1 
  2 
  3 
  4 
  5 
  6 
  7 
  8 
  9 
 10 
 11 
 12 
 13 
 14 
 15 
 16 
 17 
 18 
 19 
 20 
 21 
 22 
 23 
 24 
 25 
 26 
 27 
 28 
 29 
 30 
 31 
 32 
 33 
 34 
 35 
 36 
 37 
 38 
 39 
 40 
 41 
 42 
 43 
 44 
 45 
 46 
 47 
 48 
 49 
 50 
 51 
 52 
 53 
 54 
 55 
 56 
 57 
 58 
 59 
 60 
 61 
 62 
 63 
 64 
 65 
 66 
 67 
 68 
 69 
 70 
 71 
 72 
 73 
 74 
 75 
 76 
 77 
 78 
 79 
 80 
 81 
 82 
 83 
 84 
 85 
 86 
 87 
 88 
 89 
 90 
 91 
 92 
 93 
 94 
 95 
 96 
 97 
 98 
 99 
100 
101 
102 
103 
104 
105 
106 
107 
108 
109 
110 
111 
112 
113 
114 
115 
116 
117 
118 
119 
120 
121 
122 
123 
124 
125 
126 
127 
128 
129 
130 
131 
132 
133 
134 
135 
136 
137 
138 
139 
140 
141 
142 
143 
144 
145 
146 
147 
148 
149 
150 
151 
152 
153 
154 
155 
156 
157 
158 
159 
160 
161 
162 
163 
164 
165 
166 
167 
168 
169 
170 
171 
172 
173 
174 
175 
176 
177 
178 
179 
180 
181 
182 
183 
184 
185 
186 
187 
188 
189 
190 
191 
192 
193 
194 
195 
196 
197 
198 
199 
200 
201 
202 
203 
204 
205 
206 
207 
208 
209 
210 
211 
212 
213 
214 
215 
216 
217 
218 
219 
220 
221 
222 
223 
224 
225 
226 
227 
228 
229 
230 
231 
232 
233 
234 
235 
236 
237 
238 
239 
240 
241 
242 
243 
244 
245 
246 
247 
248 
249 
250 
251 
252 
253 
254 
255 
256 
257 
258 
259 
260 
261 
262 
263 
264 
265 
266 
267 
268 
269 
270 
271 
272 
273 
274 
275 
276 
277 
278 
279 
280 
281 
282 
283 
284 
285 
286 
287 
288 
289 
290 
291 
292 
293 
294 
295 
296 
297 
298 
299 
300 
301 
302 
 
 | 
 
      SUBROUTINE SGBMV(TRANS,M,N,KL,KU,ALPHA,A,LDA,X,INCX,BETA,Y,INCY) 
*     .. Scalar Arguments .. 
      REAL ALPHA,BETA 
      INTEGER INCX,INCY,KL,KU,LDA,M,N 
      CHARACTER TRANS 
*     .. 
*     .. Array Arguments .. 
      REAL A(LDA,*),X(*),Y(*) 
*     .. 
* 
*  Purpose 
*  ======= 
* 
*  SGBMV  performs one of the matrix-vector operations 
* 
*     y := alpha*A*x + beta*y,   or   y := alpha*A**T*x + beta*y, 
* 
*  where alpha and beta are scalars, x and y are vectors and A is an 
*  m by n band matrix, with kl sub-diagonals and ku super-diagonals. 
* 
*  Arguments 
*  ========== 
* 
*  TRANS  - CHARACTER*1. 
*           On entry, TRANS specifies the operation to be performed as 
*           follows: 
* 
*              TRANS = 'N' or 'n'   y := alpha*A*x + beta*y. 
* 
*              TRANS = 'T' or 't'   y := alpha*A**T*x + beta*y. 
* 
*              TRANS = 'C' or 'c'   y := alpha*A**T*x + beta*y. 
* 
*           Unchanged on exit. 
* 
*  M      - INTEGER. 
*           On entry, M specifies the number of rows of the matrix A. 
*           M must be at least zero. 
*           Unchanged on exit. 
* 
*  N      - INTEGER. 
*           On entry, N specifies the number of columns of the matrix A. 
*           N must be at least zero. 
*           Unchanged on exit. 
* 
*  KL     - INTEGER. 
*           On entry, KL specifies the number of sub-diagonals of the 
*           matrix A. KL must satisfy  0 .le. KL. 
*           Unchanged on exit. 
* 
*  KU     - INTEGER. 
*           On entry, KU specifies the number of super-diagonals of the 
*           matrix A. KU must satisfy  0 .le. KU. 
*           Unchanged on exit. 
* 
*  ALPHA  - REAL            . 
*           On entry, ALPHA specifies the scalar alpha. 
*           Unchanged on exit. 
* 
*  A      - REAL             array of DIMENSION ( LDA, n ). 
*           Before entry, the leading ( kl + ku + 1 ) by n part of the 
*           array A must contain the matrix of coefficients, supplied 
*           column by column, with the leading diagonal of the matrix in 
*           row ( ku + 1 ) of the array, the first super-diagonal 
*           starting at position 2 in row ku, the first sub-diagonal 
*           starting at position 1 in row ( ku + 2 ), and so on. 
*           Elements in the array A that do not correspond to elements 
*           in the band matrix (such as the top left ku by ku triangle) 
*           are not referenced. 
*           The following program segment will transfer a band matrix 
*           from conventional full matrix storage to band storage: 
* 
*                 DO 20, J = 1, N 
*                    K = KU + 1 - J 
*                    DO 10, I = MAX( 1, J - KU ), MIN( M, J + KL ) 
*                       A( K + I, J ) = matrix( I, J ) 
*              10    CONTINUE 
*              20 CONTINUE 
* 
*           Unchanged on exit. 
* 
*  LDA    - INTEGER. 
*           On entry, LDA specifies the first dimension of A as declared 
*           in the calling (sub) program. LDA must be at least 
*           ( kl + ku + 1 ). 
*           Unchanged on exit. 
* 
*  X      - REAL             array of DIMENSION at least 
*           ( 1 + ( n - 1 )*abs( INCX ) ) when TRANS = 'N' or 'n' 
*           and at least 
*           ( 1 + ( m - 1 )*abs( INCX ) ) otherwise. 
*           Before entry, the incremented array X must contain the 
*           vector x. 
*           Unchanged on exit. 
* 
*  INCX   - INTEGER. 
*           On entry, INCX specifies the increment for the elements of 
*           X. INCX must not be zero. 
*           Unchanged on exit. 
* 
*  BETA   - REAL            . 
*           On entry, BETA specifies the scalar beta. When BETA is 
*           supplied as zero then Y need not be set on input. 
*           Unchanged on exit. 
* 
*  Y      - REAL             array of DIMENSION at least 
*           ( 1 + ( m - 1 )*abs( INCY ) ) when TRANS = 'N' or 'n' 
*           and at least 
*           ( 1 + ( n - 1 )*abs( INCY ) ) otherwise. 
*           Before entry, the incremented array Y must contain the 
*           vector y. On exit, Y is overwritten by the updated vector y. 
* 
*  INCY   - INTEGER. 
*           On entry, INCY specifies the increment for the elements of 
*           Y. INCY must not be zero. 
*           Unchanged on exit. 
* 
*  Further Details 
*  =============== 
* 
*  Level 2 Blas routine. 
*  The vector and matrix arguments are not referenced when N = 0, or M = 0 
* 
*  -- Written on 22-October-1986. 
*     Jack Dongarra, Argonne National Lab. 
*     Jeremy Du Croz, Nag Central Office. 
*     Sven Hammarling, Nag Central Office. 
*     Richard Hanson, Sandia National Labs. 
* 
*  ===================================================================== 
* 
*     .. Parameters .. 
      REAL ONE,ZERO 
      PARAMETER (ONE=1.0E+0,ZERO=0.0E+0) 
*     .. 
*     .. Local Scalars .. 
      REAL TEMP 
      INTEGER I,INFO,IX,IY,J,JX,JY,K,KUP1,KX,KY,LENX,LENY 
*     .. 
*     .. External Functions .. 
      LOGICAL LSAME 
      EXTERNAL LSAME 
*     .. 
*     .. External Subroutines .. 
      EXTERNAL XERBLA 
*     .. 
*     .. Intrinsic Functions .. 
      INTRINSIC MAX,MIN 
*     .. 
* 
*     Test the input parameters. 
* 
      INFO = 0 
      IF (.NOT.LSAME(TRANS,'N') .AND. .NOT.LSAME(TRANS,'T') .AND. 
     +    .NOT.LSAME(TRANS,'C')) THEN 
          INFO = 1 
      ELSE IF (M.LT.0) THEN 
          INFO = 2 
      ELSE IF (N.LT.0) THEN 
          INFO = 3 
      ELSE IF (KL.LT.0) THEN 
          INFO = 4 
      ELSE IF (KU.LT.0) THEN 
          INFO = 5 
      ELSE IF (LDA.LT. (KL+KU+1)) THEN 
          INFO = 8 
      ELSE IF (INCX.EQ.0) THEN 
          INFO = 10 
      ELSE IF (INCY.EQ.0) THEN 
          INFO = 13 
      END IF 
      IF (INFO.NE.0) THEN 
          CALL XERBLA('SGBMV ',INFO) 
          RETURN 
      END IF 
* 
*     Quick return if possible. 
* 
      IF ((M.EQ.0) .OR. (N.EQ.0) .OR. 
     +    ((ALPHA.EQ.ZERO).AND. (BETA.EQ.ONE))) RETURN 
* 
*     Set  LENX  and  LENY, the lengths of the vectors x and y, and set 
*     up the start points in  X  and  Y. 
* 
      IF (LSAME(TRANS,'N')) THEN 
          LENX = N 
          LENY = M 
      ELSE 
          LENX = M 
          LENY = N 
      END IF 
      IF (INCX.GT.0) THEN 
          KX = 1 
      ELSE 
          KX = 1 - (LENX-1)*INCX 
      END IF 
      IF (INCY.GT.0) THEN 
          KY = 1 
      ELSE 
          KY = 1 - (LENY-1)*INCY 
      END IF 
* 
*     Start the operations. In this version the elements of A are 
*     accessed sequentially with one pass through the band part of A. 
* 
*     First form  y := beta*y. 
* 
      IF (BETA.NE.ONE) THEN 
          IF (INCY.EQ.1) THEN 
              IF (BETA.EQ.ZERO) THEN 
                  DO 10 I = 1,LENY 
                      Y(I) = ZERO 
   10             CONTINUE 
              ELSE 
                  DO 20 I = 1,LENY 
                      Y(I) = BETA*Y(I) 
   20             CONTINUE 
              END IF 
          ELSE 
              IY = KY 
              IF (BETA.EQ.ZERO) THEN 
                  DO 30 I = 1,LENY 
                      Y(IY) = ZERO 
                      IY = IY + INCY 
   30             CONTINUE 
              ELSE 
                  DO 40 I = 1,LENY 
                      Y(IY) = BETA*Y(IY) 
                      IY = IY + INCY 
   40             CONTINUE 
              END IF 
          END IF 
      END IF 
      IF (ALPHA.EQ.ZERO) RETURN 
      KUP1 = KU + 1 
      IF (LSAME(TRANS,'N')) THEN 
* 
*        Form  y := alpha*A*x + y. 
* 
          JX = KX 
          IF (INCY.EQ.1) THEN 
              DO 60 J = 1,N 
                  IF (X(JX).NE.ZERO) THEN 
                      TEMP = ALPHA*X(JX) 
                      K = KUP1 - J 
                      DO 50 I = MAX(1,J-KU),MIN(M,J+KL) 
                          Y(I) = Y(I) + TEMP*A(K+I,J) 
   50                 CONTINUE 
                  END IF 
                  JX = JX + INCX 
   60         CONTINUE 
          ELSE 
              DO 80 J = 1,N 
                  IF (X(JX).NE.ZERO) THEN 
                      TEMP = ALPHA*X(JX) 
                      IY = KY 
                      K = KUP1 - J 
                      DO 70 I = MAX(1,J-KU),MIN(M,J+KL) 
                          Y(IY) = Y(IY) + TEMP*A(K+I,J) 
                          IY = IY + INCY 
   70                 CONTINUE 
                  END IF 
                  JX = JX + INCX 
                  IF (J.GT.KU) KY = KY + INCY 
   80         CONTINUE 
          END IF 
      ELSE 
* 
*        Form  y := alpha*A**T*x + y. 
* 
          JY = KY 
          IF (INCX.EQ.1) THEN 
              DO 100 J = 1,N 
                  TEMP = ZERO 
                  K = KUP1 - J 
                  DO 90 I = MAX(1,J-KU),MIN(M,J+KL) 
                      TEMP = TEMP + A(K+I,J)*X(I) 
   90             CONTINUE 
                  Y(JY) = Y(JY) + ALPHA*TEMP 
                  JY = JY + INCY 
  100         CONTINUE 
          ELSE 
              DO 120 J = 1,N 
                  TEMP = ZERO 
                  IX = KX 
                  K = KUP1 - J 
                  DO 110 I = MAX(1,J-KU),MIN(M,J+KL) 
                      TEMP = TEMP + A(K+I,J)*X(IX) 
                      IX = IX + INCX 
  110             CONTINUE 
                  Y(JY) = Y(JY) + ALPHA*TEMP 
                  JY = JY + INCY 
                  IF (J.GT.KU) KX = KX + INCX 
  120         CONTINUE 
          END IF 
      END IF 
* 
      RETURN 
* 
*     End of SGBMV . 
* 
      END 
 
 |