[CMake] Re-add -fno-rtti to llvm-config --cxxflags #174084
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Fix for LLVM_CONFIG_HAS_RTTI.
nikic
reviewed
Jan 1, 2026
Contributor
nikic
left a comment
nikic
left a comment
There was a problem hiding this comment.
I'm somewhat confused by how this code works, including for the existing LLVM_CONFIG_HAS_RTTI flag. If LLVM_REQUIRES_RTTI is used somewhere, does this depend on whether the last target for which llvm_update_compile_flags is called happens to have that set or not?
Should llvm-config only be using the global LLVM_ENABLE_RTTI instead of this second LLVM_CONFIG_HAS_RTTI variable that also uses LLVM_REQUIRES_RTTI? And if so, can we extract determining the corresponding compiler flag into a separate function that gets explicitly called from the llvm-config cmake instead of relying on the side-effect here?
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
Contributor
Author
|
CI failure was a spurious link error, worked fine on second try. |
Collaborator
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/157/builds/43393 Here is the relevant piece of the build log for the reference |
Collaborator
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/95/builds/19609 Here is the relevant piece of the build log for the reference |
ronlieb
added a commit
to ROCm/llvm-project
that referenced
this pull request
Jan 2, 2026
…)" This reverts commit d1b88ca.
ronlieb
added a commit
to ROCm/llvm-project
that referenced
this pull request
Jan 2, 2026
Contributor
Author
|
Buildbot failures seem unrelated: first is just a timeout, second is spurious (tests pass in next build). @ronlieb you noted that this breaks your build -- what's happening and is this a pure downstream thing? |
Collaborator
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/76/builds/13697 Here is the relevant piece of the build log for the reference |
Collaborator
|
@aengelke This is spamming MSVC builds with exception override warnings https://lab.llvm.org/buildbot/#/builders/46/builds/28456 |
Contributor
Author
|
Thanks for noticing, hopefully fixed with 2b903df. |
Collaborator
Thanks - that fixed it |
Contributor
seems to be purely downstream, i think i am seeing an additionaly flag "-fno-rtti" during our build of offload |
Member
|
Same, I can see failures downstream even with the commit above, which causes: |
ronlieb
pushed a commit
to ROCm/llvm-project
that referenced
this pull request
Jan 2, 2026
Fix for llvm#173869. If there's no strong reason, we should get rid of per-target RTTI later.
Contributor
Author
Member
to me this indicates something's off with cmake, let me try a clean build, maybe these rtti changes messed up the cache etc. |
kuhar
added a commit
to iree-org/llvm-project
that referenced
this pull request
Jan 2, 2026
…)" This reverts commit d1b88ca.
kuhar
added a commit
to iree-org/iree
that referenced
this pull request
Jan 2, 2026
New reverts: * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932.
Abhishek-Varma
pushed a commit
to iree-org/llvm-project
that referenced
this pull request
Jan 5, 2026
…)" This reverts commit d1b88ca.
Yu-Zhewen
pushed a commit
to iree-org/iree
that referenced
this pull request
Jan 10, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) Signed-off-by: Abhishek Varma <[email protected]> Signed-off-by: Abhishek Varma <[email protected]>
Member
|
I also hit a linking error on my downstream project. I have to add |
Abhishek-Varma
pushed a commit
to iree-org/llvm-project
that referenced
this pull request
Jan 11, 2026
…)" This reverts commit d1b88ca.
Abhishek-Varma
added a commit
to Abhishek-Varma/iree
that referenced
this pull request
Jan 11, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) Signed-off-by: Abhishek Varma <[email protected]>
aengelke
added a commit
that referenced
this pull request
Jan 12, 2026
This should help users who use AddLLVM.cmake without HandleLLVMOptions. Note that remains MSVC is unlikely to work. Follow up of #174084.
krzysz00
pushed a commit
to iree-org/iree
that referenced
this pull request
Jan 12, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) Signed-off-by: Abhishek Varma <[email protected]> Signed-off-by: Abhishek Varma <[email protected]>
krzysz00
added a commit
to iree-org/iree
that referenced
this pull request
Jan 12, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. Reverts dropped: * llvm/llvm-project#174084 and followups - it works now
Abhishek-Varma
added a commit
to Abhishek-Varma/iree
that referenced
this pull request
Jan 13, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) Signed-off-by: Abhishek Varma <[email protected]>
krzysz00
added a commit
to iree-org/iree
that referenced
this pull request
Jan 13, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. Reverts dropped: * llvm/llvm-project#174084 and followups - it works now Local workarounds dropped: * Remove copying around some Python DLLs since upstream seems to put them in the right place now
Contributor
|
I do wonder whether omitting the exception flags from llvm-config is really the right choice. While not strictly necessary for ABI compatibility, it does seem pretty likely that code directly interfacing with LLVM also wants to use I ran into this change due to a build failure in LLVM bindings on MacOS, because C++17 with deployment target 10.12 is only supported with Manually setting |
Contributor
|
I'm not fully confident on whether we should include the EH flags or not, but I've put up #176195 anyway, before I start introducing workarounds downstream. |
nikic
added a commit
that referenced
this pull request
Jan 16, 2026
#173869 accidentally dropped rtti and eh flags from `llvm-config --cxxflags`. Then #174084 restored the rtti flags. The eh flags were not included with the rationale that they are not ABI relevant. This PR restores the eh flags as well. While they are not strictly necessary, I believe that code that directly interfaces with LLVM almost certainly does not want to build with exceptions if LLVM is not built with exceptions. Building in the peculiar `-fexceptions -fno-rtti` configuration is rarely useful and likely not intended. On MacOS, this is also relevant because it's not possible to use C++17 headers without `-fno-exceptions` when using older deployment targets. In that configuration, `-fno-exceptions` is required to interact with LLVM.
llvm-sync bot
pushed a commit
to arm/arm-toolchain
that referenced
this pull request
Jan 16, 2026
…(#176195) llvm/llvm-project#173869 accidentally dropped rtti and eh flags from `llvm-config --cxxflags`. Then llvm/llvm-project#174084 restored the rtti flags. The eh flags were not included with the rationale that they are not ABI relevant. This PR restores the eh flags as well. While they are not strictly necessary, I believe that code that directly interfaces with LLVM almost certainly does not want to build with exceptions if LLVM is not built with exceptions. Building in the peculiar `-fexceptions -fno-rtti` configuration is rarely useful and likely not intended. On MacOS, this is also relevant because it's not possible to use C++17 headers without `-fno-exceptions` when using older deployment targets. In that configuration, `-fno-exceptions` is required to interact with LLVM.
kkwli
pushed a commit
to kkwli/llvm-project
that referenced
this pull request
Jan 16, 2026
llvm#173869 accidentally dropped rtti and eh flags from `llvm-config --cxxflags`. Then llvm#174084 restored the rtti flags. The eh flags were not included with the rationale that they are not ABI relevant. This PR restores the eh flags as well. While they are not strictly necessary, I believe that code that directly interfaces with LLVM almost certainly does not want to build with exceptions if LLVM is not built with exceptions. Building in the peculiar `-fexceptions -fno-rtti` configuration is rarely useful and likely not intended. On MacOS, this is also relevant because it's not possible to use C++17 headers without `-fno-exceptions` when using older deployment targets. In that configuration, `-fno-exceptions` is required to interact with LLVM.
Priyanshu3820
pushed a commit
to Priyanshu3820/llvm-project
that referenced
this pull request
Jan 18, 2026
This should help users who use AddLLVM.cmake without HandleLLVMOptions. Note that remains MSVC is unlikely to work. Follow up of llvm#174084.
Priyanshu3820
pushed a commit
to Priyanshu3820/llvm-project
that referenced
this pull request
Jan 18, 2026
llvm#173869 accidentally dropped rtti and eh flags from `llvm-config --cxxflags`. Then llvm#174084 restored the rtti flags. The eh flags were not included with the rationale that they are not ABI relevant. This PR restores the eh flags as well. While they are not strictly necessary, I believe that code that directly interfaces with LLVM almost certainly does not want to build with exceptions if LLVM is not built with exceptions. Building in the peculiar `-fexceptions -fno-rtti` configuration is rarely useful and likely not intended. On MacOS, this is also relevant because it's not possible to use C++17 headers without `-fno-exceptions` when using older deployment targets. In that configuration, `-fno-exceptions` is required to interact with LLVM.
c-rhodes
pushed a commit
to llvmbot/llvm-project
that referenced
this pull request
Jan 19, 2026
llvm#173869 accidentally dropped rtti and eh flags from `llvm-config --cxxflags`. Then llvm#174084 restored the rtti flags. The eh flags were not included with the rationale that they are not ABI relevant. This PR restores the eh flags as well. While they are not strictly necessary, I believe that code that directly interfaces with LLVM almost certainly does not want to build with exceptions if LLVM is not built with exceptions. Building in the peculiar `-fexceptions -fno-rtti` configuration is rarely useful and likely not intended. On MacOS, this is also relevant because it's not possible to use C++17 headers without `-fno-exceptions` when using older deployment targets. In that configuration, `-fno-exceptions` is required to interact with LLVM. (cherry picked from commit 9bbea75)
llvm-sync bot
pushed a commit
to arm/arm-toolchain
that referenced
this pull request
Jan 19, 2026
…(#176195) llvm/llvm-project#173869 accidentally dropped rtti and eh flags from `llvm-config --cxxflags`. Then llvm/llvm-project#174084 restored the rtti flags. The eh flags were not included with the rationale that they are not ABI relevant. This PR restores the eh flags as well. While they are not strictly necessary, I believe that code that directly interfaces with LLVM almost certainly does not want to build with exceptions if LLVM is not built with exceptions. Building in the peculiar `-fexceptions -fno-rtti` configuration is rarely useful and likely not intended. On MacOS, this is also relevant because it's not possible to use C++17 headers without `-fno-exceptions` when using older deployment targets. In that configuration, `-fno-exceptions` is required to interact with LLVM. (cherry picked from commit 9bbea75)
BStott6
pushed a commit
to BStott6/llvm-project
that referenced
this pull request
Jan 22, 2026
llvm#173869 accidentally dropped rtti and eh flags from `llvm-config --cxxflags`. Then llvm#174084 restored the rtti flags. The eh flags were not included with the rationale that they are not ABI relevant. This PR restores the eh flags as well. While they are not strictly necessary, I believe that code that directly interfaces with LLVM almost certainly does not want to build with exceptions if LLVM is not built with exceptions. Building in the peculiar `-fexceptions -fno-rtti` configuration is rarely useful and likely not intended. On MacOS, this is also relevant because it's not possible to use C++17 headers without `-fno-exceptions` when using older deployment targets. In that configuration, `-fno-exceptions` is required to interact with LLVM.
keshavvinayak01
pushed a commit
to iree-org/iree
that referenced
this pull request
Jan 27, 2026
New reverts: * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. Signed-off-by: Keshav Vinayak Jha <[email protected]>
keshavvinayak01
pushed a commit
to iree-org/iree
that referenced
this pull request
Jan 27, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) Signed-off-by: Abhishek Varma <[email protected]> --------- Signed-off-by: Abhishek Varma <[email protected]> Co-authored-by: Jakub Kuderski <[email protected]> Signed-off-by: Keshav Vinayak Jha <[email protected]>
keshavvinayak01
pushed a commit
to iree-org/iree
that referenced
this pull request
Jan 27, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) Signed-off-by: Abhishek Varma <[email protected]> --------- Signed-off-by: Abhishek Varma <[email protected]> Signed-off-by: Keshav Vinayak Jha <[email protected]>
keshavvinayak01
pushed a commit
to iree-org/iree
that referenced
this pull request
Jan 27, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) -- Along with the above, IREE header fixes were done to accomodate changes introduced via llvm/llvm-project#82190 for `Affine/Transforms/Passes.h`. NOTE: [Twine fix commit](iree-org/llvm-project@2eaa29c) was dropped from our fork since llvm/llvm-project#175077 got merged upstream. Signed-off-by: Abhishek Varma <[email protected]> Signed-off-by: Abhishek Varma <[email protected]> Signed-off-by: Keshav Vinayak Jha <[email protected]>
keshavvinayak01
pushed a commit
to iree-org/iree
that referenced
this pull request
Jan 27, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) Signed-off-by: Abhishek Varma <[email protected]> Signed-off-by: Abhishek Varma <[email protected]> Signed-off-by: Keshav Vinayak Jha <[email protected]>
keshavvinayak01
pushed a commit
to iree-org/iree
that referenced
this pull request
Jan 27, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. * llvm/llvm-project#174084 -- causes linker errors related to rtti in stablehlo code * llvm/llvm-project@2b903df -- follow up fix to the above (doesn't fix it for us) Signed-off-by: Abhishek Varma <[email protected]> Signed-off-by: Abhishek Varma <[email protected]> Signed-off-by: Keshav Vinayak Jha <[email protected]>
keshavvinayak01
pushed a commit
to iree-org/iree
that referenced
this pull request
Jan 27, 2026
Existing local reverts carried forward: * Local revert of llvm/llvm-project#169614 due to llvm/llvm-project#172932. Reverts dropped: * llvm/llvm-project#174084 and followups - it works now Local workarounds dropped: * Remove copying around some Python DLLs since upstream seems to put them in the right place now Signed-off-by: Keshav Vinayak Jha <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
10 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix for #173869.
If there's no strong reason, we should get rid of per-target RTTI later.