[llvm] [AMDGPU] Folding imm offset in more cases for scratch access (PR #70634)

Tue Nov 7 07:02:23 PST 2023

================
@@ -1860,7 +1905,7 @@ bool AMDGPUDAGToDAGISel::SelectScratchSVAddr(SDNode *N, SDValue Addr,
           CurDAG->getTargetConstant(RemainderOffset, SDLoc(), MVT::i32));
         VAddr = SDValue(VMov, 0);
----------------
ruiling wrote:

If bit 31 was set, then we will *only* have problem if the hardware interprets it as signed value. I think this is not the case for now?

https://github.com/llvm/llvm-project/pull/70634
