[PATCH] D90001: [llvm/DebugInfo] Emit DW_OP_implicit_value when tuning for LLDB

[Med Ismail Bennani via Phabricator via llvm-commits]

mib added a comment.

Also it seems that the HWASAN pre-check test is also failing in D89844 <https://reviews.llvm.org/D89844> (https://buildkite.com/llvm-project/premerge-checks/builds/14519) ... I also tested on Linux and the test passes so I guess this is an issue with the bot.

  $ /usr/bin/python3.7 /Users/docker_user/Developer/open-source/llvm.org/build/Modular/./bin/llvm-lit -svv --filter=TestCases/sizes.cpp -- /Users/docker_user/Developer/open-source/llvm.org/build/Modular/projects/compiler-rt/test/hwasan/X86_64
  Testing Time: 5.52s
    Excluded: 50
    Passed  :  1
  
  $ uname -s 
  Linux



================
Comment at: llvm/lib/CodeGen/AsmPrinter/DwarfDebug.cpp:2483-2484
   } else if (Value.isConstantFP()) {
-    if (AP.getDwarfVersion() >= 4 && AP.getDwarfDebug()->tuneForGDB()) {
+    if (AP.getDwarfVersion() >= 4 && (AP.getDwarfDebug()->tuneForGDB() ||
+                                      AP.getDwarfDebug()->tuneForLLDB())) {
       DwarfExpr.addConstantFP(Value.getConstantFP()->getValueAPF(), AP);
----------------
JDevlieghere wrote:
> 
I'm not 100% sure `GDB || LLDB <-> !SCE`, this is why I put it this way ... What if the `-debugger-tune` option is not set all ?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D90001/new/

https://reviews.llvm.org/D90001