markcmiller86 commented on issue visit-dav/visit#19952.
Why not compute in both weighted and unweighted and produce both results? …
JustinPrivitera commented on issue visit-dav/visit#19952.
I like this idea. …
tgamblin commented on issue spack/spack#47481.
@spackbot rebuild everything…
artv3 commented on issue mfem/mfem#4576.
One of the CI builds has an issue, but overall this helps our application…
tzanio commented on issue mfem/mfem#4476.
Re-merged in next for testing……
v-dobrev commented on issue mfem/mfem#4484.
Removing the internal boundaries (either with Mesh::RemoveInternalBoundaries or from mesh files) will do the job, however, in the case of periodic meshes (such as the ones used in the above-mentioned sample runs) it will be nice to keep the periodic boundaries (which are internal boundaries), so that they will be drawn when visualized in GLVis.
…
uvilla commented on issue hippylib/hippylib#78.
This is the MCMC algorithm. …
BradWhitlock commented on issue LLNL/axom#1438.
I added a script that will make BASH scripts to run the mir_concentric_circles example program on various platforms and gather the caliper data results. I plotted the MIR runtime and some things look good, some very bad at the moment. I think SEQ mode could be optimized more to bring it down, likely bringing down time in the other modes too. OMP has inexplicably bad performance. I’ve seen some hints of this before during development but focused more on GPU so far. I’ll need to profile OMP and see what’s the matter. …
cyrush commented on issue visit-dav/visit#20022.
@BenWibking thanks! We will try to resolve this during our release process for 3.4.2….
hppritcha commented on issue uofl-capstone-open-mpi/prrte#6.
One idea to try folks is to modify the .github/workflows yml files to build against a specific sha of openpmix. …
vladotomov commented on issue mfem/mfem#4570.
Continue in #4572….
dylan-copeland commented on issue mfem/mfem#4038.
Hi @adam-sim-dev, thank you for testing this branch. Since this is a new feature, it is not surprising that you may find things not supported or working correctly yet with parallel p-refinement. I have listed some things not supported at the top of this PR, and when we find more, they should either be fixed in this PR or documented for a future PR. Please let me know if you are able to identify exactly what is wrong with your application. …
tgamblin commented on issue hpsfoundation/tac#17.
Tom: …
garlick commented on issue flux-framework/flux-core#6414.
Maybe a new RPC in kvs-watch?…
chu11 commented on issue flux-framework/flux-core#6414.
> Since watching a key whose value is being repeatedly overwritten (the original watch design point) is a different use case than watching one that is being appended to, a client flag to select one or the other seems justified. …
grondo commented on issue flux-framework/flux-core#6423.
@chu11 noticed I dropped the actual conversion of the dmesg timestamp to local time here. Setting WIP until I fix that….
IanLee1521 commented on issue LLNL/toss-stig#56.
Added 4a2dbbe to remove a few other ancillary things from the repo….
artv3 commented on issue LLNL/camp#158.
Awesome!!!! …
CRobeck commented on issue triton-lang/triton#5067.
@fywkevin @Jokeren can you take another pass? I think I got everything. …
v-dobrev commented on issue mfem/web#290.
Hi @gabsillis, I added you to the Contributors team, so you should be able to push your branch in this repo. Let me know if that still does not work. Thank you for your contribution!…
grondo commented on issue flux-framework/rfc#423.
BTW, I just recalled that when we discussed this proposal of a minimum duration one possible solution was to still use the preemptible flag, but to have the job-manager post the flag to the eventlog as soon as minimum runtime expired. We should explore if that makes sense, or if we leave it up to the scheduler to look for any minimum runtime and implement preemption internally without need of any kind of flags. …
garlick commented on issue flux-framework/rfc#423.
In #5378, @ryanday36 mentioned an alternate idea to a standby flag: allow a duration range in jobspec, where a standby job would specify a min < max duration and the job would be eligible for preemption after the min duration expires. For traditional standby, min = 0, but it would allow for a job to request to get a minimum amount of work done before preemption. It seems like we discussed this and decided it was a better idea than the flag but I can’t find anything to cite here. …
milroy commented on issue flux-framework/flux-sched#1314.
Thanks for the reviews. Setting MWP….
IanLee1521 commented on issue LLNL/llnl.github.io#691.
Thanks for being on top of this @hauten !…
bgunnar5 commented on issue LLNL/axom#1378.
@LLNL/axom I think this is finally ready to hit the merge button…
jedbrown commented on issue openjournals/joss#1383.
I think proposing it in Slack would also connect with an audience beyond this issue, and help prime a discussion at the next monthly meeting….
samuelkgutierrez commented on issue hpc/quo-vadis#288.
Thank you, @GuillaumeMercier. …
garlick commented on issue flux-framework/flux-security#194.
Great! I put this change on my test cluster and my test also works now! …
cmoussa1 commented on issue flux-framework/flux-accounting#528.
The test continued to pass because the grep check was lazy (and my fault); grep -w "username: user5015\|bank: F\|default_bank: F" new_default_bank.out was checking for a whole-word match for one of the patterns instead of all three. I changed this grep check to be more strict and check that all three were present.
…
balos1 commented on issue LLNL/sundials#541.
> > Are the changes in the .gitlab folder intended? …
cyrush commented on issue Alpine-DAV/ascent#1416.
I believe that mfem 4.7 will work with Ascent 0.9.3, based on: https://github.com/Alpine-DAV/ascent/issues/1393 …
adrienbernede commented on issue mfem/mfem#4431.
@v-dobrev @johnbowen42 latest pipeline run: …
trws commented on issue flux-framework/spack#245.
It’s possible, autoconf would look for both but if something set PYTHON=python or something like that it could do it….
jameshcorbett commented on issue flux-framework/flux-coral2#201.
Closing this because it seems I misunderstood what was desired and discussion was restarted on https://github.com/NearNodeFlash/NearNodeFlash.github.io/issues/199….
trws commented on issue spack/spack#47365.
I guess I was wondering if it would be worth having a way to say what the extra path is so the library would resolve, rather than ignoring it. Either gets us through I guess….
mplegendre commented on issue hpc/Spindle#56.
But then you’ve got to decide how long to buffer. And you’re still paying the internal overhead of logging. …
adrienbernede commented on issue LLNL/Caliper#593.
Hello @daboehme the easiest way to reproduce the issue is by using the in-log reproducer. …
v-dobrev commented on issue GLVis/glvis#319.
In parallel, to resolve the issue you see, you can start GLVis with the flag -a/--real-attributes – otherwise the boundary attributes are replaced by processor boundaries….
hppritcha commented on issue open-mpi/ompi#12901.
its a problem with the way the jenkins CI build script handles errors. Like I said above it just starts echoing all the logs rather than just exiting itself. …
v-dobrev commented on issue mfem/mfem#4548.
Re-merged in next for testing……
grondo commented on issue flux-framework/flux-security#188.
Removed WIP now. This seems to be working as designed now, and is much more robust than the flux-imp kill method. Probably would be good if @garlick verifies on his cluster before this gets merged though….
milroy commented on issue flux-framework/flux-sched#1292.
Thanks for the review! Setting MWP….
chu11 commented on issue flux-framework/flux-core#6370.
doh, took me like 3 seconds to realize the bug …
balos1 commented on issue LLNL/sundials#554.
Fixed by #560….
cyrush commented on issue visit-dav/visit#19957.
Resolved for 3.4.2 on 3.4RC in #19969 …
CRobeck commented on issue triton-lang/triton#5025.
cc: @Jokeren …
vsoch commented on issue milroy/flux-sched#8.
Test is done, and I cleaned up the Go testing file (and added that final test). Link to PR for future reference. The bug was that I needed to include the full path of the cluster -> rack in the graph to add (which I’m fairly sure you told me) and then one of my identifiers was wrong - a “file already exists” error. Fixing those two things resulted in a working grow API! :hamburger: …
vsoch commented on issue flux-framework/fluxion-go#12.
Note that the tests can’t pass here because they aren’t using the PR branch. They pass locally….
garlick commented on issue flux-framework/flux-security#186.
I grabbed your snapshot from last night and put it on my test cluster. This morning I made some minimal changes to job-exec on my issue#6406 branch to send SIGUSR1 instead of imp-kill when shell termination methods fail. Then I ran a test on my system by starting a long running job, sending the flux-shell a SIGSTOP, then canceling the job. Looks good! Flux on rank 0 logged:
…
trws commented on issue flux-framework/flux-sched#1292.
Looks good to me @milroy, thanks for all the work on this!…
grondo commented on issue flux-framework/flux-core#6370.
> it appears the system coverage builder has failed on the 60m timeout 3 times in a row. Unclear to me if this is due to the extra tests, which are “beefier” with all of the 50 meg file copies. Could be slowing things down a lot there or eating up too much disk. …
jameshcorbett commented on issue flux-framework/flux-coral2#233.
Thanks! Setting MWP….
JustinPrivitera commented on issue LLNL/conduit#1332.
We should enhance the documentation for partitioning and provide more and better examples….
v-dobrev commented on issue GLVis/glvis#319.
You can use F9/F10 to cycle through the boundary attributes – the output in the terminal will show which bdr attribute is currently shown. Another way is to load the mesh in the mesh-explorer miniapp in miniapps/meshing and then use menu option b to view the boundary attributes in different colors….
hppritcha commented on issue open-mpi/prrte#48.
sure why not. always nice to have a little excitement….
cmoussa1 commented on issue flux-framework/flux-accounting#522.
Thanks! Setting MWP here …
IanLee1521 commented on issue elastic/kibana#195829.
@Heenawter - yes, sorry, I meant if Andrea changes that options location, we’d still want this functionality to be available….
jedbrown commented on issue openjournals/joss-reviews#6936.
@vsangelidakis :wave: Hi, would you be available to review this submission for JOSS?…
jandrej commented on issue mfem/mfem#4476.
Thank you for the contribution!…
acfisher commented on issue mfem/mfem#4547.
For future testing I put together a minimal mesh with all of the element types in 3D. You can find it in the mfem repo data/tinyzoo-3d.mesh….
artv3 commented on issue mfem/mfem#4408.
Thanks all! …
IanLee1521 commented on issue elastic/integrations#11218.
Thanks for adding these @tehbooom ! #11234 doesn’t actually address the nginx part of this request though, so hoping to inquire further about that log specifically?…
jedbrown commented on issue EnzymeAD/Enzyme#2035.
I see the Bazel job is failing on main too….
cyrush commented on issue Alpine-DAV/ascent#1360.
We can use pipelines with the new transform filter to accomplish this. …
vladotomov commented on issue mfem/mfem#4552.
Did you download METIS from https://github.com/mfem/tpls?
…
jedbrown commented on issue EnzymeAD/rust#183.
Thanks. Seems to work (accepts the option), but rustbook/samples-loose-types is currently failing…
trws commented on issue spack/spack#47136.
FWIW, fedora actually builds with the system llvm if the version is sufficient: https://src.fedoraproject.org/rpms/rust/blob/rawhide/f/rust.spec …
tzanio commented on issue mfem/mfem#4550.
Can you give me permissions so I can push to your branch?…
tzanio commented on issue mfem/mfem#4539.
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….
tzanio commented on issue GLVis/glvis#318.
Discussion continued in https://github.com/mfem/mfem/pull/4539…
jeremylt commented on issue CEED/libCEED#1709.
I’d double check the other Sycl JiT files …
jandrej commented on issue mfem/mfem#4483.
> For me, a much cleaner solution would be to make some virtual method of IterativeSolver, let us say CheckConvergence(), which would do the classical check of the norm, but the user could override it in a derived class from the solver to include some custom stopping criterion
cjekel commented on issue cjekel/piecewise_linear_fit_py#115.
closes https://github.com/cjekel/piecewise_linear_fit_py/issues/107…
correaa commented on issue boostorg/test#429.
Thanks @LuminaScript , but I don’t think I have the power to assign issues. I am not part of the project….
bwpriest commented on issue LLNL/MuyGPyS#245.
Thanks for the comments! The HeteroscedasticNoise class is the right tool for your use case. The current implementation does not play nicely with the MPI backend, so it is still considered an experimental feature. You are right it would be good to have an example for users. We will try to add one when there is time….
JustinPrivitera commented on issue visit-dav/visit#19876.
will need logic for both blueprint and mfem plugins…
cyrush commented on issue visit-dav/visit#19876.
This issue is specific to periodic meshes. …
cyrush commented on issue spack/spack#47161.
Thanks, I don’t foresee any issues with using 1.14. …
alecbcs commented on issue spack/spack#46765.
Merging in develop to trigger a new CI run based on updated binaries from develop….
v-dobrev commented on issue mfem/mfem#4523.
@helloworld922, I was not able to reproduce the issues you mentioned with HYPRE+HIP+UVM builds….
cyrush commented on issue mfem/mfem#4547.
@mlstowell can you provide the output file you saved from mesh explorer? Then I can test with newer VisIt and debug if necessary. …
garlick commented on issue flux-framework/rfc#427.
Thanks! I vote we cross that bridge when we come to it. Setting MWP….
wihobbs commented on issue flux-framework/flux-docs#285.
Thanks @jameshcorbett! I’ll set MWP….
wihobbs commented on issue flux-framework/flux-core#6391.
Sorry I misunderstood! Just wanted to throw that sort of niche edge case in there….
garlick commented on issue flux-framework/flux-core#6391.
Hmm, that’s definitely possible to check at connect time, although two concerns: …
chu11 commented on issue flux-framework/flux-core#6370.
as an aside, I looked into using “negative credits” to get the flux-exec stdin going faster. For the moment, decided not to add it. It would require some “did I send the first wave of data” flags, and didn’t think it super worth it….
grondo commented on issue flux-framework/flux-core#6390.
Thanks @wihobbs! I actually just pushed one more doc update here: A note/warning about calling start() multiple times.
…
emily-howell commented on issue Alpine-DAV/ascent#1397.
@cyrush To answer your above questions: …
samuelkgutierrez commented on issue openpmix/openpmix#3424.
You bet!…
alizma commented on issue openjournals/joss-reviews#6838.
@crvernon Fixed, please see here….
milroy commented on issue flux-framework/flux-sched#1292.
> Would you agree @milroy? …
trws commented on issue flux-framework/flux-sched#1292.
It sounds like the complexity favors doing the unconditional full cancel, and the performance doesn’t really push away from that much either. Would you agree @milroy? If so, then lets do that. I have a feeling we should keep this in mind as we think through what we want to do when evolving the planner setup….
cmoussa1 commented on issue flux-framework/flux-accounting#512.
Thanks! Setting MWP here…
balos1 commented on issue LLNL/uberenv#136.
Pinging @adrienbernede ….
jedbrown commented on issue CEED/libCEED#1706.
We wanted libCEED to be usable with unusual toolchains and with more user control, but without needing to edit files. CMake requires platform/toolchain files for this purpose, and both it and autotools run lots of checks at configure time. You could set OPT=$(CFLAGS) and unset CFLAGS if you want the convenience of not specifying -fPIC….
CamStan commented on issue spack/spack#45737.
> There are pipeline failures, for instance: https://gitlab.spack.io/spack/spack/-/jobs/12236018 …
artv3 commented on issue mfem/mfem#4408.
> I noticed that lor-transfer-p.cpp uses “CPU time” (sum of all ranks times), which I think is what you report in the initial pull request?
…
jameshcorbett commented on issue flux-framework/flux-docs#285.
It depends on how much work there is to be done. If you think you could do it without a ton more effort that would be nice, because the proper way to review this PR as it is now would be to manually verify that all the links work which is tedious…
grondo commented on issue flux-framework/flux-docs#284.
It appears to have been dropped in commit 0d9d2b567fb26852f1864912f176cd145c80daf2, potentially by accident….
cyrush commented on issue Alpine-DAV/ascent#1404.
@nicolemarsaglia …
trws commented on issue spack/spack#47136.
Quite fair, and also we gain an important benefit combining the llvms, the ability to do cross-language LTO across all of c, c++ and rust. Like I said, all for that part as long as it
jedbrown commented on issue openjournals/joss#1283.
Note that Elsevier and Springer Nature (along with many professional organizations and national councils) are signatories of the Declaration on Research Assessment (DORA), which vows to avoid metrics like impact factor and focus on scientific content. My department incorporated some of this language in revised promotion and tenure guidelines, and were able to hire on that basis. I understand the process takes time, but change is achievable. …
samuelkgutierrez commented on issue hpc/quo-vadis#283.
Physical numbering should not be used. If it is, that’s a regression in behavior. …
corbett5 commented on issue ITensor/ITensors.jl#1546.
> …
trws commented on issue spack/spack#46910.
Ok, everything is passing but codecov. @tgamblin, @eugeneswalker any comments/concerns?…
v-dobrev commented on issue mfem/mfem#4523.
@helloworld922, the issue you see is related to this issue: #4351. I’m testing a fix for that and I’ll push it when ready. …
jakubcerveny commented on issue jakubcerveny/gilbert#16.
Hi, I agree that a better ordered flatten could be useful for AI, image compression, and maybe more.
…
vsoch commented on issue oras-project/oras-py#165.
@tarilabs I understand your logic - the question is if a user is ever going to call get_auth_backend outside of the provider. In that case, they could do something like:
…
hppritcha commented on issue open-mpi/ompi#12872.
Could you try running …
tzanio commented on issue mfem/mfem#4436.
ping: @bslazarov …
trws commented on issue flux-framework/flux-sched#1309.
This might be a case for keeping the
grondo commented on issue flux-framework/flux-core#6380.
> Is it possible we’ll need other “prefix options” in the future, like sort:foo,bar foooption:blah baroption:blah {t_submit} {id} {user} …, etc. …
corbett5 commented on issue ITensor/ITensors.jl#1546.
> …
cyrush commented on issue visit-dav/visit#19770.
FYI: this does not cleanly apply to develop. …
tgamblin commented on issue spack/spack#46973.
> I am still a bit concerned about the case where C overrides D, and B extends D. …
milroy commented on issue milroy/flux-sched#6.
LGTM; thanks!…
jameshcorbett commented on issue milroy/flux-sched#6.
Done!…
samuelkgutierrez commented on issue hpc/quo-vadis#283.
This is my understanding, but I would double check this (and please correct me if I’m wrong). …
garlick commented on issue flux-framework/rfc#430.
OK then, I’ll go ahead and set MWP. Apparently azure cloud or whatever can’t get to the venti paper link even though I can from home, so the link check is failing. I’ll wait a bit in restart that builder….
grondo commented on issue flux-framework/rfc#430.
> Repushed with those changes. I guess the big question is whether we think it’s reasonable to do away wtih flux imp kill in favor of the IMP lingering and forwarding signals in the run case. …
milroy commented on issue flux-framework/flux-sched#1309.
@jameshcorbett pointed out that even the LOG_WARNING will fill the logs on systems with rabbits or brokerless ssds. I think this is a compelling argument for making it a LOG_DEBUG….
cmoussa1 commented on issue flux-framework/flux-accounting#517.
This should be fixed by #518 now! Closing….
grondo commented on issue flux-framework/flux-accounting#519.
If it is helpful, there’s examples of how to subclass the flux.util.OutputFormat class to easily add a -o, --format option to a command in the flux-core codebase.
…
jameshcorbett commented on issue LLNL/scr#596.
I tested with scr_run srun ... on Ruby, I think this is good to go….
cyrush commented on issue LLNL/conduit#1328.
This looks great! …
JustinPrivitera commented on issue LLNL/conduit#1322.
Document all of it where Arlie has documented here: #1328…
BradWhitlock commented on issue LLNL/axom#1451.
This merged PR initializes m_executeOnGPU. I ran into that with valgrind on rzansel….
corbett5 commented on issue GEOS-DEV/GEOS#3273.
@rrsettgast @wrtobin I think you guys should give this a look over. I commented a few of the pain points. It passes most of the integrated tests on ruby (except a few FPE runtime errors and restart checks I think are unrelated to this PR) and I have been unable to run them on Lassen. …
cyrush commented on issue Alpine-DAV/ascent#1389.
@TobyFlynn this should be resolved on ascent develop. …
JustinPrivitera commented on issue Alpine-DAV/ascent#1404.
I know there are issues with building VisIt docs on BlueOS systems. Maybe same problem for Ascent?…
adrienbernede commented on issue spack/spack#47044.
Definitely a problem on my side……
v-dobrev commented on issue spack/spack#47044.
Hi @adrienbernede, hipsparse is a required dependency under the specified conditions: @4.4.0:+rocm. The assumption in the mfem package is that all ROCm/HIP packages (hip, hipsparse, hipblas, rocblas, rocrand, etc) are all either external or they are all installed by Spack – anything else does not really make sense to me. Therefore, the logic for hipsparse checks if hip is external and handles the two cases based on that.
…
tgamblin commented on issue hpsfoundation/tac#12.
TACC voted on 2024-10-03 to approve Trilinos at the “established” level
jeremylt commented on issue freeCodeCamp/freeCodeCamp#54767.
Maybe no so users can play with the difference between the two?…
wihobbs commented on issue flux-framework/rfc#414.
Ideas that were discussed, but not included in the RFC: …
garlick commented on issue flux-framework/flux-security#184.
> Perhaps we should add this to the RFC, then implement once we’re happy with the results? …
jameshcorbett commented on issue flux-framework/flux-sched#1292.
> @jameshcorbett this PR is almost ready for review. First, I’d like to integrate the tests you wrote that reproduced the behavior and helped me fix it. Can you make a PR to my fork on the issue-1284 branch (https://github.com/milroy/flux-sched/tree/issue-1284)? …
chu11 commented on issue flux-framework/flux-core#6353.
re-pushed, re-working the PR per discussion above…
chu11 commented on issue flux-framework/flux-accounting#518.
> If we’re past that then agreed we can probably just drop the tests. …
garlick commented on issue flux-framework/flux-accounting#518.
My concern was that we don’t necessarily want to lose coverage for reading and parsing the job-archive database if we still need to migrate some production systems to the new job table. If we’re past that then agreed we can probably just drop the tests….
wihobbs commented on issue flux-framework/flux-accounting#514.
Some longer term things you could consider doing: …
agcapps commented on issue LLNL/axom#1441.
Thank you for this work, @kennyweiss . So far it looks good. What docs do you plan to add?…
adrienbernede commented on issue LLNL/RAJAPerf#480.
@rhornung67 Is the floating point model something that can be changed for RAJA only, or should it propagate to the project using RAJA as a dependency ? In the former case, we could apply the fly via CMake or via Spack. In the latter case… well I hope it’s not the latter……
jandrej commented on issue EnzymeAD/Enzyme#2062.
ping @wsmoses …
eliasboegel commented on issue CEED/libCEED#1694.
I currently keep my small change in https://github.com/eliasboegel/libCEED. I use this as for the Cargo dependency, and disable static with default-features = false.
…
jedbrown commented on issue CEED/libCEED#1694.
Okay, thanks. Would you mind putting what you’re thinking in a branch? I still wonder if there is some mixing of Apple and gcc/clang upstream linking tools that is interfering with weak symbols, but perhaps it just doesn’t work with static libs?…
jeremylt commented on issue CEED/libCEED#1696.
Not in this PR, no….
dylan-copeland commented on issue mfem/mfem#4371.
@hughcars The series of refinements you described is actually not possible, which is a different problem. I modified ref321 to perform these refinements, and the 1/3 Y-refinement of the right element does not introduce a new vertex. Instead, it identifies the vertices for the left and right refinements, putting them at the same location. I will have to think about how to handle this case….
jandrej commented on issue mfem/mfem#4483.
pong @tzanio …
artv3 commented on issue mfem/mfem#4408.
Hi all, thanks for the review. Just got back from vacation, will start addressing comments/feedback this week. …
wihobbs commented on issue flux-framework/flux-core#6373.
Ooh I wonder how hard it would be to make the old link point to the new link instead of a 404, in case people have it bookmarked. …
jameshcorbett commented on issue flux-framework/flux-core#6371.
That might be nice!…
dylan-copeland commented on issue GLVis/glvis#318.
@JonLangdon can you please let us know if branch ncmesh-attr works for you? This sets the element attributes in NCMesh. Since this is an MFEM bug, not glvis, you should only need to rebuild MFEM, not glvis.
…
trws commented on issue spack/spack#46910.
@scottwittenburg, any idea how I could be getting an error that it couldn’t find a spack.lock in the generate phase for the stack? https://gitlab.spack.io/spack/spack/-/jobs/12974881…
alizma commented on issue openjournals/joss-reviews#6838.
Hi @HaoZeke, apologies for the delay, life got a bit busy. I made a new release (v1.0.3 that syncs all the changes from the reviews and uploaded to a Zenodo archive, please find it here. Please let me know if there’s anything else to complete on our end. Thank you for your help and feedback!…
trws commented on issue neovim/neovim#30749.
Ok, here’s a version that does the check using otool -hv to dump the Mach-O header, then checks to see if that contains “BUNDLE” in which case it prints an informative error message….
vsoch commented on issue hanwen/go-fuse#535.
The first thing I tried is to try and inherit the functions, e.g., …
jeremylt commented on issue freeCodeCamp/freeCodeCamp#56642.
You still are not following the agreed upon solution….
chu11 commented on issue flux-framework/flux-core#6367.
> raising the priority of the prep watcher still has it running in essentially the same place, after all other watchers except prep watchers, which generally are not doing “real work” anyway (so being first won’t buy much). …
garlick commented on issue flux-framework/flux-core#6367.
Got it. I guess my two points were: …
cjekel commented on issue cjekel/piecewise_linear_fit_py#113.
Thank you! I’m tracking to try to get this in a new release this weekend! LGTM….
markcmiller86 commented on issue visit-dav/visit#19898.
@alhom…in your submission you explain that an HTG reader “could be useful” or “would be rather nice”. You go on to explain that “we have our own file format, and a VisIt plugin for it”. …
tgamblin commented on issue spack/spack#46924.
Oh ok, I see what you are saying. I think we shouldn’t bother raising if the splice wouldn’t even be executed b/c the roots don’t depend on the spliced spec….
jedbrown commented on issue rust-lang/rust#124874.
I rebased and updated this PR to include the f16 and f128 intrinsics (thanks @saethlin), and I can use them when I enable the corresponding features, with or without hardware support (e.g., vfmadd213sh are issued for f16 with target-cpu=sapphirerapids).
…
dylan-copeland commented on issue mfem/mfem#4371.
> High level question: how can this approach reconcile a conformal interface? …
tzanio commented on issue mfem/mfem#4461.
ping: @sohailreddy …
chu11 commented on issue flux-framework/rfc#427.
> I hate to make every client register a credit callback. …
garlick commented on issue flux-framework/rfc#427.
Maybe we should establish a minimum buffer size of 4K or so and say that if a subprocess client wants to ignore credits (not register the callback) they can behave as before, but if they send more than 4K of data, it will not be reliable? …
tpatki commented on issue flux-framework/PerfFlowAspect#171.
Hi @slabasan, I think we should close this PR as well, given our discussion. Let me know….
vsoch commented on issue expfactory/experiments#61.
It looks great! :space_invader: …
IanLee1521 commented on issue elastic/kibana#195829.
Thanks Chris. cc/ @davidherst …
vsoch commented on issue converged-computing/fluxnetes#20.
Note that we had the same bug in fluence: https://github.com/flux-framework/flux-k8s/pull/85…
jandrej commented on issue EnzymeAD/Enzyme#2062.
ping @wsmoses …
jeremylt commented on issue CEED/libCEED#1646.
@jrwrigh CI has been updated, so HONEE can be updated now…
JustinPrivitera commented on issue visit-dav/visit#19168.
No longer an issue; perhaps was fixed in #19864…
vsoch commented on issue urlstechie/urlchecker-python#93.
Thank you @mabraham!…
astrophpeter commented on issue openjournals/joss-reviews#7342.
Ok, I think everything is fixed now @warrickball . Thanks for your help. …
jedbrown commented on issue openjournals/joss-reviews#6099.
I think the updates (with stronger static guarantees) are a novel and interesting contribution, and am fine with pausing this for a few weeks so you can include it. When ready, if you can point to release notes or otherwise share highlights here, I’d like reviewers to be aware of what changed….
hppritcha commented on issue open-mpi/ompi#12723.
Okay i’ll put the mtl code back in….
v-dobrev commented on issue mfem/mfem#3480.
The nightly tests show a number of failures from sample runs in ex9, ex10, ex18, ex10p, ex18p – the new ODE selection mechanism does not seem to be backward-compatible. I see some changes in the numbers of the -s/--ode-solver option in some examples, however, it looks like more changes are needed in the examples that are failing.
…
grondo commented on issue flux-framework/rfc#428.
Sorry if this was discussed elsewhere, but should we take a step back before breaking the protocol? Would it make sense to keep the current rexec flags and add a new rexec_flags member or mask the subprocess flags into the current flags member, rather than renaming?
…
wihobbs commented on issue flux-framework/rfc#414.
I need to reread some of the previous comments before asking for another set of reviews. This is a pretty substantial update to the proposed design. …
wihobbs commented on issue flux-framework/flux-sched#1140.
Sorry @trws I completely forgot about testing this. Just did and it worked exactly as expected. Thanks!…
cmoussa1 commented on issue flux-framework/flux-docs#280.
This should be fixed by #281; closing this!…
grondo commented on issue flux-framework/flux-core#6082.
Yeah, I was struggling with NO_SETPGRP too, but the simplicity is indeed appealing. I think this PR is probably fine as is….
grondo commented on issue flux-framework/flux-accounting#501.
Check .readthedocs.yaml in the flux-core source (which also includes a link to the rtd documentation)
…
bgunnar5 commented on issue LLNL/merlin#490.
Thanks for looking at this @koning …
tpatki commented on issue ovis-hpc/ldms#1447.
Hi @tom95858 : This is draft PR at the moment. Kathleen (@kshoga1) will need to test this at scale internally. We plan to do that in this month. Once we have tested at scale, we’ll address Chris’ comments, update it as ready for review, and it can be merged….
hppritcha commented on issue open-mpi/prrte#36.
@gagarc Jeff and I decided that it would be better to retarget this PR https://github.com/uofl-capstone-open-mpi/prrte/capstone-devel…
JustinPrivitera commented on issue LLNL/conduit#1326.
I would argue that conduit should only read number_of_elements elements instead of reading until it hits a null terminator…
cyrush commented on issue LLNL/conduit#1326.
When i see number of elements ==1, there maybe an expectation that is a null termed string. …
JustinPrivitera commented on issue Alpine-DAV/ascent#1396.
Strange. Glad you are no longer running into your issue though! Feel free to re-open if it comes up again….
adrienbernede commented on issue mfem/mfem#4431.
@v-dobrev @johnbowen42 latest pipeline run: …
jameshcorbett commented on issue flux-framework/flux-sched#1255.
On Hetchy, JGF for jobs appears to be around 15K per node….
jameshcorbett commented on issue flux-framework/flux-accounting#498.
Sounds like this has been resolved offline @cmoussa1 ? In which case feel free to close….
agcapps commented on issue LLNL/conduit#1316.
@cyrush , I’d appreciate your further review. I think I got all the points you mentioned….
adrienbernede commented on issue LLNL/Umpire#911.
@davidbeckingsale This PR is ready !…
adrienbernede commented on issue LLNL/RAJA#1740.
@rhornung67 provided the test pass after updating with develop (likely) this is ready to go….
bwpriest commented on issue LLNL/MuyGPyS#244.
Thanks @esheldon for investigating. There is a known incompatibility with recent versions of JAX in Python >= 3.9 arising from their config objects. We can fix this in a future release, but in the meantime thank you for identifying a compatible version of JAX. …
adrienbernede commented on issue LLNL/Caliper#588.
Superseded by #593 …
adrienbernede commented on issue LLNL/CHAI#283.
@adayton1 This PR is ready….
adrienbernede commented on issue LLNL/CARE#293.
@adayton1 This PR is ready….
jedbrown commented on issue CEED/libCEED#1683.
Let’s do it on CUDA only. Do you know about how much of the slower ROCm time is attributable to the weaker GPU versus slower RTC?…
cyrush commented on issue visit-dav/visit#19879.
I think we would want this both on RC and develop if possible. …
jedbrown commented on issue CGNS/CGNS#778.
FWIW, that is still a race condition in serial. It might not trigger frequently because people aren’t super likely to have a concurrent job (possibly on a different machine that accesses the same network file system) moving/replacing files, but that does happen. Even if the file still exists between the probe and open call, it could have been replaced by one with a different file type (ADF vs HDF5). It’s kind of unfortunate there is no H5F open with file descriptor to avoid such races. …
tgamblin commented on issue spack/spack#46608.
@alecbcs @adamjstewart was this supposed to be merged? Last comment says …
JustinPrivitera commented on issue visit-dav/visit#19859.
merge w/ develop in #19880…
hppritcha commented on issue open-mpi/ompi#12723.
Any objections or suggestions for changes?…
grondo commented on issue flux-framework/flux-sched#1307.
Setting MWP….
garlick commented on issue flux-framework/flux-sched#1307.
This fixes it for me!…
vsoch commented on issue awslabs/soci-snapshotter#1389.
I got it working! But I need to figure out if I can automate it, will report back….
gardner48 commented on issue LLNL/sundials#588.
Current PRs do not need to adopt this style but future additions to the docs should follow this guide. As time allows we can migrate existing documentation to this style. …
jeremylt commented on issue Eudicods/outworlds-wastes#115.
It would be substantially easier to review only corrections for typos and grammar errors rather than a full rewrite of everything….
v-dobrev commented on issue xsdk-project/xsdk-issues#258.
> BTW: Would there be a mfem@4.7.1 release - with these updates? …
trws commented on issue spack/spack#46784.
I don’t think it is. My guess would be that because we pass the makeflags down through the environment, so the jobserver can work, the --keep-going flag is getting propagated through spack and causing the output to change. This seems reasonable to me, any issues @haampie?…
adrienbernede commented on issue spack/spack#42701.
@simonpintarelli @msimberg …
tzanio commented on issue mfem/mfem#4123.
Merged in next for testing……
cjvogl commented on issue mfem/mfem#3480.
@IdoAkkerman , sorry that this appears to have fallen off the collective radar of the MFEM team. I’ll bring it up at the next meeting (and will see if I can’t help figure out why the single precision test is failing)…
jeremylt commented on issue freeCodeCamp/freeCodeCamp#56518.
If someone makes a PR, this is what I’d say…
balos1 commented on issue LLNL/sundials#581.
Fixed in #584. …
jeremylt commented on issue CEED/libCEED#1681.
With this diff (we should maybe apply here …
cyrush commented on issue visit-dav/visit#19501.
@markcmiller86 if you run with -dump, you can get visibility with what happening with respect to the ghost/facelist filtering. …
alecbcs commented on issue spack/spack#46418.
@spackbot run pipeline…
jedbrown commented on issue sarah-ek/faer-rs#153.
Would you mind updating the figures for nanogemm? …
v-dobrev commented on issue mfem/mfem#4517.
There is a comment about this issue just after the line you edited in defaults.cmake: https://github.com/mfem/mfem/blob/8ed11a5c9a1f8af4401562dfd2df354304b36145/config/defaults.cmake#L158-L164
…
jameshcorbett commented on issue flux-framework/flux-sched#1284.
Ok @milroy I think I have an improved reproducer at https://github.com/jameshcorbett/flux-sched/tree/issue-1284…
wihobbs commented on issue flux-framework/flux-core#6341.
Oh, I think I see: …
cmoussa1 commented on issue flux-framework/flux-accounting#226.
I’m not sure if I’ve been able to reproduce this issue in my local Docker environment or on our test cluster, so I think I’ll close this for now, but can always re-open if we see this issue again….
adrienbernede commented on issue LLNL/Caliper#593.
@daboehme It appears that recent changes in Caliper main branch fixed the issues we were seeing with cce compilers. …
corbett5 commented on issue GEOS-DEV/LvArray#329.
> > Adding hyperbolic sine because it’s needed in GEOS-DEV/GEOS#3280 …
CRobeck commented on issue triton-lang/triton#4836.
> I’d suggest adding sort and percentage in different PRs. …
vsoch commented on issue singularityhub/singularity-hpc#673.
hmm :/ @marcodelapierre what do you think?…
jedbrown commented on issue rsmpi/rsmpi#197.
panic! must be doing some file descriptor tinkering or something. Replacing it with the two lines
…
vsoch commented on issue pydicom/deid#268.
another idea is to have a set of helper scripts - not integrated into the main client, but optional to install (and support this extra functionality)….
jedbrown commented on issue openjournals/joss-reviews#6099.
Sorry about my delay. I was unclear if it was considered ready, though I checked with Sarah and it looks good. I submitted a PR with some bib/copy edits and a couple questions. Once resolved, we’ll be ready to move forward as far as I’m concerned….
uvilla commented on issue hippylib/hippylib#75.
Thank you for your prompt reply. …
BradWhitlock commented on issue LLNL/axom#1434.
I added a test that approximates what I was doing in my MIR branch. There was some weirdness with RelWithDebInfo but not a crash. The test calls a new function axom::isDeviceAllocator() to exercise it….
jedbrown commented on issue CEED/libCEED#1677.
There could be a libceed-builder crate (to be used by build.rs) that would automate locating and building such things. The README comment is about shared library paths if you disable the static feature. If you’re having Cargo build libCEED, there isn’t much reason to disable static.
…
jandrej commented on issue mfem/mfem#4527.
I would try invoking nvcc by passing the options -Xcompiler -std=gnu++14, which will set the host compiler c++’s dialect version to match nvcc’s toolchain.
…
CRobeck commented on issue iree-org/iree#18347.
@josemonsalve2 did you get this figured out?…
wihobbs commented on issue flux-framework/flux-docs#279.
It must’ve gotten confused with the add/remove but I’m surprised refresh didn’t fix that. Oh well…
jameshcorbett commented on issue flux-framework/flux-docs#279.
@wihobbs not sure what happened to mergify can you merge this for me?…
chu11 commented on issue flux-framework/flux-core#6328.
Could a --list option be a good compromise here? So we keep the default output with well known / defined keys. Tweak the output to say --list for full list and that option outputs the recursive true list?…
eliasboegel commented on issue CEED/libCEED#1677.
Perhaps the larger problem is informing cc about the include directory for libCEED such that ceed.h is accessible in the QFunction source file. That seems especially difficult (or impossible?) if libCEED was built by the bindings. Essentially, cc would need to be supplied in the users build.rs with a path to target/{profile}/build/libceed-sys-{hash}/out/include/ but I don’t know any way to automatically retrieve the right hash, nevermind a way that is simple enough that a user can be reasonably expected to figure it out. I saw in the libceed-sys readme that you’ve had a similar problem in the past?…
cyrush commented on issue Alpine-DAV/ascent#1391.
It seems like the output is malformed – extra text should not be there. Not sure why that is happening. The pages are created to avoid need for a webserver, so it should not matter if server is used or not. We will check out cinema output routines. …
artv3 commented on issue mfem/mfem#4408.
Hi folks, just a friendly reminder about this PR. …
trws commented on issue flux-framework/flux-sched#1302.
I think that’s the list, will go over the JGF simplification shortly, I just set MWP on the Dense resource type PR, because we’re actually testing the error report in an old test, will add more to that but I think we need that fix in ASAP and I’m not sure I can get to the addition in a reasonable time. …
IanLee1521 commented on issue elastic/kibana#194240.
Thanks all appreciate it and expect that will indeed help us out (curse our O(100s Billions) of documents!…
bwpriest commented on issue LLNL/MuyGPyS#236.
Apologies, this was due to a dependency in the then-current version of hnswlib around the time of the v0.8.2 release. This has been fixed in the latest release….
cjekel commented on issue cjekel/similarity_measures#38.
@nucccc Nice catch! I can see how this being vectorized will be substantially faster. Thanks! …
cjekel commented on issue cjekel/piecewise_linear_fit_py#113.
@fredrikhellman Thanks for this addition! …
tgamblin commented on issue spack/spack#42931.
@spackbot Fix style…
garlick commented on issue flux-framework/flux-core#6323.
Great point. …
tpatki commented on issue flux-framework/PerfFlowAspect#130.
@slabasan …
gardner48 commented on issue LLNL/sundials#541.
It would be good to add a test similar to test/unit_tests/ark_test_dahlquist_erk.cpp that verifies there are no extraneous RHS evaluations…
adayton1 commented on issue LLNL/axom#1429.
/style…
eliasboegel commented on issue CEED/libCEED#1677.
> The JiT backends know Q and that is a compile time constant for them. The number of components is compile time constant from the basis. …
jeremylt commented on issue CEED/libCEED#1677.
The JiT backends know Q and that is a compile time constant for them. The number of components is compile time constant from the basis….
cyrush commented on issue visit-dav/visit#19858.
For some databases, there is no concept of a root file - they expect to open a directory and look for data in files in that directory to assemble the mesh metadata and data. …
alecbcs commented on issue spack/spack#23224.
@skosukhin @scheibelp is this PR still something you all want to merge or has this been fixed elsewhere?…
adrienbernede commented on issue spack/spack#45648.
@trws I need you to hit the “approve” button. ;)…
hppritcha commented on issue open-mpi/prrte#20.
Yes the bot is complaining about the cherry-picked commit not being in this repo. Do you know if there’s a way to customize the string generated by git when using this “-x” option? google isn’t helping……
dylan-copeland commented on issue mfem/mfem#4514.
There is some documentation on the MFEM website, relevant to your questions: …
v-dobrev commented on issue mfem/PyMFEM#253.
One option, as discussed in #57, may be to derive your own class from mfem.MatrixPyCoefficient as illustrated in https://github.com/mfem/PyMFEM/blob/master/examples/wrapper_example/matrix_coefficient.py.
…
correaa commented on issue llvm/llvm-project#102817.
How the algorithm know that is can exploit segmented iterators now?…
trws commented on issue flux-framework/flux-sched#1298.
We should get the PR out ASAP. There must be some resource type string that isn
grondo commented on issue flux-framework/flux-sched#1298.
I think the rank 0 broker on tuolumne hit this issue and crashed last night: …
wihobbs commented on issue flux-framework/flux-core#6314.
It’s friday and clearly my brain is turning to mush. @grondo suggested earlier today that for non-fatal exceptions, continue to print the exception but not the “canceling” message, and proceed to “waiting for prolog.” That’s cleaner….
chu11 commented on issue flux-framework/flux-core#6322.
Note #3066 …
balos1 commented on issue LLNL/radiuss-shared-ci#38.
@rhornung67 I was thinking that the other projects copy the subscribed-pipelines.yml into their own repository (rather than include it from here). So other projects could decide to comment out/delete the Dane pipeline if they don’t want it. This seems to be what RAJA is doing (see https://github.com/LLNL/RAJA/blob/develop/.gitlab/subscribed-pipelines.yml). It would be good to know though if other projects are in fact including the file rather than using it as a template and making their own copy of it. Regardless, its great to test on more machines I suppose too! Happy to contribute.
…
adrienbernede commented on issue LLNL/radiuss-shared-ci#38.
Since Dane is a new machine, then I suspect flux to be the default scheduler and slurm to be a wrapper around it. If that’s the case, then I think Dane shared CI should use flux, as it’s the case for tioga and corona for example….
adayton1 commented on issue LLNL/RAJA#1744.
@kennyweiss, I will be on parental leave for the next two weeks. But I put up a branch for you to try: https://github.com/LLNL/RAJA/pull/1746…
artv3 commented on issue LLNL/RAJA#1728.
@johnbowen42 @rhornung67 can I get another review? I just pushed up the changes I thought I had pushed up. …
cyrush commented on issue Alpine-DAV/ascent#1389.
@TobyFlynn thanks for this concise reproducer! We will work on resolving this….
markcmiller86 commented on issue visit-dav/visit#19826.
@biagas GFM has no options for coloring text. I thought you might be able to if you fell back to raw html as in…<span style="color: red;">This text is red.</span> but that doesn’t work.
…
trws commented on issue spack/spack#45648.
I’m just getting back from Australia so I’m not sure I’ll be faster, but I can try. What do you still need?…
adrienbernede commented on issue mfem/mfem#4431.
Latest pipeline run: …
eliasboegel commented on issue helix-editor/helix#1662.
I very much like this idea. Having an empty line that I rarely look at below the status line permanently is annoying. Having it pop in above the status like when a command is being written seems like a great compromise. …
garlick commented on issue flux-framework/flux-sched#1301.
Have to leave right now but one thing that seems wrong is flux job taskmap shows the same map for both of those cases.
…
trws commented on issue flux-framework/flux-core#6303.
Awesome, thanks @grondo! …
grondo commented on issue flux-framework/flux-coral2#223.
Well that list is just hardcoded in the flux job info command. I wonder, though, if instead the command with no args could list available keys that the current user can fetch? Maybe open an issue in flux-core if that idea is appealing….
garlick commented on issue flux-framework/flux-coral2#222.
I think the taskmaps are actually correct and I was confused. Fluxion is packing 4 ranks onto the first node in both cases, and 3 on the second, but for some reason when -N is specified, the order of nodes is reversed….
vsoch commented on issue eksctl-io/eksctl#6870.
Thank you! And no worries about stale bot - it can be very helpful. I’m subscribed to the thread and am good to ping when it needs to be reopened….
JustinPrivitera commented on issue LLNL/conduit#1111.
How is specmix_quad.silo able to have multiple materials defined? Am I not remembering how these things work?…
adrienbernede commented on issue LLNL/RAJAPerf#477.
@rhornung67 This PR was merged it #480….
vsoch commented on issue hpc-social/hpc-social.github.io#78.
Closed with #79 …
agcapps commented on issue LLNL/axom#1424.
@bmhan12 , thank you for your attention to detail on this….
jedbrown commented on issue CGNS/CGNS#778.
access is always racy if the calling code has logic depending on the result. Suppose an archival job is running in the background, moving files to slower storage after a certain age. Then your production job checks access and sees the file, but it’s gone by the time it tries to open it. It is likely that access will be formally deprecated in a future version of POSIX. I think CGNS is important enough that it should not rely on such race conditions, and instead correctly handle errors from the likes of H5Fopen….
cjvogl commented on issue mfem/mfem#4513.
> @cjvogl , that makes sense, but where would you put it? It is just an interface, so FluxFunction::ComputeAvgFlux() has just MFEM_ABORT(). The only implementations are in AdvectionFlux and BurgersFlux, which are both scalar by construction
jameshcorbett commented on issue flux-framework/flux-sched#1298.
I can try. Alternatively, Tom said he should be back Friday….
jedbrown commented on issue CEED/libCEED#1662.
I would always put in a specified ordering when saving to disk, rather than relying on implicit matching or attaching metadata to reconstruct a specific backend ordering….
v-dobrev commented on issue spack/spack#46537.
Looking at the failure, it looks like -lsci_cray needs -ldl – the lld linker does not allow linking with unresolved symbols.
…
jakubcerveny commented on issue jakubcerveny/gilbert#15.
Thanks!…
cmoussa1 commented on issue flux-framework/flux-accounting#490.
Sounds great @ryanday36! Thanks for taking a look at it. I’ll mark this as ready for review then….
corbett5 commented on issue GEOS-DEV/LvArray#333.
I don’t think we want to do this in LvArray, plus if you put it there then anytime LvArray::internal::getArrayManager is called call backs will be disabled even if the user wanted to enable them. I would put it somewhere high up in GEOS (perhaps directly in main? https://github.com/GEOS-DEV/GEOS/blob/f9a64bb2d1fe34eab99adc25ea1d4be822eca129/src/main/main.cpp#L35) and then remove it from ProblemManager https://github.com/GEOS-DEV/GEOS/blob/f9a64bb2d1fe34eab99adc25ea1d4be822eca129/src/coreComponents/mainInterface/ProblemManager.cpp#L142
…
tgamblin commented on issue spack/spack#46478.
We call socket.getfqdn() when writing the database, as well. It’s used in the tempfile name. I wonder if that is also causing a performance penalty on macOS….
tzanio commented on issue mfem/mfem#4480.
ping: @kmittal2 …
jameshcorbett commented on issue NearNodeFlash/NearNodeFlash.github.io#159.
A closely related Flux issue (although it’s not obvious how it’s related): https://github.com/flux-framework/flux-sched/issues/1255…
JustinPrivitera commented on issue visit-dav/visit#19836.
Things have changed since my last failed build. I may have messed up with the build invocation as well. All is working as intended now….
markcmiller86 commented on issue visit-dav/visit#19501.
@markcmiller86 has data now….
cyrush commented on issue visit-dav/visit#19768.
may be related to #19544 …
tpatki commented on issue ovis-hpc/ldms#1447.
@morrone …
jandrej commented on issue mfem/mfem#4480.
Probably better this way. Looks good….
cjvogl commented on issue mfem/mfem#4480.
Looks like it was a bug: the Picard iteration works fine so long as KINSolver::EnableAndersonAcc isn’t called. I’ll push a fix momentarily (with a bit of refactoring to avoid combinations of features that cause issues)….
garlick commented on issue flux-framework/flux-core#6291.
That seems expected to me since the consumer is in charge of the rate of date transfer….
chu11 commented on issue flux-framework/flux-core#6291.
hmmmm, …
wihobbs commented on issue flux-framework/flux-coral2#218.
wowemage of Bahroo approves…
IanLee1521 commented on issue elastic/logstash#16468.
Definitely anything that can be done to make it easier to preview, save, replay, simulate, etc logs and what the processing will do to them as they go through Logstash would be a huge win. This for us is something that we have started looking at Cribl as a potential solution for. They have a very nice method for doing this capture, and live preview of logs as they are being processed….
markcmiller86 commented on issue HDFGroup/hdf5#4849.
Just one other observation to share here and that is that the libver bounds property is associated with file access, not file creation properties. Therefere, I would NOT expect it to have affected the file creation. Furthermore, the documentation regarding setting of libver bounds (in props or on an open file) does not speak specifically to file creation…only object creation in a file….
v-dobrev commented on issue GLVis/glvis#316.
> Btw, is Ctrl+A documented? I do not see it anywhere…
jeremylt commented on issue CEED/libCEED#1668.
The issue appeared after I got this commit and later, I think (haven’t bisected, just going off two rebuilds of PETSc for CI on back to back days) …
cyrush commented on issue Alpine-DAV/ascent#1385.
@indra098124 can you share your invocation? …
adrienbernede commented on issue spack/spack#45648.
@adayton1 please note the migration of CARE package to CachedCMakePackage (based on implementation in radiuss-spack-configs) …
tgamblin commented on issue spack/packages.spack.io#27.
Oops! Thanks!…
jedbrown commented on issue rust-lang/rust#124874.
Hi :wave: Just checking in about when this can be reviewed. I’m unclear if it needs all the listed reviewers or just a subset since it’s been pared down….
tgamblin commented on issue hpsfoundation/tac#2.
I think @bradcray could feel free to make a proposal….
grondo commented on issue flux-framework/flux-coral2#215.
Note that if you are using test execution then it is a lot less complicated. Just take the t/scripts/sign-as.py script from flux-core, and use this function:
…
garlick commented on issue flux-framework/flux-coral2#215.
It’s unfortunately a bit complicated to fake that, but have a look at https://github.com/flux-framework/flux-core/blob/master/t/t2404-job-exec-multiuser.t…
vsoch commented on issue converged-computing/performance-study#59.
Ping @asarkar-parsys this is your :baby: ! Oh no wait, no it’s not. The other other kind of baby. The one you throw out with the bath water. Just kidding :P :water_polo: …
markcmiller86 commented on issue betterscientificsoftware/bssw.io#2160.
@bernhold, @curfman and @rinkug I think this is ready to publish….
artv3 commented on issue LLNL/RAJA#1737.
I was able to address some of the issues in this PR: https://github.com/LLNL/RAJA/pull/1738. …
alecbcs commented on issue spack/spack#46447.
This might also need to rebase on to of https://github.com/spack/spack/pull/46445 to get CI to pass….
jedbrown commented on issue rsmpi/rsmpi#197.
Note that a barrier does not affect behavior of a correct MPI program (you can delete them all and get correct results, unless you’re communicating through side channels such as the file system, and even that is race-prone). Barriers are usually only used for a modicum of timing consistency (at the expense of performance). …
trws commented on issue flux-framework/flux-core#3927.
In case you want to play with it, here’s the script: …
cmoussa1 commented on issue flux-framework/flux-accounting#490.
@ryanday36 would you like to give this PR a high-level look and let me know if this might be a useful feature? I thought this might be a first-crack at tackling #352 to make it easier to gather jobs by project….
chu11 commented on issue chaos/powerman#208.
oops, had to adjust a test expected output. should hopefully pass CI now….
trws commented on issue spack/spack#46385.
I love the idea, and definitely want to do this on a few repos. Does this use the action to use their CLI or is this still using the legacy upload mechanism? I haven’t worked through their new workflow yet but thought they were pushing pretty hard for that now….
dylan-copeland commented on issue mfem/mfem#4507.
In parallel, this is the MPI rank owning the element. NCMesh::Print writes out Element::rank which has the comment (see ncmesh.hpp)
…
wihobbs commented on issue flux-framework/flux-core#3927.
Here’s a little primer on running containers in LC: https://hpc.llnl.gov/services/cloud/containers. Some of the “gotchas” are probably different with the new way of managing sub[u/g]ids….
asarkar-parsys commented on issue converged-computing/performance-study#47.
I had run into issues running mt-gemm on cyclecloud. The following is from the notes I had made available before leaving. …
gardner48 commented on issue LLNL/sundials#574.
Changes already in develop…
bgunnar5 commented on issue LLNL/merlin#490.
@lucpeterson Yeah I’ve been working on this on and off when I got some free time since last December ish :)…
vsoch commented on issue converged-computing/performance-study#40.
For quick access (and if it helps) they were added under this commit on August 22nd….
vsoch commented on issue oras-project/oras-py#155.
@xarses you will need to sign the DCO to merge this fully - click on the little “Details” next to the Robot that says “DCO” in the checks section….
grondo commented on issue flux-framework/flux-core#6275.
Looking back, it is a bit confusing that non-empty sets that overlap are combined (e.g. all,debug) and not the empty set. Though admittedly an empty set is a special case since it has no nodes the nodelist for all queues matches. …
garlick commented on issue flux-framework/flux-core#6283.
The thing that’s bugging me about this solution is it adds a synchronous RPC (as well as a flux_open() / flux_close() to the parent broker on every rank. They will go in parallel so it will scale but it adds a fixed startup latency to every job.
…
CRobeck commented on issue triton-lang/triton#4726.
cc: @ThomasRaoux …
adrienbernede commented on issue spack/spack#45648.
Note: ascent being built with vtk-m 2.2 in Spack CI, the commit above had no effect (it’s only requiring vtk-m > 2.1). …
jedbrown commented on issue rsmpi/rsmpi#197.
I’d be curious to test if you have a reproducer with the same MPI implementation, but I don’t think rsmpi has anything (directly) to do with this behavior. In general, there isn’t a guarantee about orderly termination. It’s possible that the errors in C are calling MPI_Abort while Rust errors are not….
vsoch commented on issue kubernetes-sigs/jobset#672.
No I don
chu11 commented on issue flux-framework/flux-core#6281.
Hmm, I don’t think so. The Value too large for defined data type suggests it might be an issue more similar to https://github.com/flux-framework/flux-core/issues/6256…
wihobbs commented on issue flux-framework/flux-core#6282.
Ah, ok, thanks!…
trws commented on issue flux-framework/flux-core#3927.
Yeah, OpenMPI should work, but if we want the library to be loaded we would have to bind mount it into the container and do a great deal of extra intrusive work to do that. Even domain sockets would need to be bind mounted in, but at least wouldn
astrophpeter commented on issue conda-forge/staged-recipes#27582.
@conda-forge/help-python This is ready for review, thanks!…
jedbrown commented on issue RDycore/RDycore#211.
If we were to implement this upstream in PETSc, what would be a good name for the time interval instead of step interval?…
balos1 commented on issue LLNL/sundials#561.
I dont think this is a bug as much as a mismatch in the Petsc and SUNDIALS configuration. The size/type of sunindextype can be changed between 32 and 64-bit with the CMake option SUNDIALS_INDEX_SIZE. PetscInt can also be configured to be 32 or 64-bit. The libraries should be configured in a compatible way. …
tpatki commented on issue LLNL/msr-safe#171.
@slabasan I don’t have a preference either way, let me know if Barry or you prefer it to be at the top-level (easy change for me to make)….
agcapps commented on issue LLNL/axom#1407.
Thank you, Jacob!…
tzanio commented on issue GLVis/glvis#316.
Ok, it seems to work with https://github.com/GLVis/glvis/pull/316/commits/031ca5c1c63f24d12b167b908299500d40c07249…
eliasboegel commented on issue CEED/libCEED#1662.
Awesome, thanks Jed. …
jedbrown commented on issue CEED/libCEED#1662.
Note that the ordering (with the innermost index over points, which may span a batch of elements) is due to memory coalescing and the way SIMT vectorizes, as well as the way CPU SIMD instructions work. If you change it, the compiler and hardware would have to do lots of transposition, and in many cases would result in quite poor code. …
markcmiller86 commented on issue visit-dav/visit#19366.
> @visit-dav/visit-developers @markcmiller86 are these fixes going to be there for next release? …
alecbcs commented on issue spack/spack-gantry#94.
That’s…kinda gross on the part of the OU Runners, but nice a simple fix. I like it!…
cyrush commented on issue spack/spack#45648.
@adrienbernede That shows what we have tested & we can’t really guarantee Ascent will work with over versions. If we had testing in spack we could try to loosen, but I don’t think we have good testing to try an experiment with any version….
alecbcs commented on issue spack/spack#46350.
Ah I see the URL has to be updated to /git/emacs….
tgamblin commented on issue spack/spack#44425.
@alalazo this is probably ready for another review. I’ll add some more tests when the coverage is done….
imguoguo commented on issue rv2036/rvspoc-S2428-k3s#1.
> @imguoguo
cjvogl commented on issue mfem/mfem#4499.
Works greats, thanks @pazner! …
mej commented on issue mej/nhc#151.
> I’ve started doing all my service checks via systemctl, e.g. …
milroy commented on issue flux-framework/flux-sched#1284.
After our discussion with Tom, I’m almost certain this issue is related to the ssds not being mapped to a broker rank. …
jameshcorbett commented on issue flux-framework/flux-multi-cluster-utilities#2.
OK @tpatki I pushed up some changes. Things are still broken but I think they’re getting farther. …
cmoussa1 commented on issue flux-framework/flux-accounting#293.
This should now be closed by #443 as projects should now be annotated with an association’s jobs. I can re-open more specific issues if need be….
garlick commented on issue chaos/powerman#205.
FWIW …
chu11 commented on issue chaos/powerman#205.
> We do have aliases so blade1477 and blade1477 could just map to the same blade name. …
correaa commented on issue STEllAR-GROUP/hpx#6541.
Godbolt already includes some version of Boost. …
markcmiller86 commented on issue LLNL/Silo#313.
If I write ZFP compressed data with newer Silo/HDF5, the older Silo cannot read it. This is expected because the older Silo library is not set up to use ZFP and there was no generalized approach to specifying an HDF5 filter plugin in older Silo. This is something that should probably be changed….
aperloff commented on issue LLNL/SSAPy#22.
@rusu24edward, are the tests you couldn’t fix in #18? If not, please document the issues you couldn’t fix here….
jeremylt commented on issue CEED/libCEED#1662.
Issues work fine for me. …
bgunnar5 commented on issue BarrySlyDelgado/merlin#3.
In the results_backend and broker sections of your app.yaml file, try setting cert_reqs: none. This is typically what’s done when setting this connection up at the lab too so that’s nothing out of the ordinary. You may also need to set name: rediss (with 2 s’s).
…
bwpriest commented on issue LLNL/saltatlas#78.
It would also be nice to pass a YGM container to dnnd::query() and dnnd::query_with_features(), although this would require either (a) an internal copy of the contents to a temporary vector, or (b) a significant overhaul of dndetail::dknn_batch_query_kernel, neither of which seems desirable at this time….
cyrush commented on issue LLNL/axom#1408.
@ldowen thank you, does this happen when your program closes? …
adrienbernede commented on issue LLNL/Umpire#902.
@davidbeckingsale this is ready….
adrienbernede commented on issue LLNL/RAJA#1712.
@rhornung67 this is ready. Your approval being a month old I
cyrush commented on issue LLNL/Caliper#577.
@daboehme I still plan to help with this, but doens’t look like I will free energy until October :-(…
adrienbernede commented on issue LLNL/CARE#281.
@adayton1 This is ready….
eliasboegel commented on issue helix-editor/helix#2869.
This looks like a great feature @lazytanuki - thanks for all the work! …
trws commented on issue git/git#1719.
I’m sorry I dropped, I was out of the country for several weeks and lost the thread. Did at least the doc fix get merged at some point? If not, I can do some followup to make that happen, or re-work this with a different flag name or both….
wihobbs commented on issue flux-framework/flux-sched#1294.
@trws Sure am, made it easier once I realized the configure.ac is really just a shell script with a different suffix. …
cyrush commented on issue Alpine-DAV/ascent#1383.
Applies replay fix tested in frontier branch. …
garlick commented on issue worlickwerx/pi-parport#68.
Added a small change to include the 40-pin connector area in the ground fill with 0.2mm keepouts around each NPTH. I’ll squash that down once approved, before merging….
JustinPrivitera commented on issue visit-dav/visit#19793.
@cyrush this is ready to go now. I made pretty radical changes to the scripts but I think these changes ultimately will make the scripts much easier to use, understand, and much more maintainable….
CRobeck commented on issue triton-lang/triton#4669.
@giuseros …
tgamblin commented on issue spack/spack#44425.
* @becker33 This is ready for another review, and the description is updated. …
vsoch commented on issue oras-project/oras-py#147.
@tarilabs I don’t have bandwidth at the moment to work on thees components, but I can fully support you and provide review for what you are able to contribute. Thank you!…
vsoch commented on issue converged-computing/performance-study#27.
Answering some of my comments above: …
v-dobrev commented on issue xsdk-project/xsdk-issues#258.
SUNDIALS 7+ is not supported in MFEM as of 09/05/2024. I’ve been planning to work on this for some time now, however I haven’t found time to do it yet….
cyrush commented on issue visit-dav/visit#19798.
this issue: …
markcmiller86 commented on issue visit-dav/visit#2605.
Dup of #2605…
cyrush commented on issue spack/spack#45648.
@adrienbernede here are the details for Ascent 0.9.3: …
adrienbernede commented on issue spack/spack#45648.
@cyrush I
aschaf commented on issue mfem/mfem#4479.
Thank you. I will look into it and keep you updated….
tzanio commented on issue mfem/mfem#4480.
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….
v-dobrev commented on issue mfem/mfem#4479.
This requires some form of re-partitioning for the ParMesh which is currently possible only if you convert the serial mesh to nonconforming mesh, e.g. using EnsureNCMesh(). You can then use a custom partitioning array to distribute the serial mesh to 4 processors assigning 0 elements to the remaining 16-4=12 ranks. After parallel refinement, you can re-partition the mesh to 16 ranks using Rebalance(const Array<int> &partition).
…
tzanio commented on issue mfem/PyMFEM#251.
do you mind posting the solution?…
vsoch commented on issue kubernetes-sigs/jobset#672.
This would be great! I agree to not try to create a workflow tool, but allowing this basic dependency structure is something that workflow tools can use. +1 from me….
trws commented on issue flux-framework/flux-sched#1294.
Looks good to me! Are you doing a similar one on core?…
grondo commented on issue flux-framework/flux-sched#1229.
Users are seeing this error again in a slightly different scenario. Example: …
wihobbs commented on issue flux-framework/flux-sched#1294.
Oh, grr, git fetch tags needs to be git fetch --tags …
jameshcorbett commented on issue flux-framework/flux-sched#1292.
> What are the scheduler and queue policies set to in the coral2 tests? …
chu11 commented on issue flux-framework/flux-core#6268.
> I’d suggest in this PR to then just make the maximum safe value the default for user instances. Another PR (or at least another commit) could lower it if we think that’s a good idea. …
trws commented on issue flux-framework/flux-core#6259.
Yeah, I still need to get the updated clang-format file merged and set to only apply where we’ve already applied it. It’s definitely possible I missed some things though, I’m not good at catching them (part of why I always try to automate that)….
grondo commented on issue flux-framework/flux-core#6259.
Some trivial cleanup still needed (just formatting afaics, and sorry we don’t have a valid clang format for our style :disappointed:) …
garlick commented on issue flux-framework/flux-core#6268.
Should we make it an error if the user tries to configure a limit that may trigger #6256? …
markcmiller86 commented on issue betterscientificsoftware/bssw.io#2152.
Based on the observations above comment, I’ve removed the tables. I think it looks better. …
adrienbernede commented on issue LLNL/Caliper#588.
@daboehme The cce jobs (18.0.0 and 17.0.1) are failing because …
jeremylt commented on issue CEED/libCEED#1646.
Ok, all backends should be covered, including SYCL (not tested) …
cyrush commented on issue Alpine-DAV/ascent#1380.
Turns out, we had a #ifdef issue ASCENT_MPI_ENABLED is not defined in replay for the MPI case, but ASCENT_REPLAY_MPI is.
…
jandrej commented on issue mfem/mfem#4470.
See #4487…
mej commented on issue mej/nhc#151.
The check_ps_service() check has an option, -m, which allows you to specify your own match string in lieu of the default behavior (which is to match any command whose argv[0] ends with the name of the specified service – hence the *sshd it’s using in your example above).
…
eliasboegel commented on issue lilyinstarlight/nixos-cosmic#290.
Yes, I have the same behavior….
milroy commented on issue flux-framework/flux-sched#1292.
> Some of my flux-coral2 tests are suggesting to me that the rabbit resources aren’t freed, even though the error message goes away. …
jameshcorbett commented on issue flux-framework/flux-coral2#207.
Thanks! Done and setting MWP….
IanLee1521 commented on issue elastic/integrations#10909.
Thank you!…
adayton1 commented on issue LLNL/CHAI#275.
I think nvcc preprocesses std::function incorrectly. This isn’t the first time I’ve seen issues with nvcc and the
adrienbernede commented on issue LLNL/CHAI#275.
@adayton1 regarding the failure in the gcc 11.2.1 + cuda 11.8 build, do you think I should ignore the failure or wait for a fix?…
v-dobrev commented on issue GLVis/glvis#311.
@najlkin, is this issue under Windows with WSL? …
emily-howell commented on issue Alpine-DAV/ascent#1380.
Ah, good to know. Would you like to make a MR to merge in your branch with the fix?…
trws commented on issue spack/spack#46192.
It’s trying to compile sched with gcc 9.4, despite 11 being available. 11 might work, but we only officially support 12+….
adayton1 commented on issue LLNL/blt#671.
Thanks for pushing this through, @white238 !…
markcmiller86 commented on issue LLNL/Silo#313.
I’ve run some tests to check compatability between HDF5 versions. I created multi_ucd3d.h5 with Silo 4.11.3RC and HDF5-1.14.0 and then used browser from an old install of Silo 4.10 on CZ to read it and it read everything fine. I need to test checksumming and compression under the same conditions. But, this does indicate older silo/hdf5 combinations will read files produced by newer silo/hdf5 combinations….
cmoussa1 commented on issue flux-framework/flux-accounting#488.
Thanks @jameshcorbett! Going to set MWP here…
adayton1 commented on issue LLNL/spheral#299.
Good point! I’ll see if the axom include can be moved to the .cc file…
artv3 commented on issue LLNL/RAJA#1711.
@artv3 TODO: add to documentation …
garlick commented on issue worlickwerx/pi-parport#68.
Thanks! I tacked on an additional commit that adds a 1K pullup on the EEPROM WP per the HAT+ spec, then rebased on master after #69 was merged. …
jandrej commented on issue EnzymeAD/Enzyme#2062.
To clarify, current main branch has this https://github.com/EnzymeAD/Enzyme/blob/b006eb21388dc1f0c13e7f4584847805767f95aa/enzyme/Enzyme/CMakeLists.txt#L214-L218 where PUBLIC_HEADER DESTINATION is wrong. The changes in this PR don’t address this yet….
grondo commented on issue flux-framework/flux-sched#1171.
Oh, nice! Thanks for that!…
trws commented on issue flux-framework/flux-sched#1171.
Found it, there was an extra traversal we don’t need with a bit of adjustment. This test now runs at a more expected rate, actually much better than I expected: …
grondo commented on issue flux-framework/flux-core#6252.
@jameshcorbett: how does this solution sound?…
vsoch commented on issue converged-computing/performance-study#18.
I don’t think there is any way for me to actually review this, so I’m just going to merge….
milroy commented on issue converged-computing/performance-study#18.
I fixed the conflicts, so this is ready to look at….
JustinPrivitera commented on issue LLNL/conduit#1322.
so we want set methods that go between all 4 types: data/execution arrays/accessors …
corbett5 commented on issue GEOS-DEV/GEOS#3312.
It might solve it, you never know with these problems. But I’m pretty sure the error is coming from the creation of the LvArray::Array itself so logically if we fill in the values later it shouldn’t make a difference….
JustinPrivitera commented on issue visit-dav/visit#19787.
merge w/ develop in #19788…
markcmiller86 commented on issue visit-dav/visit#19787.
> shoot I committed directly to 3.4RC :( …
hppritcha commented on issue openpmix/prrte#2004.
> Problematic: the issue here is that specifying an MPI for srun will automatically make Slurm think that the daemons are MPI procs, which has implications for how they are run. What “mpi” option are you thinking of trying?
…
tzanio commented on issue mfem/mfem#4416.
Re-merged in next for testing……
correaa commented on issue include-what-you-use/include-what-you-use#1606.
I don’t know python either, but more fundamental I don’t understand how the library detects automatically the mappings from the libc++ code alone. …
garlick commented on issue flux-framework/flux-core#6235.
My admittedly light testing on my test cluster did not turn up any issues. You have a green light from me :-)…
correaa commented on issue correaa/boost-multi#67.
customization of common_reference is now commented and seems to work. …
bgunnar5 commented on issue LLNL/axom#1378.
Ok @kennyweiss I think I’ve made (almost) all of the changes you suggested. The only one I haven’t touched yet is the Sina versioning but we should continue the discussion on that in the thread above. Other than that everything should be good to go here now (so long as all of the tests pass again)….
tzanio commented on issue GLVis/pyglvis#49.
Oh, I see now. You are correct…
tzanio commented on issue GLVis/glvis#314.
I can confirm Q works now. Thanks @justinlaughlin ~…
JustinPrivitera commented on issue Alpine-DAV/ascent#1376.
silo/overlink extract will be added in #1377…
bgunnar5 commented on issue ryannova/merlin#2.
Marking this as ready for review but haven’t done the merlin monitor -> merlin manager changes yet. This just has the changes suggested in LLNL#489….
jameshcorbett commented on issue flux-framework/flux-sched#1284.
The issue seems to have been introduced between 0.36.1 and 0.37.0….
jameshcorbett commented on issue flux-framework/flux-core#6245.
Sounds great to me! Would simplify the configuration of rabbit clusters quite a bit….
cmoussa1 commented on issue flux-framework/flux-accounting#487.
thanks! setting MWP here…
gardner48 commented on issue erf-model/ERF#1767.
I think new gold files are needed for ABL_MYNN_PBL. The CPU tests happen to pass on Lassen but fail on other systems. …
chu11 commented on issue chaos/whatsup#20.
@andrew-boatrocker could you checkout the code in PR #22 and see if it fixes your problem. The new config option is “ping_socket_receive_buffer BYTECOUNT”. Here’s the patch file if you just want do apply it. …
adayton1 commented on issue LLNL/spheral#297.
> This looks good to me, what does the integration.ats edit do? …
jandrej commented on issue mfem/mfem#4419.
This interface is terrible….
cjvogl commented on issue mfem/mfem#4419.
> > > > Is there a way to test these new functions? Perhaps something being added to an example or is this mainly being driven from application code? …
wihobbs commented on issue flux-framework/flux-core#5531.
That did it, thanks @grondo! …
imguoguo commented on issue fedora-riscv/kojihub#1.
The old third-party docker image(imbearchild/fedora-riscv) has been deprecated and replaced with a new docker image(fedorariscv/base). …
IanLee1521 commented on issue elastic/kibana#179322.
This is awesome, thank you!!…
cyrush commented on issue visit-dav/visit#19781.
@BradWhitlock thanks for this fix – we will want get this on the 3.4RC. …
tgamblin commented on issue spack/spack#46004.
some of that is done here but the gitlab support could probably still use a refactoring: https://github.com/spack/spack/pull/42605…
jedbrown commented on issue rust-lang/rust#124874.
Sorry about the big delay getting back to this. I’ve updated this PR to remove the float_mul_add_fast feature and associated safe functions, keeping the intrinsics::fmuladdf{32,64} with associated tests and Miri support (thanks Ralf, for the guidance). I think it’s reviewable now….
trws commented on issue flux-framework/flux-core#6236.
> Safe mode being read only mode? Minimally no new jobs would be able to be submitted due to ENOSPC. Would current running jobs fail as well? I guess this is something to mess around with, as we don’t quite know the fallout of such a decision. …
cmoussa1 commented on issue flux-framework/flux-coral2#203.
OK sounds good, let me know what you might require! …
jameshcorbett commented on issue flux-framework/flux-coral2#200.
> If it helps, there’s a proposal to add a -S short option for –setattr so users could use -S dw=small-xfs. …
jbschroder commented on issue XBraid/xbraid#97.
Cool! I will squash and merge then….
cyrush commented on issue LLNL/conduit#1321.
This exists, I was blocked on a typo. sorry for the noise. …
eliasboegel commented on issue CEED/libCEED#1650.
Thanks!…
jedbrown commented on issue CEED/libCEED#1650.
Hi, thanks for the question. We haven’t added cuda/hip/sycl features to the libceed-sys package, but you can build libceed.so any way you like and then use the system feature to use that pre-built library instead of having libceed-sys/build.rs build its own. We’d be happy to accept a pull request that added such features, though they may be somewhat brittle because one needs build.rs to find the toolchain (CUDA) that the user intends….
alecbcs commented on issue spack/spack-tutorial#389.
@dependabot recreate…
alecbcs commented on issue spack/spack#24907.
@dylex out of general curiosity did you all move to a different package manager ecosystem or end up modify your workflows to where you didn’t need a package manager?…
markcmiller86 commented on issue betterscientificsoftware/bssw.io#2146.
@betterscientificsoftware/bssw-maint this is ready to publish …
JustinPrivitera commented on issue visit-dav/visit#4027.
I am scheduling this because I ran into this today….
CRobeck commented on issue iree-org/iree#18347.
The PR for getting this into Triton is here: …
adammoody commented on issue hpc/mpifileutils#579.
Thanks for the report, @samlor . dbz2 writes to a single shared file from multiple processes. For correctness, it requires a POSIX-compliant parallel file system like Lustre or IBM’s Spectrum Scale. In particular, many NFS file systems are not POSIX-compliant. …
milroy commented on issue converged-computing/performance-study#6.
Updated the PR per your recommendations! …
cyrush commented on issue visit-dav/visit#19492.
Resolve by @brugger1 in #19552 and related develop PR….
CRobeck commented on issue triton-lang/triton#4568.
@Jokeren missed one line…
hppritcha commented on issue spack/spack#29180.
I think this PR is no longer an relevant and can be closed….
alecbcs commented on issue spack/spack#29494.
@haralmha @haampie is this PR still relevant or has this since been fixed for keyutils?…
tzanio commented on issue mfem/mfem#4464.
Merged in next for testing……
jandrej commented on issue mfem/mfem#4470.
No requirement on having a nodeset (not supported in mfem AFAIK). You have to assign all faces that are “in contact with void” to a sideset….
trws commented on issue flux-framework/flux-core#3927.
We still need a solution for ensuring users have additional UID mappings in the LC environment, ideally matching the slurm --userns option we have as a plugin on toss4 slurm systems….
wihobbs commented on issue flux-framework/flux-core#3927.
> Can we do that experiment and report what, if anything, is needed from Flux to support launching jobs in containers? I think that is the question that motivated this issue. The other cool namespacey things we can do if allowed can have their own issues as we dream them up. …
grondo commented on issue flux-framework/flux-coral2#198.
I’m more just thinking out loud here, but --requires=hosts:rabbit203 would only work (I think) if constraining the resource graph to only vertexes that match a “hostname” of rabbit203 would result in something that could match the rest of the job request. (Given that I have no idea how the datawarp/rabbit matches work in Fluxion, I could be way off here)
…
cmoussa1 commented on issue flux-framework/flux-accounting#485.
TIL! Thanks for the commit message tip. Glad we got it figured out. Setting MWP here…
chu11 commented on issue chaos/powerman#197.
> Suggestion: pull in the unit tests from lsd-tools. …
garlick commented on issue chaos/powerman#197.
Suggestion: pull in the unit tests from lsd-tools. …
grondo commented on issue chaos/powerman#197.
Second one seems simpler :sweat_smile:…
markcmiller86 commented on issue betterscientificsoftware/bssw.io#2138.
Thanks @ryanmrichard for the contribution…
jandrej commented on issue ROCm/omniperf#325.
Filtering is still an issue with rocprofv2 …
cyrush commented on issue LLNL/blt#695.
Got it, blt make it very easy to set the target properties to change the dir when you create an exe, but MFEM isn’t using BLT itself. …
artv3 commented on issue LLNL/RAJA#1718.
@rahulb1218 can you post the performance numbers, also I just realized it might be worth looking at the kernels through the NVIDIA profiler. As @MrBurmark RAJA has a high initial kernel launch overhead due to the way streams are setup. …
emily-howell commented on issue Alpine-DAV/ascent#1367.
@cyrush Let me know if you want to do these name changes on their own or along with the addition of tests….
markcmiller86 commented on issue visit-dav/visit#19721.
This has been resolved by removing md5 checksums from the largedata site in this commit…
bgunnar5 commented on issue ryannova/merlin#2.
@ryannova do we want to have celery worker entries deleted from the redis database when we stop workers? I’m worried that for redis servers not hosted by the container (i.e. LaunchIT) we’ll end up having way too many worker entries over a long period of time. …
trws commented on issue flux-framework/flux-sched#1171.
I have an idea why this might be happening. Normally first is quick because it considers resources in order, sorted by most available components under them first. I’m guessing that we consider down nodes to have all their resources available, so even though we quickly skip them we consider them when we shouldn’t. A quick fix for this might be to mark all the resources unavailable when a resource is marked down, or adding up/down as part of the key for that ordered map. It shouldn’t normally make a huge difference, but treating the resources as “allocated” until the end of time in the planner might let us simpllicy some of the logic too. @milroy, you know the planners a whole lot more than I do, can you think of a reason that would cause problems?…
grondo commented on issue flux-framework/flux-core#6226.
Sounds good to me! I’ll set MWP….
vsoch commented on issue flux-framework/flux-core#6226.
> I’d like to get more feedback if this example should just go in an EXAMPLES section of flux-submit(1) directly or something. …
garlick commented on issue flux-framework/flux-core#6217.
> goes down hard meaning broker shall exit? Edit: and bring down full instance? …
agcapps commented on issue LLNL/conduit#1316.
> @agcapps Thanks for putting this together. …
markcmiller86 commented on issue LLNL/Silo#264.
Resolved in #403 …
corbett5 commented on issue JuliaSparse/SparseArrays.jl#557.
Yes if we simply want to use eps(Float64) in all cases then this is good to go….
jedbrown commented on issue openjournals/joss-reviews#6766.
@editorialbot recommend-accept…