File:  [local] / rpl / lapack / blas / dgemm.f
Revision 1.4: download - view: text, annotated - select for diffs - revision graph
Sat Aug 7 13:22:08 2010 UTC (13 years, 9 months ago) by bertrand
Branches: MAIN
CVS tags: HEAD
Mise à jour globale de Lapack 3.2.2.

    1:       SUBROUTINE DGEMM(TRANSA,TRANSB,M,N,K,ALPHA,A,LDA,B,LDB,BETA,C,LDC)
    2: *     .. Scalar Arguments ..
    3:       DOUBLE PRECISION ALPHA,BETA
    4:       INTEGER K,LDA,LDB,LDC,M,N
    5:       CHARACTER TRANSA,TRANSB
    6: *     ..
    7: *     .. Array Arguments ..
    8:       DOUBLE PRECISION A(LDA,*),B(LDB,*),C(LDC,*)
    9: *     ..
   10: *
   11: *  Purpose
   12: *  =======
   13: *
   14: *  DGEMM  performs one of the matrix-matrix operations
   15: *
   16: *     C := alpha*op( A )*op( B ) + beta*C,
   17: *
   18: *  where  op( X ) is one of
   19: *
   20: *     op( X ) = X   or   op( X ) = X',
   21: *
   22: *  alpha and beta are scalars, and A, B and C are matrices, with op( A )
   23: *  an m by k matrix,  op( B )  a  k by n matrix and  C an m by n matrix.
   24: *
   25: *  Arguments
   26: *  ==========
   27: *
   28: *  TRANSA - CHARACTER*1.
   29: *           On entry, TRANSA specifies the form of op( A ) to be used in
   30: *           the matrix multiplication as follows:
   31: *
   32: *              TRANSA = 'N' or 'n',  op( A ) = A.
   33: *
   34: *              TRANSA = 'T' or 't',  op( A ) = A'.
   35: *
   36: *              TRANSA = 'C' or 'c',  op( A ) = A'.
   37: *
   38: *           Unchanged on exit.
   39: *
   40: *  TRANSB - CHARACTER*1.
   41: *           On entry, TRANSB specifies the form of op( B ) to be used in
   42: *           the matrix multiplication as follows:
   43: *
   44: *              TRANSB = 'N' or 'n',  op( B ) = B.
   45: *
   46: *              TRANSB = 'T' or 't',  op( B ) = B'.
   47: *
   48: *              TRANSB = 'C' or 'c',  op( B ) = B'.
   49: *
   50: *           Unchanged on exit.
   51: *
   52: *  M      - INTEGER.
   53: *           On entry,  M  specifies  the number  of rows  of the  matrix
   54: *           op( A )  and of the  matrix  C.  M  must  be at least  zero.
   55: *           Unchanged on exit.
   56: *
   57: *  N      - INTEGER.
   58: *           On entry,  N  specifies the number  of columns of the matrix
   59: *           op( B ) and the number of columns of the matrix C. N must be
   60: *           at least zero.
   61: *           Unchanged on exit.
   62: *
   63: *  K      - INTEGER.
   64: *           On entry,  K  specifies  the number of columns of the matrix
   65: *           op( A ) and the number of rows of the matrix op( B ). K must
   66: *           be at least  zero.
   67: *           Unchanged on exit.
   68: *
   69: *  ALPHA  - DOUBLE PRECISION.
   70: *           On entry, ALPHA specifies the scalar alpha.
   71: *           Unchanged on exit.
   72: *
   73: *  A      - DOUBLE PRECISION array of DIMENSION ( LDA, ka ), where ka is
   74: *           k  when  TRANSA = 'N' or 'n',  and is  m  otherwise.
   75: *           Before entry with  TRANSA = 'N' or 'n',  the leading  m by k
   76: *           part of the array  A  must contain the matrix  A,  otherwise
   77: *           the leading  k by m  part of the array  A  must contain  the
   78: *           matrix A.
   79: *           Unchanged on exit.
   80: *
   81: *  LDA    - INTEGER.
   82: *           On entry, LDA specifies the first dimension of A as declared
   83: *           in the calling (sub) program. When  TRANSA = 'N' or 'n' then
   84: *           LDA must be at least  max( 1, m ), otherwise  LDA must be at
   85: *           least  max( 1, k ).
   86: *           Unchanged on exit.
   87: *
   88: *  B      - DOUBLE PRECISION array of DIMENSION ( LDB, kb ), where kb is
   89: *           n  when  TRANSB = 'N' or 'n',  and is  k  otherwise.
   90: *           Before entry with  TRANSB = 'N' or 'n',  the leading  k by n
   91: *           part of the array  B  must contain the matrix  B,  otherwise
   92: *           the leading  n by k  part of the array  B  must contain  the
   93: *           matrix B.
   94: *           Unchanged on exit.
   95: *
   96: *  LDB    - INTEGER.
   97: *           On entry, LDB specifies the first dimension of B as declared
   98: *           in the calling (sub) program. When  TRANSB = 'N' or 'n' then
   99: *           LDB must be at least  max( 1, k ), otherwise  LDB must be at
  100: *           least  max( 1, n ).
  101: *           Unchanged on exit.
  102: *
  103: *  BETA   - DOUBLE PRECISION.
  104: *           On entry,  BETA  specifies the scalar  beta.  When  BETA  is
  105: *           supplied as zero then C need not be set on input.
  106: *           Unchanged on exit.
  107: *
  108: *  C      - DOUBLE PRECISION array of DIMENSION ( LDC, n ).
  109: *           Before entry, the leading  m by n  part of the array  C must
  110: *           contain the matrix  C,  except when  beta  is zero, in which
  111: *           case C need not be set on entry.
  112: *           On exit, the array  C  is overwritten by the  m by n  matrix
  113: *           ( alpha*op( A )*op( B ) + beta*C ).
  114: *
  115: *  LDC    - INTEGER.
  116: *           On entry, LDC specifies the first dimension of C as declared
  117: *           in  the  calling  (sub)  program.   LDC  must  be  at  least
  118: *           max( 1, m ).
  119: *           Unchanged on exit.
  120: *
  121: *  Further Details
  122: *  ===============
  123: *
  124: *  Level 3 Blas routine.
  125: *
  126: *  -- Written on 8-February-1989.
  127: *     Jack Dongarra, Argonne National Laboratory.
  128: *     Iain Duff, AERE Harwell.
  129: *     Jeremy Du Croz, Numerical Algorithms Group Ltd.
  130: *     Sven Hammarling, Numerical Algorithms Group Ltd.
  131: *
  132: *  =====================================================================
  133: *
  134: *     .. External Functions ..
  135:       LOGICAL LSAME
  136:       EXTERNAL LSAME
  137: *     ..
  138: *     .. External Subroutines ..
  139:       EXTERNAL XERBLA
  140: *     ..
  141: *     .. Intrinsic Functions ..
  142:       INTRINSIC MAX
  143: *     ..
  144: *     .. Local Scalars ..
  145:       DOUBLE PRECISION TEMP
  146:       INTEGER I,INFO,J,L,NCOLA,NROWA,NROWB
  147:       LOGICAL NOTA,NOTB
  148: *     ..
  149: *     .. Parameters ..
  150:       DOUBLE PRECISION ONE,ZERO
  151:       PARAMETER (ONE=1.0D+0,ZERO=0.0D+0)
  152: *     ..
  153: *
  154: *     Set  NOTA  and  NOTB  as  true if  A  and  B  respectively are not
  155: *     transposed and set  NROWA, NCOLA and  NROWB  as the number of rows
  156: *     and  columns of  A  and the  number of  rows  of  B  respectively.
  157: *
  158:       NOTA = LSAME(TRANSA,'N')
  159:       NOTB = LSAME(TRANSB,'N')
  160:       IF (NOTA) THEN
  161:           NROWA = M
  162:           NCOLA = K
  163:       ELSE
  164:           NROWA = K
  165:           NCOLA = M
  166:       END IF
  167:       IF (NOTB) THEN
  168:           NROWB = K
  169:       ELSE
  170:           NROWB = N
  171:       END IF
  172: *
  173: *     Test the input parameters.
  174: *
  175:       INFO = 0
  176:       IF ((.NOT.NOTA) .AND. (.NOT.LSAME(TRANSA,'C')) .AND.
  177:      +    (.NOT.LSAME(TRANSA,'T'))) THEN
  178:           INFO = 1
  179:       ELSE IF ((.NOT.NOTB) .AND. (.NOT.LSAME(TRANSB,'C')) .AND.
  180:      +         (.NOT.LSAME(TRANSB,'T'))) THEN
  181:           INFO = 2
  182:       ELSE IF (M.LT.0) THEN
  183:           INFO = 3
  184:       ELSE IF (N.LT.0) THEN
  185:           INFO = 4
  186:       ELSE IF (K.LT.0) THEN
  187:           INFO = 5
  188:       ELSE IF (LDA.LT.MAX(1,NROWA)) THEN
  189:           INFO = 8
  190:       ELSE IF (LDB.LT.MAX(1,NROWB)) THEN
  191:           INFO = 10
  192:       ELSE IF (LDC.LT.MAX(1,M)) THEN
  193:           INFO = 13
  194:       END IF
  195:       IF (INFO.NE.0) THEN
  196:           CALL XERBLA('DGEMM ',INFO)
  197:           RETURN
  198:       END IF
  199: *
  200: *     Quick return if possible.
  201: *
  202:       IF ((M.EQ.0) .OR. (N.EQ.0) .OR.
  203:      +    (((ALPHA.EQ.ZERO).OR. (K.EQ.0)).AND. (BETA.EQ.ONE))) RETURN
  204: *
  205: *     And if  alpha.eq.zero.
  206: *
  207:       IF (ALPHA.EQ.ZERO) THEN
  208:           IF (BETA.EQ.ZERO) THEN
  209:               DO 20 J = 1,N
  210:                   DO 10 I = 1,M
  211:                       C(I,J) = ZERO
  212:    10             CONTINUE
  213:    20         CONTINUE
  214:           ELSE
  215:               DO 40 J = 1,N
  216:                   DO 30 I = 1,M
  217:                       C(I,J) = BETA*C(I,J)
  218:    30             CONTINUE
  219:    40         CONTINUE
  220:           END IF
  221:           RETURN
  222:       END IF
  223: *
  224: *     Start the operations.
  225: *
  226:       IF (NOTB) THEN
  227:           IF (NOTA) THEN
  228: *
  229: *           Form  C := alpha*A*B + beta*C.
  230: *
  231:               DO 90 J = 1,N
  232:                   IF (BETA.EQ.ZERO) THEN
  233:                       DO 50 I = 1,M
  234:                           C(I,J) = ZERO
  235:    50                 CONTINUE
  236:                   ELSE IF (BETA.NE.ONE) THEN
  237:                       DO 60 I = 1,M
  238:                           C(I,J) = BETA*C(I,J)
  239:    60                 CONTINUE
  240:                   END IF
  241:                   DO 80 L = 1,K
  242:                       IF (B(L,J).NE.ZERO) THEN
  243:                           TEMP = ALPHA*B(L,J)
  244:                           DO 70 I = 1,M
  245:                               C(I,J) = C(I,J) + TEMP*A(I,L)
  246:    70                     CONTINUE
  247:                       END IF
  248:    80             CONTINUE
  249:    90         CONTINUE
  250:           ELSE
  251: *
  252: *           Form  C := alpha*A'*B + beta*C
  253: *
  254:               DO 120 J = 1,N
  255:                   DO 110 I = 1,M
  256:                       TEMP = ZERO
  257:                       DO 100 L = 1,K
  258:                           TEMP = TEMP + A(L,I)*B(L,J)
  259:   100                 CONTINUE
  260:                       IF (BETA.EQ.ZERO) THEN
  261:                           C(I,J) = ALPHA*TEMP
  262:                       ELSE
  263:                           C(I,J) = ALPHA*TEMP + BETA*C(I,J)
  264:                       END IF
  265:   110             CONTINUE
  266:   120         CONTINUE
  267:           END IF
  268:       ELSE
  269:           IF (NOTA) THEN
  270: *
  271: *           Form  C := alpha*A*B' + beta*C
  272: *
  273:               DO 170 J = 1,N
  274:                   IF (BETA.EQ.ZERO) THEN
  275:                       DO 130 I = 1,M
  276:                           C(I,J) = ZERO
  277:   130                 CONTINUE
  278:                   ELSE IF (BETA.NE.ONE) THEN
  279:                       DO 140 I = 1,M
  280:                           C(I,J) = BETA*C(I,J)
  281:   140                 CONTINUE
  282:                   END IF
  283:                   DO 160 L = 1,K
  284:                       IF (B(J,L).NE.ZERO) THEN
  285:                           TEMP = ALPHA*B(J,L)
  286:                           DO 150 I = 1,M
  287:                               C(I,J) = C(I,J) + TEMP*A(I,L)
  288:   150                     CONTINUE
  289:                       END IF
  290:   160             CONTINUE
  291:   170         CONTINUE
  292:           ELSE
  293: *
  294: *           Form  C := alpha*A'*B' + beta*C
  295: *
  296:               DO 200 J = 1,N
  297:                   DO 190 I = 1,M
  298:                       TEMP = ZERO
  299:                       DO 180 L = 1,K
  300:                           TEMP = TEMP + A(L,I)*B(J,L)
  301:   180                 CONTINUE
  302:                       IF (BETA.EQ.ZERO) THEN
  303:                           C(I,J) = ALPHA*TEMP
  304:                       ELSE
  305:                           C(I,J) = ALPHA*TEMP + BETA*C(I,J)
  306:                       END IF
  307:   190             CONTINUE
  308:   200         CONTINUE
  309:           END IF
  310:       END IF
  311: *
  312:       RETURN
  313: *
  314: *     End of DGEMM .
  315: *
  316:       END

CVSweb interface <joel.bertrand@systella.fr>