[Lldb-commits] [lldb] [lldb] Support arbitrary precision integer registers (PR #166363)

[Matej Košík via lldb-commits]

================
@@ -46,7 +46,8 @@ class RegisterValue {
     eTypeUInt16,
     eTypeUInt32,
     eTypeUInt64,
-    eTypeUInt128,
+    eTypeUIntN, /// < This value is used when the (integer) register is larger
+                /// than 64-bits.
----------------
sedymrak wrote:

> I think the comment is incorrect because you've added test cases that use UintN for < 64-bit values. I may be mistaken about that.

It is true that I have added tests that are related to integers with < 64-bit values.
However, the type of the corresponding `RegisterValue` object is not equal to `eTypeUIntN` in those cases.

To make things more clear, I have pushed a commit that updates the existing tests so that the type of the `RegisterValue` object is checked. You can see that, e.g., the `RegisterValue` object corresponding to a 7-bit integer value has type `eTypeUInt8`.

https://github.com/llvm/llvm-project/pull/166363