[GHC] #7356: Building HEAD on Mac OS X 10.5 reports Undefined "_pthread_threadid_np" ref'd from _kernelThreadId in libHSrts_thr.a(OSThreads.thr_o)

[GHC]

#7356: Building HEAD on Mac OS X 10.5 reports Undefined "_pthread_threadid_np"
ref'd from _kernelThreadId in libHSrts_thr.a(OSThreads.thr_o)
---------------------------------+------------------------------------------
 Reporter:  thorkilnaur          |          Owner:                
     Type:  bug                  |         Status:  new           
 Priority:  normal               |      Component:  Runtime System
  Version:  7.7                  |       Keywords:                
       Os:  MacOS X              |   Architecture:  x86           
  Failure:  Building GHC failed  |       Testcase:                
Blockedby:                       |       Blocking:                
  Related:                       |  
---------------------------------+------------------------------------------
Changes (by spl):

 * cc: leather@… (added)


Comment:

 Thanks for reporting this. I sent an email to ghc-cvs and Duncan Coutts on
 Oct 3 about this, but now that I have tried to search for a link to that
 email, I realize that I was using the wrong "from" email address.

 I traced the problem to the following commit:

 {{{
 commit 647ae1cfbb5ea3e2d3b1541c2bc12ea5db321134
 Author: Duncan Coutts
 Date:   Tue Jul 3 19:18:46 2012 +0100
 }}}

 I wonder if it's possible to modify kernelThreadId() to do something a bit
 more portable. I've seen references to the following options:

  1. [http://elliotth.blogspot.nl/2012/04/gettid-on-mac-os.html
 pthread_self() (using the pointer)]
  1. [http://stackoverflow.com/questions/1540603/mac-iphone-is-there-a-way-
 to-get-a-thread-identifier-without-using-objective-c
 pthread_mach_thread_np(pthread_self())]

 I'm not sure if either one is equivalent or "better." I noticed
 pthread_self() is already used for osThreadId(). I have changed it locally
 to the second option, which compiles, but I haven't attempted to test
 threads.

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/7356#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler