[llvm] [AMDGPU] Add alignment attr & propagate alignment through make.buffer.rsrc inst (PR #145278)

[Matt Arsenault via llvm-commits]

================
@@ -1381,7 +1399,7 @@ static bool runImpl(Module &M, AnalysisGetter &AG, TargetMachine &TM,
        &AAAMDMaxNumWorkgroups::ID, &AAAMDWavesPerEU::ID, &AAAMDGPUNoAGPR::ID,
        &AACallEdges::ID, &AAPointerInfo::ID, &AAPotentialConstantValues::ID,
        &AAUnderlyingObjects::ID, &AAAddressSpace::ID, &AAIndirectCallInfo::ID,
-       &AAInstanceInfo::ID});
+       &AAInstanceInfo::ID, &AAAlign::ID});
----------------
arsenm wrote:

Adding this to AMDGPUAttributor should be a separate patch, you can do this in the base attributor first 

https://github.com/llvm/llvm-project/pull/145278