[llvm] [NVPTX] fix debug register encoding of special %Depot register (PR #111596)

[via llvm-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-nvptx

Author: William G Hatch (willghatch)

<details>
<summary>Changes</summary>

cuda-gdb doesn't seem to be able to read the `%Depot` register, but because we always copy it to `%SP` in lowering, simply switching to use it fixes the problem.

---
Full diff: https://github.com/llvm/llvm-project/pull/111596.diff


1 Files Affected:

- (modified) llvm/lib/Target/NVPTX/NVPTXRegisterInfo.cpp (+5) 


``````````diff
diff --git a/llvm/lib/Target/NVPTX/NVPTXRegisterInfo.cpp b/llvm/lib/Target/NVPTX/NVPTXRegisterInfo.cpp
index c30372fed1832d..dc2584645b314c 100644
--- a/llvm/lib/Target/NVPTX/NVPTXRegisterInfo.cpp
+++ b/llvm/lib/Target/NVPTX/NVPTXRegisterInfo.cpp
@@ -176,6 +176,11 @@ void NVPTXRegisterInfo::addToDebugRegisterMap(
 int64_t NVPTXRegisterInfo::getDwarfRegNum(MCRegister RegNum, bool isEH) const {
   if (Register::isPhysicalRegister(RegNum)) {
     std::string name = NVPTXInstPrinter::getRegisterName(RegNum.id());
+    // In NVPTXFrameLowering.cpp, we do arrange for %Depot to be accessible from
+    // %SP. Using the %Depot register doesn't provide any debug info in
+    // cuda-gdb, but switching it to %SP does.
+    if (RegNum.id() == NVPTX::VRDepot)
+      name = "%SP";
     return encodeRegisterForDwarf(name);
   }
   uint64_t lookup = debugRegisterMap.lookup(RegNum.id());

``````````

</details>


https://github.com/llvm/llvm-project/pull/111596