Open Source Heartbeat: Open Source Heartbeat

Open Source Heartbeat

garlick commented on issue flux-framework/flux-core#7251.

Fixed those things, and made another pass, expanding the slurm migration section to include more information. …

View Comment

cyrush open issue Alpine-DAV/ascent#1643.

cuda vtk-m use dp issue with slice filter

Using:…View Comment

JustinPrivitera closed issue visit-dav/visit#19121.

Take advantage of MFEM 4.6 - 4.8 Features

We should take advantage of new MFEM 4.6 features after #19112 is merged. …View Comment

cyrush commented on issue visit-dav/visit#20768.

@biagas can you try out the new mp4 logic to see if that helps with #20748 ?…

View Comment

tgamblin commented on issue spack/spack#51781.

So, I was hoping to use protocols to make typing of spec data structures much easier. This doesn’t bode well for that, so I guess I’ll ask some questions: …

View Comment

hppritcha commented on issue open-mpi/ompi#13583.

@jsquyres check now. fixed typo inherited from slurm page too….

View Comment

dylan-copeland commented on issue mfem/mfem#5123.

Thank you @IdoAkkerman for your careful review. There were many important issues raised in this review, which are hard to find in the resolved comments. Here is a list to summarize some issues to revisit later. …

View Comment

gardner48 closed issue llnl/sundials#806.

[BUG] Build fails with MPI under Win, latest release 2025

When trying to build on Windows (after generating with CMake) within Visual Studio the build fails simply due to the following line in sundials_mpi_errors.c…View Comment

balos1 commented on issue llnl/sundials#806.

Thanks for the report. This header is unused, so I have removed it entirely in #807….

View Comment

cyrush commented on issue llnl/conduit#1438.

@mlohry these jobs failed b/c the windows image in ci was removed around the time of testing. If we merge develop into your branch, we can fully test…

View Comment

adayton1 commented on issue llnl/axom#1599.

This is currently a blocker for us having any CUDA builds….

View Comment

adayton1 commented on issue llnl/Umpire#1033.

> It seems there are some C++20 compatibility issues in the current version of fmt, so will see how updating to 11.1.0 works …

View Comment

artv3 commented on issue llnl/RAJAPerf#599.

How big can meshes get? I’m wondering how the meshes will impact the repo size….

View Comment

artv3 commented on issue llnl/RAJA#1964.

Hey this looks pretty cool and simpler than we were originally trying, let me try it within the RAJA PerfSuite and I can get back to you …

View Comment

jeremylt commented on issue freeCodeCamp/freeCodeCamp#64787.

How about:…

View Comment

trws commented on issue flux-framework/flux-sched#1423.

Doing it in that literal order is impossible. We may be able to get you to something like what you want, but let me lay out the situation a bit so you know where I’m coming from. In order to give a predicted start time we need to run a match against the current resource state, allocations and reservations to determine when the job would be reserved. Without doing that match, any result we give out would be even more wrong than without doing it (even doing the match is still an estimate at best due to new job arrival and constant state changes)….

View Comment

chu11 commented on issue flux-framework/flux-core#7191.

Continued prototyping with the eventlog in this format as I describe above: …

View Comment

jsmeredith closed issue tensorflow/tensorflow#55414.

Known issues with new RUN_EAGER_OP_AS_FUNCTION feature

This refers to the unified eager op/tf.function execution mode available for optional use in 2.9. …View Comment

vsoch commented on issue rseng/jobs-updater#12.

Thanks @crd477 ! …

View Comment

adrienbernede commented on issue llnl/uberenv#154.

I have the same error on my end with quandary….

View Comment

adrienbernede commented on issue llnl/radiuss-spack-configs#155.

@davidbeckingsale building went well. (Be mindful that the tests are not run). …

View Comment

uvilla commented on issue hippylib/hippylib#84.

@finsberg: Thank you for your contribution. …

View Comment

vsoch commented on issue flux-framework/flux-core#7217.

A contributor just needs to be able to find “how to contribute” quickly. The CONTRIBUTING.md is (to me) preferred over what I had here, which was a workaround to not having one. …

View Comment

grondo commented on issue flux-framework/flux-core#7241.

Thanks! I’ll set MWP and perhaps @ryanday36 could give feedback and open a new issue if there needs to be adjustments or improvements in the future….

View Comment

wihobbs commented on issue flux-framework/flux-core#7245.

James is definitely also someone who’s approved and merged my PRs as a maintainer. …

View Comment

vsoch commented on issue converged-computing/usernetes#2.

@AkihiroSuda if you are interested, we were testing calico for usernetes https://github.com/converged-computing/usernetes/tree/test-calico (hoping to not need to set net.ipv4.conf.default.rp_filter to 2) and got it deployed (calico pods running), but the connectivity between pods isn’t there yet. I think packets are being dropped. Maybe by the kernel (which could be related to that setting still being 2). We were trying to use vxlan although I’m not sure if that is the right path, and if we can test with the rp_filter setting still at 2….

View Comment

tgamblin commented on issue spack/spack-packages#2709.

@olesenm: welcome back! See: …

View Comment

becker33 commented on issue spack/spack#51422.

Can spack -E commands update-completion do what you need for this?…

View Comment

CamStan closed issue llnl/scr#605.

Bijungi Script

Bijungi is re-constructed language on Indonesia was invented by Leng Mokorang, Kavita Tengnyan and Anda Upin was invented by script horizontal was inspired by Tagbanwa alphabet the type is abugida…View Comment

becker33 closed issue llnl/llnl.github.io#735.

Update site with LLNL logo and color palette

1. Update logo on LLNL org…View Comment

adrienbernede commented on issue llnl/Umpire#1012.

@davidbeckingsale this PR is now ready….

View Comment

adrienbernede commented on issue llnl/RAJA#1935.

@rhornung67 This PR is now ready for review….

View Comment

adayton1 commented on issue llnl/CHAI#341.

> @adayton1 #343 should be merged first. …

View Comment

adrienbernede commented on issue llnl/CHAI#341.

@adayton1 https://github.com/LLNL/CHAI/pull/343 should be merged first….

View Comment

tgamblin commented on issue hpsfoundation/tac#61.

I mean if you have something like, say, Flux Operator that has a name with “Flux” in it, it needs to either be renamed to something like “Kubernetes Operator for Flux” (in compliance with the doc I linked above), or it needs to be in the Flux org as part of the official upstream Flux project….

View Comment

cmoussa1 open issue flux-framework/flux-accounting#793.

storage: long-term/centralized/durable storage solutions for flux-accounting

@trws and I discussed offline that it might be nice to think about and eventually implement a long-term storage solution for flux-accounting and its contents. As it stands now, flux-accounting utilizes SQLite to store all user, bank, association, queue, and job data to complete its functions, which is lightweight and portable. Offering a more centralized storage option such as PostgreSQL (for which there exists a Python API to interface with such a DB) might be a nice evolution for this module, perhaps as a way to offer a center-wide solution for all things flux-accounting, like managing accounts, banks, projects, etc….View Comment

alecbcs commented on issue spack/spack-packages#2365.

@tgymnich @vchuravy @wsmoses @amd-toolchain-support any objections here? …

View Comment

tzanio commented on issue mfem/mfem#5107.

This PR is now under review (see the table in the PR description). To help with the review process, please do not force push to the branch….

View Comment

v-dobrev commented on issue mfem/mfem#5122.

Both mfem+umpire and mfem~umpire should work well – try it out to see which one works best. Here’s how this is done in Laghos: https://github.com/CEED/Laghos/pull/198….

View Comment

jeremylt closed issue CEED/libCEED#1081.

examples/fluids: unstructured distance function

Computing distance from walls is necessary for some turbulence models and to create initial conditions for which the initial transient is not too extreme. The Eikonal equation offers a proper way to solve this equation, but it’s difficult to solve scalably using variational methods. This paper compares several strategies and suggests a simple method: …View Comment

vladotomov commented on issue CEED/Laghos#201.

@rfhaque Let me know when it’s done….

View Comment

markcmiller86 commented on issue visit-dav/visit#20752.

I added logic to silo plugin to provide a little more info about the file (and plugin) software versions… …

View Comment

tgamblin commented on issue llnl/axom#1740.

The axom package looks way better (160 added lines, nearly all declarative vs. 242 in the first revision with custom consistency checking). We can prioritize improving the error messages for package requirements in Spack….

View Comment

markcmiller86 open issue llnl/Silo#514.

Add ability to get driver library version info

It would be nice, without having to include the HDF5 library in the client, for a client of Silo to get library version information from the underlying driver library. Probably the only reasonable thing to do is return it as a string of some kind. Different driver libraries can use very different versioning schemes and probably a string representation is the only approach that be portable….View Comment

tgamblin closed issue hpc-social/events#9.

New Event: HPSF Community Summit 2026

### Event Submission…View Comment

garlick commented on issue flux-framework/flux-foundry#20.

@grondo any thoughts on this one? This should be sufficient to get tomlc99 amd tomltk.c out of flux-core. More work would be needed for flux-security….

View Comment

markcmiller86 open issue visit-dav/visit#20762.

Splash screen shows 3.5.9, git info shows 3.4.9

I just wanted to be sure we don’t forget to fix this before 3.5.0 release….View Comment

JustinPrivitera open issue visit-dav/visit#20761.

Viewer user text resizes on every mouse click

This text here in the bottom corner…View Comment

markcmiller86 commented on issue spack/spack-packages#2346.

@tldahlgren I noticed you had some recent activity on an MFEM PR here and I wanted to reach out and ask if your (or someone you might wanna recommend) could merge this PR for Silo?…

View Comment

v-dobrev commented on issue spack/spack-packages#143.

Please do not merge yet. I’ll review this soon – I’ll do it right after I’m done with the PR to add and test the new v4.9. (For now I just have the initial commit for that pushed: https://github.com/spack/spack-packages/tree/mfem-4.9.)…

View Comment

vsoch closed issue singularityhub/singularity-cli#228.

conversion of Dockerfile to singularity recipes can fail when remote file was redirected

## Expected Behavior…View Comment

tzanio commented on issue mfem/mfem#5153.

Merged in next for testing……

View Comment

imguoguo open issue imguoguo/Comments#43.

garlick commented on issue flux-framework/flux-core#7238.

Thanks - setting MWP…

View Comment

vsoch open issue converged-computing/flux-hierarchy#5.

Optimizations and Improvements Needed

This issue summarizes some todo items for flux-hierarchy. I am creating it for easy discussion to bring in some of the Flux devs….View Comment

garlick commented on issue chaos/diod#164.

My test setup:…

View Comment

cyrush commented on issue visit-dav/visit#20752.

@markcmiller86 I think we should try the new version, would be good to remove ancient patches and have same code used on windows as linux if possible. …

View Comment

becker33 commented on issue spack/spack#51731.

I’m not convinced “every node is a separate object” is a good assumption in abstract specs, even without this bug. But that can be a separate conversation….

View Comment

jandrej commented on issue mfem/mfem#5151.

```…

View Comment

chu11 open issue flux-framework/flux-core#7235.

new job-list client RPCs hang when connecting to older Flux instances

#7224 changed the job-list RPC to stream responses by default. The implementation maintains backward compatibility when the streaming flag is not set in the request, allowing older clients to work with newer servers. However, there is a compatibility issue in the opposite direction:…View Comment

cmelone commented on issue LLNL/hubcast#190.

> I’d vote for the 1st option, but let’s make it optional via a config. Some users might want to disable draft PRs and not get a message from Hubcast while others probably do want a place holder status check with a message. …

View Comment

BradWhitlock open issue LLNL/conduit#1485.

Failure with find_reference_node.

I ran across a strange case where find_reference_node fails when I move mesh domain node to a different node. I would expect that since the new node took ownership of the mesh data, it would serve as the root and ought to basically be the same as the copy case….View Comment

adrienbernede commented on issue LLNL/RAJAPerf#587.

@rhornung67 This PR is ready. I reduces the number of specs running in CI (shared specs in RSC have been updated) and clearly deactivates some of the shared specs to focus on specs for interest for RAJAPerf….

View Comment

cyrush commented on issue Alpine-DAV/ascent#1636.

Ok, I understand. If someone takes over they can change the version to a malicious commit. …

View Comment

trws commented on issue vsoch/hpsf-tac#1.

Depending on what you need, it also often works well to include just user-mode flux (core and sched is often enough) inside the container, then in the allocation use srun to launch a flux instance with all brokers running in the container with your optimizer running as the initial program. It requires changing the container, where the sidecar does not, but avoids the need for kubernetes to use the operator. It

hppritcha commented on issue open-mpi/ompi#13460.

bingo! now after a little bit of rebasing fixes I can get #13280 to pass mpi4py declaring OMPI is MPI 5.0 compliant and not needing to exclude the big count tests! …

View Comment

v-dobrev commented on issue mfem/mfem#5057.

Merged in next for testing….

View Comment

acfisher commented on issue mfem/mfem#4689.

@tzanio it looks like this ended up getting 2 approvals and then never pulled in. I am resurrecting it and want to shepard it through….

View Comment

bgunnar5 open issue llnl/merlin#598.

Tutorial for Merlin 2.0

There’s a good amount of new/updated features going into Merlin v2.0. We should update the entire tutorial in order to showcase everything in its current state….View Comment

uvilla commented on issue hippylib/hippylib#83.

Thank you @dc-luo. I agree with the proposed class renaming. We could …

View Comment

trws commented on issue flux-framework/rfc#487.

Got it, thanks!…

View Comment

jameshcorbett commented on issue flux-framework/flux-sched#1398.

Yes I think so!…

View Comment

wihobbs commented on issue flux-framework/flux-sched#1398.

@jameshcorbett I think we got packages into TOSS 5 but correct me if I’m wrong, ok to close this one?…

View Comment

grondo commented on issue flux-framework/flux-core#7220.

After implementing the noverify = ["core"] support, I realized this approach does not cleanly handle the case for falling verification when more resources than configured are discovered….

View Comment

cmoussa1 open issue flux-framework/flux-accounting#792.

job usage reporting: add arbitrary job usage calculation for associations, banks

One of the shortcomings with the way usage is calculated in flux-accounting is that it is only represented in one way: the association or bank’s current historical usage based on just one date range. There isn’t an easy way to calculate an association or bank’s job usage based on some arbitrary date range that would come in handy for reporting. It might be nice to offer a way for flux-accounting to natively provide the ability to customize how job usage is calculated for a user, a bank, or an association….View Comment

bgunnar5 open issue LLNL/merlin#598.

Tutorial for Merlin 2.0

There’s a good amount of new/updated features going into Merlin v2.0. We should update the entire tutorial in order to showcase everything in its current state….View Comment

JustinPrivitera open issue LLNL/conduit#1484.

See if new silo 4.12 fixes namescheme issues

> https://github.com/LLNL/Silo/releases…View Comment

cyrush commented on issue LLNL/conduit#1482.

``` …

View Comment

artv3 closed issue LLNL/RAJA#1954.

Can't build RAJA with CUDA

Building on Matrix using the following command: …View Comment

adrienbernede commented on issue LLNL/RAJA#1955.

> This also needs to be turned on in the GItLab UI, correct? …

View Comment

jeremylt closed issue CEED/libCEED#1906.

Adding an example in Julia

We have ex1_volume, ex2_surface, and ex3_volume in Python. And we have ex1-volume and ex2-surface implemented in Julia but not ex3-volume. …View Comment

emily-howell closed issue Alpine-DAV/ascent#1607.

move from vtk-m to viskores

View Comment

acfisher commented on issue mfem/web#204.

@tzanio ok I put my changes in, so I think this is ready for another reviewer….

View Comment

markcmiller86 commented on issue llnl/H5Z-ZFP#155.

@brtnfld and @jhendersonHDF thanks for submission and review. @brtnfld I think you have authority to merge at this point…or, you should. …

View Comment

milroy commented on issue flux-framework/kubeflux#1.

This came up in the Flux meeting today. We can delete this repo….

View Comment

milroy commented on issue flux-framework/flux-sched#1419.

I took a first pass on this PR, and it generally looks good. I did post a question about the function of a parameter. …

View Comment

markcmiller86 commented on issue LLNL/H5Z-ZFP#155.

@brtnfld and @jhendersonHDF thanks for submission and review. @brtnfld I think you have authority to merge at this point…or, you should. …

View Comment

v-dobrev commented on issue xsdk-project/xsdk-issues#280.

I just merged the Spack PR, so hopefully this issue is now resolved….

View Comment

jameshcorbett commented on issue vsoch/hpsf-tac#1.

> @jameshcorbett

imguoguo commented on issue sipeed/NanoKVM-USB#89.

Bug(fixed):

wihobbs commented on issue sam-maloney/flux-core#1.

Another thought I just had is perhaps the plugin should be able to slurp up json/yaml files with a resource specification in them based on RFC 14 in addition to shape. But maybe that should be a separate plugin/option….

View Comment

cjvogl commented on issue mfem/web#291.

> Along this line: for my version of hipcc I manually specified the MPI include directory for METIS and added -Wno-implicit-int and -fPIC to get it to work. …

View Comment

chu11 commented on issue flux-framework/rfc#482.

I don’t think there are any gotchas, will set MWP…

View Comment

adayton1 commented on issue LLNL/axom#1599.

> I noticed that Axom+CUDA is built in a funny mode where nvcc appears to use a host compiler. If I change the host-config to CUDA 12.2..2 and the host compiler to newer clang then it gets farther….

View Comment

samuelkgutierrez commented on issue hpc/quo-vadis#424.

Perhaps the environment variable can be a CPU mask that encodes the allowed resources. hwloc-calc can help sites generate this….

View Comment

BradWhitlock commented on issue LLNL/axom#1732.

@bmhan12 - I ran into “yaml_parser->problem is NULL (missing)” recently in some other work when my data submodule was not up to date. When I went in and made sure it had “main” checked out rather than a detached HEAD, it worked. I don’t know how it got into the detached HEAD state though….

View Comment

homijan open issue GenDiL/GenDiL#17.

Integration over a specific dimension

Hi @YohannDudouit ,…View Comment

jedbrown closed issue rsmpi/rsmpi#215.

Build process fails on Mac with this error

When running cargo build of late versions of Mac, the build script of libffi-sys a internal dep of libff1 encounters this error; …View Comment

v-dobrev commented on issue mfem/mfem#4977.

Adding the tolerance for changing dt did not actually change the result since the actual difference in dt on the last step is slightly above 2.7569e-15 (this is for the sample run I posted above). However, this is probably fine. …

View Comment

vsoch commented on issue flux-framework/rfc#481.

Actually I don’t see your feedback yet @grondo - let me know if you have thoughts on the above!…

View Comment