Ben just landed the 3.20.x upgrade to Node, and I'm noticing a non-trivial
performance regression using Persistent<>. Before it was fast to cache
objects that would be passed to Function::Call as Persistent<Object>, but
now it seems to require a Local<Object>::New(Isolate*, Persistent<Object>)
before being sent. For quick calls out of cc to js it's adding ~10% to call
time. Is there a way around this regression that I'm not seeing?


On Sat, Jul 6, 2013 at 12:25 AM, Dan Carney <[email protected]> wrote:

>
>> Is that comment in error?  Put another way: if or when those functions
>> are removed, how are you supposed to get at the handle that the
>> Persistent points to?
>>
>
> Yes, it's an error in the comment.  Those functions will not be removed.
>
>
>> Another thing is Persistent<T>::ClearAndLeak().  Here is its definition:
>>
>>   V8_INLINE(T* ClearAndLeak());
>>
>> What purpose does it serve?  You can't construct a Handle<T> from a
>> raw T*, all constructors and New() class methods that take a T* are
>> private.
>>
>
> We use it internally to embed persistents in collections.  We were
> planning on having some API function for using it, but have recently
> decided against that plan.  It will be removed.
>
>  --
> --
> v8-users mailing list
> [email protected]
> http://groups.google.com/group/v8-users
> ---
> You received this message because you are subscribed to a topic in the
> Google Groups "v8-users" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/v8-users/6kSAbnUb-rQ/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> [email protected].
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>

-- 
-- 
v8-users mailing list
[email protected]
http://groups.google.com/group/v8-users
--- 
You received this message because you are subscribed to the Google Groups 
"v8-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to