[BUG]: Memory leak in CUB's iterator_deprecated test

[alliepiper]

Is this a duplicate?

• I confirmed there appear to be no duplicate issues for this bug and that I agree to the Code of Conduct

Type of Bug

Something else

Component

CUB

Describe the bug

https://github.com/NVIDIA/cccl/actions/runs/9606424817/job/26497170531?pr=1879

  	compute-sanitizer --tool memcheck --check-device-heap yes --leak-check full --padding 512 --track-stream-ordered-races all --check-warpgroup-mma yes --require-cuda-init no --check-exit-code yes --error-exitcode 1 --nvtx true /home/coder/cccl/build/cuda12.4-gcc13/cub-cpp17/bin/cub.cpp17.test.iterator_deprecated
  ========= COMPUTE-SANITIZER
  Using device 0: Tesla V100-PCIE-32GB (PTX version 700, SM700, 80 SMs, 32086 free / 32494 total MB physmem, 898.048 GB/s @ 877000 kHz mem clock, ECC on)
  <snip>
  ========= Leaked 2097152 bytes at 0x7fb002000000
  =========     Saved host backtrace up to driver entry point at allocation time
  =========     Host Frame: [0x28e3ce]
  =========                in /usr/lib/x86_64-linux-gnu/libcuda.so.1
  =========     Host Frame:libcudart_static_5382377d5c772c9d197c0cda9fd9742ee6ad893c [0x9afad]
  =========                in /home/coder/cccl/build/cuda12.4-gcc13/cub-cpp17/bin/cub.cpp17.test.iterator_deprecated
  =========     Host Frame:libcudart_static_f74e2f2bcf2cf49bd1a61332e1d15bd1e748f9cf [0x65c32]
  =========                in /home/coder/cccl/build/cuda12.4-gcc13/cub-cpp17/bin/cub.cpp17.test.iterator_deprecated
  =========     Host Frame:cudaMalloc [0xaa504]
  =========                in /home/coder/cccl/build/cuda12.4-gcc13/cub-cpp17/bin/cub.cpp17.test.iterator_deprecated
  =========     Host Frame:cub::CUB_200500_600_700_800_NS::CachingDeviceAllocator::DeviceAllocate(int, void**, unsigned long, CUstream_st*) [clone .constprop.0] [0xe31f]
  =========                in /home/coder/cccl/build/cuda12.4-gcc13/cub-cpp17/bin/cub.cpp17.test.iterator_deprecated
  =========     Host Frame:void TestTexRef<long4, long4>() [0x5241c]
  =========                in /home/coder/cccl/build/cuda12.4-gcc13/cub-cpp17/bin/cub.cpp17.test.iterator_deprecated
  =========     Host Frame:main [0xc5c1]
  =========                in /home/coder/cccl/build/cuda12.4-gcc13/cub-cpp17/bin/cub.cpp17.test.iterator_deprecated
  =========     Host Frame: [0x23a8f]
  =========                in /usr/lib/x86_64-linux-gnu/libc.so.6
  =========     Host Frame:__libc_start_main [0x23b48]
  =========                in /usr/lib/x86_64-linux-gnu/libc.so.6
  =========     Host Frame:_start [0xc904]
  =========                in /home/coder/cccl/build/cuda12.4-gcc13/cub-cpp17/bin/cub.cpp17.test.iterator_deprecated

How to Reproduce

compute-sanitizer --tool memcheck --check-device-heap yes --leak-check full --padding 512 --track-stream-ordered-races all --check-warpgroup-mma yes --require-cuda-init no --check-exit-code yes --error-exitcode 1 --nvtx true /home/coder/cccl/build/cuda12.4-gcc13/cub-cpp17/bin/cub.cpp17.test.iterator_deprecated

Expected behavior

No diagnostics emitted.

Reproduction link

No response

Operating System

No response

nvidia-smi output

No response

NVCC version

No response

[bernhardmgruber]

I think I killed this test in #1868, so that bug should have disappeared :)

[alliepiper]

I think I killed this test in #1868, so that bug should have disappeared :)

Indeed it has!