[PATCH] D54882: [AMDGPU] Add sdwa support for ADD|SUB U64 decomposed Pseudos

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec added a comment.

Essentially this is a limited version of shrinking. So I have several questions:

1. Why not to run shrink pass before sdwa instead?
2. If we want to shrink individual instructions, should not we just add SIInstrInfo::shrink() interface and call it from SIShrinkInstructions as well?
3. A the very least checks needs to use SIInstrInfo::canShrink() instead of many and insufficient checks here. For instance this code does not check for valid register classes which SIInstrInfo::canShrink() does. We really need to stop cloning that code.
4. In general SIInstrInfo::canShrink() should be extended to handle OpSel (via SIInstrInfo::hasModifiers()).
5. SIInstrInfo::canShrink() shall call SIInstrInfo::hasVALU32BitEncoding() to check for presence of target instruction.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D54882/new/

https://reviews.llvm.org/D54882