[lldb-dev] [Bug 45920] lldb wrongly stopped at a statement within a nested for statement by si (step instruction)

[via lldb-dev]

https://bugs.llvm.org/show_bug.cgi?id=45920

Jeremy Morse <jeremy.morse.llvm at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Blocks|                            |38768
            Version|unspecified                 |trunk
          Component|All Bugs                    |LLVM Codegen
           Assignee|lldb-dev at lists.llvm.org     |unassignedclangbugs at nondot.
                   |                            |org
             Status|NEW                         |CONFIRMED
                 CC|                            |dblaikie at gmail.com,
                   |                            |jeremy.morse.llvm at gmail.com
                   |                            |, neeilans at live.com,
                   |                            |paul.robinson at am.sony.com,
                   |                            |richard-llvm at metafoo.co.uk,
                   |                            |vsk at apple.com
            Product|lldb                        |clang

--- Comment #1 from Jeremy Morse <jeremy.morse.llvm at gmail.com> ---
Thanks for the bug report -- could you include the command line used to compile
the source too please. Assuming -O0, this looks very similar to bug 45676 and
its related bug 19864 -- clang is labelling the exit of the inner loop with
what we would consider the wrong location:

> for.end:                                          ; preds = %for.cond1
>   br label %for.inc7, !dbg !51

> 51 = !DILocation(line: 8, column: 16, scope: !29)

Where line 8 is the line with "return 1;". I can replicate this with gdb, so
switching component to clang.


Referenced Bugs:

https://bugs.llvm.org/show_bug.cgi?id=38768
[Bug 38768] [meta][DebugInfo] Umbrella bug for poor debug experiences
-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/lldb-dev/attachments/20200514/369de4de/attachment.html>