[llvm] [AMDGPU] Fix an issue that wrong index is used in calculation of byte provider when the op is extract_vector_elt (PR #91697)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,126 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc -verify-machineinstrs -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 < %s | FileCheck -check-prefix=GFX9 %s
+; RUN: llc -verify-machineinstrs -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1030 < %s | FileCheck -check-prefix=GFX10 %s
+; RUN: llc -verify-machineinstrs -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 < %s | FileCheck -check-prefix=GFX11 %s
+
+define protected amdgpu_kernel void @test(ptr addrspace(1) %srcA, ptr addrspace(1) %srcB, ptr addrspace(1) %dst) {
----------------
arsenm wrote:

```suggestion
define amdgpu_kernel void @test(ptr addrspace(1) %srcA, ptr addrspace(1) %srcB, ptr addrspace(1) %dst) {
```

https://github.com/llvm/llvm-project/pull/91697