llvm dev - Apr 2014 - [LLVMdev] decompiler

On Thu, Apr 3, 2014 at 11:50 AM, Jevin Sweval <jevinsweval at gmail.com>
wrote:
> On Wed, Apr 2, 2014 at 1:57 AM, "C. Bergström" <cbergstrom at
pathscale.com> wrote:
>> Hi -
>>
>> Not sure if anyone else saw this or cares about a decompiler (not
personally
>> tested)
>> https://github.com/draperlaboratory/fracture
>>
>> I wonder if they have been in contact with anyone in the community in
>> getting this upstreamed. Does it look interesting to anyone else?
>> (thoughts/random comments/feedback)
>>
>
> I saw fracture but I was unable to figure out how to get even a small
> portion of x86 machine code decompiled to IR. That was several weeks
> ago and it looks like the project has progressed quite a bit.
>
> Other projects:
> https://github.com/libcpu/libcpu (tried this one a long time ago, also
> couldn't get any decompilation)
> http://dagger.repzret.org/ (they finally released some source! One of
> the devs presented at a LLVM meeting)
>
> I am definitely interested in seeing a production-quality decompiler!
>
> Cheers,
> Jevin
Very cool, I just tested out dagger with a small x86_64 test and it
worked! You can see my test and the decompiled IR here:
https://gist.github.com/jevinskie/9957869

Cheers,
Jevin

On 04/ 4/14 12:55 AM, Per Larsen wrote:
> For completeness, I want to mention that there is a commercial effort 
> (by SecondWrite LLC) to build a production-quality decompiler (also 
> called SecondWrite [1,2]) that targets the LLVM intermediate 
> representation. It is a very ambitious project and a high level of 
> compatibility can only be offered after substantial implementation and 
> research efforts have been made. I'm adding a SecondWrite developer, 
> Kapil Anand, to the loop in case you want to start a conversation 
> directly. As an independent researcher in compilation and security, I 
> can attest to the quality of the work and research that has gone into 
> the development of SecondWrite.
Is it open source? Do they plan to work with upstream? Commercial 
products are fun, but my original post was about a project that had 
pushed the source to github. I was hoping to have it get some attention 
or upstream to notice people are interested.

For completeness, I want to mention that there is a commercial effort (by
SecondWrite LLC) to build a production-quality decompiler (also called
SecondWrite [1,2]) that targets the LLVM intermediate representation. It is
a very ambitious project and a high level of compatibility can only be
offered after substantial implementation and research efforts have been
made. I'm adding a SecondWrite developer, Kapil Anand, to the loop in case
you want to start a conversation directly. As an independent researcher in
compilation and security, I can attest to the quality of the work and
research that has gone into the development of SecondWrite.

Cheers,
Per

[1] http://dl.acm.org/citation.cfm?doid=2465351.2465380
[2] http://dl.acm.org/citation.cfm?doid=2462156.2462165


On Thu, Apr 3, 2014 at 9:34 AM, Jevin Sweval <jevinsweval at gmail.com>
wrote:
> On Thu, Apr 3, 2014 at 11:50 AM, Jevin Sweval <jevinsweval at
gmail.com>
> wrote:
> > On Wed, Apr 2, 2014 at 1:57 AM, "C. Bergström"
<cbergstrom at pathscale.com>
> wrote:
> >> Hi -
> >>
> >> Not sure if anyone else saw this or cares about a decompiler (not
> personally
> >> tested)
> >> https://github.com/draperlaboratory/fracture
> >>
> >> I wonder if they have been in contact with anyone in the community
in
> >> getting this upstreamed. Does it look interesting to anyone else?
> >> (thoughts/random comments/feedback)
> >>
> >
> > I saw fracture but I was unable to figure out how to get even a small
> > portion of x86 machine code decompiled to IR. That was several weeks
> > ago and it looks like the project has progressed quite a bit.
> >
> > Other projects:
> > https://github.com/libcpu/libcpu (tried this one a long time ago, also
> > couldn't get any decompilation)
> > http://dagger.repzret.org/ (they finally released some source! One of
> > the devs presented at a LLVM meeting)
> >
> > I am definitely interested in seeing a production-quality decompiler!
> >
> > Cheers,
> > Jevin
>
> Very cool, I just tested out dagger with a small x86_64 test and it
> worked! You can see my test and the decompiled IR here:
> https://gist.github.com/jevinskie/9957869
>
> Cheers,
> Jevin
>
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>


-- 
http://www.ics.uci.edu/~perl/
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140403/c978e805/attachment.html>