ZGLMTS(3)
β¨ ZGLMTS(3) β¨
okay so bestie, buckle up π this one lives in the LAPACK manual β the page header is literally /home/abui...IG/zglmts.f(3), so yes, we're deep in the source tree, and no, the path is not a vibe, it's just where the file lives π
π NAME
/home/abuild/rpmbuild/BUILD/lapack-static-3.12.0-build/lapack-3.12.0/TESTING/EIG/zglmts.f
like. that's the whole name. it's giving "my parents named me after my address" π but that's literally what the page says and we respect it π«Ά
πΈ SYNOPSIS
Functions/Subroutines
subroutine ZGLMTS (n, m, p, a, af, lda, b, bf, ldb, d, df, x, u, work,
lwork, rwork, result)
ZGLMTS β that's her. that's the girl. π
π¦ Function/Subroutine Documentation
subroutine ZGLMTS (integer n, integer m, integer p, complex*16, dimension(
lda, * ) a, complex*16, dimension( lda, * ) af, integer lda, complex*16,
dimension( ldb, * ) b, complex*16, dimension( ldb, * ) bf, integer ldb,
complex*16, dimension( * ) d, complex*16, dimension( * ) df, complex*16,
dimension( * ) x, complex*16, dimension( * ) u, complex*16, dimension(
lwork ) work, integer lwork, double precision, dimension( * ) rwork,
double precision result)
ZGLMTS β¨ (yes she's listed again, she contains multitudes π)
π Purpose:
ZGLMTStestsZGGGLMβ a subroutine for solving the generalized linear model problem.
no cap, that's the whole assignment. she's a tester. she tests the solver. main character energy honestly π
π Parameters
NβNisINTEGER. the number of rows of the matricesAandB, andN >= 0. no negative rows babe, that's not a thing π«ΆMβMisINTEGER. the number of columns of the matrixA, andM >= 0.PβPisINTEGER. the number of columns of the matrixB, andP >= 0.AβAisCOMPLEX*16array, dimension(LDA,M). it's the N-by-M matrixA. iconic, no notes πAFβAFisCOMPLEX*16array, dimension(LDA,M). that's it, that's the description, that's all the page gives us π«LDAβLDAisINTEGER. the leading dimension of the arraysA,AF.LDA >= max(M,N). periodt, respect the bound βBβBisCOMPLEX*16array, dimension(LDB,P). "The N-by-P matrix A." β yes bestie, the page literally saysAhere. i am not fixing it. i am not touching it. we honor the source code's typos like sacred artifacts ππBFβBFisCOMPLEX*16array, dimension(LDB,P). again: that's the whole description. she's mysterious. she's lowkey withholding π₯LDBβLDBisINTEGER. the leading dimension of the arraysB,BF.LDB >= max(P,N). same energy asLDA, same rules πDβDisCOMPLEX*16array, dimension( N ). on input, the left hand side of the GLM. she's the goal, the target, the dream π―DFβDFisCOMPLEX*16array, dimension( N ). no further comment from the manpage. take it up with the authors πXβXisCOMPLEX*16array, dimension( M ). solution vectorXin the GLM problem πUβUisCOMPLEX*16array, dimension( P ). solution vectorUin the GLM problem π (yes, two solution vectors, it's a whole duo, they're a package deal π«Ά)WORKβWORKisCOMPLEX*16array, dimension(LWORK). the workspace. the backstage. the green room πΏLWORKβLWORKisINTEGER. the dimension of the arrayWORK. that's it. that's the rule βRWORKβRWORKisDOUBLE PRECISIONarray, dimension(M). (yes, dimensionMβ notN, don't @ me, that's what it says π)RESULTβRESULTisDOUBLE PRECISION. the test ratio:
norm( d - A*x - B*u )
RESULT = -----------------------------------------
(norm(A)+norm(B))*(norm(x)+norm(u))*EPS
like, how far off is the solve? that's the whole vibe check right there π§ small RESULT = she ate. big RESULT = something's sus π
π Author
- Univ. of Tennessee
- Univ. of California Berkeley
- Univ. of Colorado Denver
- NAG Ltd.
the whole squad showed up for this one π«Άβ¨
π Definition at line 144 of file zglmts.f.
π Author
Generated automatically by Doxygen for LAPACK from the source code. so a robot wrote this page and honestly? she did fine π€πΈ
okay that's the whole page, no crumbs left, every parameter accounted for, every bound respected, every typo preserved π
signing off with the footer tea: this is LAPACK, Version 3.12.0, from /home/abui...IG/zglmts.f(3) β go check your RESULT, bestie, and slay ππ
π¦ where this page comes from
- opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 Β· BSD-3-Clause
- opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 Β· BSD-3-Clause
- opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 Β· BSD-3-Clause
- opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 Β· BSD-3-Clause
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms