj = ....                           DO jj = ...
15!      DO ji = ....                           DO ji = ...
16!         .                   OR                 . 
17!         .                                      .
18!     END DO                                  END DO
19!   END DO                                 END DO
20!                                       END DO
21! and white-space variants thereof.
22!
23! Additionally, only loops with recognised jj and ji loops limits were treated; these were:
24! Lower limits of 1, 2 or fs_2
25! Upper limits of jpi, jpim1 or fs_jpim1 (for ji) or jpj, jpjm1 or fs_jpjm1 (for jj)
26!
27! The macro naming convention takes the form: DO_2D( B, T, L, R) where:
28!   B is the Bottom offset from the PE's inner domain;
29!   T is the Top    offset from the PE's inner domain;
30!   L is the Left   offset from the PE's inner domain;
31!   R is the Right  offset from the PE's inner domain
32!
33! So, given an inner domain of 2,jpim1 and 2,jpjm1, a typical example would replace:
34!
35!   DO jj = 2, jpj
36!      DO ji = 1, jpim1
37!         .
38!         .
39!      END DO
40!   END DO
41!
42! with:
43!
44!   DO_2D( 0, 1, 1, 0 )
45!      .
46!      .
47!   END_2D
48!
49! similar conventions apply to the 3D loops macros. jk loop limits are retained through macro arguments
50! and are not restricted. This includes the possibility of strides for which an extra set of DO_3DS
51! macros are defined.
52!
53! In the following definitions the inner PE domain is defined by start indices of (Nis0, Njs0) and end
54! indices of (Nie0, Nje0) where:
55!
56! Nis0 =   1 + nn_hls     Njs0 =   1 + nn_hls
57! Nie0 = jpi - nn_hls     Nje0 = jpj - nn_hls
58!
59#endif
60
61#define DO_2D(B, T, L, R) DO jj = Njs0-(B), Nje0+(T)   ;   DO ji = Nis0-(L), Nie0+(R)
62
63#define DO_3D(B, T, L, R, ks, ke) DO jk = ks, ke   ;   DO_2D(B, T, L, R)
64
65#define DO_3DS(B, T, L, R, ks, ke, ki) DO jk = ks, ke, ki   ;   DO_2D(B, T, L, R)
66
67#define END_2D   END DO   ;   END DO
68#define END_3D   END DO   ;   END DO   ;   END DO
Note: See TracBrowser for help on using the repository browser.