Reverse excecution and Simics testing: short report

Reverse excecution and Simics testing: short report

[Jakob Engblom]

Hi!

We have done some preliminary "smoke testing" at Virtutech to see how
gdb-main-branch revexec works with our old proven revexec engine. The results
are quite encouraging.

Simple testing using gdb-remote to talk to Simics, with Power Architecture and
ARM targets appears to work. What worked in our specially-patched gdb-6.8
version now works with gdb-trunk. We can stop and reverse and step etc.
Essentially, this indicates that the reverse execution core support is OK. 

We also tested using MI commands to drive reverse execution. We used gdb-trunk
plus our proposed MI commands, and the only problem we ran into seems to be the
same reverse-finish (or preferably, uncall) operation that Teawater and Marc
have been reporting. Once again, the system seems to be working fine overall. 

Best regards,

/jakob

_______________________________________________________

Jakob Engblom, PhD, Technical Marketing Manager

Virtutech                   Direct: +46 8 690 07 47   
Drottningholmsvägen 22      Mobile: +46 709 242 646  
11243 Stockholm             Web:    www.virtutech.com 
Sweden
________________________________________________________
  

Re: Reverse excecution and Simics testing: short report

[teawater]

That is great!  Thanks Jakob.

Hui

On Fri, Feb 6, 2009 at 20:12, Jakob Engblom <jakob@virtutech.com> wrote:
> Hi!
>
> We have done some preliminary "smoke testing" at Virtutech to see how
> gdb-main-branch revexec works with our old proven revexec engine. The results
> are quite encouraging.
>
> Simple testing using gdb-remote to talk to Simics, with Power Architecture and
> ARM targets appears to work. What worked in our specially-patched gdb-6.8
> version now works with gdb-trunk. We can stop and reverse and step etc.
> Essentially, this indicates that the reverse execution core support is OK.
>
> We also tested using MI commands to drive reverse execution. We used gdb-trunk
> plus our proposed MI commands, and the only problem we ran into seems to be the
> same reverse-finish (or preferably, uncall) operation that Teawater and Marc
> have been reporting. Once again, the system seems to be working fine overall.
>
> Best regards,
>
> /jakob
>
> _______________________________________________________
>
> Jakob Engblom, PhD, Technical Marketing Manager
>
> Virtutech                   Direct: +46 8 690 07 47
> Drottningholmsvägen 22      Mobile: +46 709 242 646
> 11243 Stockholm             Web:    www.virtutech.com
> Sweden
> ________________________________________________________
>
>
>
>