jedbrown commented on issue rust-lang/rust#138689.
I rebased. The warnings for implied features were fixed by https://github.com/rust-lang/rust/pull/140920. I updated a stale comment left behind in that refactor, and my handling of features transitively implied by base features moved to rustc_codegen_ssa. …
homeomorfismo commented on issue mfem/mfem#4769.
> I noticed that now the following two sample runs are actually identical: …
v-dobrev commented on issue mfem/mfem#4903.
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….
grondo commented on issue flux-framework/flux-core#6882.
Thanks! I’ll set MWP…
v-dobrev commented on issue xsdk-project/xsdk-issues#277.
Hi @balay,…
adrienbernede commented on issue spack/spack#50916.
For the records, here is a pre-v1 compiler compiler config that I am trying to port over to v1:…
tgamblin commented on issue spack/spack#50920.
This one’s more complete – with updated docs…
garlick commented on issue flux-framework/flux-core#6885.
I’ll set MWP - thanks!…
chu11 commented on issue flux-framework/flux-core#6880.
just pushed a commit on top that implements --tail=N in flux job attach. Everything appears to work as expected.
…
cjekel commented on issue cjekel/similarity_measures#39.
Thank you!…
JustinPrivitera commented on issue LLNL/conduit#1424.
With my two comments above I believe I have addressed all of your feedback @cyrush. I think this is ready to be merged. Please let me know if there is anything else we should discuss….
trws commented on issue LLNL/camp#179.
They could if someone wants to set up an image with one of the CPU-backed SYCL implementations like AdaptiveCPP or the clang upstream offload version (which I think now has “run in separate process” support to give you memory errors on CPUs if you mess up transfers). We don’t have that set up though. …
markcmiller86 commented on issue LLNL/Silo#432.
Already handled in f10090abcc63f5e93d52130db0b8bb0773c765d9…
tgamblin commented on issue spack/spack-packages#44.
@kwryankrattiger …
hppritcha commented on issue open-mpi/ompi#13279.
@edgargabriel i pushed the remainder of the fortran related files to the PR - hopefully….
vsoch commented on issue kubernetes-sigs/node-feature-discovery#2170.
The nfd worker generates a lot of logs (and other unexpected output) so I wasn’t sure how to cleanly do that. Whereas with writing the file explicitly, I can control exactly the contents….
trws commented on issue flux-framework/flux-sched#986.
Not substantively, we operate on the IDs as logical IDs (effectively assuming, in absence of actual structure, that lower IDs are close to other lower IDs). We still have a direct hwloc reader, which I believe is still tested, though we’re not currently using it. We would still need a way to connect the two together. Really what would likely be most useful in the short term would be to ensure that those logical IDs are translated back to physical IDs in exec or the shell. Having the actual locality in fluxion would definitely be good, but with what @ryanday36 showed in the issue description just doing that logical->physical translation even with hwloc-calc right before doing the env var assignment would fix it. Admittedly not fix it the long-term “right” way, but we’ll need that logical to physical translation anyway because we don’t want to be tied to whatever crazy variable numbering cuda or hip use from 1pm to 2pm on alternate Thursdays….
grondo commented on issue flux-framework/flux-sched#986.
This issue came up again in the flux dev meeting yesterday, so just commenting here to revive the issue….
mplegendre commented on issue flux-framework/flux-core#6883.
We’d previously discussed starting a spindle session at allocation time that lasts the whole allocation. That would also solve this problem. But the drawback is it’s hard for users to pass configuration arguments to spindle, since spindle’s session starts before the user gets control. Though I’ll go this route if we can’t do alternatives….
cmoussa1 commented on issue flux-framework/flux-accounting#667.
Rebased this PR to catch up after #672. Setting MWP here…
vsoch commented on issue expfactory-experiments/attention-network-test#2.
What sticks out to me is defining “blocks” that are then looped through for trials. So possible reduce the sections there, and within each entity added to blocks, you can also delete….
balos1 commented on issue LLNL/sundials#721.
Thank you for opening this pull request. Due its size, it will take us awhile to be able to consider and review it. …
alecbcs commented on issue LLNL/hubcast#19.
Oh good point! I think we can close this one out :)…
cmelone commented on issue LLNL/hubcast#19.
@alecbcs is this still relevant given the addition of the config handler?…
JustinPrivitera open issue Alpine-DAV/ascent#1563.
Original vs Actual Extents
Original mesh:…View Comment
JustinPrivitera open issue visit-dav/visit#20449.
VisIt crashes when rendering dataset using transparency
I am trying to visualize my x ray imaging setup, so I am making renders of two files, curv3d.silo and output.root. When I adjust opacity on my rays mesh and then click the viewer, VisIt crashes….View Comment
cmelone commented on issue spack/spack-packages#107.
If you’d like a review, try posting in the #pull-requests channel on Slack. The CI errors will need to be fixed though…
alecbcs commented on issue spack/spack#50874.
Yep. Working on a few other things today / tomorrow. I can do this Friday….
markcmiller86 commented on issue openjournals/joss-reviews#8106.
I recommend this paper for publication….
adayton1 commented on issue kokkos/kokkos#8195.
Thank you! I’ll give Kokkos_IMPL_AMDGPU_FLAGS a try!…
grondo commented on issue flux-framework/rfc#454.
Setting MWP….
garlick open issue flux-framework/flux-accounting#671.
allow urgency weight to be reconfigurable
Problem: when the flux accounting priority plugin is in use, the urgency (described in RFC 30 has a very small impact on priority, which means that users cannot meaningfully lower the priority of their jobs (other than hold) relative to others, and sys admins cannot raise the priority of jobs….View Comment
artv3 commented on issue LLNL/RAJA#1853.
Consider moving the plugins into src/plugins – keep as cpp files. …
jeremylt closed issue CEED/libCEED#1822.
GPU AtPoints Assembly
Following up from https://github.com/CEED/libCEED/pull/1799…View Comment
emily-howell commented on issue Alpine-DAV/ascent#1502.
## TLDR: …
markcmiller86 commented on issue visit-dav/visit#20442.
> Here’s a question: the query output object for Spatial Extents queries is a single entry with key = ‘extents’, and value an array/vector of the values. With the changes to the output string, should the output object (XmlResult) be modified to use the XMin/XMax, etc keys instead of just ‘extents’? If so, this would possibly be more of a problem, even if it is the better way to report. …
imguoguo commented on issue osbuild/bootc-image-builder#963.
> Thank you! This looks nice! Ideally we would have a proper test for this but given that its not an official fedora arch yet it fine. …
jandrej commented on issue mfem/mfem#4898.
> @jandrej, could I suggest a couple small changes that I noticed in the DFEM code? …
tzanio commented on issue mfem/mfem#4898.
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….
vsoch commented on issue kubernetes/website#51164.
@graz-dev do you need any more clarification from us? …
samuelkgutierrez commented on issue hpc/quo-vadis#342.
Is something up with Github?…
adayton1 commented on issue LLNL/Umpire#967.
I think you are going to need to update the MemoryResourceType enum in MemoryResourceTypes.hpp. Otherwise, “SHARED” could map to the wrong implementation….
JustinPrivitera commented on issue LLNL/Silo#448.
This would be a nice feature!…
balos1 commented on issue spack/spack-packages#219.
@everythingfunctional Can you try adding depends_on("blas", when="+lapack") to the sundials package.py to see if that fixes this. …
trws commented on issue flux-framework/flux-core#6492.
I don’t recall seeing this when it was being done, but it was mentioned in one of our recent meetings so I came looking. There are at least some updates here, in that uv_ref/unref now exist at least, and the child handling could be handled one of a couple of ways depending on the requirements (either uv_process_t if we can, or uv_signal_t if not). The others I think are still at least different, and the close item is vaguely interesting, I would think we’d be able to tolerate that even though it is a bit annoying. It’s probably safer for io_uring style setups as well as windows, but I’m not 100% sure about that….
gardner48 commented on issue LLNL/sundials#713.
Fixing this typo in another PR….
adrienbernede commented on issue LLNL/radiuss-spack-configs#129.
@tdrwenski Could you open a PR in Quandary that checks whether your CI works well with the present PR in RSC?…
camierjs commented on issue mfem/mfem#4326.
There were some changes to bring from doftrans-three-arg, done in 2bef31b and 4a46614.
…
grondo commented on issue flux-framework/flux-core#6873.
> I wonder if we could add an exec_bypass flag and let this plugin use both flags and post alloc (adding a fake or empty R), free, start, and finish (with job result) and have the proxy job track the state of the actual job? …
garlick commented on issue flux-framework/flux-core#6873.
This is pretty neat. However it seems like its usefulness will be limited as long as the result of the proxy job is always reported as failure. I see why a fatal exception was used to get the job from DEPEND to CLEANUP state in the prototype, since there’s no other way to do that currently, but I wonder if we can think up a better way that allows the proxy job result to reflect the result of the actual job. That way workflow tools, not to mention Flux native dependency schemes like afterok ,would not need to treat delegated jobs specially when they need to fetch the job result.
…
BradWhitlock open issue LLNL/conduit#1444.
Partitioner fails when multiple topologies provide adjsets.
I wrote a new test case to test adjset spatial sorting, which makes a mesh and generates lines and corners for it and validates the adjsets. Later, I decided to hook up partitioning. I made the partitioned mesh after the other meshes. It caused the partitioner to crash because topologies (other than the one being partitioned) had adjsets already defined. The logic loops over all adjsets in the domain, which is not right….View Comment
vsoch commented on issue vsoch/oci-python#24.
Thanks! …
markcmiller86 commented on issue visit-dav/visit#4914.
Here is a potential confusion. When listing the dimensions of an multi-dimensional mesh (e.g. 2DCoRectMesh as above), the dimensions are specified as ZYX, not XYZ. So, the mesh is 9 nodes in Y and 33 nodes in X….
JustinPrivitera commented on issue visit-dav/visit#20440.
> > One question - does it make sense to rename bv_qt6.sh to bv_qt.sh, and to change the name in associated build scripts and in modules.xml?
…
jedbrown commented on issue rsmpi/rsmpi#215.
This would be resolved by #214 . There has also been an issue with the runtime feature (cf. #203 ). Do you have an opinion about this question? https://github.com/rsmpi/rsmpi/pull/214#discussion_r2136959267…
jedbrown commented on issue openjournals/joss-reviews#7881.
@editorialbot recommend-accept…
tzanio commented on issue mfem/mfem#4841.
@dylan-copeland , can you please resolve conflicts?…
trws commented on issue flux-framework/flux-core#6315.
> > add the resource.rediscover to the sub-instance config by default…
garlick commented on issue flux-framework/flux-coral2#369.
Cool, thanks. I’ll set MWP….
cmoussa1 open issue flux-framework/flux-accounting#666.
documentation: update docs to include information on resource limits
Now that #562 has landed, the docs need to be updated with information on how per-association resource limits are enforced in the priority plugin and how they can be configured both in the database. Perhaps a separate page just on limits could be worthwhile….View Comment
vsoch commented on issue flux-framework/Tutorials#54.
It also looks like there is a “Slides coming soon!” message that can now be updated….
artv3 commented on issue LLNL/RAJA#1859.
> > > This PR changes cmake to support a minimum cxx standard of 17. …
JustinPrivitera commented on issue Alpine-DAV/ascent#1556.
Let’s have renders for optical_depth fields as well. We should attempt to have tests that capture every single thing included in the output and the inputs we provide in tests should cover as many of the edge cases as possible….
emily-howell commented on issue Alpine-DAV/ascent#1526.
All of these action items were addressed in #1524 …
cyrush commented on issue visit-dav/visit#5770.
> > For AMR, we shouldn’t use VTK. For blanking out cells, we should use nesting….
hppritcha commented on issue spack/spack-packages#209.
@RonRahaman please open a PR for this, thanks….
v-dobrev commented on issue mfem/mfem#4880.
The normal on a boundary edge is not uniquely defined – typically, the edge has two adjacent faces that have, generally, different normals. How do you want to treat these cases?…
mplegendre closed issue hpc/Spindle#86.
Renames logging #defines
New logging #defines did shell exit and return respectively. Names should be changed to reflect flow control. …View Comment
grondo commented on issue flux-framework/flux-accounting#659.
It does seem like a good idea to store the priority factor weights in the database, especially if there’s going to be an easy method to view the current weights….
wihobbs open issue flux-framework/Tutorials#51.
Flux for Sysadmins
At dinner with @trws last night, we discussed the need as Flux grows for a few additional tutorials that go beyond the traditional user training we’ve had in the past: Flux for Sysadmins and Flux for Developers and Hackers….View Comment
bgunnar5 commented on issue LLNL/merlin#512.
@lucpeterson after spending hours trying to research and implement this, turns out Redis only accepts plaintext in the redis.conf file. Therefore, even if we encrypt the password with Fernet, Redis will just use the encrypted string as the password. I don’t think there’s any way around this so those last three CodeQL issues are unable to be resolved….
cyrush commented on issue Alpine-DAV/ascent#1486.
resolved in #1527 …
chu11 commented on issue flux-framework/flux-core#6871.
thanks! and I just noticed I typoed “updated” in the PR title and the commit message. Fixed that too. Will set MWP…
jameshcorbett commented on issue flux-framework/flux-core#6870.
Force-pushed to rebase with master and expand the test a little bit, setting MWP….
wihobbs commented on issue flux-framework/flux-core#6315.
@eleon has a few thoughts here shared today with @trws and me. Tagging him so he can share more….
jameshcorbett commented on issue flux-framework/flux-coral2#367.
Force-pushed to fix release date, setting MWP….
jedbrown commented on issue RDycore/RDycore#313.
These would be useful for dependencies:…
eleon commented on issue LLNL/mpibind#45.
For mpibind to work as intended, you need to tell Flux you want the full node (-x). I also suggest using as few parameters as possible in your command line since you can, inadvertently, add conflicts between Flux affinity and mpibind. …
tgamblin commented on issue spack/spack#50867.
closing as the RHEL8 UBI issue seems to have sorted itself out….
chu11 commented on issue flux-framework/flux-accounting#656.
I wasn’t able to find where in the code you configure journal_mode and synchronous modes. In content-sqlite it is set to “WAL” and “NORMAL”, for a smidge better performance….
tgamblin commented on issue spack/spack-packages#127.
I think the answers would be: …
tgamblin commented on issue spack/spack#50595.
I think we should do this but we are attempting to get #47615 in for 1.0, which also adds the hash of the spack instance to the misc_cache. …
jedbrown commented on issue rsmpi/rsmpi#203.
@mscroggs Following up on this, we get build failures on MacOS without the runtime feature. It works everywhere else. Do you know of a way to fix this?
…
tzanio commented on issue mfem/mfem#4804.
Re-merged in next for testing……
tpatki commented on issue flux-framework/flux-core#6873.
Hi @trws @garlick @grondo: …
btalamini closed issue LLNL/serac#1399.
Make sure mesh is guaranteed to be in a valid state
Even if given a parallel mfem mesh….View Comment
BradWhitlock open issue visit-dav/visit#20420.
Improve Subset wireframe rendering
Improve Subset wireframe rendering to eliminate Z-fighting and improve overall appearance….View Comment
alecbcs commented on issue spack/spack#48820.
Closing since this PR was sucessfully migrated to spack/spack-packages….
v-dobrev commented on issue spack/spack#49790.
bwpriest commented on issue grid-parity-exchange/Egret#318.
> @bwpriest I do not understand the failures … maybe I messed something up in 160a4a4? …
chu11 open issue flux-framework/flux-core#6872.
kvs-watch: support a sentinel value indicating "end of initial values"
It would be convenient if kvs-watch could inform watchers that all data that is presently available (i.e. data may appear in the future but we don’t know) has already been streamed and sent….View Comment
artv3 commented on issue LLNL/RAJAPerf#515.
> > yes, 2 is equivalent to the x index in cuda/hip …
artv3 open issue LLNL/RAJA#1856.
Warnings with GCC 12.1.1
There a number of warnings that appear with GCC 12.1.1. They look something like this:…View Comment
cyrush commented on issue Alpine-DAV/ascent#1550.
…
v-dobrev commented on issue mfem/mfem#4851.
I’m not sure why this was merged without approvals and with requested changes pending. Is the idea to review these changes as part of #4800?…
trws commented on issue flux-framework/flux-core#6858.
I really like this, had one tiny docstring tweak and a question but it looks really good @grondo….
grondo commented on issue flux-framework/flux-core#6868.
Thanks! Setting MWP….
cmoussa1 commented on issue flux-framework/flux-accounting#631.
Looks like #632 ended up fixing this issue on tuolumne at least, so I’ll go ahead and close this now….
markcmiller86 commented on issue visit-dav/visit#20396.
@JustinPrivitera the best workaround I have found so far is to just create an alias expression and then use that alias in any further expressions requiring the original variable. On my mac with current develop, I don’t get the garbage data condition then. So, in other words, define foo as an alias for ElementBlock and then define bar as foo*1 and that works fine….
alecbcs closed issue spack/spack-tutorial#191.
Add note about `spack help --spec` to Basics tutorial
The spec syntax we cover can be confusing for new users, and it’s good to know that we have a command just to describe this syntax. We should end the tutorial with a note about spack help --spec….View Comment