Removes unused SMIN from bdsqr

[weslleyspereira]

Closes #243.

xBDSQR seems to be using SMINL to test if shifting would ruin relative accuracy. Therefore, SMIN is not necessary at:

lapack/SRC/dbdsqr.f

Lines 448 to 465 in dfad0d5

	*
	* Find diagonal block of matrix to work on
	*
	IF( TOL.LT.ZERO .AND. ABS( D( M ) ).LE.THRESH )
	$ D( M ) = ZERO
	SMAX = ABS( D( M ) )
	SMIN = SMAX
	DO 70 LLL = 1, M - 1
	LL = M - LLL
	ABSS = ABS( D( LL ) )
	ABSE = ABS( E( LL ) )
	IF( TOL.LT.ZERO .AND. ABSS.LE.THRESH )
	$ D( LL ) = ZERO
	IF( ABSE.LE.THRESH )
	$ GO TO 80
	SMIN = MIN( SMIN, ABSS )
	SMAX = MAX( SMAX, ABSS, ABSE )
	70 CONTINUE

This was noted by @TarcioV at #243. Thanks!

[codecov]

Codecov Report

Patch and project coverage have no change.

Comparison is base (1fafb88) 0.00% compared to head (fb67be3) 0.00%.

❗ Current head fb67be3 differs from pull request most recent head 8b27cef. Consider uploading reports for the commit 8b27cef to get more accurate results

Additional details and impacted files

@@           Coverage Diff            @@
##           master     #832    +/-   ##
========================================
  Coverage    0.00%    0.00%            
========================================
  Files        1908     1908            
  Lines      187072   186954   -118     
========================================
+ Misses     187072   186954   -118     

Impacted Files	Coverage Δ
SRC/cbdsqr.f	0.00% <0.00%> (ø)
SRC/cgbsvx.f	0.00% <ø> (ø)
SRC/cgebal.f	0.00% <0.00%> (ø)
SRC/cgees.f	0.00% <ø> (ø)
SRC/cgeesx.f	0.00% <ø> (ø)
SRC/cgeev.f	0.00% <ø> (ø)
SRC/cgeevx.f	0.00% <ø> (ø)
SRC/cgejsv.f	0.00% <ø> (ø)
SRC/cgels.f	0.00% <ø> (ø)
SRC/cgelsd.f	0.00% <ø> (ø)
... and 227 more

... and 16 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[angsch]

xBDSQR seems to be using SMINL to test if shifting would ruin relative accuracy.

I agree, smin does not make any sense. My understanding of the algorithm is that smax and and sminl are over- and underestimators of the singular values. In view of the comment

lapack/SRC/dbdsqr.f

Line 385 in bd82aa0

* Compute approximate maximum, minimum singular values

I wonder if it makes sense to rename sminl to smin to have a matching name pair. I suspect that the name sminl was an emergency solution since smin was already taken.

In any case, it may be good to remove the variable that becomes unused from the list of "local scalars".

[weslleyspereira]

Thanks @angsch! Your comments make complete sense to me. I went ahead and applied your suggestions regarding SMIN and SMINL.