Annotation of rpl/lapack/blas/dgbmv.f, revision 1.14
1.8 bertrand 1: *> \brief \b DGBMV
2: *
3: * =========== DOCUMENTATION ===========
4: *
1.14 ! bertrand 5: * Online html documentation available at
! 6: * http://www.netlib.org/lapack/explore-html/
1.8 bertrand 7: *
8: * Definition:
9: * ===========
10: *
11: * SUBROUTINE DGBMV(TRANS,M,N,KL,KU,ALPHA,A,LDA,X,INCX,BETA,Y,INCY)
1.14 ! bertrand 12: *
1.8 bertrand 13: * .. Scalar Arguments ..
14: * DOUBLE PRECISION ALPHA,BETA
15: * INTEGER INCX,INCY,KL,KU,LDA,M,N
16: * CHARACTER TRANS
17: * ..
18: * .. Array Arguments ..
19: * DOUBLE PRECISION A(LDA,*),X(*),Y(*)
20: * ..
1.14 ! bertrand 21: *
1.8 bertrand 22: *
23: *> \par Purpose:
24: * =============
25: *>
26: *> \verbatim
27: *>
28: *> DGBMV performs one of the matrix-vector operations
29: *>
30: *> y := alpha*A*x + beta*y, or y := alpha*A**T*x + beta*y,
31: *>
32: *> where alpha and beta are scalars, x and y are vectors and A is an
33: *> m by n band matrix, with kl sub-diagonals and ku super-diagonals.
34: *> \endverbatim
35: *
36: * Arguments:
37: * ==========
38: *
39: *> \param[in] TRANS
40: *> \verbatim
41: *> TRANS is CHARACTER*1
42: *> On entry, TRANS specifies the operation to be performed as
43: *> follows:
44: *>
45: *> TRANS = 'N' or 'n' y := alpha*A*x + beta*y.
46: *>
47: *> TRANS = 'T' or 't' y := alpha*A**T*x + beta*y.
48: *>
49: *> TRANS = 'C' or 'c' y := alpha*A**T*x + beta*y.
50: *> \endverbatim
51: *>
52: *> \param[in] M
53: *> \verbatim
54: *> M is INTEGER
55: *> On entry, M specifies the number of rows of the matrix A.
56: *> M must be at least zero.
57: *> \endverbatim
58: *>
59: *> \param[in] N
60: *> \verbatim
61: *> N is INTEGER
62: *> On entry, N specifies the number of columns of the matrix A.
63: *> N must be at least zero.
64: *> \endverbatim
65: *>
66: *> \param[in] KL
67: *> \verbatim
68: *> KL is INTEGER
69: *> On entry, KL specifies the number of sub-diagonals of the
70: *> matrix A. KL must satisfy 0 .le. KL.
71: *> \endverbatim
72: *>
73: *> \param[in] KU
74: *> \verbatim
75: *> KU is INTEGER
76: *> On entry, KU specifies the number of super-diagonals of the
77: *> matrix A. KU must satisfy 0 .le. KU.
78: *> \endverbatim
79: *>
80: *> \param[in] ALPHA
81: *> \verbatim
82: *> ALPHA is DOUBLE PRECISION.
83: *> On entry, ALPHA specifies the scalar alpha.
84: *> \endverbatim
85: *>
86: *> \param[in] A
87: *> \verbatim
88: *> A is DOUBLE PRECISION array of DIMENSION ( LDA, n ).
89: *> Before entry, the leading ( kl + ku + 1 ) by n part of the
90: *> array A must contain the matrix of coefficients, supplied
91: *> column by column, with the leading diagonal of the matrix in
92: *> row ( ku + 1 ) of the array, the first super-diagonal
93: *> starting at position 2 in row ku, the first sub-diagonal
94: *> starting at position 1 in row ( ku + 2 ), and so on.
95: *> Elements in the array A that do not correspond to elements
96: *> in the band matrix (such as the top left ku by ku triangle)
97: *> are not referenced.
98: *> The following program segment will transfer a band matrix
99: *> from conventional full matrix storage to band storage:
100: *>
101: *> DO 20, J = 1, N
102: *> K = KU + 1 - J
103: *> DO 10, I = MAX( 1, J - KU ), MIN( M, J + KL )
104: *> A( K + I, J ) = matrix( I, J )
105: *> 10 CONTINUE
106: *> 20 CONTINUE
107: *> \endverbatim
108: *>
109: *> \param[in] LDA
110: *> \verbatim
111: *> LDA is INTEGER
112: *> On entry, LDA specifies the first dimension of A as declared
113: *> in the calling (sub) program. LDA must be at least
114: *> ( kl + ku + 1 ).
115: *> \endverbatim
116: *>
117: *> \param[in] X
118: *> \verbatim
119: *> X is DOUBLE PRECISION array of DIMENSION at least
120: *> ( 1 + ( n - 1 )*abs( INCX ) ) when TRANS = 'N' or 'n'
121: *> and at least
122: *> ( 1 + ( m - 1 )*abs( INCX ) ) otherwise.
123: *> Before entry, the incremented array X must contain the
124: *> vector x.
125: *> \endverbatim
126: *>
127: *> \param[in] INCX
128: *> \verbatim
129: *> INCX is INTEGER
130: *> On entry, INCX specifies the increment for the elements of
131: *> X. INCX must not be zero.
132: *> \endverbatim
133: *>
134: *> \param[in] BETA
135: *> \verbatim
136: *> BETA is DOUBLE PRECISION.
137: *> On entry, BETA specifies the scalar beta. When BETA is
138: *> supplied as zero then Y need not be set on input.
139: *> \endverbatim
140: *>
141: *> \param[in,out] Y
142: *> \verbatim
143: *> Y is DOUBLE PRECISION array of DIMENSION at least
144: *> ( 1 + ( m - 1 )*abs( INCY ) ) when TRANS = 'N' or 'n'
145: *> and at least
146: *> ( 1 + ( n - 1 )*abs( INCY ) ) otherwise.
147: *> Before entry, the incremented array Y must contain the
148: *> vector y. On exit, Y is overwritten by the updated vector y.
149: *> \endverbatim
150: *>
151: *> \param[in] INCY
152: *> \verbatim
153: *> INCY is INTEGER
154: *> On entry, INCY specifies the increment for the elements of
155: *> Y. INCY must not be zero.
156: *> \endverbatim
157: *
158: * Authors:
159: * ========
160: *
1.14 ! bertrand 161: *> \author Univ. of Tennessee
! 162: *> \author Univ. of California Berkeley
! 163: *> \author Univ. of Colorado Denver
! 164: *> \author NAG Ltd.
1.8 bertrand 165: *
1.14 ! bertrand 166: *> \date December 2016
1.8 bertrand 167: *
168: *> \ingroup double_blas_level2
169: *
170: *> \par Further Details:
171: * =====================
172: *>
173: *> \verbatim
174: *>
175: *> Level 2 Blas routine.
176: *> The vector and matrix arguments are not referenced when N = 0, or M = 0
177: *>
178: *> -- Written on 22-October-1986.
179: *> Jack Dongarra, Argonne National Lab.
180: *> Jeremy Du Croz, Nag Central Office.
181: *> Sven Hammarling, Nag Central Office.
182: *> Richard Hanson, Sandia National Labs.
183: *> \endverbatim
184: *>
185: * =====================================================================
1.1 bertrand 186: SUBROUTINE DGBMV(TRANS,M,N,KL,KU,ALPHA,A,LDA,X,INCX,BETA,Y,INCY)
1.8 bertrand 187: *
1.14 ! bertrand 188: * -- Reference BLAS level2 routine (version 3.7.0) --
1.8 bertrand 189: * -- Reference BLAS is a software package provided by Univ. of Tennessee, --
190: * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
1.14 ! bertrand 191: * December 2016
1.8 bertrand 192: *
1.1 bertrand 193: * .. Scalar Arguments ..
194: DOUBLE PRECISION ALPHA,BETA
195: INTEGER INCX,INCY,KL,KU,LDA,M,N
196: CHARACTER TRANS
197: * ..
198: * .. Array Arguments ..
199: DOUBLE PRECISION A(LDA,*),X(*),Y(*)
200: * ..
201: *
202: * =====================================================================
203: *
204: * .. Parameters ..
205: DOUBLE PRECISION ONE,ZERO
206: PARAMETER (ONE=1.0D+0,ZERO=0.0D+0)
207: * ..
208: * .. Local Scalars ..
209: DOUBLE PRECISION TEMP
210: INTEGER I,INFO,IX,IY,J,JX,JY,K,KUP1,KX,KY,LENX,LENY
211: * ..
212: * .. External Functions ..
213: LOGICAL LSAME
214: EXTERNAL LSAME
215: * ..
216: * .. External Subroutines ..
217: EXTERNAL XERBLA
218: * ..
219: * .. Intrinsic Functions ..
220: INTRINSIC MAX,MIN
221: * ..
222: *
223: * Test the input parameters.
224: *
225: INFO = 0
226: IF (.NOT.LSAME(TRANS,'N') .AND. .NOT.LSAME(TRANS,'T') .AND.
227: + .NOT.LSAME(TRANS,'C')) THEN
228: INFO = 1
229: ELSE IF (M.LT.0) THEN
230: INFO = 2
231: ELSE IF (N.LT.0) THEN
232: INFO = 3
233: ELSE IF (KL.LT.0) THEN
234: INFO = 4
235: ELSE IF (KU.LT.0) THEN
236: INFO = 5
237: ELSE IF (LDA.LT. (KL+KU+1)) THEN
238: INFO = 8
239: ELSE IF (INCX.EQ.0) THEN
240: INFO = 10
241: ELSE IF (INCY.EQ.0) THEN
242: INFO = 13
243: END IF
244: IF (INFO.NE.0) THEN
245: CALL XERBLA('DGBMV ',INFO)
246: RETURN
247: END IF
248: *
249: * Quick return if possible.
250: *
251: IF ((M.EQ.0) .OR. (N.EQ.0) .OR.
252: + ((ALPHA.EQ.ZERO).AND. (BETA.EQ.ONE))) RETURN
253: *
254: * Set LENX and LENY, the lengths of the vectors x and y, and set
255: * up the start points in X and Y.
256: *
257: IF (LSAME(TRANS,'N')) THEN
258: LENX = N
259: LENY = M
260: ELSE
261: LENX = M
262: LENY = N
263: END IF
264: IF (INCX.GT.0) THEN
265: KX = 1
266: ELSE
267: KX = 1 - (LENX-1)*INCX
268: END IF
269: IF (INCY.GT.0) THEN
270: KY = 1
271: ELSE
272: KY = 1 - (LENY-1)*INCY
273: END IF
274: *
275: * Start the operations. In this version the elements of A are
276: * accessed sequentially with one pass through the band part of A.
277: *
278: * First form y := beta*y.
279: *
280: IF (BETA.NE.ONE) THEN
281: IF (INCY.EQ.1) THEN
282: IF (BETA.EQ.ZERO) THEN
283: DO 10 I = 1,LENY
284: Y(I) = ZERO
285: 10 CONTINUE
286: ELSE
287: DO 20 I = 1,LENY
288: Y(I) = BETA*Y(I)
289: 20 CONTINUE
290: END IF
291: ELSE
292: IY = KY
293: IF (BETA.EQ.ZERO) THEN
294: DO 30 I = 1,LENY
295: Y(IY) = ZERO
296: IY = IY + INCY
297: 30 CONTINUE
298: ELSE
299: DO 40 I = 1,LENY
300: Y(IY) = BETA*Y(IY)
301: IY = IY + INCY
302: 40 CONTINUE
303: END IF
304: END IF
305: END IF
306: IF (ALPHA.EQ.ZERO) RETURN
307: KUP1 = KU + 1
308: IF (LSAME(TRANS,'N')) THEN
309: *
310: * Form y := alpha*A*x + y.
311: *
312: JX = KX
313: IF (INCY.EQ.1) THEN
314: DO 60 J = 1,N
1.12 bertrand 315: TEMP = ALPHA*X(JX)
316: K = KUP1 - J
317: DO 50 I = MAX(1,J-KU),MIN(M,J+KL)
318: Y(I) = Y(I) + TEMP*A(K+I,J)
319: 50 CONTINUE
1.1 bertrand 320: JX = JX + INCX
321: 60 CONTINUE
322: ELSE
323: DO 80 J = 1,N
1.12 bertrand 324: TEMP = ALPHA*X(JX)
325: IY = KY
326: K = KUP1 - J
327: DO 70 I = MAX(1,J-KU),MIN(M,J+KL)
328: Y(IY) = Y(IY) + TEMP*A(K+I,J)
329: IY = IY + INCY
330: 70 CONTINUE
1.1 bertrand 331: JX = JX + INCX
332: IF (J.GT.KU) KY = KY + INCY
333: 80 CONTINUE
334: END IF
335: ELSE
336: *
1.7 bertrand 337: * Form y := alpha*A**T*x + y.
1.1 bertrand 338: *
339: JY = KY
340: IF (INCX.EQ.1) THEN
341: DO 100 J = 1,N
342: TEMP = ZERO
343: K = KUP1 - J
344: DO 90 I = MAX(1,J-KU),MIN(M,J+KL)
345: TEMP = TEMP + A(K+I,J)*X(I)
346: 90 CONTINUE
347: Y(JY) = Y(JY) + ALPHA*TEMP
348: JY = JY + INCY
349: 100 CONTINUE
350: ELSE
351: DO 120 J = 1,N
352: TEMP = ZERO
353: IX = KX
354: K = KUP1 - J
355: DO 110 I = MAX(1,J-KU),MIN(M,J+KL)
356: TEMP = TEMP + A(K+I,J)*X(IX)
357: IX = IX + INCX
358: 110 CONTINUE
359: Y(JY) = Y(JY) + ALPHA*TEMP
360: JY = JY + INCY
361: IF (J.GT.KU) KX = KX + INCX
362: 120 CONTINUE
363: END IF
364: END IF
365: *
366: RETURN
367: *
368: * End of DGBMV .
369: *
370: END
CVSweb interface <joel.bertrand@systella.fr>