[CMake] Don't wrap command in `ROOTTEST_COMPILE_MACRO` in CMake target #19780

guitargeek · 2025-08-30T15:06:58Z

Instead of running the root ... <macro>.C+ build step as the command
associated with a custom CMake target, just run the command directly in
the test added with add_test.

This simplification is possible because for the custom CMake target, no
dependencies are ever specified. This makes building it not different
from just running the underlying command directly.

Do some further CMake code simplifications in a second commit (more detail in the commit description).

github-actions · 2025-08-30T17:10:51Z

Test Results

19 files 19 suites 3d 9h 1m 51s ⏱️
3 658 tests 3 656 ✅ 0 💤 2 ❌
67 869 runs 67 767 ✅ 100 💤 2 ❌

For more details on these failures, see this check.

Results for commit c273e07.

♻️ This comment has been updated with latest results.

hahnjo

Wrapping to command inside a Cmake target doesn't seem to be necessary, because in the test, the target is forced to be built without dependencies anyway (e.g. cmake --build . --target my_macro_test/fast --always-make when using makefiles).

This is only partially true, for Unix Makefiles. With Ninja, we don't use /fast nor --always-make AFAICT

cmake/modules/RootMacros.cmake

guitargeek · 2025-08-31T10:26:35Z

The intention of the original macro is apparently to only build this one target that runs the command. The fact that this was not done correctly for Ninja doesn't mean that rebuilding all dependencies was the intended behavior. I suppose it was just not correctly implemented.

guitargeek · 2025-09-01T10:13:27Z

cmake/modules/RootMacros.cmake

@@ -2340,36 +2340,18 @@ macro(ROOTTEST_COMPILE_MACRO filename)
                            WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR})

  if(ARG_DEPENDS)
-    set(deps ${ARG_DEPENDS})
+    message(WARNING "Specifying DEPENDS in ROOTTEST_COMPILE_MACRO has no effect.")


Note that this warning is never hit. Nowhere in the codebase to we specify DEPENDS in a ROOTTEST_COMPILE_MACRO. Together with the dependencies being ignored via the \fast target when using Makefiles, that makes it clear to me that it was never the intention to have dependencies for these macro tests.

The problem is that even with no dependencies, a re-configuration will still be triggered if the CMake code was changed, which is the reason thy the RUN_SERIAL true workaround was needed.

I would like to suggest removing DEPENDS from the allowed arguments altogether.

hahnjo · 2025-09-04T07:08:17Z

In principle makes sense to me; maybe rephrase the first commit message because it still sounds wrong that dependencies would be ignored "anyway".

Also adding @hageboeck because it's related to #19702

hageboeck

I find the explanation confusing. I'll come discuss this in person.

hageboeck · 2025-09-05T12:59:23Z

cmake/modules/RootMacros.cmake

@@ -2340,36 +2340,18 @@ macro(ROOTTEST_COMPILE_MACRO filename)
                            WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR})

  if(ARG_DEPENDS)
-    set(deps ${ARG_DEPENDS})
+    message(WARNING "Specifying DEPENDS in ROOTTEST_COMPILE_MACRO has no effect.")


I would like to suggest removing DEPENDS from the allowed arguments altogether.

cmake/modules/RootMacros.cmake

Instead of running the `root ... <macro>.C+` build step as the command associated with a custom CMake target, just run the command directly in the test added with `add_test`. This simplification is possible because for the custom CMake target, no dependencies are ever specified. This makes building it not different from just running the underlying command directly.

* Remove preprocessor variables that were always defined and hence carried no information * Add include path and library path via environment variables * Remove redundant `gROOT->SetMacroPath()` call with path to source directory, because the `${realfp}` variable passed to the `build.C` script already contains the full path

hageboeck

Much clearer, thank you!

pcanal · 2025-09-05T18:21:26Z

roottest/scripts/build.C

@@ -75,9 +75,6 @@ void build(const char *filename, const char *lib = 0, const char *obj = 0, const
   }
   // fprintf(stderr,"from %s to %s\n",filename,fname.Data());
   int result = gSystem->CompileMacro(fname,"kc");
-#elif defined(CMakeEnvironment) && defined(CMakeBuildDir)
-   // fprintf(stderr,"CmakeBuildDir: %s filename: %s", CMakeBuildDir, filename);
-   int result = gSystem->CompileMacro(filename,"kc-", libname, CMakeBuildDir);


This seems to be an unintentional behavior change. If I read correctly, before this PR, ROOTTEST_COMPILE_MACRO was always setting CMakeEnvironment and CMakeBuildDir and thus always passing "kc-", while it now passes "only" "kc" thus removing the - and changing the output directory layout.

pcanal · 2025-09-05T18:29:51Z

Simplify ROOTTEST_COMPILE_MACRO to shorten test command
Add include path and library path via environment variables

How does switching from an argument to an environment variable concretely help? If one ever tries to reproduce the failing compilation 'exactly', one now needs to copy paste (the extended version of):

      ROOT_LIBRARY_PATH="${CMAKE_CURRENT_BINARY_DIR}"
      ROOT_INCLUDE_PATH="${CMAKE_CURRENT_BINARY_DIR}"
      ${ROOT_root_CMD}
      -e "gSystem->SetBuildDir(\"${CMAKE_CURRENT_BINARY_DIR}\", true)"
      ${RootMacroDirDefines}
      -q -l -b ....

instead of the extended version of:

      ${ROOT_root_CMD}
      -e "gSystem->AddDynamicPath(\"${CMAKE_CURRENT_BINARY_DIR}\")"
      -e "gInterpreter->AddIncludePath(\"-I${CMAKE_CURRENT_BINARY_DIR}\")"
      -e "gSystem->SetBuildDir(\"${CMAKE_CURRENT_BINARY_DIR}\", true)"
      ${RootMacroDirDefines}
      -q -l -b ...

guitargeek self-assigned this Aug 30, 2025

guitargeek requested a review from bellenot as a code owner August 30, 2025 15:06

guitargeek added in:Build System clean build Ask CI to do non-incremental build on PR labels Aug 30, 2025

guitargeek mentioned this pull request Aug 30, 2025

Revert "[CMake] Never force RUN_SERIAL in roottest when using Ninja" #19779

Merged

hahnjo reviewed Aug 31, 2025

View reviewed changes

cmake/modules/RootMacros.cmake Outdated Show resolved Hide resolved

guitargeek force-pushed the ninja_fixup branch from 94f401c to f215e1f Compare September 1, 2025 10:05

guitargeek mentioned this pull request Sep 1, 2025

Speed up CTest, particularly for RDataFrame #19702

Draft

guitargeek commented Sep 1, 2025

View reviewed changes

guitargeek force-pushed the ninja_fixup branch from f215e1f to c4ce0be Compare September 2, 2025 14:48

guitargeek requested a review from dpiparo as a code owner September 2, 2025 14:48

hahnjo requested a review from hageboeck September 4, 2025 07:07

hageboeck reviewed Sep 5, 2025

View reviewed changes

guitargeek added 2 commits September 5, 2025 15:57

guitargeek force-pushed the ninja_fixup branch from c4ce0be to c273e07 Compare September 5, 2025 14:09

hageboeck approved these changes Sep 5, 2025

View reviewed changes

pcanal reviewed Sep 5, 2025

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[CMake] Don't wrap command in `ROOTTEST_COMPILE_MACRO` in CMake target #19780

[CMake] Don't wrap command in `ROOTTEST_COMPILE_MACRO` in CMake target #19780

guitargeek commented Aug 30, 2025 •

edited

Loading

Uh oh!

github-actions bot commented Aug 30, 2025 •

edited

Loading

Uh oh!

hahnjo left a comment

Uh oh!

Uh oh!

guitargeek commented Aug 31, 2025

Uh oh!

guitargeek Sep 1, 2025

Uh oh!

hageboeck Sep 5, 2025

Uh oh!

hahnjo commented Sep 4, 2025

Uh oh!

hageboeck left a comment

Uh oh!

hageboeck Sep 5, 2025

Uh oh!

Uh oh!

hageboeck left a comment

Uh oh!

pcanal Sep 5, 2025

Uh oh!

pcanal commented Sep 5, 2025

Uh oh!

Uh oh!

[CMake] Don't wrap command in ROOTTEST_COMPILE_MACRO in CMake target #19780

Are you sure you want to change the base?

[CMake] Don't wrap command in ROOTTEST_COMPILE_MACRO in CMake target #19780

Conversation

guitargeek commented Aug 30, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Aug 30, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Test Results

Uh oh!

hahnjo left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

guitargeek commented Aug 31, 2025

Uh oh!

guitargeek Sep 1, 2025

Choose a reason for hiding this comment

Uh oh!

hageboeck Sep 5, 2025

Choose a reason for hiding this comment

Uh oh!

hahnjo commented Sep 4, 2025

Uh oh!

hageboeck left a comment

Choose a reason for hiding this comment

Uh oh!

hageboeck Sep 5, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

hageboeck left a comment

Choose a reason for hiding this comment

Uh oh!

pcanal Sep 5, 2025

Choose a reason for hiding this comment

Uh oh!

pcanal commented Sep 5, 2025

Uh oh!

Uh oh!

[CMake] Don't wrap command in `ROOTTEST_COMPILE_MACRO` in CMake target #19780

[CMake] Don't wrap command in `ROOTTEST_COMPILE_MACRO` in CMake target #19780

guitargeek commented Aug 30, 2025 •

edited

Loading

github-actions bot commented Aug 30, 2025 •

edited

Loading