You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@tuscany.apache.org by Luciano Resende <lu...@gmail.com> on 2010/10/15 23:17:09 UTC
Reverting changes from r931936 to re-enable Eclipse Compiler
I did a quick experiment to check the build time difference between
the current default compiler compared to the eclipse one we used to
use before changes from r931936. Based on the results below, I'd like
to go back to use the eclipse compiler.
Regular compiler
[INFO] ------------------------------------------------------------------------
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESSFUL
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 22 minutes 5 seconds
[INFO] Finished at: Fri Oct 15 11:49:02 PDT 2010
[INFO] Final Memory: 282M/507M
[INFO] ------------------------------------------------------------------------
Reverting the changes from r931936 to enable eclipse compiler
[INFO] ------------------------------------------------------------------------
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESSFUL
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 9 minutes 17 seconds
[INFO] Finished at: Fri Oct 15 14:12:29 PDT 2010
[INFO] Final Memory: 289M/576M
[INFO] ------------------------------------------------------------------------
Thoughts ?
--
Luciano Resende
http://people.apache.org/~lresende
http://twitter.com/lresende1975
http://lresende.blogspot.com/
Re: Reverting changes from r931936 to re-enable Eclipse Compiler
Posted by ant elder <an...@gmail.com>.
On Fri, Oct 15, 2010 at 10:17 PM, Luciano Resende <lu...@gmail.com> wrote:
> I did a quick experiment to check the build time difference between
> the current default compiler compared to the eclipse one we used to
> use before changes from r931936. Based on the results below, I'd like
> to go back to use the eclipse compiler.
>
> Regular compiler
>
> [INFO] ------------------------------------------------------------------------
> [INFO] ------------------------------------------------------------------------
> [INFO] BUILD SUCCESSFUL
> [INFO] ------------------------------------------------------------------------
> [INFO] Total time: 22 minutes 5 seconds
> [INFO] Finished at: Fri Oct 15 11:49:02 PDT 2010
> [INFO] Final Memory: 282M/507M
> [INFO] ------------------------------------------------------------------------
>
>
> Reverting the changes from r931936 to enable eclipse compiler
>
> [INFO] ------------------------------------------------------------------------
> [INFO] ------------------------------------------------------------------------
> [INFO] BUILD SUCCESSFUL
> [INFO] ------------------------------------------------------------------------
> [INFO] Total time: 9 minutes 17 seconds
> [INFO] Finished at: Fri Oct 15 14:12:29 PDT 2010
> [INFO] Final Memory: 289M/576M
> [INFO] ------------------------------------------------------------------------
>
> Thoughts ?
>
If it really does give significant speed improvements that would be
great and we should look at it, but are you sure that difference
wasn't just due to the local environment? Eg if i run two builds the
second is always much faster, due to disk caching or something like
that i guess. When we we went back to the regular compiler i did check
if for any speed changes but didn't see any significant change either
on my local machine or on Hudson. Theres also the build instability
problem of the Tuscany compiler which was why we took it out. How
about for now adding a separate build profile that uses it so we can
test it out without breaking anything.
...ant