[Lldb-commits] [lldb] [lldb][RISCV] Fix x8 register aliasing for gdb-remote targets (PR #209070)

Sam Elliott via lldb-commits lldb-commits at lists.llvm.org
Sun Jul 12 22:07:55 PDT 2026


================
@@ -780,9 +780,9 @@ def haltReason(self):
         self.match("register read x5", ["t0 = 0x0807060504030201"])
         self.match("register read x6", ["t1 = 0x0807060504030201"])
         self.match("register read x7", ["t2 = 0x0807060504030201"])
-        # Register x8 is probably not working because it has two aliases fp, s0
-        # See issue #127900
-        # self.match("register read x8", ["fp = 0x0807060504030201"])
+        self.match("register read fp", ["s0 = 0x0807060504030201"])
----------------
lenary wrote:

I think techically `s0` is more accurate, as the register is always the lowest-numbered (callee-)saved register, but is only the FP if you actually have one.

That said, It might be worth adding as a release note, if the LLDB maintainers think that's important.

https://github.com/llvm/llvm-project/pull/209070


More information about the lldb-commits mailing list