[Lldb-commits] [lldb] r190812 - Fixes symbol resolution for a function with a tail call because the PC

Thirumurthi, Ashok ashok.thirumurthi at intel.com
Fri Sep 20 08:54:34 PDT 2013


Thanks for the fix, Greg!

My commit in r190812 was too naïve to handle the case where debug libraries were available.  In this case, for frame 3, sc.function gets resolved but sc.symbol doesn't until the pc is slid.  The attached patch introduces eSymbolContextTailCall so that ResolveSymbolContextForAddress can get this right.  At the same time, callers can decide if they want this scope of symbol resolution,

- Ashok

-----Original Message-----
From: carpeddiem at gmail.com [mailto:carpeddiem at gmail.com] On Behalf Of Ed Maste
Sent: Wednesday, September 18, 2013 5:43 PM
To: Greg Clayton
Cc: Thirumurthi, Ashok; lldb-commits at cs.uiuc.edu
Subject: Re: [Lldb-commits] [lldb] r190812 - Fixes symbol resolution for a function with a tail call because the PC

On 18 September 2013 16:07, Greg Clayton <gclayton at apple.com> wrote:
> There we some logic errors in the Module::ResolveSymbolContextForAddress fix which I just fixed with:
>
> % svn commit
> Sending        source/Core/Module.cpp
> Transmitting file data .
> Committed revision 190955.
>
> Let me know if this fixes anything for you?

It looks like the test still fails for me on FreeBSD after this change.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: inferior-assert-3.patch
Type: application/octet-stream
Size: 6938 bytes
Desc: inferior-assert-3.patch
URL: <http://lists.llvm.org/pipermail/lldb-commits/attachments/20130920/7b263437/attachment.obj>


More information about the lldb-commits mailing list