Re: [PATCH] Change configure to check if pthreads are usable without any extra flags

From: Max Horn <max@quendi.de>
Date: Jul 9, 2012, 21:44 UTC
Message-ID: <84D9664E-3249-4D80-980A-C01B02DF5E43@quendi.de>
In-Reply-To: <7vtxxgkac0.fsf@alter.siamese.dyndns.org>

On 09.07.2012, at 21:23, Junio C Hamano wrote:

Show 13 quoted lines

> Max Horn <max@quendi.de> writes:
> 
>>> would it be feasible for the purpose of
>>> the check to tweak the definition of "works" used in the loop so that
>>> it considers the warning as "not working"?
>> 
>> That would be possible, and probably a good idea. But it is also
>> completely orthogonal to my patch. Indeed, if done without my
>> patch,...
> 
> No, I was suggesting it as a possible way to make the addition of ""
> order independent (which you said is impossible in your initial
> reply).

This would make things "order independent" for the specific subset of pthread implementations git supports right now. But there are systems where e.g. both -lpthreads and -lpthread work (link correctly, produce no warnings), but only one provides a POSIX compliant pthread implementation. For such systems, order will play a role, no matter what. Granted, git does not yet support such systems (with regards to pthreads, at least) at all.

But all in all, I don't understand why this order independence seems to be so important? To me it seems merely an illusion anyway, not worth the extra effort.