Code commits
Casa6
-
Rui Xue 8f5e9581750c4df973277e17e3e5d999f318e7f7
Pull request #1030: CAS-14758: Optimize image processing by caching tool handles in fill_summary_minor to avoid redundant open/close operations
Merge in CASA/casa6 from CAS-14758 to master
* commit 'a46b0fa46be4985558dfe705e9f2eccf577b4aed':
CAS-14758: Handle missing residual images when constructing minor cycle summary
CAS-14758: Optimize image processing by caching tool handles in fill_summary_minor to avoid redundant open/close operations -
Rui Xue 44d9e7387ac3f53b7841b93a583e1354e361c06f
Pull request #1040: CAS-14622
Merge in CASA/casa6 from CAS-14622 to master
* commit '6599d358e517e2dc3f94260a0d8770dd0cc64964':
CAS-14622: Fix out-of-bounds MPI segmentation faults in BriggsCubeWeightor
CAS-14622: Fix ncores-dependent tclean segmentation fault in `BriggsCubeWeightor` -
Akeem Wells 6599d358e517e2dc3f94260a0d8770dd0cc64964
Merge branch 'master' into CAS-14622
-
Bob Garwood 3f49f7503b516f27ec0cf65423a7a9e51c238085
Pull request #1049: CAS-14792
Merge in CASA/casa6 from CAS-14792 to master
* commit 'c4206a386e84fa23a749f5b95753860d36396bd7':
added more text to describe the measures data install step
readme.md uses www.astron.nl now for measures data -
Akeem Wells c4206a386e84fa23a749f5b95753860d36396bd7
Merge branch 'master' into CAS-14792
-
Kumar Golap 4d6918c6411688aeb8ab0b8b75ff610cb4885d89
Pull request #1048: when resetting image center the coordinate system was not being reset
Merge in CASA/casa6 from CAS-14747 to master
* commit '3c5bde23464661fe96ff365766465083f95adafd':
when resetting image center the coordinate system was not being reset -
Bob Garwood 3493dc9839d8a596467d534b28d465c141bc35b2
added more text to describe the measures data install step
-
Bob Garwood 632ee87ba92665433856b99c66e70e590c1fff99
readme.md uses www.astron.nl now for measures data
-
Kumar Golap 0673f3ad43b910fe023c96aea061c23f81f0fb38
Pull request #1047: It was not the precision of converting fromRecord or to Record but numpy print precision.
Merge in CASA/casa6 from CAS-13338 to master
* commit 'dfb7e5241859986be77d9b1c19a09c8e2e5c3747':
It was not the precision of converting fromRecord or to Record but numpy print precision. That I believe fixes the precision of phasecenter for parallel continuum -
Srikrishna Sekhar 50e2cf84c451f0970e69a3aaf15ddd26166eb3de
Pull request #1046: CAS-13818
Merge in CASA/casa6 from CAS-13818 to master
* commit 'b0973d16124b120e2baf377a91488649c0034986':
CAS-13818 : Skip failing deconvolve tests
CAS-13818 : Fix mask cache conflict on tclean restart
CAS-13818: Add test to check for full stokes mask
Added a skip to test_mtmfsimgval_startmodel_axesmismatch in test_task_deconvolve for MacOS due to Fatal Python error. The failure and solution are being investigated on CAS-14117
CAS-13818 : Disable test_imgval_startmodel_axesmismatch on MacOS
CAS-13818 : Fix for failing test
CAS-13818 : Fix for image.tt0 pb mask -
Ville Suoranta 1d8db3ef8539fcf1f25aac6429130548c3a6304c
Pull request #1045: CAS-14769
Merge in CASA/casa6 from CAS-14769 to master
* commit '85227bc5735154327b1a1394763b90f6bd60ac0a':
Update casacore reference
Revert "Point casacore to dev branch"
Remove redefition of PtrBlock in TransformMachines
Point casacore to dev branch -
Akeem Wells 96543550841e6adfc298ea4bf932d3a6befa0157
Pull request #1044: CAS-14768
Merge in CASA/casa6 from CAS-14768 to master
* commit '1011c5933ca639144a5ae3c253086f038da63442':
Update xmlrunner_to_junit to work with unittest-xml-reporting module -
Srikrishna Sekhar b0973d16124b120e2baf377a91488649c0034986
Merge remote-tracking branch 'origin/master' into CAS-13818
-
Srikrishna Sekhar a46b0fa46be4985558dfe705e9f2eccf577b4aed
Merge remote-tracking branch 'origin/master' into CAS-14758
-
Kumar Golap dfb7e5241859986be77d9b1c19a09c8e2e5c3747
It was not the precision of converting fromRecord or to Record but numpy print precision. That I believe fixes the precision of phasecenter for parallel continuum
-
Ville Suoranta 85227bc5735154327b1a1394763b90f6bd60ac0a m
Update casacore reference
-
Ville Suoranta 5401a2510043db196722fda194d292eae1fbe0e4 m
Revert "Point casacore to dev branch"
This reverts commit cf53b61404c57a80cfcef14f4df9392985b57443. -
Akeem Wells 1011c5933ca639144a5ae3c253086f038da63442
Merge branch 'master' into CAS-14768
-
Kumar Golap 3c5bde23464661fe96ff365766465083f95adafd m
when resetting image center the coordinate system was not being reset
-
Ville Suoranta 76a61842ff2f6c1e457900680d14157c96c2e59e m
Remove redefition of PtrBlock in TransformMachines
-
Ville Suoranta cf53b61404c57a80cfcef14f4df9392985b57443 m
Point casacore to dev branch
-
Akeem Wells fe4695a054c35ccb9b648d91e0871c322bc267ad m
Merge branch 'master' into CAS-14622
-
Akeem Wells 624b1aca4d9570ee193ab3b18ddac0db9fc2bf37 m
Update xmlrunner_to_junit to work with unittest-xml-reporting module
-
Rui Xue ce691fac04de68e393cb9495357f72e9271ff892 m
CAS-14758: Handle missing residual images when constructing minor cycle summary
-
Rui Xue f03ba5194e6eaca95fb699a2fa3cccd2c7658cbf m
CAS-14758: Optimize image processing by caching tool handles in fill_summary_minor to avoid redundant open/close operations
-
Rui Xue 8fbe09615738a7a02c5c218a2ba4598911874183 m
CAS-14622: Fix out-of-bounds MPI segmentation faults in BriggsCubeWeightor
The SynthesisImager uses BriggsCubeWeightor to compute weights for the image cube. During distributed MPI execution (`ncores > 1`), the output cube's spectral axis is partitioned across multiple nodes into smaller local subcubes.
When evaluating MS data channels, `GridFT::channelMap` maps the visibilities' frequencies into the local subcube's channel indices. If an MS contains dataset frequencies (e.g., from large Doppler swings or wide ephemeris velocity changes) that land *above* the local MPI subcube's maximum channel, `channelMap` returns an index exceeding the local array bounds.
Previously, BriggsCubeWeightor only checked the lower bound (`chanMap(chn) > -1`) before using the calculated index to access `wgtDensity` and the `f2_p`/`d2_p` arrays. Accessing these arrays with indices belonging to higher MPI partitions resulted in unconstrained out-of-bounds memory accesses, causing silent heap corruption or explicit Segmentation Faults, heavily dependent on the MPI slicing and dataset footprint.
This commit resolves the issue by introducing strict upper-bound channel footprint constraints (`chanMap(chn) < shape()[3]`) in both `weightUniform` and `getWeightUniform`. Data channels that mathematically fall outside the upper bounds of the local MPI partition are now securely ignored. -
Rui Xue 3000ee0fe103736833f139a6c7dbdf3eca4abb82 m
CAS-14622: Fix ncores-dependent tclean segmentation fault in `BriggsCubeWeightor`
The SynthesisImager uses `BriggsCubeWeightor` to compute weights for the image cube. During this calculation, `BriggsCubeWeightor::cube2Matrix` extracts raw storage pointers `pcube` and `pflags` from `casacore::Array::getStorage(deleteIt)`.
Previously, the code incremented these exact pointers (`*pflags = *pcube++`) during array traversal. At the end of the function, the incremented pointers were passed directly to `freeStorage()` and `putStorage()`.
When array memory isn't contiguous (which occurs when data is sliced across MPI ranks), `getStorage()` allocates a temporary buffer and sets `deleteIt = true`. Passing the shifted pointer into `freeStorage()` invokes `delete[]` on an invalid memory address, causing heap corruption and a Segmentation fault randomly depending on the data slicing.
This commit resolves the issue by securely caching the original pointers returned from `getStorage` into distinct variables. The traversal loops now use separate copies of these pointers, enabling the original allocation addresses to be correctly evaluated and freed by `freeStorage()` and `putStorage()`. -
Srikrishna Sekhar ca5cf74ea4826b7c981a5931f405467735b895bc m
CAS-13818 : Skip failing deconvolve tests
These two tests are skipped in master, but were not skipped here even
after a merge from master.- casatasks/tests/tasks/test_task_deconvolve.py (version ca5cf74ea4826b7c981a5931f405467735b895bc) (diffs)
- casatools/src/code/synthesis/ImagerObjects/SIImageStore.cc (version ca5cf74ea4826b7c981a5931f405467735b895bc) (diffs)
- casatools/src/code/synthesis/ImagerObjects/SIImageStoreMultiTerm.cc (version ca5cf74ea4826b7c981a5931f405467735b895bc) (diffs)
-
Srikrishna Sekhar b8764caa1a4106dddb60f554b01670367722607b m
Merge remote-tracking branch 'origin/master' into CAS-13818
-
Srikrishna Sekhar 1dcf3787bf43a9bf7622907c35d8971ad01d7087 m
CAS-13818 : Fix mask cache conflict on tclean restart
As documented, the fix for the original issue triggered a mask cache
conflict on restart. In the end the fix turned out to be fairly simple -
check if the mask exists and is valid, and do not touch it if it is.
So in the first instance, when the mask does not exist, it will do the
right thing. And it will not try to do anything on the restart either.
It ended up being a simple inversion of the original logic, with
appropriate traps in place so it works for both the original bug and the
derived bug. -
Srikrishna Sekhar f12d89e8155d92b786530e0078ed983406ba2287 m
Merge remote-tracking branch 'origin/master' into CAS-13818
-
Srikrishna Sekhar e9c363510f36d63230e74a9075a06e65eb4ee6b2 m
CAS-13818: Add test to check for full stokes mask
Added a check within test_stokes_mtmfs_IQUV within test_task_tclean to test the
mask when using mtmfs + full Stokes. -
Akeem Wells 9fd98280248fddc92b692d14aab4b5fb9a5c6dd3 m
Added a skip to test_mtmfsimgval_startmodel_axesmismatch in test_task_deconvolve for MacOS due to Fatal Python error. The failure and solution are being investigated on CAS-14117
-
Srikrishna Sekhar 40c02d7f0905f0b1a5fe4160e14a37fef39b3154 m
CAS-13818 : Disable test_imgval_startmodel_axesmismatch on MacOS
Added a skip to test_imgval_startmodel_axesmismatch in test_task_deconvolve for
MacOS. This test fails for unknown reasons, unrelated to the code on this
branch.
The failure and solution are being investigated on CAS-14117 and will be
re-enabled on that ticket. -
Srikrishna Sekhar 1095b56a0e0b446e4733b3a0afa1a5a7d8b1b123 m
CAS-13818 : Fix for failing test
The previous commit fixed the issue from the parent ticket (no mask for
multi-stokes MT-MFS images) in the wrong location, causing a single tclean test
to fail. This commit moves the fix to the right spot, i.e., into
SDAlgorithMSMFS::restore which works.
All the tclean + deconvolve tests now pass locally. -
Srikrishna Sekhar 24294615507801e5a512818b85aa9956a9a1518e m
CAS-13818 : Fix for image.tt0 pb mask
Fixed the issue, the full Stokes MT-MFS restored images now have the PB mask
included correctly. This has been tested with the script provided to reproduce
the original issue.
GitHub Casacore
-
Marcel Loose <loose@astron.nl> f6c6c6a228b7b84e6cf354a071123e03e5fb89e2
Update GitHub actions and docker build (#1480)
* Update GitHub workflow actions
* Update Docker builds
Dropped builds for Python 3.8 and 3.9. Added build for Python 3.14.
Switched to more recent manylinux image.
* Update URL to Boost download site
* Update docker image builds
Removed docker builds for Ubuntu 22.04; added docker builds for Ubuntu 26.04.- .github/workflows/docker.yml (version f6c6c6a228b7b84e6cf354a071123e03e5fb89e2)
- .github/workflows/linux.yml (version f6c6c6a228b7b84e6cf354a071123e03e5fb89e2)
- .github/workflows/osx.yml (version f6c6c6a228b7b84e6cf354a071123e03e5fb89e2)
- docker/make_images.sh (version f6c6c6a228b7b84e6cf354a071123e03e5fb89e2)
- docker/py_wheel.docker (version f6c6c6a228b7b84e6cf354a071123e03e5fb89e2)
- docker/ubuntu2004_gcc.docker (version f6c6c6a228b7b84e6cf354a071123e03e5fb89e2)
- docker/ubuntu2204_clang.docker (version f6c6c6a228b7b84e6cf354a071123e03e5fb89e2)
- docker/ubuntu2604_clang.docker (version f6c6c6a228b7b84e6cf354a071123e03e5fb89e2)
- docker/ubuntu2604_gcc.docker (version f6c6c6a228b7b84e6cf354a071123e03e5fb89e2)