Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Tester fixes #183

Merged
merged 3 commits into from
May 27, 2024
Merged

Tester fixes #183

merged 3 commits into from
May 27, 2024

Conversation

mgates3
Copy link
Collaborator

@mgates3 mgates3 commented May 27, 2024

In run_tests, renames cond to condest, and add cond option.
Fix testers for herk, syrk, her2k, syr2k to take n-by-k. It was passing m-by-n.

For xsmall matrix sizes, CholQR needs a smaller condition number to pass. Previously failing runs:

./run_tests.py --xsmall --type s,d,z
...
--------------------------------------------------------------------------------
./tester  --origin s --target t --ref n --nb 5,8 --lookahead 1 --dim 10 --dim 20x10 --trans n,c --method-gels cholqr --matrix svd --cond 1e3 --type s,c gels
% SLATE version 2023.11.05, id de5d2ec2
% input: ./tester --origin s --target t --ref n --nb 5,8 --lookahead 1 --dim 10 --dim 20x10 --trans n,c --method-gels cholqr --matrix svd --cond 1e3 --type s,c gels
% 2024-05-27 11:44:42, 1 MPI ranks, CPU-only MPI, 4 OpenMP threads per MPI rank
                                                                                                                                                                                                          
 A   B   C  type  origin  target  cholQR    gels    trans       m       n    nrhs    nb  ib    p    q  la  pt  leastsqr  min norm  residual   time (s)       gflop/s  ref time (s)   ref gflop/s  status  
 1   2   2     s  scalpk    task    auto  CholQR  notrans      10      10      10     5  32    1    1   1   2  2.55e-06        NA  5.34e-07   0.000303        0.0161            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR  notrans      10      10      10     8  32    1    1   1   2  7.14e-06        NA  1.14e-06   0.000263        0.0186            NA            NA  FAILED  
 1   2   2     s  scalpk    task    auto  CholQR  notrans      20      10      10     5  32    1    1   1   2  3.50e-07        NA  3.34e-07   0.000266        0.0413            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR  notrans      20      10      10     8  32    1    1   1   2  1.55e-06        NA  6.43e-07   0.000290        0.0379            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR     conj      10      10      10     5  32    1    1   1   2  9.53e-07        NA  5.38e-07   0.000179        0.0273            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR     conj      10      10      10     8  32    1    1   1   2  6.91e-07        NA  1.85e-07   0.000343        0.0142            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR     conj      20      10      10     5  32    1    1   1   2        NA  8.31e-07  1.24e-07   0.000296        0.0371            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR     conj      20      10      10     8  32    1    1   1   2        NA  2.80e-07  1.63e-07   0.000278        0.0395            NA            NA  pass    

 1   2   2     c  scalpk    task    auto  CholQR  notrans      10      10      10     5  32    1    1   1   2  3.09e-06        NA  5.99e-07   0.000275        0.0734            NA            NA  FAILED  
 1   2   2     c  scalpk    task    auto  CholQR  notrans      10      10      10     8  32    1    1   1   2  1.84e-06        NA  4.45e-07   0.000203        0.0994            NA            NA  pass    
 1   2   2     c  scalpk    task    auto  CholQR  notrans      20      10      10     5  32    1    1   1   2  1.26e-06        NA  5.09e-07   0.000216         0.207            NA            NA  pass    
 1   2   2     c  scalpk    task    auto  CholQR  notrans      20      10      10     8  32    1    1   1   2  1.17e-06        NA  3.46e-07   0.000261         0.172            NA            NA  pass    
 1   2   2     c  scalpk    task    auto  CholQR     conj      10      10      10     5  32    1    1   1   2  3.34e-06        NA  1.05e-06   0.000204        0.0989            NA            NA  FAILED  
 1   2   2     c  scalpk    task    auto  CholQR     conj      10      10      10     8  32    1    1   1   2  2.61e-06        NA  4.55e-07   0.000185         0.109            NA            NA  pass    
 1   2   2     c  scalpk    task    auto  CholQR     conj      20      10      10     5  32    1    1   1   2        NA  1.87e-06  2.14e-07   0.000191         0.234            NA            NA  pass    
 1   2   2     c  scalpk    task    auto  CholQR     conj      20      10      10     8  32    1    1   1   2        NA  1.44e-06  4.01e-07   0.000227         0.197            NA            NA  pass    

% Matrix kinds:
%  1: svd, cond = 1000
%  2: rand, cond unknown

% 3 tests FAILED: gels
FAILED: gels, exit code 3
--------------------------------------------------------------------------------
./tester  --origin s --target t --ref n --nb 5,8 --lookahead 1 --dim 10 --dim 20x10 --trans n,c --method-gels cholqr --matrix svd --cond 1e3 --type d,z gels
% SLATE version 2023.11.05, id de5d2ec2
% input: ./tester --origin s --target t --ref n --nb 5,8 --lookahead 1 --dim 10 --dim 20x10 --trans n,c --method-gels cholqr --matrix svd --cond 1e3 --type d,z gels
% 2024-05-27 11:44:42, 1 MPI ranks, CPU-only MPI, 4 OpenMP threads per MPI rank
                                                                                                                                                                                                          
 A   B   C  type  origin  target  cholQR    gels    trans       m       n    nrhs    nb  ib    p    q  la  pt  leastsqr  min norm  residual   time (s)       gflop/s  ref time (s)   ref gflop/s  status  
 1   2   2     d  scalpk    task    auto  CholQR  notrans      10      10      10     5  32    1    1   1   2  1.49e-15        NA  3.43e-16   0.000292        0.0167            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR  notrans      10      10      10     8  32    1    1   1   2  2.97e-15        NA  6.80e-16   0.000286        0.0171            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR  notrans      20      10      10     5  32    1    1   1   2  4.70e-16        NA  2.24e-16   0.000239        0.0459            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR  notrans      20      10      10     8  32    1    1   1   2  7.68e-16        NA  2.78e-16   0.000269        0.0408            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR     conj      10      10      10     5  32    1    1   1   2  1.14e-15        NA  6.25e-16   0.000218        0.0224            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR     conj      10      10      10     8  32    1    1   1   2  2.28e-15        NA  4.91e-16   0.000245        0.0199            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR     conj      20      10      10     5  32    1    1   1   2        NA  1.59e-15  4.44e-16   0.000285        0.0385            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR     conj      20      10      10     8  32    1    1   1   2        NA  1.37e-15  4.10e-16   0.000234        0.0469            NA            NA  pass    

 1   2   2     z  scalpk    task    auto  CholQR  notrans      10      10      10     5  32    1    1   1   2  6.01e-15        NA  1.42e-15   0.000255        0.0791            NA            NA  FAILED  
 1   2   2     z  scalpk    task    auto  CholQR  notrans      10      10      10     8  32    1    1   1   2  4.24e-15        NA  9.86e-16   0.000220        0.0917            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR  notrans      20      10      10     5  32    1    1   1   2  3.14e-15        NA  5.72e-16   0.000266         0.168            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR  notrans      20      10      10     8  32    1    1   1   2  1.44e-15        NA  6.90e-16   0.000206         0.217            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR     conj      10      10      10     5  32    1    1   1   2  5.58e-15        NA  8.71e-16   0.000191         0.106            NA            NA  FAILED  
 1   2   2     z  scalpk    task    auto  CholQR     conj      10      10      10     8  32    1    1   1   2  1.17e-14        NA  2.13e-15   0.000192         0.105            NA            NA  FAILED  
 1   2   2     z  scalpk    task    auto  CholQR     conj      20      10      10     5  32    1    1   1   2        NA  1.35e-15  2.89e-16   0.000193         0.232            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR     conj      20      10      10     8  32    1    1   1   2        NA  2.61e-15  5.14e-16   0.000209         0.214            NA            NA  pass    

% Matrix kinds:
%  1: svd, cond = 1000
%  2: rand, cond unknown

% 3 tests FAILED: gels
FAILED: gels, exit code 3

Runs now pass:

./run_tests.py --xsmall --type s,d,z
...
--------------------------------------------------------------------------------
./tester  --origin s --target t --ref n --nb 5,8 --type s,d,z --lookahead 1 --dim 10 --dim 20x10 --trans n,c --cond 1e2 --method-gels cholqr --matrix svd gels
% SLATE version 2023.11.05, id de5d2ec2
% input: ./tester --origin s --target t --ref n --nb 5,8 --type s,d,z --lookahead 1 --dim 10 --dim 20x10 --trans n,c --cond 1e2 --method-gels cholqr --matrix svd gels
% 2024-05-27 11:43:33, 1 MPI ranks, CPU-only MPI, 4 OpenMP threads per MPI rank
                                                                                                                                                                                                          
 A   B   C  type  origin  target  cholQR    gels    trans       m       n    nrhs    nb  ib    p    q  la  pt  leastsqr  min norm  residual   time (s)       gflop/s  ref time (s)   ref gflop/s  status  
 1   2   2     s  scalpk    task    auto  CholQR  notrans      10      10      10     5  32    1    1   1   2  4.30e-07        NA  9.36e-08    0.00135       0.00361            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR  notrans      10      10      10     8  32    1    1   1   2  7.25e-07        NA  1.49e-07   0.000307        0.0159            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR  notrans      20      10      10     5  32    1    1   1   2  5.13e-08        NA  3.42e-08   0.000341        0.0322            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR  notrans      20      10      10     8  32    1    1   1   2  3.19e-08        NA  2.36e-08   0.000286        0.0384            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR     conj      10      10      10     5  32    1    1   1   2  8.93e-08        NA  2.14e-08   0.000267        0.0183            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR     conj      10      10      10     8  32    1    1   1   2  3.94e-07        NA  6.77e-08   0.000250        0.0195            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR     conj      20      10      10     5  32    1    1   1   2        NA  6.27e-08  2.19e-08   0.000267        0.0411            NA            NA  pass    
 1   2   2     s  scalpk    task    auto  CholQR     conj      20      10      10     8  32    1    1   1   2        NA  6.94e-08  9.00e-09   0.000216        0.0508            NA            NA  pass    

 1   2   2     d  scalpk    task    auto  CholQR  notrans      10      10      10     5  32    1    1   1   2  6.51e-16        NA  1.41e-16   0.000300        0.0163            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR  notrans      10      10      10     8  32    1    1   1   2  1.24e-15        NA  1.24e-16   0.000244        0.0200            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR  notrans      20      10      10     5  32    1    1   1   2  6.66e-17        NA  3.30e-17   0.000249        0.0441            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR  notrans      20      10      10     8  32    1    1   1   2  1.79e-16        NA  6.66e-17   0.000268        0.0410            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR     conj      10      10      10     5  32    1    1   1   2  8.34e-16        NA  1.01e-16   0.000192        0.0254            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR     conj      10      10      10     8  32    1    1   1   2  3.98e-16        NA  1.61e-16   0.000290        0.0168            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR     conj      20      10      10     5  32    1    1   1   2        NA  1.43e-16  2.44e-17   0.000295        0.0372            NA            NA  pass    
 1   2   2     d  scalpk    task    auto  CholQR     conj      20      10      10     8  32    1    1   1   2        NA  8.80e-17  1.70e-17   0.000171        0.0642            NA            NA  pass    

 1   2   2     z  scalpk    task    auto  CholQR  notrans      10      10      10     5  32    1    1   1   2  3.23e-16        NA  6.96e-17   0.000218        0.0926            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR  notrans      10      10      10     8  32    1    1   1   2  7.27e-16        NA  1.23e-16   0.000154         0.131            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR  notrans      20      10      10     5  32    1    1   1   2  1.00e-16        NA  4.67e-17   0.000165         0.271            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR  notrans      20      10      10     8  32    1    1   1   2  1.16e-16        NA  3.60e-17   0.000154         0.291            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR     conj      10      10      10     5  32    1    1   1   2  3.29e-16        NA  1.46e-16   0.000151         0.134            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR     conj      10      10      10     8  32    1    1   1   2  3.67e-16        NA  1.34e-16   0.000120         0.168            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR     conj      20      10      10     5  32    1    1   1   2        NA  3.20e-16  4.47e-17   0.000152         0.295            NA            NA  pass    
 1   2   2     z  scalpk    task    auto  CholQR     conj      20      10      10     8  32    1    1   1   2        NA  2.47e-16  5.25e-17   0.000132         0.339            NA            NA  pass    

% Matrix kinds:
%  1: svd, cond = 100
%  2: rand, cond unknown

% All tests passed: gels
passed: gels

Copy link
Collaborator Author

@mgates3 mgates3 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Self-reviewed.

@mgates3
Copy link
Collaborator Author

mgates3 commented May 27, 2024

Example failed with seemingly unrelated error:

2024-05-27T16:10:24.3382179Z --------------------------------------------------------------------------------
2024-05-27T16:10:24.3382535Z mpirun -np 4 ./ex06_linear_system_lu s d c z
2024-05-27T16:10:24.3382796Z mpi_size 4, grid_p 2, grid_q 2
2024-05-27T16:10:24.3383048Z rank 0: void test_lu() [with scalar_type = float]
2024-05-27T16:10:24.3383354Z rank 0: void test_lu_factor() [with scalar_type = float]
2024-05-27T16:10:24.3383763Z rank 0: void test_lu_inverse() [with scalar_type = float]
2024-05-27T16:10:24.3384284Z terminate called after throwing an instance of 'std::out_of_range'
2024-05-27T16:10:24.3384596Z   what():  map::at
2024-05-27T16:10:24.3384823Z [6a9f601703b0:1471142] *** Process received signal ***
2024-05-27T16:10:24.3385097Z [6a9f601703b0:1471142] Signal: Aborted (6)
2024-05-27T16:10:24.3385332Z [6a9f601703b0:1471142] Signal code:  (-6)
2024-05-27T16:10:24.3385589Z [6a9f601703b0:1471142] [ 0] /lib64/libc.so.6(+0x54db0)[0x7fe937b81db0]
2024-05-27T16:10:24.3385904Z [6a9f601703b0:1471142] [ 1] /lib64/libc.so.6(+0xa154c)[0x7fe937bce54c]
2024-05-27T16:10:24.3386203Z [6a9f601703b0:1471142] [ 2] /lib64/libc.so.6(raise+0x16)[0x7fe937b81d06]
2024-05-27T16:10:24.3386515Z [6a9f601703b0:1471142] [ 3] /lib64/libc.so.6(abort+0xd3)[0x7fe937b557f3]
2024-05-27T16:10:24.3387161Z [6a9f601703b0:1471142] [ 4] /spack/opt/spack/linux-rocky9-x86_64/gcc-11.4.1/gcc-11.3.0-zpqh477vskt2stxzsmhukoscwoxjazzj/lib64/libstdc++.so.6(+0xa9ada)[0x7fe937f27ada]
2024-05-27T16:10:24.3388097Z [6a9f601703b0:1471142] [ 5] /spack/opt/spack/linux-rocky9-x86_64/gcc-11.4.1/gcc-11.3.0-zpqh477vskt2stxzsmhukoscwoxjazzj/lib64/libstdc++.so.6(+0xb50aa)[0x7fe937f330aa]
2024-05-27T16:10:24.3388932Z [6a9f601703b0:1471142] [ 6] /spack/opt/spack/linux-rocky9-x86_64/gcc-11.4.1/gcc-11.3.0-zpqh477vskt2stxzsmhukoscwoxjazzj/lib64/libstdc++.so.6(+0xb4109)[0x7fe937f32109]
2024-05-27T16:10:24.3389832Z [6a9f601703b0:1471142] [ 7] /spack/opt/spack/linux-rocky9-x86_64/gcc-11.4.1/gcc-11.3.0-zpqh477vskt2stxzsmhukoscwoxjazzj/lib64/libstdc++.so.6(__gxx_personality_v0+0x87)[0x7fe937f32837]
2024-05-27T16:10:24.3390726Z [6a9f601703b0:1471142] [ 8] /spack/opt/spack/linux-rocky9-x86_64/gcc-11.4.1/gcc-11.3.0-zpqh477vskt2stxzsmhukoscwoxjazzj/lib64/libgcc_s.so.1(+0x12464)[0x7fe937d48464]
2024-05-27T16:10:24.3391579Z [6a9f601703b0:1471142] [ 9] /spack/opt/spack/linux-rocky9-x86_64/gcc-11.4.1/gcc-11.3.0-zpqh477vskt2stxzsmhukoscwoxjazzj/lib64/libgcc_s.so.1(_Unwind_Resume+0x12e)[0x7fe937d48e9e]
2024-05-27T16:10:24.3392379Z [6a9f601703b0:1471142] [10] /tmp/slate/slate/install/lib/libslate.so(_ZN5slate10BaseMatrixIfE9tileBcastILNS_6TargetE72EEEvllRKS1_N4blas6LayoutEi+0xe0c)[0x7fe93fbdfe6c]
2024-05-27T16:10:24.3393003Z [6a9f601703b0:1471142] [11] /tmp/slate/slate/install/lib/libslate.so(+0xb1d24d)[0x7fe94030d24d]
2024-05-27T16:10:24.3393669Z [6a9f601703b0:1471142] [12] /spack/opt/spack/linux-rocky9-x86_64/gcc-11.4.1/gcc-11.3.0-zpqh477vskt2stxzsmhukoscwoxjazzj/lib64/libgomp.so.1(+0x180db)[0x7fe937d690db]
2024-05-27T16:10:24.3394522Z [6a9f601703b0:1471142] [13] /spack/opt/spack/linux-rocky9-x86_64/gcc-11.4.1/gcc-11.3.0-zpqh477vskt2stxzsmhukoscwoxjazzj/lib64/libgomp.so.1(+0x205b8)[0x7fe937d715b8]
2024-05-27T16:10:24.3395303Z [6a9f601703b0:1471142] [14] /spack/opt/spack/linux-rocky9-x86_64/gcc-11.4.1/gcc-11.3.0-zpqh477vskt2stxzsmhukoscwoxjazzj/lib64/libgomp.so.1(+0x1daf2)[0x7fe937d6eaf2]
2024-05-27T16:10:24.3395822Z [6a9f601703b0:1471142] [15] /lib64/libc.so.6(+0x9f802)[0x7fe937bcc802]
2024-05-27T16:10:31.8842470Z [6a9f601703b0:1471142] [16] /lib64/libc.so.6(+0x3f450)[0x7fe937b6c450]
2024-05-27T16:10:31.8842936Z [6a9f601703b0:1471142] *** End of error message ***
2024-05-27T16:10:31.8843609Z --------------------------------------------------------------------------

@mgates3 mgates3 merged commit deb1858 into icl-utk-edu:master May 27, 2024
8 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant