arichardson added inline comments.

================
Comment at: clang/lib/CMakeLists.txt:24
 add_subdirectory(IndexSerialization)
-if(CLANG_ENABLE_STATIC_ANALYZER)
-  add_subdirectory(StaticAnalyzer)
----------------
arichardson wrote:
> thakis wrote:
> > hans wrote:
> > > Why does removing the condition here work?
> > As far as I understand, it just impacts which targets CMake generates. 
> > clang/lib/FrontendTool/CMakeLists.txt only adds the dep on 
> > clangStaticAnalyzerFrontend if CLANG_ENABLE_STATIC_ANALYZER is set, so this 
> > doesn't change what gets built for "clang". If you build all targets, this 
> > will now always build the analyzer sources and I suppose it makes it a bit 
> > easier to accidentally depend on clangStaticAnalyzerFrontend , but I don't 
> > know of another way to be able to link this into clang-tidy when it's not 
> > built at all over here.
> I just noticed that my builds (just a plain `ninja`) are compiling all static 
> analyzer sources. I am explicitly passing 
> `-DCLANG_ENABLE_STATIC_ANALYZER=OFF` to cmake (and not building any 
> clang-tools-extra).
> I feel like this was not happening before so it's possible there was some 
> CMake change more recently that is now causing this behaviour.
> 
> I tried setting EXCLUDE_FROM_ALL 
> (https://cmake.org/cmake/help/latest/prop_tgt/EXCLUDE_FROM_ALL.html) on the 
> directories and targets but that didn't fix the issue for me.
> 
> How about changing the condition to
> `if (CLANG_TIDY_ENABLE_STATIC_ANALYZER OR CLANG_ENABLE_STATIC_ANALYZER)`? Or 
> will that not work since CLANG_TIDY_ENABLE_STATIC_ANALYZER isn't defined yet?
Possible solution: D109611


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D87118/new/

https://reviews.llvm.org/D87118

_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
  • [PATCH] D87118: Add a... Alexander Richardson via Phabricator via cfe-commits
    • [PATCH] D87118: ... Alexander Richardson via Phabricator via cfe-commits

Reply via email to