612a924ea3
Reformat spoutput.c
Replace tabs by spaces
1 year ago
46e4a67079
Some cosmetics for readability:
replace tabs by spaces
1 year ago
243bdff377
Formatting: tabs to spaces
1 year ago
72d70e3471
bool definition is in stdbool.h, no local definition required
1 year ago
deb3cd9809
Replace all BOOLEAN, BOOL, _Bool by bool
Remove all #undef bool (set in conjunction with #iclude <Windows.h>)
1 year ago
54865ed39d
bring the KLU_malloc() in right order, ticket #717
1 year ago
7b815c743e
intel cc see obsolete }; as empty declaration
1 year ago
3528156d14
Fix bug #680 . Check that src/dest memcpy arguments are non-NULL.
2 years ago
f495431509
Add a comment
2 years ago
83e8cb3715
Remove VS compiler warning
2 years ago
d01868b725
Cleanup gcc compiler warnings.
2 years ago
adaa3d39fc
no use for grouping in cx_fft and cx_ifft
2 years ago
8aa20cfc98
fft window functions back to correct scaling - no need need for post scaling step
2 years ago
9e41383dd9
window weightings between 0 and 1 for fft vector function and command
2 years ago
aafd7bbb42
correct bartlett/triangle fft window formulae
2 years ago
0141473aa4
Don't derefence Matrix->SMPkluMatrix->KLUmatrixCommon if it is NULL.
Test for NULL moved upwards in front of dereferencing.
2 years ago
f4c560ba71
Add more KLU headers to enable their distribution
2 years ago
3cf8fe8c41
Stop after last data point, preventing read overrun of nscale.
2 years ago
d82f948832
Fix a crash found by Brian Taylor: when .plot attempts to plot digital
node history, interpolation may produce an infinite value at digital edges.
Remove vertical edges when interpolating and make some other improvements:
do not calculate a polynomial approximation for unused frames;
center the target x-value in the frame; and do not propogate a reduction
in degree to later frames.
2 years ago
0735b8d0a5
fix previous commit on repeating error messages
2 years ago
55ba927973
If nan occurs during op calculation, issue the warning
no more than 10 times.
2 years ago
2cf2693706
few code cleanings
2 years ago
0a087809d6
If both node names are the same, just print them once in the error message.
2 years ago
a6d5ce8ea6
Some KLU warnings are useless for the normal user, as she or he
does not have any means to further analyze or repair the issue:
Warning: KLU ReFactor failed. Factoring again...
Warning (ReFactor Complex): KLU Matrix is SINGULAR
Numerical Rank: %d\n
Singular Node: %d\n
So print these messages only in debug mode.
3 years ago
4864f0a118
Remove crash in CIDER example meclgate.cir
3 years ago
05aa9f0927
Better fix for previous commit
3 years ago
13e74751cc
Fixed a MLK for XSPICE and commented out an unused array
3 years ago
cab5e8cd37
Added the case of VSRCport to VSRC model for KLU
3 years ago
b08eb2266a
Added initial KLU support for node collapsing
3 years ago
470cd4a514
Fixed KLU SINGULAR matrix handling and messaging
3 years ago
d68e83be78
Fixed VBIC model for KLU
Fixed SMPfindElt messages
3 years ago
8123fde6f7
Handled properly KLU messages when XSPICE is purely digital
3 years ago
4200362f42
Fixed Empty Matrix handling in KLU
3 years ago
937325e2ce
Handled an empty circuit with KLU
3 years ago
6b6e9bb7cb
Enable log output (warning and error messages) on the Windows GUI
4 years ago
caf1ee7c47
Fixed a crash when SPARSE is used
6 years ago
3c15993fe5
Fixed the case when DiagGmin has to not be loaded again and the case when refactoring returns 1
6 years ago
12288f7be6
Fixed KLU memory leak during SMPdestroy
6 years ago
4639e936a3
Fixed the dimension of the COO Trash Can
6 years ago
613ae18db9
Fixed a memory leak
6 years ago
d5434a2df2
fix a memory leak
ckt->CKTmatrix is now malloced, so has to bee freed as well
6 years ago
967ee13940
Added a comment in NIiter
6 years ago
1110019ff2
Changed the PivTol in SMPreorder with PivRel
6 years ago
3b1f5e6122
Changed the Error message in klusmp.c to Information
6 years ago
fe89da2b2c
Getting rid of SPARSE for Initial Circuit Matrix creation. Now KLU is totally independent from SPARSE. This opens up the interface for other solvers as well
SPARSE can still used through '.options sparse', but it isn't possible to switch from KLU to SPARSE and viceversa at run-time
6 years ago
49f6444957
Added "klu_memgrow_factor" option to control the KLU memory grow factor
6 years ago
5513f6e5f6
Handled the SINGULAR matrix case in a more clever way!
6 years ago
fbd8d94cdf
Fixed again the NIiter loop for KLU
6 years ago
710341bbfd
Fixed KLU returns values so that NIiter can Factor again, if the matrix is numerically singular from the refactorization
Fixed the NIiter code to perform a factorization without reloading the matrix
6 years ago
77fa5c06a9
Fixed the KLU returns values for Factorization and ReFactorization.
If the matrix is Numerically Singular, continue the factorization till the end
6 years ago
Holger Vogt
dwarning
dwarning
Brian Taylor
Giles Atkinson
Francesco Lannutti
Francesco Lannutti