On Monday, January 7, 2002, at 02:59 PM, Perry E. Metzger wrote:

>
> Adam Lebsack <lebsack@mac.com> writes:
>> On Monday, January 7, 2002, at 02:14 PM, Perry E. Metzger wrote:
>>> Adam Lebsack <lebsack@mac.com> writes:
>>>> Well, I removed the mit-pthreads distfile and tried again.  No go.
>>>
>>> Use something other than mit-pthreads. It isn't very good. There are
>>> other threads packages in pkgsrc.
>>
>> Well, that's not a very productive way of looking at it, "it's broken,
>> so let's not fix it and use something else."  :)
>
> It isn't fixable -- it would need a rewrite. Also, NetBSD is going to
> have a native pthreads implementation using kernel assists (scheduler
> activations in this case) pretty soon, so the effort of a rewrite
> doesn't seem like a good idea.
>

OK, then.  Would you happen to know why it broke in the first place?  I 
find it hard to believe that it's not fixable unless there was something 
done to NetBSD's core codebase that was intentional.

>> Also, mysql-server has mit-pthreads grafted into it, so unless I were
>> to hack up the mysql-server build to use something else, mit-threads
>> should be fixed, and I am just wondering if anybody has any insight on
>> the problem.
>
> I'd recommend someone make it use something else. :)
>

So, what would you recommend?  Perhaps I can tackle this.  :)

Adam Lebsack
lebsack@mac.com