|
Boost : |
From: Martin Wille (mw8329_at_[hidden])
Date: 2003-11-19 12:48:32
Peter Dimov wrote:
> Martin Wille wrote:
>
>>David Abrahams wrote:
>>
>>>"Peter Dimov" <pdimov_at_[hidden]> writes:
>>
>>>>The right thing to do here is to just fix thread_specific_ptr
>>>>(starting with its specification, if it precludes the alternative
>>>>implementation).
>>
>>FWIW, I didn't mean the alternative implementation to fix anything,
>>since I didn't consider boost::thread_specific_ptr broken when I
>>started the alternative.
>>
>>The only thing that was supposed to be "fixed" was the limitation
>>on the number of thread specific data items imposed by operating
>>systems.
>
>
> FWIW, the thread_dev branch appears to contain an implementation that
> doesn't have the limit
This is good news. Sadly, I didn't know of that fact earlier.
> (and it also has the nice alternative cleanup handler
> feature, something that I need).
Custom deleter functions would have been the next feature to suggest
for me :)
> It still has few issues, though. I wonder
> what the Boost.Threads "roadmap" is at this point.
So do I. Maybe, my suggestions for the interface specification can
be integrated into that thread_dev effort.
Regards,
m
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk