tpcon(3)
πΈ tpcon(3) β from the Library Functions Manual, section 3 πΈ
okay so bestie, welcome to tpcon, and the vibe is condition number estimate β¨. That's the whole name, that's the whole tagline: tpcon β tpcon: condition number estimate. Short, sweet, iconic, and she knows exactly what she is π
.
This page is a long one so it's split into parts β you're on part 1 right now, so don't go anywhere π«Ά.
β¨ NAME β¨
tpcon β tpcon: condition number estimate
like, literally what it says on the tin, no cap π
β¨ SYNOPSIS β¨
π Functions
okay so there are exactly four of these bad boys, one per precision/flavour, and here's the full roster β names and arguments kept exactly as they are, because we do not editorialise the signatures, periodt π₯
subroutine ctpcon (norm, uplo, diag, n, ap, rcond, work, rwork, info)βCTPCONπ thecone, withrworkin the mix.subroutine dtpcon (norm, uplo, diag, n, ap, rcond, work, iwork, info)βDTPCONπ thedone, and note she's rockin'iworkrather thanrwork.subroutine stpcon (norm, uplo, diag, n, ap, rcond, work, iwork, info)βSTPCONπ thesone, sameiworksituation as herdsibling.subroutine ztpcon (norm, uplo, diag, n, ap, rcond, work, rwork, info)βZTPCONπ thezone, back on therworkteam.
so yeah, four subroutines, all taking the same shape of argument list with that one iwork/rwork plot twist in the middle π β norm, uplo, diag, n, ap, rcond, work, then the work-array variant, then info. Slay.
subroutine ctpcon (norm, uplo, diag, n, ap, rcond, work, rwork, info)
subroutine dtpcon (norm, uplo, diag, n, ap, rcond, work, iwork, info)
subroutine stpcon (norm, uplo, diag, n, ap, rcond, work, iwork, info)
subroutine ztpcon (norm, uplo, diag, n, ap, rcond, work, rwork, info)
β¨ Detailed Description β¨
okay so this is the part where the detailed description would go π and on this pageβ¦ it's giving empty. Nothing here yet, bestie. No prose, no breakdown, no nothing β just the heading sitting there with main character energy and zero lines under it π.
So fr fr, that's all part 1 has to give us: the name, the four subroutine signatures, and a Detailed Description that's living its best minimalist life. More parts are coming, so keep scrolling when they land π¦
π Function Documentation
Okay bestie, so tpcon isn't just one subroutine β it's a whole friend group π. Four of them, actually: one for each real/complex and single/double combo. They all do the exact same glow-up (estimate that reciprocal condition number of a packed triangular matrix, 1-norm or infinity-norm, you know the drill), they just speak different precision languages. Let's meet the whole squad, one by one πβ¨
πΈ ctpcon β the complex single-precision queen π
subroutine ctpcon (character norm, character uplo, character diag, integer n, complex, dimension( * ) ap, real rcond, complex, dimension( * ) work, real, dimension( * ) rwork, integer info)
β¨ Purpose
Okay so CTPCON does the thing: it estimates the reciprocal of the condition number of a packed triangular matrix A, and it'll do it in either the 1-norm or the infinity-norm β your call, bestie π«Ά.
The way it works: it computes norm(A), then gets an estimate for norm(inv(A)), and finally the reciprocal of the condition number comes out as:
RCOND = 1 / ( norm(A) * norm(inv(A)) ).
No cap, that's the whole vibe π
π Parameters
NORMβCHARACTER*1. Tells it which condition number you want, 1-norm or infinity-norm:= '1' or 'O':1-norm π₯= 'I':Infinity-norm β
UPLOβCHARACTER*1.= 'U':Ais upper triangular πΈ= 'L':Ais lower triangular π¦
DIAGβCHARACTER*1.= 'N':Ais non-unit triangular= 'U':Ais unit triangular
NβINTEGER. The order of the matrixA. Constraint:N >= 0β¨APβCOMPLEXarray, dimension(N*(N+1)/2). This is the upper or lower triangular matrixA, packed columnwise into one linear array. Thej-th column ofAlives inAPlike this:- if
UPLO = 'U',AP(i + (j-1)*j/2) = A(i,j)for1<=i<=j; - if
UPLO = 'L',AP(i + (j-1)*(2n-j)/2) = A(i,j)forj<=i<=n.
- if
And if DIAG = 'U', the diagonal elements of A are literally not referenced β they're just assumed to be 1 π (iconic shortcut, honestly).
RCONDβREAL. The reciprocal of the condition number of the matrixA, computed asRCOND = 1/(norm(A) * norm(inv(A)))πWORKβCOMPLEXarray, dimension(2*N)π§RWORKβREALarray, dimension(N)πΈINFOβINTEGER.= 0:successful exit β¨< 0:ifINFO = -i, thei-th argument had an illegal value π¨
π Author
Univ. of Tennessee
Univ. of California Berkeley
Univ. of Colorado Denver
NAG Ltd.
Definition at line 128 of file ctpcon.f.
π¦ dtpcon β double precision, double the drama π
subroutine dtpcon (character norm, character uplo, character diag, integer n, double precision, dimension( * ) ap, double precision rcond, double precision, dimension( * ) work, integer, dimension( * ) iwork, integer info)
β¨ Purpose
Same energy, double precision π
. DTPCON estimates the reciprocal of the condition number of a packed triangular matrix A, in either the 1-norm or the infinity-norm.
It computes norm(A), estimates norm(inv(A)), and then:
RCOND = 1 / ( norm(A) * norm(inv(A)) ).
Periodt π
π Parameters
NORMβCHARACTER*1. Which condition number, bestie?= '1' or 'O':1-norm π₯= 'I':Infinity-norm β
UPLOβCHARACTER*1.= 'U':Ais upper triangular πΈ= 'L':Ais lower triangular π¦
DIAGβCHARACTER*1.= 'N':Ais non-unit triangular= 'U':Ais unit triangular
NβINTEGER. The order of the matrixA. Constraint:N >= 0β¨APβDOUBLE PRECISIONarray, dimension(N*(N+1)/2). The upper or lower triangular matrixA, packed columnwise into a linear array. Thej-th column ofAis stored inAPas follows:- if
UPLO = 'U',AP(i + (j-1)*j/2) = A(i,j)for1<=i<=j; - if
UPLO = 'L',AP(i + (j-1)*(2n-j)/2) = A(i,j)forj<=i<=n.
- if
If DIAG = 'U', the diagonal elements of A are not referenced and are assumed to be 1 π
RCONDβDOUBLE PRECISION. The reciprocal of the condition number of the matrixA, computed asRCOND = 1/(norm(A) * norm(inv(A)))πWORKβDOUBLE PRECISIONarray, dimension(3*N)π§IWORKβINTEGERarray, dimension(N)πINFOβINTEGER.= 0:successful exit β¨< 0:ifINFO = -i, thei-th argument had an illegal value π¨
π Author
Univ. of Tennessee
Univ. of California Berkeley
Univ. of Colorado Denver
NAG Ltd.
Definition at line 128 of file dtpcon.f.
π stpcon β single precision, still eating π½οΈ
subroutine stpcon (character norm, character uplo, character diag, integer n, real, dimension( * ) ap, real rcond, real, dimension( * ) work, integer, dimension( * ) iwork, integer info)
β¨ Purpose
STPCON estimates the reciprocal of the condition number of a packed triangular matrix A, in either the 1-norm or the infinity-norm β because why pick one when the spec says you get both π.
Computes norm(A), estimates norm(inv(A)), and boom:
RCOND = 1 / ( norm(A) * norm(inv(A)) ).
Slay π
π Parameters
NORMβCHARACTER*1. Pick your norm:= '1' or 'O':1-norm π₯= 'I':Infinity-norm β
UPLOβCHARACTER*1.= 'U':Ais upper triangular πΈ= 'L':Ais lower triangular π¦
DIAGβCHARACTER*1.= 'N':Ais non-unit triangular= 'U':Ais unit triangular
NβINTEGER. The order of the matrixA. Constraint:N >= 0β¨APβREALarray, dimension(N*(N+1)/2). The upper or lower triangular matrixA, packed columnwise in a linear array. Thej-th column ofAis stored inAPas follows:- if
UPLO = 'U',AP(i + (j-1)*j/2) = A(i,j)for1<=i<=j; - if
UPLO = 'L',AP(i + (j-1)*(2n-j)/2) = A(i,j)forj<=i<=n.
- if
If DIAG = 'U', the diagonal elements of A are not referenced and are assumed to be 1 π
RCONDβREAL. The reciprocal of the condition number of the matrixA, computed asRCOND = 1/(norm(A) * norm(inv(A)))πWORKβREALarray, dimension(3*N)π§IWORKβINTEGERarray, dimension(N)πINFOβINTEGER.= 0:successful exit β¨< 0:ifINFO = -i, thei-th argument had an illegal value π¨
π Author
Univ. of Tennessee
Univ. of California Berkeley
Univ. of Colorado Denver
NAG Ltd.
Definition at line 128 of file stpcon.f.
π ztpcon β complex double, main character energy ππ
subroutine ztpcon (character norm, character uplo, character diag, integer n, complex*16, dimension( * ) ap, double precision rcond, complex*16, dimension( * ) work, double precision, dimension( * ) rwork, integer info)
β¨ Purpose
ZTPCON estimates the reciprocal of the condition number of a packed triangular matrix A, in either the 1-norm or the infinity-norm. The full double-complex package, no cap π«Ά.
The norm of A gets computed, an estimate is obtained for norm(inv(A)), and then the reciprocal of the condition number is:
RCOND = 1 / ( norm(A) * norm(inv(A)) ).
Literal perfection β¨
π Parameters
NORMβCHARACTER*1. Which one do you need?= '1' or 'O':1-norm π₯= 'I':Infinity-norm β
UPLOβCHARACTER*1.= 'U':Ais upper triangular πΈ= 'L':Ais lower triangular π¦
DIAGβCHARACTER*1.= 'N':Ais non-unit triangular= 'U':Ais unit triangular
NβINTEGER. The order of the matrixA. Constraint:N >= 0β¨APβCOMPLEX*16array, dimension(N*(N+1)/2). The upper or lower triangular matrixA, packed columnwise in a linear array. Thej-th column ofAis stored in the arrayAPas follows:- if
UPLO = 'U',AP(i + (j-1)*j/2) = A(i,j)for1<=i<=j; - if
UPLO = 'L',AP(i + (j-1)*(2n-j)/2) = A(i,j)forj<=i<=n.
- if
If DIAG = 'U', the diagonal elements of A are not referenced and are assumed to be 1 π
RCONDβDOUBLE PRECISION. The reciprocal of the condition number of the matrixA, computed asRCOND = 1/(norm(A) * norm(inv(A)))πWORKβCOMPLEX*16array, dimension(2*N)π§RWORKβDOUBLE PRECISIONarray, dimension(N)πΈINFOβINTEGER.= 0:successful exit β¨< 0:ifINFO = -i, thei-th argument had an illegal value π¨
π Author
Univ. of Tennessee
Univ. of California Berkeley
Univ. of Colorado Denver
NAG Ltd.
Definition at line 128 of file ztpcon.f.
π«Ά Author
Generated automatically by Doxygen for LAPACK from the source code. (Yes bestie, a robot wrote this one β and it ate π€β¨)
And that's the whole page, thank you for coming to my TED talk π This little masterpiece is LAPACK Version 3.12.0, and you're reading tpcon(3). Go forth and estimate your condition numbers responsibly, queen πππ
π¦ where this page comes from
- arch rolling: lapack-doc 3.12.1-2 Β· custom
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms