#2364: avoid opening and closing HDF5 files multiple times during output
Reporter: Roland Haas
Status: open
Milestone:
Version:
Type: enhancement
Priority: minor
Component:
Comment (by Roland Haas):
Please review. Or close and reject the pull request.
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2364/avoid-opening-and…
#2837: Building the Einstein Toolkit
Reporter: Suo-Ning Wang
Status: open
Milestone:
Version: development version
Type: bug
Priority: major
Component: EinsteinToolkit website
Comment (by Roland Haas):
`libsz.so` \(but not eg `libsz.so.1`\) should also be sufficient I think.
To remove PETSc and Silo, yes please remove the respective lines that contain `ExternalLibraries/Silo` and `ExternalLibraries/PETSc` \(the latter may be commented out using a `#DISABLED` . I’d remove it anyway because there is some functionality in simfactory that can enable `#DISABLED` thorns.\)
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2837/building-the-eins…
#2845: support (again) filesystem in std::experimental in Silo output
Reporter: Roland Haas
Status: open
Milestone:
Version:
Type: enhancement
Priority: major
Component: EinsteinToolkit thorn
Changes (by Roland Haas):
status: open (was new)
Comment (by Roland Haas):
Please review
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2845/support-again-fil…
#2845: support (again) filesystem in std::experimental in Silo output
Reporter: Roland Haas
Status: new
Milestone:
Version:
Type: enhancement
Priority: major
Component: EinsteinToolkit thorn
CarpetX’s Silo output \(and only it\) uses `std::filesystem`which is defined in C\+\+17 but not all compilers fully support it, RoCM on Frontier uses `std::experimental::filesystem` instead.
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2845/support-again-fil…
#2844: HelloWorldCuda
Reporter: Anthony Shoup
Status: submitted
Milestone: ET_2024_11
Version:
Type: bug
Priority: minor
Component: EinsteinToolkit thorn
The HelloWorldCuda thorn does not seem to work in my build of the Einstein Toolkit, ET\_2024\_11 release. The routine, HelloWorldCUDA\_evol, that launches the cuda kernel uses cctk\_lsh grid size values that are very large \(x-dir > 1x10^9\), which then tries to launch too many kernels. Below is the code:
extern "C"
void HelloWorldCUDA\_evol\(CCTK\_ARGUMENTS\)
\{
DECLARE\_CCTK\_ARGUMENTS;
```
const int val1 = cctk_iteration;
const int val2 = 3;
int res = 42; // poison
const dim3 blockDim(4, 4, 4);
```
// const dim3 gridDim\(\(cctk\_lsh\[0\] \+ blockDim.x - 1\) / blockDim.x, // Commented out by ALS
// \(cctk\_lsh\[1\] \+ blockDim.y - 1\) / blockDim.y,
// \(cctk\_lsh\[2\] \+ blockDim.z - 1\) / blockDim.z\);
```
// This code does work - ALS
const dim3 gridDim((8 + blockDim.x - 1) / blockDim.x,
(8 + blockDim.y - 1) / blockDim.y,
(8 + blockDim.z - 1) / blockDim.z);
add<<<gridDim, blockDim>>>(val1, val2, &res);
cudaError_t cerr = cudaDeviceSynchronize(); // Added by ALS
check_error(cerr, "Could not do cuda device synchronize"); // Added by ALS
CCTK_VInfo(CCTK_THORNSTRING, "CUDA says(testing): %d + %d = %d", val1, val2, res);
```
\}
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2844/helloworldcuda