[PATCH] D104622: [AMDGPU] Add 224-bit vector types and link 192-bit types to MVTs

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/SIRegisterInfo.td:739-746
+  // This occurs because SGPR_160 and SReg_160 classes are equivalent in size
+  // meaning their enumeration order is dependent on alphanumeric ordering of
+  // their names.  The superclass for inherence is the last one in topological
+  // order (i.e. enumeration order), hence SReg_160 is selected.
+  // Potential workarounds involve renaming SGPR_160, adding another class
+  // which is ordered last and hence used for inheritance, or adding more
+  // registers to SReg_160 to cause it to be moved earlier in the superclass
----------------
foad wrote:
> Can't we get rid of the dual SGPR_*/SReg_* classes for all sizes > 64? I thought the only difference between them was that Sreg_* added non-SGPR registers like EXEC and M0, but for the larger sizes, if there are no non-SGPR registers to add, what is the point of having two different classes?
No, because the TTMP_* registers also form the large tuples


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D104622/new/

https://reviews.llvm.org/D104622