You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@spamassassin.apache.org by Justin Mason <jm...@jmason.org> on 2004/12/15 19:47:30 UTC

Storable and hyperthreading

OK, so on the spamd hang bugs, we have:

- a set of people reporting hangs predominantly (all?) when running spamd
  on hyperthreaded CPUs
- not all HT CPUs are acting up
- a hang traced into Storable::dclone() (thanks Dallas!)

so I think we may have run into a perl thread-safety bug, possibly
in Storable, possibly at a lower level, and running on HT cpus causes
this bug to manifest itself.

Another reason to get rid of our use of Storable, in my opinion.

--j.

Re: Storable and hyperthreading

Posted by Michael Parker <pa...@pobox.com>.
On Wed, Dec 15, 2004 at 11:05:31AM -0800, Justin Mason wrote:
> 
> well, I have a patch which radically changes Conf for 3.1.0, but it's held
> up at the moment.

Yeah, that is what I was thinking, so I hadn't really done anything
with it, hoping that would shake loose sometime soon :)

> in the short term though -- I don't think there's any need to use
> a serialization format at all; just copy the data a la
> 
>     %{$copy->{blah}} = %{$self->{blah}};
> 

Hmmm....does this provide a deep enough copy?

Michael

Re: Storable and hyperthreading

Posted by Theo Van Dinter <fe...@kluge.net>.
On Wed, Dec 15, 2004 at 11:05:31AM -0800, Justin Mason wrote:
> in the short term though -- I don't think there's any need to use
> a serialization format at all; just copy the data a la
> 
>     %{$copy->{blah}} = %{$self->{blah}};

The issue is that you want to deal with references properly.  The above
doesn't deference all the values, just the initial {blah} ...

-- 
Randomly Generated Tagline:
"Just 'cuz I don't have an IP doesn't mean I'm not on your network."
                                 - Mike Poor

Re: Storable and hyperthreading

Posted by Michael Parker <pa...@pobox.com>.
On Wed, Dec 15, 2004 at 10:47:30AM -0800, Justin Mason wrote:
> 
> Another reason to get rid of our use of Storable, in my opinion.
> 

I was contemplating replacing Storable with YAML.

However, do we have a possible solution that does away with the need
all together?  If so, then it might not be worth the effort.

Michael