FailedChanges

Summary

  1. [DAG] getNode(ISD::EXTRACT_SUBVECTOR) Drop unnecessary N2C null check - we assert that this isn't null and have already used the pointer. NFCI. (details)
  2. [InstCombine] foldOrOfICmps - use m_Specific instead of explicit comparisons. NFCI. (details)
  3. [Clang] Update newpm pipeline test in clang after D87322. (details)
  4. [clang testing] Fix a read-only source build system failure (details)
  5. [llvm-mca] Add test for cortex-a57 memory instructions (details)
  6. [llvm] Use early exits and get rid of if-return-else-return pattern; NFC (details)
  7. [LIT] error if directly named test won't be run indirectly (details)
  8. [LIT] fixed up accidentally committed EOF problems. (details)
  9. [SystemZ] Mark unsaved argument R6 as live throughout function. (details)
  10. [libc++] Include <__config_site> from <__config> (details)
  11. [CostModel] remove cost-kind predicate for memcpy cost (details)
  12. [libc++] Revert "Include <__config_site> from <__config>" (details)
  13. Only run when `arm` is registered. NFC. (details)
  14. [DebugInfo] Follow up c521e44defb5 with an API improvement (details)
  15. [nfc] [lldb] Fix harmless slicing of DWARFDIE (details)
  16. [ARM] [TableGen] Clean up !if(!eq(boolean, 1) and related booleans (details)
  17. Remove .svn from exclude list as we moved to git (details)
  18. [lldb] Port lldb gdb-server to libOption (details)
  19. Fix "Unknown arguments specified" to if in lldb (details)
  20. [take 2] [libc++] Include <__config_site> from <__config> (details)
Commit 88523f6f4bd10287455f058b386306e11a518b3d by llvm-dev
[DAG] getNode(ISD::EXTRACT_SUBVECTOR) Drop unnecessary N2C null check - we assert that this isn't null and have already used the pointer. NFCI.

Fixes cppcheck + null dereference warning.
The file was modifiedllvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
Commit 7b4a828452f4f59adf1561697b853e78efd006a9 by llvm-dev
[InstCombine] foldOrOfICmps - use m_Specific instead of explicit comparisons. NFCI.
The file was modifiedllvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp
Commit c50f0d239dc367b087ef7b06d413ae1df0750758 by flo
[Clang] Update newpm pipeline test in clang after D87322.

This fixes a test failure because a LLVM pipeline test file in clang/
did not get updated in 88241ffb5636.
The file was modifiedclang/test/CodeGen/thinlto-distributed-newpm.ll
Commit 87f6de72bcd346bbbf468e9f9a0e9d1bbf0630a9 by dave
[clang testing] Fix a read-only source build system failure
The file was modifiedclang/test/CodeGen/basic-block-sections.c
Commit 9f5ece63ce62253321a8e8cdd3e052b5b5270b8e by eleviant
[llvm-mca] Add test for cortex-a57 memory instructions
The file was addedllvm/test/tools/llvm-mca/ARM/cortex-a57-memory-instructions.s
Commit 96685faf6dd9b044394af6f7a9d8b10fadb327b5 by kbobyrev
[llvm] Use early exits and get rid of if-return-else-return pattern; NFC

https://llvm.org/docs/CodingStandards.html#use-early-exits-and-continue-to-simplify-code

Reviewed By: kadircet

Differential Revision: https://reviews.llvm.org/D89857
The file was modifiedllvm/lib/Support/Path.cpp
Commit 81c0d36a1836c9be7c34a6d8198310ad7ea9bb53 by Ben.Dunbobbin
[LIT] error if directly named test won't be run indirectly

Currently, a LIT test named directly (on the command line) will
be run even if the name of the test file does not meet the rules
to be considered a test in the LIT test configuration files for
its test suite. For example, if the test does not have a
recognised file extension.

This makes it relatively easy to write a LIT test that won't
actually be run. I did in: https://reviews.llvm.org/D82567

This patch adds an error to avoid users doing that. There is a
small performance overhead for this check. A command line option
has been added so that users can opt into the old behaviour.

Differential Revision: https://reviews.llvm.org/D83069
The file was modifiedllvm/utils/lit/lit/cl_arguments.py
The file was modifiedllvm/utils/lit/lit/discovery.py
The file was modifiedllvm/utils/lit/lit/LitTestCase.py
The file was modifiedllvm/utils/lit/tests/discovery.py
The file was modifiedllvm/utils/lit/lit/main.py
The file was modifiedllvm/utils/lit/tests/unit/TestRunner.py
The file was addedllvm/utils/lit/tests/Inputs/discovery/test.not-txt
Commit 6a5c19dd3956b6fe4607c8dd00a550bf3ead25b0 by Ben.Dunbobbin
[LIT] fixed up accidentally committed EOF problems.
The file was modifiedllvm/utils/lit/lit/LitTestCase.py
The file was modifiedllvm/utils/lit/tests/unit/TestRunner.py
The file was modifiedllvm/utils/lit/lit/discovery.py
The file was modifiedllvm/utils/lit/tests/discovery.py
Commit 1606755da0e4fb362c88fdf02373f0dc3e4eef8f by paulsson
[SystemZ] Mark unsaved argument R6 as live throughout function.

For historical reasons, the R6 register is a callee-saved argument
register. This means that if it is used to pass an argument to a function
that does not clobber it, it is live throughout the function.

This patch makes sure that in this special case any kill flags of it are
removed.

Review: Ulrich Weigand, Eli Friedman

Differential Revision: https://reviews.llvm.org/D89451
The file was addedllvm/test/CodeGen/SystemZ/frame-26.mir
The file was modifiedllvm/lib/Target/SystemZ/SystemZFrameLowering.cpp
Commit 1e46d1aa3f9b618e8db32af5c855fbc386a950aa by Louis Dionne
[libc++] Include <__config_site> from <__config>

Prior to this patch, we would generate a fancy <__config> header by
concatenating <__config_site> and <__config>. This complexifies the
build system and also increases the difference between what's tested
and what's actually installed.

This patch removes that complexity and instead simply installs <__config_site>
alongside the libc++ headers. <__config_site> is then included by <__config>,
which is much simpler. Doing this also opens the door to having different
<__config_site> headers depending on the target, which was impossible before.

It does change the workflow for testing header-only changes to libc++.
Previously, we would run `lit` against the headers in libcxx/include.
After this patch, we run it against a fake installation root of the
headers (containing a proper <__config_site> header). This makes use
closer to testing what we actually install, which is good, however it
does mean that we have to update that root before testing header changes.
Thus, we now need to run `ninja check-cxx-deps` before running `lit` by
hand.

Differential Revision: https://reviews.llvm.org/D89041
The file was modifiedlibcxx/CMakeLists.txt
The file was modifiedlibcxxabi/src/CMakeLists.txt
The file was modifiedlibcxx/include/CMakeLists.txt
The file was modifiedlibcxx/utils/libcxx/test/config.py
The file was modifiedlibcxx/include/__config
The file was modifiedlibcxx/docs/TestingLibcxx.rst
The file was modifiedlibcxx/test/configs/legacy.cfg.in
Commit 01ea93d85d6e2240d70daf52e84772aba310bc19 by spatel
[CostModel] remove cost-kind predicate for memcpy cost

The default implementation base returns TCC_Expensive (currently
set to '4'), so that explains the test diff. This probably does
not make sense for most callers, but at least now the costs will
be consistently wrong instead of mysteriously wrong.

The ARM target has an override that tries to model codegen expansion,
and that should likely be adapted for general usage.

This probably does not affect anything because the vectorizers are
the primary users of the throughput cost, but memcpy is not listed
as a trivially vectorizable intrinsic.
The file was modifiedllvm/include/llvm/CodeGen/BasicTTIImpl.h
The file was modifiedllvm/test/Analysis/CostModel/X86/intrinsic-cost-kinds.ll
Commit eb60c48744f4edbdc72a44250c84796f3bf5f262 by Louis Dionne
[libc++] Revert "Include <__config_site> from <__config>"

This temporarily reverts commit 1e46d1aa until I find a solution to fix
the libc++abi and libunwind test suites with that change.
The file was modifiedlibcxx/CMakeLists.txt
The file was modifiedlibcxx/include/__config
The file was modifiedlibcxx/test/configs/legacy.cfg.in
The file was modifiedlibcxx/docs/TestingLibcxx.rst
The file was modifiedlibcxx/include/CMakeLists.txt
The file was modifiedlibcxx/utils/libcxx/test/config.py
The file was modifiedlibcxxabi/src/CMakeLists.txt
Commit 1bcec29afb321976cdcaa632ee6a47567dd651a7 by michael.hliao
Only run when `arm` is registered. NFC.
The file was modifiedclang/test/Driver/arm-float-abi.c
Commit 537f0fbe82049b8d5b6c700ecc4ab166c350b0c6 by jeremy.morse
[DebugInfo] Follow up c521e44defb5 with an API improvement

As mentioned post-commit in D85749, the 'substituteDebugValuesForInst'
method added in c521e44defb5 would be better off with a limit on the
number of operands to substitute. This handles the common case of
"substitute the first operand between these two differing instructions",
or possibly up to N first operands.
The file was modifiedllvm/lib/CodeGen/MachineFunction.cpp
The file was modifiedllvm/include/llvm/CodeGen/MachineFunction.h
Commit 7bf066a20f4bfd52a79ae7650632bb3925171104 by jan.kratochvil
[nfc] [lldb] Fix harmless slicing of DWARFDIE

Differential Revision: https://reviews.llvm.org/D89875
The file was modifiedlldb/source/Plugins/SymbolFile/DWARF/DWARFUnit.h
Commit dfd6b69e018c88c752e89b504c427718d99e6c8f by paul
[ARM] [TableGen] Clean up !if(!eq(boolean, 1) and related booleans

Differential Revision: https://reviews.llvm.org/D89822
The file was modifiedllvm/lib/Target/ARM/ARMInstrMVE.td
Commit 0784e17f1b4ac6b613ebf1fb1fb9b0dc9d0776ec by sylvestre
Remove .svn from exclude list as we moved to git

Reviewed By: emaste

Differential Revision: https://reviews.llvm.org/D89859
The file was modifiedlldb/cmake/modules/LLDBConfig.cmake
The file was modifiedpolly/CMakeLists.txt
The file was modifiedclang/tools/libclang/CMakeLists.txt
The file was modifiedllvm/cmake/modules/CMakeLists.txt
The file was modifiedlld/CMakeLists.txt
The file was modifiedpolly/lib/External/CMakeLists.txt
The file was modifiedclang/CMakeLists.txt
The file was modifiedllvm/CMakeLists.txt
Commit fa5fa63fd140f2d4bad0357839378606a583b32c by pavel
[lldb] Port lldb gdb-server to libOption

The existing help text was very terse and was missing several important
options. In the new version, I add a short description of each option
and a slightly longer description of the tool as a whole.

The new option list does not include undocumented no-op options:
--debug and --verbose. It also does not include undocumented short
aliases for long options, with two exceptions: -h, because it's
well-known; and -S (--setsid), as it's used in one test. Using these
options will now produce an error. I believe that is acceptable as users
aren't generally invoking lldb-server directly, and the only way to
learn about the short aliases was by looking at the source.

Differential Revision: https://reviews.llvm.org/D89477
The file was modifiedlldb/tools/lldb-server/CMakeLists.txt
The file was modifiedlldb/include/lldb/Utility/Args.h
The file was addedlldb/tools/lldb-server/LLGSOptions.td
The file was modifiedlldb/source/Utility/Args.cpp
The file was modifiedlldb/tools/lldb-server/lldb-gdbserver.cpp
The file was addedlldb/test/Shell/lldb-server/TestErrorMessages.test
Commit 26459e6d8eeb3c2e61c70d207de7e10a00b4ed1c by ctetreau
Fix "Unknown arguments specified" to if in lldb

Reviewed By: labath

Differential Revision: https://reviews.llvm.org/D89807
The file was modifiedlldb/cmake/modules/FindPythonAndSwig.cmake
Commit 5d796645d6c8cadeb003715c33e231a8ba05b6de by Louis Dionne
[take 2] [libc++] Include <__config_site> from <__config>

Prior to this patch, we would generate a fancy <__config> header by
concatenating <__config_site> and <__config>. This complexifies the
build system and also increases the difference between what's tested
and what's actually installed.

This patch removes that complexity and instead simply installs <__config_site>
alongside the libc++ headers. <__config_site> is then included by <__config>,
which is much simpler. Doing this also opens the door to having different
<__config_site> headers depending on the target, which was impossible before.

It does change the workflow for testing header-only changes to libc++.
Previously, we would run `lit` against the headers in libcxx/include.
After this patch, we run it against a fake installation root of the
headers (containing a proper <__config_site> header). This makes use
closer to testing what we actually install, which is good, however it
does mean that we have to update that root before testing header changes.
Thus, we now need to run `ninja check-cxx-deps` before running `lit` by
hand.

This commit was originally applied in 1e46d1aa3 and reverted in eb60c487
because it broke the libc++abi and libunwind test suites. This has now
been fixed.

Differential Revision: https://reviews.llvm.org/D89041
The file was modifiedlibcxx/test/configs/legacy.cfg.in
The file was modifiedlibcxx/docs/TestingLibcxx.rst
The file was modifiedlibcxxabi/test/lit.site.cfg.in
The file was modifiedlibcxxabi/CMakeLists.txt
The file was modifiedlibcxx/utils/libcxx/test/config.py
The file was modifiedlibcxxabi/test/libcxxabi/test/config.py
The file was modifiedlibcxx/include/__config
The file was modifiedlibcxx/include/CMakeLists.txt
The file was modifiedlibcxxabi/src/CMakeLists.txt
The file was modifiedlibcxx/CMakeLists.txt
The file was modifiedlibunwind/test/libunwind/test/config.py