[llvm] [RISCV] Add a pass to remove ADDI by reassociating to fold into load/store address. (PR #127151)

[Craig Topper via llvm-commits]

================
@@ -1167,8 +1167,10 @@ declare void @f(ptr)
 define i32 @crash() {
 ; RV32I-LABEL: crash:
 ; RV32I:       # %bb.0: # %entry
-; RV32I-NEXT:    lui a0, %hi(g+401)
-; RV32I-NEXT:    lbu a0, %lo(g+401)(a0)
+; RV32I-NEXT:    lui a0, %hi(g)
+; RV32I-NEXT:    addi a0, a0, %lo(g)
+; RV32I-NEXT:    add a0, a0, zero
+; RV32I-NEXT:    lbu a0, 401(a0)
----------------
topperc wrote:

> I just want to note that these legitimately are regressions. We're folding the constant offset into the using load, but before we'd folded it with the lo part of the symbol, and then that into the load. After your change, we still end up needing the add for that (at least, before linker relaxation). It seems like we should be able to undo this, are we missing a generalization somewhere else?

I had this fixed previously by looking for the COPY that replaced the ADDI in RISCVMergeBaseOffset. The ReplaceRegWith broke that so I reverted the change I had made to RISCVMergeBaseOffset. Note this a slightly weird.

Note this is a slightly weird case. There's a small constant offset, but it lives in another basic block instead of being constant folded into the GEP. This prevented it from being visible to SelectionDAG where it would have been selected into the load offset already.

> Also, separate issue, but why the heck do we still have the ADD a0, a0, zero here?

Because the ADDI that got deleted was addi a0, zero, 1. The replaceRegWith moved the X0 to the user. There's no ADD constant folder later in the pipeline.

Constant folding the ADD in the FoldMemOffset pass would fix both issues. Alternatively, I might be able to ignore ADDIs with X0 base in FoldMemOffset. They don't show up in the motivating pattern.

https://github.com/llvm/llvm-project/pull/127151