In D61754#1500658, @hfinkel wrote:

LGTM.

If this test case proves too fragile in the future, we might replace it with an MIR test.

address spacing issues.

Added option to specify cutoff for CR bit definition search with default of 100.

Update comments and renamed new ppc ISD nodes.

Address review comments.

@delcypher I am having a problem with my internal build after this change have gone in. Can you please help to determine what went wrong? Previously I was able to do the following:

$ cmake -G Ninja -DCMAKE_BUILD_TYPE=Release -DLLVM_ENABLE_ASSERTIONS=On '-DLLVM_ENABLE_PROJECTS=clang;llvm;compiler-rt;clang-tools-extra;openmp;lld' -DCMAKE_C_COMPILER=/home/llvm/gcc/7.3.0/bin/gcc -DCMAKE_CXX_COMPILER=/home/llvm/gcc/7.3.0/bin/g++ -DLLVM_BINUTILS_INCDIR=/usr/include $LLVM_SRC
$ ninja -j 20
$ ninja check-libomptarget check-openmp

PR39124 has a reduced IR that seem to be the same issue this fixes. However I think Chri's suggested fix in PR28102 to remove the MBB references from the jump tables is a better way to go. Can you please give that a try?

Any other concerns about committing this patch?

Warning message updated.

For PPC this needs to be done for both tsan and msan.

This is causing the buildbot failures for powerpc: http://lab.llvm.org:8011/builders/clang-ppc64le-linux

In rL342657#289221, @calixte wrote:

According to the logs, there is an issue in the # of Runs for each file which should be fixed in clearing directory build/projects/compiler-rt/test/profile/Profile-x86_64/Output/ on build machines.
@lei how can I do that ?

This has broken ALL of the following build bots and they have been broken for more than 3 days. Please fix this or pull the changeset:

• http://lab.llvm.org:8011/builders/clang-ppc64le-linux
• http://lab.llvm.org:8011/builders/clang-ppc64le-linux-lnt
• http://lab.llvm.org:8011/builders/clang-ppc64be-linux
• http://lab.llvm.org:8011/builders/clang-ppc64be-linux-lnt
• http://lab.llvm.org:8011/builders/clang-ppc64be-linux-multistage

LGTM

merged into D47552

I am not sure why this has changed since ISA 2.07. But I have verified the new encoding on pwr8/pwr9 and that the old encoding produces an illegal instruction.

LGTM. Please feel free to fix the nit comments before committing.

LGTM

The indentation in the td files are still wrong. You should take a look at the existing pattern indentations and follow ...

Added hadP9Vector to calling convention def for f128.
I also merged the patch D47573 for extload of const into here since it's a 2 line fix and the test case associated depended on this patch.

address review comments.

This patch has a dependency on: https://reviews.llvm.org/D47552