harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Tim Ellison <t.p.elli...@gmail.com>
Subject Re: [performance] quick sort is 4x slower on Harmony
Date Fri, 04 Jan 2008 10:36:46 GMT
Andrew Zhang wrote:
> On Jan 4, 2008 12:07 PM, Endre StĂžlsvik <Endre@stolsvik.com> wrote:
>>> The main reason is that I didn't want to mention all copyright and
>> license
>>> issues here. It's GPL licensed code, and db4objects owns the copyright.
>> I
>>> don't think everyone here is allowed to read GPL code at will.
>> What a sick world.
>> Too bad you posted it two posts later, then - hopefully those folks cut
>> their eyes out right away..
> If you are/have been working for some big companies, you should know the
> policy. If I posted the code here, maybe I got blame:
> How can you post a GPL-licensed code here?

There is an assumption that code you send to the dev list is a 
contribution, and thereby a possibility that it may end up in a test 
case for example.  So we ask that you only contribute your own code, or 
make it clear in the posting that the code is not a contribution.

See the monthly reminders sent to this list [1].

>>> The code on the blog entry is indeed slower on Harmony. Harmony is
>> young,
>>> and moving fast. Do we need to hide problem if there's a problem? I
>> think
>>> the archive of this mailing list can be also searched by many search
>>> engines. Shall we also forbid to use any 'bad' words here?
>> No - you must not have read my email.
>>> But if you think the blog entry is not objective, it's another story.
>> The
>>> code in the blog is indeed slower and it's deterministic.
>> In one month time it will probably (hopefully!) not be accurate, but
>> your blog entry will still be there, without any context - people will
>> find it, and wonder - "say, Harmony must suck really hard..". Which
>> probably is the sane thing to think, given that there is no context to
>> view the statement in.
> I just wondered to add Harmony revision to the blog post, but since we're so
> concerned about the "slow" and "immature" of Harmony, I deleted the evil
> post.

I don't think you need to delete the blog entry, or that it is "evil". 
We are as open about our problems as we are about our accomplishments.

>> I probably should stop repeating myself now.
> I would not like to discuss this topic either. Sorry that my reproducible
> test doesn't help DRVLM guys to tune the performance, but slanders Harmony.
> We have lots of other more important things to do than finding out the
> reason/purpose of a blog post of a Harmony contributor.

I see lots of evidence that the VM/JIT folk are interested in 
world-class performance.  I'm sure they would appreciate seeing a case 
where there is room for improvement.



View raw message