[llvm] [AMDGPU] Fix high vgpr printing with true16 (PR #160209)

Stanislav Mekhanoshin via llvm-commits llvm-commits at lists.llvm.org
Tue Sep 23 01:22:59 PDT 2025

Previous message: [llvm] [DAG] foldShiftToAvg - recognize sub(x, xor(y, -1)) >> 1 as avgceil[su] (#147946) (PR #156239)
Next message: [llvm] [AMDGPU] Fix high vgpr printing with true16 (PR #160209)
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

rampitec wrote:

Remember, in June we had a discussion if MSB lowering works correctly with t16? Well, it does not. This is the fix.

https://github.com/llvm/llvm-project/pull/160209

Previous message: [llvm] [DAG] foldShiftToAvg - recognize sub(x, xor(y, -1)) >> 1 as avgceil[su] (#147946) (PR #156239)
Next message: [llvm] [AMDGPU] Fix high vgpr printing with true16 (PR #160209)
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

More information about the llvm-commits mailing list