[PATCH] D83206: [PATCH] [ARM] Add Cortex-A78 and Cortex-X1 Support for Clang and LLVM

[Dave Green via Phabricator via cfe-commits]

dmgreen added a comment.

I would expect this to be very similar to https://reviews.llvm.org/rG8bf99f1e6f0f9b426d6060361ea6d9d47c1868d1, but some parts seems to be missing. Can you make sure that everything is included and in a sensible order.



================
Comment at: llvm/include/llvm/Support/AArch64TargetParser.def:131
+AARCH64_CPU_NAME("cortex-a78", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+                 (AArch64::AEK_RAS | AArch64::AEK_DOTPROD | AArch64::AEK_RCPC | AArch64::AEK_SSBS))
+AARCH64_CPU_NAME("cortex-x1", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
----------------
This no longer has FP16?

AEK_RAS I believe should be included in ARMV8_2A,


================
Comment at: llvm/include/llvm/Support/ARMTargetParser.def:301
+             (ARM::AEK_RAS | ARM::AEK_DOTPROD))
+ARM_CPU_NAME("cortex-a78",ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+             (ARM::AEK_RAS | ARM::AEK_DOTPROD))
----------------
All these can go in a better order, please. A78 can go next to A77. Same everywhere else.


================
Comment at: llvm/lib/Target/AArch64/AArch64Subtarget.cpp:196
+  case CortexX1:
+  case CortexA78:
+    PrefFunctionLogAlignment = 4;
----------------
These can go with the other CortexAXX cpu's, which seem to set the same PrefFunctionLogAlignment. Same for the ARM equivalent.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D83206/new/

https://reviews.llvm.org/D83206