Hi Staffan,
If hearsay is useful to you to support his change, I haven't seen or
heard of anybody using the SA's Bugspot on non-Java processes/cores, and
we haven't publicised that it was possible...
Regards
Kevin
On 16/04/12 13:23, Staffan Larsen wrote:
Here is an early webrev of the changes I'm working on.
http://cr.openjdk.java.net/~sla/remove-bugspot/webrev/
<http://cr.openjdk.java.net/%7Esla/remove-bugspot/webrev/>
This isn't a formal code review yet, but I'll be happy for any
feedback on the changes.
Thanks,
/Staffan
On 16 apr 2012, at 10:51, Krystal Mok wrote:
Hi Staffan,
Alright, then I'm ok with this :-)
Thanks,
Kris
On Mon, Apr 16, 2012 at 4:22 PM, Staffan Larsen
<[email protected] <mailto:[email protected]>> wrote:
Hi Krystal,
Those tools can all use HotSpotAgent instead without any loss of
functionality.
Regards,
/Staffan
On 16 apr 2012, at 10:21, Krystal Mok wrote:
Hi Staffan,
There are quite a few tools depending on BugSpotAgent for
attaching. At least all tools that inherit from
sun.jvm.hotspot.tools.Tool do.
Removing the BugSpot frontend shouldn't affect those tools, or
would it?
Regards,
Kris
On Sat, Apr 14, 2012 at 6:32 PM, Staffan Larsen
<[email protected] <mailto:[email protected]>>
wrote:
All,
I'd like to remove the BugSpot front end from SA in an
effort to clean up the SA code and reduce the number of
implementations. My understanding is that BugSpot was
intended as a more general native debugger not only focused
on the JVM, but that it was never fully implemented. I don't
think there is a need for a general-purpose debugger in SA.
The features that will change are:
* BugSpot graphical UI will be removed
* PStack and PMap will no longer be able to operate on
executables other than the JVM
* Some code only referenced from BugSpot will be removed
Does anyone know of a reason this is not a good idea? I
don't have the full history of SA nor all the use cases so I
would appreciate feedback.
Thanks,
/Staffan
