@ -30,37 +30,30 @@ Each directory has the name of the computational kernel routines described in th
In this test suite were considered only computational routines implemented by PSBLAS, according to the version 3.9 of the documentation. In the following table are reported all the kernels, their implementation and wheter or not they were tested yet.
|**Kernel**| **PSBLAS Subroutine**|**Description**|**Single Process Test**|**Multi-Process Test**|**Complex Test**|**GPU Test**|
|**General Dense Matrix Sum**| `psb_geaxpby`| This subroutine is an interface to the computational kernel for dense matrix sum: $Y \leftarrow \alpha X + \beta Y$|Work in progress :hammer_and_wrench:|Work in progress :hammer_and_wrench:|No ❌|No ❌|
| **Dot product**|`psb_gedot`|This function computes dot product between two vectors x and y.$dot \leftarrow x^T y$If x and y are real vectors it computes dot-product as:$dot \leftarrow x^H y$|Yes ✅|Yes ✅|No ❌|No ❌|
|**General Dense Matrix Sum**| `psb_geaxpby`| This subroutine is an interface to the computational kernel for dense matrix sum: $Y \leftarrow \alpha X + \beta Y$|Work in progress :hammer_and_wrench:|Work in progress :hammer_and_wrench:|No ❌|No ❌|
| **Dot product**|`psb_gedot`|This function computes dot product between two vectors x and y.$dot \leftarrow x^T y$If x and y are real vectors it computes dot-product as:$dot \leftarrow x^H y$|Yes ✅|Yes ✅|No ❌|No ❌|
| **Generalized Dot Product** |`psb_gedots`|This subroutine computes a series of dot products among the columns of two dense matrices x and y:$res(i) \leftarrow x(:,i)^T y(:,i)$If the matrices are complex, then the usual convention applies, i.e. the conjugate transpose of x is used. If x and y are of rank one, then res is a scalar, else it is a rank one array.|No ❌|No ❌|No ❌|No ❌|
|**Infinity-Norm of Vector**|`psb_normi`/`psb_geamax`|This function computes the infinity-norm of a vector x. If x is a real vector it computes infinity norm as:$amax \leftarrow max \mid x_i \mid$else if x is a complex vector then it computes the infinity-norm as:$amax \leftarrow max(\mid re(x_i) \mid + \mid im(x_i) \mid)$|No ❌|No ❌|No ❌|No ❌|
|**Generalized Infinity Norm**|`psb_geamaxs`|This subroutine computes a series of infinity norms on the columns of a dense matrix x:$res(i) \leftarrow max_k \mid x(k,i) \mid$|No ❌|No ❌|No ❌|No ❌|
| **1-Norm of Vector**| `psb_norm1` / `psb_geasums`|This function computes the 1-norm of a vector x. If x is a real vector it computes 1-norm as:$asum \leftarrow \mid \mid x_i \mid \mid$else if x is a complex vector then it computes 1-norm as:$asum \leftarrow \mid \mid re(x) \mid \mid_1 + \mid \mid im(x) \mid \mid_1$|No ❌|No ❌|No ❌|No ❌|
|**Generalized 1-Norm of Vector**|`psb_geasums`|This subroutine computes a series of 1-norms on the columns of a dense matrix x:$res(i) \leftarrow max_k \mid x(k,i) \mid$This function computes the 1-norm of a vector x. If x is a real vector it computes 1-norm as:$res(i) \leftarrow \mid \mid x_i \mid \mid$else if x is a complex vector then it computes 1-norm as:$res(i) \leftarrow \mid \mid re(x) \mid \mid_\ + \mid \mid im(x) \mid \mid_1$|No ❌|No ❌|No ❌|No ❌|
| **2-Norm of Vector**|`psb_norm2` / `psb_genrm2`| This function computes the 2-norm of a vector x. If x is a real vector it computes 2-norm as:$nrm2 \leftarrow \sqrt{x^T x}$else if x is a complex vector then it computes 2-norm as:$nrm2 \leftarrow \sqrt{x^H x}$|No ❌|No ❌|No ❌|No ❌|
|**Generalized 2-Norm of Vector**|`psb_genrm2s` / `psb_spnrm1` |This subroutine computes a series of 2-norms on the columns of a dense matrix x:$res(i) \leftarrow \mid \mid x(:,i) \mid \mid_2$|No ❌|No ❌|No ❌|No ❌|
|**1-Norm of Sparse Matrix**|`psb_norm1`|This function computes the 1-norm of a matrix A:$nrm1 \leftarrow \mid \mid A \mid \mid_1$where A represents the global matrix A|No ❌|No ❌|No ❌|No ❌|
|**Infinity Norm of Sparse Matrix**|`psb_normi` / `psb_spnrmi`|This function computes the infinity-norm of a matrix A:$nrmi \leftarrow \mid \mid A \mid \mid_{\infty}$where: A represents the global matrix A|No ❌|No ❌|No ❌|No ❌|
|**Sparse Matrix by Dense Matrix Product**| `psb_spmm`|This subroutine computes the Sparse Matrix by Dense Matrix Product:$y \leftarrow \alpha A x + \beta y$$y \leftarrow \alpha A^T x + \beta y$$y \leftarrow \alpha A^H x + \beta y$where: <br> x is the global dense matrix x_{:,:} <br> y is the global dense matrix y_{:,:} <br> A is the global sparse matrix A|Work in progress :hammer_and_wrench:|No ❌|No ❌|No ❌|
|**Triangular System Solve**|`psb_spsm`|This subroutine computes the Triangular System Solve:$y \leftarrow \alpha T^{-1} x + \beta y$$y \leftarrow \alpha D^{-1} x + \beta y$$y \leftarrow \alpha T^{-1} D x + \beta y$$y \leftarrow \alpha T^{-T} x + \beta y$$y \leftarrow \alpha D T^{-T} x + \beta y$$y \leftarrow \alpha T^{-T} D x + \beta y$$y \leftarrow \alpha T^{-H} x + \beta y$$y \leftarrow \alpha D T^{-H} x + \beta y$$y \leftarrow \alpha T^{-H} D x + \beta y$where: <br> x is the global dense matrix x_{:,:} <br> y is the global dense matrix y_{:,:} <br> T is the global sparse block triangular submatrix T<br> D is the scaling diagonal matrix|No ❌|No ❌|No ❌|No ❌|
|**Entrywise Product**|`psb_gemlt`|This function computes the entrywise product between two vectors x and y$dot \leftarrow x(i)y(i)$|No ❌|No ❌|No ❌|No ❌|
|**Entrywise Division**|`psb_gediv`|This function computes the entrywise division between two vectors x and y$div \leftarrow \frac{x(i)}{y(i)}$|No ❌|No ❌|No ❌|No ❌|
|**Entrywise Inversion**|`psb_geinv`|This function computes the entrywise inverse of a vector x and puts it into y$inv \leftarrow \frac{1}{x(i)}$|No ❌|No ❌|No ❌|No ❌|
|**Infinity-Norm of Vector**|`psb_normi`/`psb_geamax`|This function computes the infinity-norm of a vector x. If x is a real vector it computes infinity norm as:$amax \leftarrow max \mid x_i \mid$else if x is a complex vector then it computes the infinity-norm as:$amax \leftarrow max(\mid re(x_i) \mid + \mid im(x_i) \mid)$|No ❌|No ❌|No ❌|No ❌|
|**Generalized Infinity Norm**|`psb_geamaxs`|This subroutine computes a series of infinity norms on the columns of a dense matrix x:$res(i) \leftarrow max_k \mid x(k,i) \mid$|No ❌|No ❌|No ❌|No ❌|
| **1-Norm of Vector**| `psb_norm1` / `psb_geasums`|This function computes the 1-norm of a vector x. If x is a real vector it computes 1-norm as:$asum \leftarrow \mid \mid x_i \mid \mid$else if x is a complex vector then it computes 1-norm as:$asum \leftarrow \mid \mid re(x) \mid \mid_1 + \mid \mid im(x) \mid \mid_1$|No ❌|No ❌|No ❌|No ❌|
|**Generalized 1-Norm of Vector**|`psb_geasums`|This subroutine computes a series of 1-norms on the columns of a dense matrix x:$res(i) \leftarrow max_k \mid x(k,i) \mid$This function computes the 1-norm of a vector x. If x is a real vector it computes 1-norm as:$res(i) \leftarrow \mid \mid x_i \mid \mid$else if x is a complex vector then it computes 1-norm as:$res(i) \leftarrow \mid \mid re(x) \mid \mid_\ + \mid \mid im(x) \mid \mid_1$|No ❌|No ❌|No ❌|No ❌|
| **2-Norm of Vector**|`psb_norm2` / `psb_genrm2`| This function computes the 2-norm of a vector x. If x is a real vector it computes 2-norm as:$nrm2 \leftarrow \sqrt{x^T x}$else if x is a complex vector then it computes 2-norm as:$nrm2 \leftarrow \sqrt{x^H x}$|No ❌|No ❌|No ❌|No ❌|
|**Generalized 2-Norm of Vector**|`psb_genrm2s` / `psb_spnrm1` |This subroutine computes a series of 2-norms on the columns of a dense matrix x:$res(i) \leftarrow \mid \mid x(:,i) \mid \mid_2$|No ❌|No ❌|No ❌|No ❌|
|**1-Norm of Sparse Matrix**|`psb_norm1`|This function computes the 1-norm of a matrix A:$nrm1 \leftarrow \mid \mid A \mid \mid_1$where A represents the global matrix A|No ❌|No ❌|No ❌|No ❌|
|**Infinity Norm of Sparse Matrix**|`psb_normi` / `psb_spnrmi`|This function computes the infinity-norm of a matrix A:$nrmi \leftarrow \mid \mid A \mid \mid_{\infty}$where: $A$ represents the global matrix|No ❌|No ❌|No ❌|No ❌|
|**Sparse Matrix by Dense Matrix Product**| `psb_spmm`|This subroutine computes the Sparse Matrix by Dense Matrix Product:$y \leftarrow \alpha A x + \beta y$$y \leftarrow \alpha A^T x + \beta y$$y \leftarrow \alpha A^H x + \beta y$where: <br>$x$ is the global dense matrix $x_{:,:}$<br> y is the global dense matrix $y_{:,:}$<br>$A$ is the global sparse matrix |Work in progress :hammer_and_wrench:|No ❌|No ❌|No ❌|
|**Triangular System Solve**|`psb_spsm`|This subroutine computes the Triangular System Solve:$y \leftarrow \alpha T^{-1} x + \beta y$$y \leftarrow \alpha D^{-1} x + \beta y$$y \leftarrow \alpha T^{-1} D x + \beta y$$y \leftarrow \alpha T^{-T} x + \beta y$$y \leftarrow \alpha D T^{-T} x + \beta y$$y \leftarrow \alpha T^{-T} D x + \beta y$$y \leftarrow \alpha T^{-H} x + \beta y$$y \leftarrow \alpha D T^{-H} x + \beta y$$y \leftarrow \alpha T^{-H} D x + \beta y$where: <br>$x$ is the global dense matrix $x_{:,:}$<br>$y$ is the global dense matrix $y_{:,:}$<br>$T$ is the global sparse block triangular submatrix <br> D is the scaling diagonal matrix|No ❌|No ❌|No ❌|No ❌|
|**Entrywise Product**|`psb_gemlt`|This function computes the entrywise product between two vectors $x$ and $y$ $dot \leftarrow x(i)y(i)$|No ❌|No ❌|No ❌|No ❌|
|**Entrywise Division**|`psb_gediv`|This function computes the entrywise division between two vectors $x$ and $y$ $div \leftarrow \frac{x(i)}{y(i)}$|No ❌|No ❌|No ❌|No ❌|
|**Entrywise Inversion**|`psb_geinv`|This function computes the entrywise inverse of a vector x and puts it into $y$ $inv \leftarrow \frac{1}{x(i)}$|No ❌|No ❌|No ❌|No ❌|
## Developer Notes
In order to keep compliant the excecution of the bash script used to automate the teest excecution, remember to create a new directory to put new tests and to use the name convention of psb_test_ signature for utilities functions and psb_kernel_test for tests used for new routines.
## TODO
- Merge all the output logs
- Finish the directories description
- Check memory occupancy of parallel/ serial/ and vectors/ directories (Maybe not the best way for lots of rputines?)
## Questions
- Is it correct to use psb_gather even for a single process running?
- Is it correct to shift in 0,xxxx type of notation to compare with the correct number of significand digits?
- Does it make sense to compare the parallel solution with the serial one?
## References
<aid="testing">[1].</a> Higham, Nicholas J. Testing linear algebra software. Springer US, 1997
This is a directory developed by Luca Pepè Sciarria and Simone Staccone froma Tor Vergata University to start to create some unit tests for PSBLAS 3.9, in particular for ```psb_gedot``` routine.
This is a directory developed by Luca Pepè Sciarria and Simone Staccone from Tor Vergata University to start to create some unit tests for PSBLAS 3.9, in particular for ```psb_gedot``` routine.
## Getting started
Steps to reproduce the tests:
- Compile the code using ``` make ``` (Optional)
- Launch the script ./autotest.sh or with source ./autotest.sh if you want to add modules to the .bashrc file permenently.
- Check the output log file psblas_gedot_test.log to collect results
- Launch the script ```./autotest.sh``` or with ```source ./autotest.sh``` if you want to add modules to the .bashrc file permanently.
- Check the output log file ```psb_gedot_test.log``` to collect results and check for errors. In any case a summarization of tests passed should be shown in stdout.
NOTE: If the code is changed and a new compilation is needed to show the changes, the autotest.sh script isn't aware of this scenario, therefore it is necessary to manually recompile the code.
NOTE: If the code is changed and a new compilation is needed to show the changes, the ```autotest.sh``` script isn't aware of this scenario, therefore it is necessary to manually recompile the code. Moreover, if you are running manually the script and not launching the main ```test.sh``` script, be careful to use the last compiled version of the utils.
## Test Suite
### Overall Analysys
The ```psb_gedot```. The signature of the function is:
The ```psb_gedot``` is a function that performs the scalar product between two vectors giving a value as result. The signature of the function is:
```fortran
psb_gedot(x, y, desc_a, info [,global])
```
In the comparison 7 significand digits means having a notation like 0,$d_1 d_2 d_3 ... d_7$*10^7 also.
The strategy to validate the correctness of the computation is to compare single precision result and double precision result in the test cases in which the test should not give an error. In this way it is possible to have a correctness check of the computation comparing the two results considering a number of significant digits which is tuned on the single precision computation.
### Parameters Values
**x** vectors are located in the vectors/ directory. They are generated randomly using the same seed and then saved on different files based on their characteristics. The size of the vector is choosen accordingly to the size of the matrix column space considered for the single test instance.
**y** vectors are located in the vectors/ directory. They are generated randomly using the same seed and then saved on different files based on their characteristics. The size of the vector is choosen accordingly to the size of the matrix rows space considered for the single test instance.
**global** logical value indicating if the computation is global among all processes or it is only local, so that a global reduction is needed afterwards.
|Variable|Value|Value Description|
|:-:|:-:|:-:|
|$\beta_1$|1.0|Positive value|
|$\beta_2$|-1.0|Negative value|
|$\beta_3$|0.0|Null value|
## Output
The ouput files generated by the test are automatically compared by the autotest.sh script, but if it is needed to manually run the test here it is the naming convenction used.
The results of the computation will be saved on different files based on the instance of the test considered. In particular the naming conventiona format the output file as sol_x#_y#_a#_b#.mtx, where each # is a number choosen w.r.t. the test instance. (Ex. sol_x1_y1_a1_b1.mtx is the solution computed using the first x vector file , the first y vector file, alpha = 1.0 and beta = 1.0). Moreover, the files will be saved in the serial/ directory if the program is launched using 1 process or in parrallel/ directory if the program is launched with more than one process.
|$global_1$|True|Global computation is required implicitly|
|$global_2$|False|Global computation is required explicitly, so only a local computation is done|
## TODO
- Use also global in different ways
List of things still to add in the test:
- Add computation with broken descriptor and catch the errore result
- Test using complex data ($dot \leftarrow x^H \cdot y$)
- Test also GPU excecution
- Try multiple distributions
- Fix result_check handling, it should not be an entire vector
Stack-1
10 months ago