<div dir="ltr"><div>Hello.</div><div><br></div><div>My backend supports only i32 stores. Given the following IR:</div><div><br></div><div> %2 = alloca [2 x i8], align 1<br><br> %4 = getelementptr inbounds [2 x i8], [2 x i8]* %2, i64 0, i64 0<br> store i8 1, i8* %4, align 1<br> %5 = getelementptr inbounds [2 x i8], [2 x i8]* %2, i64 0, i64 1<br> store i8 2, i8* %5, align 1</div><div><br></div><div>is
it possible to convince LLVM codegen to first load 4 bytes, then blend
it with the value being stored using "or" and then store 4 bytes back?<br></div><div><br></div><div>Or maybe it should be performed on the IR level?</div><div><br></div><div>Thanks in advance.</div></div>