Please backport nvidia-cuda-toolkit 5.0.35-4 (multiverse) from raring

[Graham Inggs]

Please backport nvidia-cuda-toolkit 5.0.35-4 (multiverse) from raring to precise, quantal.

Reason for the backport:
========================
CUDA 5 is the latest release and introduces many new features.

Testing:
========
Mark off items in the checklist [X] as you test them, but please leave the checklist so that backporters can quickly evaluate the state of testing.

You can test-build the backport in your PPA with backportpackage:
$ backportpackage -u ppa:<lp username>/<ppa name> -s raring -d precise nvidia-cuda-toolkit
$ backportpackage -u ppa:<lp username>/<ppa name> -s raring -d quantal nvidia-cuda-toolkit

Backported packages available here:
https://launchpad.net/~ginggs/+archive/backports

* precise:
[X] Package builds without modification
    (requires nvidia-current 295.40ubuntu1.3, see LP: #950963)
[X] nvidia-profiler installs cleanly and runs (nvprof)
[X] libnvtoolsext5.0 installs cleanly
[X] libcupti-doc installs cleanly
[X] libcudart5.0 installs cleanly
[X] nvidia-cuda-dev installs cleanly
[X] libcupti-dev installs cleanly
[X] libcuinj32-5.0 installs cleanly
[X] libcupti5.0 installs cleanly
[X] libcusparse5.0 installs cleanly
[X] nvidia-cuda-doc installs cleanly
[X] libcuinj64-5.0 installs cleanly
[X] libnpp5.0 installs cleanly
[X] libcublas5.0 installs cleanly
[X] nvidia-visual-profiler installs cleanly and runs (nvvp)
[X] libcufft5.0 installs cleanly
[X] nvidia-cuda-gdb installs cleanly and runs (cuda-gdb)
[X] libcurand5.0 installs cleanly
[X] nvidia-opencl-dev installs cleanly
[X] nvidia-nsight installs cleanly and runs (nsight)
[X] nvidia-cuda-toolkit installs cleanly
* quantal:
[X] Package builds without modification
[X] nvidia-profiler installs cleanly and runs (nvprof)
[X] libnvtoolsext5.0 installs cleanly
[X] libcupti-doc installs cleanly
[X] libcudart5.0 installs cleanly
[X] nvidia-cuda-dev installs cleanly
[X] libcupti-dev installs cleanly
[X] libcuinj32-5.0 installs cleanly
[X] libcupti5.0 installs cleanly
[X] libcusparse5.0 installs cleanly
[X] nvidia-cuda-doc installs cleanly
[X] libcuinj64-5.0 installs cleanly
[X] libnpp5.0 installs cleanly
[X] libcublas5.0 installs cleanly
[X] nvidia-visual-profiler installs cleanly and runs (nvvp)
[X] libcufft5.0 installs cleanly
[X] nvidia-cuda-gdb installs cleanly and runs (cuda-gdb)
[X] libcurand5.0 installs cleanly
[X] nvidia-opencl-dev installs cleanly
[X] nvidia-nsight installs cleanly and runs (nsight)
[X] nvidia-cuda-toolkit installs cleanly

Reverse dependencies:
=====================
The following reverse-dependencies need to be tested against the new version of nvidia-cuda-toolkit. For reverse-build-dependencies (-Indep), please test that the package still builds against the new nvidia-cuda-toolkit. For reverse-dependencies, please test that the version of the package currently in the release still works with the new nvidia-cuda-toolkit installed. Reverse- Recommends, Suggests, and Enhances don't need to be tested, and are listed for completeness-sake.

libnvtoolsext5.0
----------------

libcuinj32-5.0
--------------

libcudart5.0
------------

nvidia-cuda-dev
---------------
* opencl-headers
  [ ] quantal (Reverse-Breaks)
  [ ] quantal (Reverse-Replaces)
* libcuinj4
  [ ] quantal (Reverse-Breaks)
  [ ] quantal (Reverse-Replaces)

libcupti-dev
------------

libcupti-doc
------------

libcuinj64-5.0
--------------

libcusparse5.0
--------------

nvidia-nsight
-------------

nvidia-cuda-doc
---------------
* python-pycuda-doc
  [ ] quantal (Reverse-Recommends)

libcupti5.0
-----------

libnpp5.0
---------

libcublas5.0
------------

nvidia-visual-profiler
----------------------

libcufft5.0
-----------

nvidia-profiler
---------------

libcurand5.0
------------

nvidia-opencl-dev
-----------------

nvidia-cuda-gdb
---------------

nvidia-cuda-toolkit
-------------------
* libthrust-dev
  [ ] quantal (Reverse-Suggests)
* pycuda
  [*] quantal (Reverse-Build-Depends)
  [*] quantal (Reverse-Build-Depends)
* python-pycuda
  [X] quantal (Reverse-Depends)
* starpu-contrib
  [*] quantal (Reverse-Build-Depends)
  [*] quantal (Reverse-Build-Depends)

* These packages build locally with nvidia-experimental-310 installed, but currently not on the PPA builder.

[Felix Geyer (debfx)]

Have you tested that python-pycuda works with the new nvidia-cuda-toolkit on quantal?

[Graham Inggs (ginggs)]

I was able to run example/hello_gpu.py on quantal with python-pycuda 2012.1-1ubuntu2 from the archives and nvidia-cuda-toolkit 5.0.35-4~quantal1~ppa1 from my PPA.