[SelectionDAG][VP] Fix VP memory ops' EVL/Mask indices

[frasercrmck]

Summary

This patch corrects the auto-generated EVL and Mask index positions of
the VP_LOAD/VP_STORE/VP_GATHER/VP_SCATTER nodes.

[frasercrmck]

I'm not sure there's currently a way of testing this. I found it while working on some other VP patches.

[simoll]

Apart from that LGTM. We don't currently test the SDNode layer directly.. only indirectly through isel

llvm/include/llvm/IR/VPIntrinsics.def

213–214

Could you add a comment here that describes what the SDNode operands are? Same for the others below.

[frasercrmck]

• add SDNode operand comments

[frasercrmck]

llvm/include/llvm/IR/VPIntrinsics.def

213–214

Aye good idea - done.

llvm/include/llvm/IR/VPIntrinsics.def

 
 #undef HELPER_REGISTER_BINARY_FP_VP
 
 ///// } Floating-Point Arithmetic
 
 ///// Memory Operations {
 // llvm.vp.store(ptr,val,mask,vlen)
 BEGIN_REGISTER_VP_INTRINSIC(vp_store, 2, 3)
-BEGIN_REGISTER_VP_SDNODE(VP_STORE, 0, vp_store, 3, 4)
+// chain = VP_STORE chain,val,base,offset,mask,evl
+BEGIN_REGISTER_VP_SDNODE(VP_STORE, 0, vp_store, 4, 5)
 HANDLE_VP_TO_OPC(Store)
 HANDLE_VP_TO_INTRIN(masked_store)
 HANDLE_VP_IS_MEMOP(vp_store, 1, 0)
 END_REGISTER_VP(vp_store, VP_STORE)
 
 // llvm.vp.scatter(ptr,val,mask,vlen)
-BEGIN_REGISTER_VP(vp_scatter, 2, 3, VP_SCATTER, 0)
+BEGIN_REGISTER_VP_INTRINSIC(vp_scatter, 2, 3)
+// chain = VP_SCATTER chain,val,base,indices,scale,mask,evl
+BEGIN_REGISTER_VP_SDNODE(VP_SCATTER, -1, vp_scatter, 5, 6)
 HANDLE_VP_TO_INTRIN(masked_scatter)
 HANDLE_VP_IS_MEMOP(vp_scatter, 1, 0)
 END_REGISTER_VP(vp_scatter, VP_SCATTER)
 
 // llvm.vp.load(ptr,mask,vlen)
 BEGIN_REGISTER_VP_INTRINSIC(vp_load, 1, 2)
-BEGIN_REGISTER_VP_SDNODE(VP_LOAD, -1, vp_load, 2, 3)
+// val,chain = VP_LOAD chain,base,offset,mask,evl
+BEGIN_REGISTER_VP_SDNODE(VP_LOAD, -1, vp_load, 3, 4)
 HANDLE_VP_TO_OPC(Load)
 HANDLE_VP_TO_INTRIN(masked_load)
 HANDLE_VP_IS_MEMOP(vp_load, 0, None)
 END_REGISTER_VP(vp_load, VP_LOAD)
 
 // llvm.vp.gather(ptr,mask,vlen)
-BEGIN_REGISTER_VP(vp_gather, 1, 2, VP_GATHER, -1)
+BEGIN_REGISTER_VP_INTRINSIC(vp_gather, 1, 2)
+// val,chain = VP_GATHER chain,base,indices,scale,mask,evl
+BEGIN_REGISTER_VP_SDNODE(VP_GATHER, -1, vp_gather, 4, 5)
 HANDLE_VP_TO_INTRIN(masked_gather)
 HANDLE_VP_IS_MEMOP(vp_gather, 0, None)
 END_REGISTER_VP(vp_gather, VP_GATHER)
 
 ///// } Memory Operations
 
 ///// Reductions {