XIO

xWire

Posts from every feed QUEST has found across the web — blogs, news, podcasts, channels and more.

Chapel Programming Language - Latest posts

Latest posts

[Chapel GA Tests] 2026-11-20

The following GitHub Actions jobs failed: full-checks / Chapel paratest on darwin / Aggregate results (darwin), step: "aggregate and report results" failure: Process completed with exit code 1. 2 failed test(s) in darwin: [Error matching compiler output for library/packages/DynamicLoading/simpleChplLib/callAddEm-mistake] [Error matching compiler output for library/packages/DynamicLoading/simpleChplLib/callAddEm] Chapel Build Log Test Results PRs merged since the last nightly run: Since…

[Chapel Merge] Pass `char*` to a `strchr` call in the runtime

Branch: fix-arg-handling-strchr Revision: b357d684a1d0509e4e6d6eddfed89bc381889950 Author: dlongnecke-cray Merged at: 2026-11-20T02:42:08Z Link: Pass `char*` to a `strchr` call in the runtime by dlongnecke-cray · Pull Request #29719 · chapel-lang/chapel · GitHub Log Message: Pass char* to a strchr call in the runtime Fix a compiler error when building the runtime with later versions of glibc. Adjust runtime code to work around changes to strchr. In C23, strchr now has two overloads. The…

[Chapel Merge] Adjust mentions of 'chpl_preUserCodeHook' -> 'chpl_preUserCodeSync'

Branch: fix-pre-user-code-hook-mentions Revision: b9db5111a28630e7d3325f10f7fd90f4b8666a01 Author: dlongnecke-cray Merged at: 2026-11-20T00:39:20Z Link: Adjust mentions of 'chpl_preUserCodeHook' -> 'chpl_preUserCodeSync' by dlongnecke-cray · Pull Request #29022 · chapel-lang/chapel · GitHub Log Message: Adjust mentions of 'chpl_preUserCodeHook' -> 'chpl_preUserCodeSync' Adjust two mentions of the old symbol chpl_preUserCodeHook to point to the new one chpl_preUserCodeSync. I didn't catch these…

[Chapel Merge] Run paratest with quickstart in merge queue

Branch: move-quickstart-to-queue Revision: 307a3374685421d63dec965742b2e096490b71a2 Author: jabraham17 Merged at: 2026-11-19T22:32:42Z Link: Run paratest with quickstart in merge queue by jabraham17 · Pull Request #29718 · chapel-lang/chapel · GitHub Log Message: Run paratest with quickstart in merge queue Moves the paratest with quickstart from full checks to extended checks. This whole workflow runs in about 35 minutes, so it should be well under our limit for the merge queue. This is…

[Chapel Merge] Expand simulated GPU CI

Branch: expand-gpu-ci Revision: 981f02db877a5ebc58cbcc4686d41bc60729df7c Author: jabraham17 Merged at: 2026-11-19T18:40:20Z Link: Expand simulated GPU CI by jabraham17 · Pull Request #29709 · chapel-lang/chapel · GitHub Log Message: Expand simulated GPU CI Expands Chapels support for GPU testing in the CI Builds on Add basic NVIDIA GPU correctness tests by jabraham17 · Pull Request #29708 · chapel-lang/chapel · GitHub Testing now covers the full GPU test suite (modulo some issues with virtual…

[Chapel Merge] Make sure users get errors for bad LLVM and ROCm 6.3

Branch: fix-gpu-eror Revision: b9cd70da44352dfe871efc517fd8b0e32866585a Author: jabraham17 Merged at: 2026-11-19T18:36:07Z Link: Make sure users get errors for bad LLVM and ROCm 6.3 by jabraham17 · Pull Request #29716 · chapel-lang/chapel · GitHub Log Message: Make sure users get errors for bad LLVM and ROCm 6.3 Recently someone hit an unexpected error when trying to build Chapel with the bundled LLVM and ROCm 6.3, which is not supported. Adjust chpl_gpu.py to make sure they get a nice error.…

[Chapel GA Tests] 2026-11-19

The python-3.11 issue is the same as CUDA 13.1, we are getting the dependencies just really really slowly (look at the network rates in the log). It can be fixed in a similar way as Adds an explicit apt install timeout by jabraham17 · Pull Request #29650 · chapel-lang/chapel · GitHub, just make installing the deps timeout if they run too slowly. Its not as clean as GCC since we don't cache those deps, but at least it won't consume 6 hours -Jade

[Chapel GA Tests] 2026-11-19

@dal , the DynamicLoading failures on darwin are likely from your PR merged yesterday. @jabraham , I think you recently merged a suppressif for library/packages/Socket/networkoptimization. Can you take a look at this quickstart failure? Also @jabraham , I'll assume the newly expanded CUDA 13.1 timeout is in your court. I'm guessing the python-3.11 failure is a one-off issue. Strange that we'd try installing the dependencies for 6 hours. Do we have a limit on retries? -Ben

[Chapel GA Tests] 2026-11-19

The following GitHub Actions jobs failed: extended-checks / python-version-tests (3.11, gasnet), step: "Install system dependencies" failure: The job has exceeded the maximum execution time of 6h0m0s failure: The operation was canceled. full-checks / test-simulated-gpu / Test CUDA 13.1, step: "Install LLVM and Chapel build dependencies" failure: The job has exceeded the maximum execution time of 6h0m0s failure: The operation was canceled. full-checks / Chapel paratest with quickstart…

[Chapel Merge] dlopen: Use runtime build markers to check if it is a dynamic library or not

Branch: dynamic-loading-59 Revision: 8ecfb3a5ec02edf6faf3d5bca4b2b3b449d62fd1 Author: dlongnecke-cray Merged at: 2026-11-19T06:54:46Z Link: dlopen: Use runtime build markers to check if it is a dynamic library or not by dlongnecke-cray · Pull Request #29363 · chapel-lang/chapel · GitHub Log Message: dlopen: Use runtime build markers to check if it is a dynamic library or not This PR improves checks when dynamic loading. It uses the built-in runtime build marker to detect if the runtime…

[Chapel Merge] Add nightly email text when aggregate paratest fails due to missing job

Branch: more-email-info Revision: b633b5b5042d4083bc080c1fc38c3ea821164b32 Author: jabraham17 Merged at: 2026-11-19T06:42:36Z Link: Add nightly email text when aggregate paratest fails due to missing job by jabraham17 · Pull Request #29715 · chapel-lang/chapel · GitHub Log Message: Add nightly email text when aggregate paratest fails due to missing job Adds text to the nightly failure email explaining why an aggregate paratest failed, ie if its missing jobs due to a node failures [Not reviewed…

[Chapel Merge] Add skipifs for missing extern block support

Branch: fix-skipif2 Revision: 732cf95e1d870063b5be878a8ba9f7ed0642645f Author: jabraham17 Merged at: 2026-11-19T01:56:49Z Link: Add skipifs for missing extern block support by jabraham17 · Pull Request #29712 · chapel-lang/chapel · GitHub Log Message: Add skipifs for missing extern block support Adds skipifs to 2 tests that rely on extern blocks to work [Not reviewed - trivial] Compare: Add skipifs for missing extern block support by jabraham17 · Pull Request #29712 · chapel-lang/chapel ·…

[Chapel Merge] Support multi-locale gpu programs in linux packages

Branch: expand-packages-again Revision: 778298f5af99669f8af6eb5c83944524cb2d24c1 Author: jabraham17 Merged at: 2026-11-18T22:51:53Z Link: Support multi-locale gpu programs in linux packages by jabraham17 · Pull Request #29713 · chapel-lang/chapel · GitHub Log Message: Support multi-locale gpu programs in linux packages Adds support for multi-locale gpu programs in the ubuntu26 package [Reviewed by @dlongnecke-cray] Compare: Support multi-locale gpu programs in linux packages by jabraham17 ·…

[Chapel Merge] Run gpu-cpu tests in extended checks

Branch: more-cpu-cpu Revision: 2f5f357a7e4e80fcb55bfb91824d5bf95da51132 Author: jabraham17 Merged at: 2026-11-18T22:51:52Z Link: Run gpu-cpu tests in extended checks by jabraham17 · Pull Request #29710 · chapel-lang/chapel · GitHub Log Message: Run gpu-cpu tests in extended checks Adds gpu-cpu tests to extended checks since they run really fast and cover an important merge gate Also adds gpu-cpu over gasnet [Reviewed by @dlongnecke-cray] Compare: Run gpu-cpu tests in extended checks by…

[Chapel Merge] Split linux32 build and test

Branch: linux32-cache Revision: 2dd0a3e9a989fa07eca18341122cf7e0b0d3cea6 Author: jabraham17 Merged at: 2026-11-18T22:03:18Z Link: Split linux32 build and test by jabraham17 · Pull Request #29711 · chapel-lang/chapel · GitHub Log Message: Split linux32 build and test Splits linxu32 testing build and test steps This allows us to save the cache for the build step, so if the cache is evicted and we run the full rebuild of the compiler (2+ hrs) and then hit the 6 hr timeout, we should still hit the…

[Chapel GA Tests] 2026-11-18

The following GitHub Actions jobs failed: full-checks / test-examples-linux32, step: "run examples tests in linux32 container" failure: The job has exceeded the maximum execution time of 6h0m0s failure: The operation was canceled. notice: "The ubuntu-latest label will migrate to Ubuntu 26 beginning October 19, 2026. For more information, see [Ubuntu] `ubuntu-latest` label will use Ubuntu 26.04 in November 2026 · Issue #14748 · actions/runner-images · GitHub" full-checks / Chapel paratest on…

[Chapel Merge] Add basic NVIDIA GPU correctness tests

Branch: add-basic-gpu-tests Revision: c23447d7e8df4570d3d6c71f903ff8fd8d0e1032 Author: jabraham17 Merged at: 2026-11-18T02:31:06Z Link: Add basic NVIDIA GPU correctness tests by jabraham17 · Pull Request #29708 · chapel-lang/chapel · GitHub Log Message: Add basic NVIDIA GPU correctness tests Adds basic GPU correctness tests using a simulated GPU. This initial work just runs hello-gpu.chpl Future work expand the subset of tests run AMD GPUs test the nvidia linux package using this approach…

[Chapel Merge] Restrict optimize to non sanitizer workloads for sort tests

Branch: fix-sortStrings Revision: 72ffd5a8551f8b4d5b434d3a108f81d96c59d10c Author: jabraham17 Merged at: 2026-11-18T02:11:45Z Link: Restrict optimize to non sanitizer workloads for sort tests by jabraham17 · Pull Request #29704 · chapel-lang/chapel · GitHub Log Message: Restrict optimize to non sanitizer workloads for sort tests Restrict backend optimizations to non sanitizer workloads for sort tests, as with sanitizers they timeout during compilation Introduced by Add baseline testing by…

[Chapel Merge] Improve Subprocess waits

Branch: spawn-improve Revision: 6ad524e5e1a96e57d2957e81e7eea68b7464b557 Author: jabraham17 Merged at: 2026-11-17T21:44:04Z Link: Improve Subprocess waits by jabraham17 · Pull Request #29695 · chapel-lang/chapel · GitHub Log Message: Improve Subprocess waits Improves waits when spawning subprocesses in Chapel Subprocess waits currently poll aggressively, wasting CPU and increasing contention in subprocess-heavy workloads. This recently became an issue in our CI, but this could affect users as…

[Chapel Merge] Fix perf playground

Branch: fix-playground Revision: 7a368b003333f6f824b6a0c4dc4c03364733b58b Author: jabraham17 Merged at: 2026-11-17T20:52:51Z Link: Fix perf playground by jabraham17 · Pull Request #29706 · chapel-lang/chapel · GitHub Log Message: Fix perf playground Fix perf playground [Not reviewed - trivial] Compare: Fix perf playground by jabraham17 · Pull Request #29706 · chapel-lang/chapel · GitHub Diff: M util/cron/test-perf.chapcs.playground.bash https://github.com/chapel-lang/chapel/pull/29706.diff

[Chapel Merge] Restrict cccl workaround to cuda 13.1-13.4

Branch: fix-cuda Revision: aa513b73f859b99937016bd0ca2c435e63587492 Author: jabraham17 Merged at: 2026-11-17T20:36:38Z Link: Restrict cccl workaround to cuda 13.1-13.4 by jabraham17 · Pull Request #29705 · chapel-lang/chapel · GitHub Log Message: Restrict cccl workaround to cuda 13.1-13.4 Restricts a workaround added in Add CHPL_GPU=nvidia to linux packages + fixes by jabraham17 · Pull Request #29681 · chapel-lang/chapel · GitHub to CUDA 13.1 [Not reviewed - trivial] Compare: Restrict cccl…

[Chapel Merge] Add mimalloc and no compiler driver testing

Branch: add-mimalloc-and-no-comp-drive Revision: ef8dc3e89c18c61f2eff0229153b4ef137c9dc77 Author: jabraham17 Merged at: 2026-11-17T20:20:24Z Link: Add mimalloc and no compiler driver testing by jabraham17 · Pull Request #29696 · chapel-lang/chapel · GitHub Log Message: Add mimalloc and no compiler driver testing Adds basic GA testing for mimalloc and --no-compiler-driver [Reviewed by @benharsh] Compare: Add mimalloc and no compiler driver testing by jabraham17 · Pull Request #29696 ·…

Running Chapel programs on a Canadian VPS

These seem like reasonable steps. I'm not a big expert on using Chapel in containers in the true distributed sense, but I know it can be done. You might want to look at https://github.com/Bears-R-Us/arkouda/tree/main/docker/hpe-cray-ex, which explains how to do this for the Arkouda project and is specific to HPE Cray EX machines. You might want to see how thats done to do it with GASNet. A few notes Tasks: set CHPL_RT_NUM_THREADS_PER_LOCALE to the vCPU count and check how it handles shared…

New Issue: GA workflow to update bundled LLVM version

29703, "arifthpe", "GA workflow to update bundled LLVM version", "2026-11-17T16:43:49Z" Open issue chapel-lang/chapel GA workflow to update bundled LLVM version opened 04:43PM - 17 Nov 26 UTC arifthpe area: BTR In the same vein as https://github.com/chapel-lang/chapel/issues/29702, but this… is a way larger update (and intentionally updates code) so the benefit is greater. Our current process for updating the bundled LLVM version is running [`third-party/l…