samdporter

The problem is still present - see latest commit to [this](https://github.com/samdporter/SPECTissues) repo Please note that this is only true when `keep_all_views_in_cache = True`

I'm not using it either (although having a read about it, it looks like I probably should?)

Hey all, This still needs a few changes. Currently the `BlockOperator.direct` method still returns a `BlockDataContainer` with shape (1,1) rather than a `DataContainer`. I'll do a few more tests to...

==85870== 64,167,432 bytes in 126 blocks are still reachable in loss record 7,248 of 7,250 ==85870== at 0x483C583: operator new[](unsigned long) (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so) ==85870== by 0x30673740: stir::ProjMatrixByBinSPECTUB::set_up(std::shared_ptr const&, std::shared_ptr const&)...

results from gbd: Thread 34 "python3" received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x7fc781aad700 (LWP 261805)] __gnu_cxx::__atomic_add_dispatch (__val=1, __mem=0x249) at /usr/include/c++/9/ext/atomicity.h:96 96 __atomic_add(__mem, __val); #0 __gnu_cxx::__atomic_add_dispatch (__val=1, __mem=0x219) at...

[valgrind log](https://github.com/samdporter/SPECTissues/blob/main/valgrind-out.txt) This is the valgrind output with the debugger on (I've checked it's on in cmake, but it looks the same to me)

Looks like it is an issue with multi-threading. Everything runs fine with openMP off

Hi, The problem seems to arise after creating a CIL gradient operator with a C backend see: [mimimal.ipynb](https://github.com/samdporter/SPECTissues.git) The final cell fails The algorithm runs fine when using Numpy. Should...

Yes exactly

Sorry the repository was private. Try now