jakirkham commented on issue LLNL/zfp#210.
If we are ok fixing the Python test suite for NumPy 2 in a ZFP_VERSION_TWEAK (note - this only affects Python tests as show above: https://github.com/LLNL/zfp/issues/210#issuecomment-2463210726 ), then it would be great to include.
…
cyrush commented on issue visit-dav/visit#20060.
It looks like this could create a commit for every push to a PR, which is not the same as the test suite results (max of one commit per day to the develop branch). …
haampie commented on issue spack/spack#47720.
This is not enough for closing #42417 because the longest paths to llvm and llvm-amdgpu from hip have equal length:
…
Jordan474 commented on issue spack/spack#47217.
I believe this pull request is in a good state and ready for review. See the end of this comment to compare with cpython’s ast.py. …
vucoda commented on issue spack/spack#47712.
In spack 0.23.0 when I bootstrap from source on some systems where /usr/bin/python = python 2 and /usr/bin/python3 is python 3, the python-venv spack package using find_first() may pick up /usr/bin/python as the executable, then calling /usr/bin/python -m venv will fail. the above commit makes sure to find python3.xx or python3 before python…
svengoldberg commented on issue spack/spack#47710.
@mrfh92 I am sorry, I cannot really help you here… …
adamjstewart commented on issue spack/spack#46506.
^ @Chrismarsh @tgamblin @haampie …
spackbot-app[bot] commented on issue spack/spack#47710.
…
ClaudiaComito commented on issue spack/spack#47710.
@spackbot help…
mrfh92 commented on issue spack/spack#47710.
@melven @svengoldberg as our spack-experts, do you immediately see whats the problem here? (if not, we will figure out ourselves)…
mabraham commented on issue spack/spack#47679.
> @mabraham can we merge this? …
codecov[bot] commented on issue flux-framework/flux-core#6452.
## Codecov Report…
grondo commented on issue flux-framework/flux-core#6453.
> Tweak suggestions welcome …
lindstro commented on issue LLNL/zfp#210.
I expect such a post release to be an exception to the norm, so I’m fine with bumping ZFP_VERSION_TWEAK to 1 and doing a post release of zfpy, which in this case synchronizes versions. But if additional minor changes are needed for NumPy 2.0, we’d have to do another post release. From what I can tell, few if any changes are needed, so should we not try to knock that out now and do only one post release?
…
adrienbernede commented on issue LLNL/CHAI#288.
Test complete…
markcmiller86 commented on issue visit-dav/visit#20060.
> I don’t know that we want even more robo commits. …
biagas commented on issue visit-dav/visit#20070.
@markcmiller86 what do you think of modifying the silo source tarball used by build_visit to remove those pesky windows line endings?…
ictmay commented on issue hypre-space/hypre#1095.
@victorapm All of the shallow copy code is present in our external code here: …
WagnerMatheus commented on issue hypre-space/hypre#628.
I tested the src/test/ams_driver and src/test/ij solvers on an RX6600 after changing the warp size related defines. The warp size mostly affects solvers that use BoomerAMG.
…
da-wad commented on issue LLNL/zfp#210.
Thanks for pointing me towards those, #237 fixes the bug introduced in https://github.com/LLNL/zfp/commit/63eeefa32adffe71217f3e48ae5011f8b5f75de1 and referenced by #233 …
rhornung67 commented on issue LLNL/RAJA#1661.
Thanks @adayton1 …
gzagaris commented on issue LLNL/RAJA#1558.
Hi @rhornung67, yes the question on this issue has been addressed and the issue should be closed. Thank you again for all your help. …
codecov[bot] commented on issue flux-framework/flux-sched#1296.
## Codecov Report…
adayton1 commented on issue LLNL/RAJA#1661.
I think the only thing we might want to do is add openmp testing to windows (or at least a build)….
lucpeterson commented on issue LLNL/merlin#489.
Ping on this. Is the plan to merge the new test stuff in and then keep going?…
JustinPrivitera commented on issue LLNL/conduit#1332.
#1333…
adrienbernede commented on issue LLNL/CARE#302.
@adayton1, …
daboehme commented on issue LLNL/Caliper#624.
I haven’t dug through the logs yet but I suspect it’s because Lassen and Tioga need to run MPI jobs on a compute node via lrun / flux run, which the Caliper CI script doesn’t do. It just runs an MPI program as-is since it only needs one rank. For now I’d turn off the MPI tests on Tioga/Lassen (there’s a CMake switch to do it, -DRUN_MPI_TESTS=Off) although in the long run it would be good to handle this properly….
garlick commented on issue flux-framework/flux-sched#1034.
See flux-framework/flux-core#5739 …
robertu94 commented on issue LLNL/zfp#245.
I would also appreciate it if you can release this change as 1.0.2…
kminemur commented on issue LLNL/zfp#244.
cmake out log is as follows: …
rw-anderson commented on issue LLNL/hatchet#123.
I struggled with a similar issue. Eventually I tracked it down to picking up an older installed version of cali-query and using that in conjunction with my locally built caliper library. Once I matched those versions, this error went away….
michaelmckinsey1 commented on issue LLNL/hatchet#152.
Matching a single node with name my_node
…
tmarrinan commented on issue LLNL/conduit#1332.
Hello! …
cameronrutherford commented on issue LLNL/hiop#698.
Some final notes: …
gzagaris commented on issue LLNL/Umpire#912.
Hi @davidbeckingsale – Just wanted to get a pulse on this. Any luck with getting things working on your end?…
agriyakhetarpal commented on issue LLNL/zfp#210.
Hi all, I’m also trying to update some dependencies for numcodecs for a WebAssembly build of it in zarr-developers/numcodecs#529 and I stumbled across this issue. I am fully cognisant of how concerted an effort is required to officially support NumPy 2.0 based on the previous conversation here; may I ask if it’s possible to publish a source distribution for zfpy to PyPI in the meantime? It would be beneficial for Pyodide’s in-tree recipes (pyodide/pyodide#5169) where I tried to package zfpy but there wasn’t a source distribution available (https://pypi.org/project/zfpy/#files). We’re also planning to release a new version of Pyodide soon which will have all NumPy-dependent packages rebuilt for NumPy v2 (pyodide/pyodide#4925), so we can try out an “unofficial” WASM build with NumPy v2 in the time this issue gets resolved.
…
rountree commented on issue hpc/Spindle#56.
Point taken wrt scaling and file pollution. I can imagine a solution where messages (with their priority) are buffered until initialization occurs, then parsed to see if they’re logged or discarded. It’s not wholly unlike trying to make printf() work in _start() before your file descriptors are set up, and, perhaps like that, it’s not a problem worth solving (or not worth solving this week).
…
drmichaeltcvx commented on issue LLNL/blt#707.
Hi, thanks for the response! It was actually a misconfiguration of the s/w stack running on our environment that at the configure was not enabling the proper MPI. …
seberg commented on issue LLNL/zfp#210.
I actually noticed two places (I think in the tests) that use array(..., copy=False) which need to be asarray(...) instead.
…
PhilipDeegan commented on issue LLNL/SAMRAI#93.
closed by #277…
EverettGrethel commented on issue LLNL/maestrowf#447.
Yes, the ability to pass optional args for the launcher in addition to the allocation is imperative, in my opinion. That would be great. I’ve written up a solution for the allocation optional args, which I can submit a pull request for you to check out if you’d like. …
