[llvm-dev] [MCJIT] messy call stack debug on x64 code in VisualStudio

[Vivien Millet via llvm-dev]

Thank you David for forwarding my question.
There is no official support for PDB with MCJIT, I developed one myself as
part of a project and with the help of Zachary Turner, I will soon make it
public on my github.
(If the community is interested for an official support of a JIT-PDB
feature, I can discuss with the persons in charge for helping with its
development).

Anyway, to give a little more information on my issue :
- adding "nounwind" attribute to jitted functions solved the problem for
some functions but I still have issues on others (I still have to find
why.. and this is where I need help).
- it seems that the prolog register pushes "mess up" the stack walk because
when breaking just at the beginning of the prolog or after epilog, I can
see the callling function correctly under the current one (RBP problem ?).
- when stepping inside a function, the mess in the callstack diplay
evolves, like if visual studio was trying to walk the stack based on
something not static (RSP ?)
- I don't understand how the walk could be complicated because frame
pointers are preserved in the assembly(RBP is pushed)... except if RBP is
used for something else but that would be weird.

Regards,
Vivien

Le sam. 29 févr. 2020 à 23:29, David Blaikie <dblaikie at gmail.com> a écrit :

> +Lang Hames for JIT things
> +Reid Kleckner <rnk at google.com> for Windows things
>
> (in case they've got any thoughts on this issue - no guarantees though, I
> didn't even know there was any option to use PDBs through MCJIT & I know
> MCJIT is mostly unmaintained in favor of ORC JIT)
>
> On Sat, Feb 29, 2020 at 8:14 AM Vivien Millet via llvm-dev <
> llvm-dev at lists.llvm.org> wrote:
>
>> Hi,
>>
>> I'm using IR and MCJIT to compile a script language. I debug it with on
>> the fly generated .pdb files. During debugging, almost each time I step
>> into a function, I loose information about calling function inside the
>> visual studio callstack view or I have a bunch of pure addresses in the
>> callstack in between the current function and the calling function, for
>> example :
>>
>> MyJit.dll!MyCurrentFunction()
>> [0x1234567887654321]
>> [0x8765432112345678]
>> MyJit.dll!MyCallingFunction()
>> ...
>>
>> It looks like visual studio get lost while walking up stack.
>> Does anyone know where it could come from ?
>>
>> I have disabled all optimisations (among them is the omit-frame-pointer).
>>
>> I have seen this bug here : https://bugs.llvm.org/show_bug.cgi?id=24233 which
>> is quite similar but it is quite old now, and since the proposed patch has
>> been posted, the code in RuntimeDyldCOFFX86_64.h has changed and it is
>> difficult for me to know if it has really been fixed since or not.
>>
>> Could it be related to the way IR CreateAlloca are used to build local
>> variables ? Could it be related to missing informations inside the PDB ? (I
>> don't know if there is stack related information inside PDB files to ensure
>> good stack walking).
>>
>> Thanks.
>>
>> Vivien
>> _______________________________________________
>> LLVM Developers mailing list
>> llvm-dev at lists.llvm.org
>> https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20200229/2e7633bf/attachment.html>