[PATCH] Set a debugger "target" to guide DWARF choices

[Paul Robinson]

================
Comment at: lib/CodeGen/AsmPrinter/DwarfDebug.cpp:211-219
@@ -211,1 +210,11 @@
 
+  // Make sure we know our "debugger tuning."
+  if (Asm->TM.Options.DebuggerTuning == DebuggerKind::Default) {
+    if (IsDarwin)
+      Asm->TM.Options.DebuggerTuning = DebuggerKind::LLDB;
+    else if (Triple(A->getTargetTriple()).isPS4CPU())
+      Asm->TM.Options.DebuggerTuning = DebuggerKind::SCE;
+    else
+      Asm->TM.Options.DebuggerTuning = DebuggerKind::GDB;
+  }
+
----------------
echristo wrote:
> I might have lost this part of the discussion but I thought the default was "no tuning"?
> 
> At least not in the backend.
That did come up, and it was not clear what it should mean, so I fell back on preserving the prior behavior, which effectively defaults to GDB except for Darwin and PS4.

So, what would no-tuning mean?  "No non-standard things" seems reasonably clear, but should it include all the optional-but-standard things that LLVM knows how to do?  Which I guess would be pubnames + aranges + type units (are there more?).
It would smoke out all the tests that have been relying on the triple to get the right effect, though, which could be viewed as a Good Thing.

Presumably you'd want Clang to tell LLVM what tuning to use, though?  Having Clang default to "no tuning" seems user-hostile.


================
Comment at: lib/CodeGen/AsmPrinter/DwarfDebug.cpp:224
@@ -214,3 +223,3 @@
   if (DwarfAccelTables == Default)
     HasDwarfAccelTables = IsDarwin;
   else
----------------
Should this become tuneDebugForLLDB()?

================
Comment at: lib/CodeGen/AsmPrinter/DwarfDebug.cpp:234
@@ -224,3 +233,3 @@
   if (DwarfPubSections == Default)
-    HasDwarfPubSections = !IsDarwin && !IsPS4;
+    HasDwarfPubSections = !IsDarwin && !tuneDebugForSCE();
   else
----------------
echristo wrote:
> This should be simplified to tuneDebugForGDB().
Okay.  Not being Darwin-aware or an LLDB user, I was a bit reluctant to convert all the Darwin decision-making right away.

================
Comment at: lib/CodeGen/AsmPrinter/DwarfDebug.cpp:244
@@ -236,1 +243,3 @@
+  // and GDB always wants the GNU opcode.
+  UseGNUTLSOpcode = DwarfVersion < 3 || tuneDebugForGDB();
 
----------------
echristo wrote:
> I don't think this is a tuning option, and that's the problem. This is a correctness issue in that it just won't work rather than being inefficient or superfluous.
Yeah, trying to work around a GDB bug.  GDB can't tolerate the standard opcode; SCE can't tolerate the GNU opcode; I don't know what LLDB thinks, but Adrian liked having the standard opcode.
Do you have an alternate suggestion?  Leaving it target-based seems worse, except for the historical consistency.


================
Comment at: test/DebugInfo/X86/dwarf-public-names.ll:7-10
@@ -6,1 +6,6 @@
 ; RUN: llvm-dwarfdump -debug-dump=pubnames %t.o | FileCheck --check-prefix=NOPUB %s
+; RUN: llc -mtriple=x86_64-pc-linux-gnu -debugger-tune=sce -filetype=obj -o %t.o < %s
+; RUN: llvm-dwarfdump -debug-dump=pubnames %t.o | FileCheck --check-prefix=NOPUB %s
+; RUN: llc -mtriple=x86_64-scei-ps4 -debugger-tune=gdb -filetype=obj -o %t.o < %s
+; RUN: llvm-dwarfdump -debug-dump=pubnames %t.o | FileCheck --check-prefix=LINUX %s
+
----------------
echristo wrote:
> I think I'd like this split out into a different test file.
Ironically, this is one of the tests that would break if LLVM defaulted to no-tuning; the triples would need to be replaced by tuning options.  And if tuning isn't premised on the triple, then the non-default triple/tuning combos wouldn't need testing, and the new RUN lines would just go away.

http://reviews.llvm.org/D8506

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/