Clean up stdexec-only paths and update affected tests by guptapratykshh · Pull Request #7123 · TheHPXProject/hpx

guptapratykshh · 2026-03-28T05:53:16Z

Proposed Changes

removedHPX_HAVE_STDEXECguarded cleanup paths and kept the stdexec-only implementation as the single active path
update the affected execution, execution_base, executor, algorithm, async MPI/CUDA, and synchronization tests to match the unified stdexec-only code path
cleared up related CMake/stdexec configuration so the build reflects the same single-path setup consistently

Any background context you want to provide?

Checklist

Not all points below apply to all pull requests.

I have added a new feature and have added tests to go along with it.
I have fixed a bug and have added a regression test.
I have added a test using random numbers; I have made sure it uses a seed, and that random numbers generated are valid inputs for the tests.

StellarBot · 2026-03-28T05:55:03Z

Can one of the admins verify this patch?

hkaiser · 2026-03-28T11:11:50Z

@isidorostsa Please have a look

hkaiser · 2026-03-28T14:28:31Z

BTW, how does this relate to #6708? Does this PR superseds it?

hkaiser · 2026-03-28T14:29:24Z

@guptapratykshh thanks for working on this. Could you please proactively keep an eye on the CI results. Currently there seem to be some compilation issues with this PR.

guptapratykshh · 2026-03-28T14:36:57Z

BTW, how does this relate to #6708? Does this PR superseds it?

this PR is a narrower follow up in the same direction as #6708, updated on top of the current master rather than a wholesale replacement for it. It keeps the stdexec-only cleanup goal, but its scope here is mainly removing the remaining dead internal paths and fixing the resulting test issues while preserving backward compatibility.

hkaiser · 2026-03-28T14:56:35Z

BTW, how does this relate to #6708? Does this PR superseds it?

this PR is a narrower follow up in the same direction as #6708, updated on top of the current master rather than a wholesale replacement for it. It keeps the stdexec-only cleanup goal, but its scope here is mainly removing the remaining dead internal paths and fixing the resulting test issues while preserving backward compatibility.

Thanks for the explanations. Do we still need #6708? What does it have in addition to your PR?

guptapratykshh · 2026-03-28T15:15:49Z

BTW, how does this relate to #6708? Does this PR superseds it?

this PR is a narrower follow up in the same direction as #6708, updated on top of the current master rather than a wholesale replacement for it. It keeps the stdexec-only cleanup goal, but its scope here is mainly removing the remaining dead internal paths and fixing the resulting test issues while preserving backward compatibility.

Thanks for the explanations. Do we still need #6708? What does it have in addition to your PR?

after merging #6708 into my branch and updating it to work with current master, I do not think we still need #6708 as separate PR

hkaiser · 2026-03-28T15:34:34Z

Thanks for the explanations. Do we still need #6708? What does it have in addition to your PR?

after merging #6708 into my branch and updating it to work with current master, I do not think we still need #6708 as separate PR

Ok, please let me know once you think that we can close the other PR.

isidorostsa

Thank you for taking on this task! Removing old senders would be helpful

To make it possible to review this please contain unnecessary or pedantic edits as much as possible.

Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

Copilot

Pull request overview

Copilot reviewed 147 out of 167 changed files in this pull request and generated 2 comments.

Comments suppressed due to low confidence (1)

libs/core/execution_base/tests/unit/CMakeLists.txt:1

This disables the entire execution_base unit test suite for Clang/AppleClang, which significantly reduces CI signal and increases the risk of undetected regressions on those compilers. If only a subset fails (e.g., due to deprecation warnings or specific ICEs), prefer disabling only the affected targets (via list(REMOVE_ITEM ...) or per-test conditions) or adding targeted compile options, rather than excluding the full suite.

# Copyright (c) 2019 Thomas Heller

…flow Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

Copilot

Pull request overview

Copilot reviewed 146 out of 167 changed files in this pull request and generated 4 comments.

hkaiser · 2026-04-16T19:57:29Z

@@ -23,3 +23,4 @@ tim
 uint
 vill
 wit
+endwhile


The words in this list must be alphabetically ordered.

Now you have the new word in the list twice.

hkaiser · 2026-04-16T21:40:17Z

+tests.examples.transpose.transpose_smp_block
 tests.examples.async_io.async_io_simple
 tests.examples.modules.algorithms.run_on_all
 tests.examples.modules.checkpoint.1d_stencil_4_checkpoint


I believe this is a leftover file and should have been removed. If you want to exclude certain tests from the CI, please add those here: https://github.com/TheHPXProject/hpx/blob/master/.github/workflows/exclude.targets

hkaiser · 2026-04-16T21:41:08Z

+    # stdexec's stop_token header currently declares _mm_pause without the C
+    # linkage used by MSVC's intrinsic headers, which breaks module builds.
+    string(REPLACE "extern void _mm_pause();" "extern \"C\" void _mm_pause();"
+                   _patched_content "${_patched_content}"


Why did you resolve this comment (and similar comments below)?

Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

Copilot

Pull request overview

Copilot reviewed 147 out of 167 changed files in this pull request and generated 2 comments.

Comments suppressed due to low confidence (2)

libs/core/execution/tests/unit/CMakeLists.txt:1

This change disables the entire execution unit test suite for Clang/AppleClang, which is a major loss of CI signal and can mask regressions. If only a subset of tests (or specific Clang versions) are problematic, prefer excluding just those targets (e.g., list(REMOVE_ITEM ...)) or adding versioned guards (similar to the approach in foreach_scheduler.cpp) so the rest of the suite still runs on Clang.

# Copyright (c) 2014-2026 Hartmut Kaiser

libs/core/execution_base/tests/unit/CMakeLists.txt:1

Like the execution tests, this disables the full execution_base unit test suite for Clang/AppleClang. Consider narrowing the exclusion to the specific failing tests or to specific Clang versions; otherwise Clang builds will miss coverage for a core module that was heavily modified in this PR.

# Copyright (c) 2019 Thomas Heller

Copilot

Pull request overview

Copilot reviewed 147 out of 167 changed files in this pull request and generated 6 comments.

Copilot · 2026-04-20T05:18:21Z

+if(NOT HPX_WITH_STDEXEC)
+  hpx_error(
+    "HPX_WITH_STDEXEC=OFF is no longer supported. HPX now requires stdexec."
+  )
+endif()


The option help text still suggests there are two supported modes (stdexec vs native), but the configuration now hard-errors when HPX_WITH_STDEXEC=OFF. Please update the option description to reflect that stdexec is required (or consider removing the option entirely if it is no longer a meaningful configuration knob).

I think we can remove the option altogether.

Copilot

Pull request overview

Copilot reviewed 147 out of 167 changed files in this pull request and generated 4 comments.

Copilot

Pull request overview

Copilot reviewed 147 out of 168 changed files in this pull request and generated 1 comment.

Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

Copilot

Pull request overview

Copilot reviewed 147 out of 167 changed files in this pull request and generated 1 comment.

…exec PR#2035) Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

hkaiser · 2026-04-21T18:53:12Z

@guptapratykshh Thanks so much for your patience with this. This PR is very large, so it is of utmost importance to separate all unrelated changes into their own PRs. Please make sure that the changes to the build system you applied are strictly related to this PR. I think you should not disable unrelated tests, even if those fail on the CIs, for instance. Also, please pay attention to the copilot comments (and not simply resolve them) as usually there is some truth to what it suggest (even if the suggested solution might not be appropriate).

hkaiser · 2026-04-21T18:54:43Z

 if(HPX_WITH_STATIC_LINKING)
  set(disabled_tests ${disabled_tests} hello_world_1)
 endif()
+if(CMAKE_CXX_COMPILER_ID MATCHES "Clang|AppleClang")


If disabling tests, please make sure to specify the latest compiler version that we know to fail. That will allow for the tests to be re-evaluated if a newer version will be available.

hkaiser · 2026-04-21T18:55:14Z

                return util::detail::algorithm_result<ExPolicy, FwdIter2>::get(
-                    shift_right_helper(policy, first, last, new_first));
+                    shift_right_helper(
+                        policy, first, last, new_first, result_first));


The changes to this file seem to be unrelated.

hkaiser · 2026-04-21T18:57:07Z


 void for_each_test_execute_on_sender()
 {
+#if !defined(HPX_CLANG_VERSION)


Please add a compiler version to the condition

hkaiser · 2026-04-21T18:58:08Z

-    test_uninitialized_copy_n_sender(
-        hpx::launch::async, par(task), IteratorTag());
-    test_uninitialized_copy_n_sender(
-        hpx::launch::async, par_unseq(task), IteratorTag());


Why did you remove these test cases?

hkaiser · 2026-04-21T18:58:22Z

-    test_uninitialized_copy_n_exception_sender(
-        hpx::launch::async, par(task), IteratorTag());
-    test_uninitialized_copy_n_exception_sender(
-        hpx::launch::async, par_unseq(task), IteratorTag());


Same here, why did you remove these testcases. The same applies to more places in other files.

hkaiser · 2026-04-21T19:06:05Z

@@ -23,3 +23,4 @@ tim
 uint
 vill
 wit
+endwhile


Now you have the new word in the list twice.

hkaiser · 2026-04-21T19:06:53Z

+if(NOT HPX_WITH_STDEXEC)
+  hpx_error(
+    "HPX_WITH_STDEXEC=OFF is no longer supported. HPX now requires stdexec."
+  )
+endif()


I think we can remove the option altogether.

guptapratykshh requested a review from hkaiser as a code owner March 28, 2026 05:53

guptapratykshh force-pushed the cleanup/stdexec-pr branch 4 times, most recently from 7ca8cbe to a24be1a Compare March 28, 2026 08:45

hkaiser added category: senders/receivers Implementations of the p0443r14 / p2300 + p1897 proposals type: enhancement type: compatibility issue type: cleanup labels Mar 28, 2026

guptapratykshh force-pushed the cleanup/stdexec-pr branch from c180673 to 82b73f8 Compare March 28, 2026 15:31

guptapratykshh force-pushed the cleanup/stdexec-pr branch 8 times, most recently from 3863598 to 62d14bc Compare March 29, 2026 13:59

isidorostsa suggested changes Mar 29, 2026

View reviewed changes

guptapratykshh force-pushed the cleanup/stdexec-pr branch 3 times, most recently from 49ba52e to d94eaf8 Compare March 30, 2026 13:29

guptapratykshh force-pushed the cleanup/stdexec-pr branch from 06654c7 to 2ce2ba3 Compare April 15, 2026 17:15

remove orphaned #endif in mismatch_tests.hpp

dddf55e

Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

Copilot AI review requested due to automatic review settings April 16, 2026 05:20

Copilot AI reviewed Apr 16, 2026

View reviewed changes

Comment thread libs/core/async_cuda/include/hpx/async_cuda/transform_stream.hpp

Comment thread libs/core/execution_base/tests/unit/basic_schedule.cpp

guptapratykshh added 2 commits April 16, 2026 12:27

shift_right parallel path returns first+n and passes executor to data…

59ce7b7

…flow Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

exclude transpose_smp_block from Windows CI due to stack overrun crash

a43f187

Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

Copilot AI review requested due to automatic review settings April 16, 2026 08:42

Copilot AI reviewed Apr 16, 2026

View reviewed changes

Comment thread libs/core/execution_base/tests/unit/CMakeLists.txt

Comment thread libs/core/execution/tests/unit/CMakeLists.txt

Comment thread libs/core/execution/CMakeLists.txt

Comment thread cmake/HPX_SetupStdexec.cmake Outdated

guptapratykshh requested a review from hkaiser April 16, 2026 16:12

hkaiser reviewed Apr 16, 2026

View reviewed changes

guptapratykshh added 2 commits April 17, 2026 11:03

move CI exclusions to exclude.targets and sort codespell whitelist

e3183de

Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

remove patches fixed upstream in NVIDIA/stdexec PR#2030

32683b8

Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

Copilot AI review requested due to automatic review settings April 19, 2026 04:17

Copilot AI reviewed Apr 19, 2026

View reviewed changes

Comment thread libs/core/execution_base/tests/unit/CMakeLists.txt

Comment thread cmake/HPX_SetupStdexec.cmake Outdated

Copilot AI review requested due to automatic review settings April 20, 2026 05:13

Copilot AI reviewed Apr 20, 2026

View reviewed changes

Copilot AI review requested due to automatic review settings April 20, 2026 07:05

guptapratykshh force-pushed the cleanup/stdexec-pr branch from ce48130 to f6b707d Compare April 20, 2026 07:05

Copilot AI reviewed Apr 20, 2026

View reviewed changes

Copilot AI review requested due to automatic review settings April 20, 2026 07:26

guptapratykshh force-pushed the cleanup/stdexec-pr branch from 4e3595b to d8193f7 Compare April 20, 2026 07:26

Copilot AI reviewed Apr 20, 2026

View reviewed changes

Comment thread libs/core/execution/tests/unit/CMakeLists.txt Outdated

Copilot AI review requested due to automatic review settings April 20, 2026 11:50

guptapratykshh force-pushed the cleanup/stdexec-pr branch from 63b569e to 32683b8 Compare April 20, 2026 11:50

remove stdexec::tags re-export (namespace removed from upstream stdexec)

8b4f6fc

Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

Copilot AI reviewed Apr 20, 2026

View reviewed changes

Comment thread libs/core/execution_base/tests/unit/CMakeLists.txt

remove stdexec digit separator patching (fixed upstream in NVIDIA/std…

ff262f9

…exec PR#2035) Signed-off-by: guptapratykshh <pratykshgupta9999@gmail.com>

guptapratykshh requested a review from hkaiser April 21, 2026 09:16

hkaiser reviewed Apr 21, 2026

View reviewed changes

Uh oh!

Conversation

guptapratykshh commented Mar 28, 2026

Proposed Changes

Any background context you want to provide?

Checklist

Uh oh!

StellarBot commented Mar 28, 2026

Uh oh!

hkaiser commented Mar 28, 2026

Uh oh!

hkaiser commented Mar 28, 2026

Uh oh!

hkaiser commented Mar 28, 2026

Uh oh!

guptapratykshh commented Mar 28, 2026

Uh oh!

hkaiser commented Mar 28, 2026

Uh oh!

guptapratykshh commented Mar 28, 2026

Uh oh!

hkaiser commented Mar 28, 2026

Uh oh!

isidorostsa left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

hkaiser Apr 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI Apr 20, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

hkaiser Apr 16, 2026 •

edited

Loading