[PATCH] D24661: More processors support under AARch64 state for auto detection.

[Renato Golin via llvm-commits]

rengolin requested changes to this revision.
rengolin added a comment.
This revision now requires changes to proceed.

Oh, I just noticed again, sorry (swapping a bit). This code doesn't even compile.

You have three patches here:

1. AArch64 changes, adding XGene1: You need to add the model and descriptions on all related table-gen files. You're defining it but not using it.
2. ARM changes, adding XGene1: You need to define in the same way you did for AArch64, because the usage can't find the declaration, which is in the AArch64.td.
3. Clang changes: the tests rely on the targets being there, so need to be committed *after* the LLVM patch.

For 1 and 2, you'll need two separate patches, with similar changes and appropriate unit tests.

For 3, you'll need both ARM and AArch64 tests (-target armv8a and -target aarch64), and make sure to mention the patch *depends* on the LLVM ones.

Sorry for the confusion, and please, before submitting the three patches, make sure that they build and pass "check-all" individually, as well as altogether.

cheers,
--renato


https://reviews.llvm.org/D24661