Hi,

I am trying to get Mono 2.4 running on a SGI Altix 4700 (96-way Itanium 2
box) running SUSE Linux Enterprise Server 10.  It compiles successfully
using default configuration settings however upon testing it fails at simple
locking (n threads incrementing a shared counter produces race conditions). 
I have tried;

  -> Changing "--with-tls" between both "__thread" and "pthread"
  -> Changing CFLAGS, CCASFLAGS, CXXFLAGS and FFLAGS to exclude
optimizations (removed -O2)
  -> Altering versions of glib (between glib-2.16.5 and glib-2.20.5) - there
is a gthread dependence but I am not sure how that impacts matters.

Unfortunately none of the above made any difference.  Is anybody out there
running Mono on anything similar hardware wise (IA-64 at least)?  Any ideas? 
Any help would be greatly appreciated.

It should be noted that this all works fine on a 8-way AMD64 box with the
test program producing correct results (to exclude test program as a
possible problem).

Some more details:
  Using: glib-2.20.5, libunwind-0.99, gcc version 4.1.2 20070115 (SUSE
Linux)

Test App "core" (C#) - n threads executing Worker() followed by join:

public static void Worker()
{
  int steps = increments / n_threads;
  for (int i = 0; i < steps; i++)
  {
    lock (shared_lock)
    {
       shared_counter++;
    }
  }
}

Test App results:

u...@host:~/code/tests> mono SharedCounter.exe 1000000 10
1000000 increments over 10 threads
Counter value: 999023

u...@host:~/code/tests> mono SharedCounter.exe 1000000 10
1000000 increments over 10 threads
Counter value: 999087

u...@host:~/code/tests> mono SharedCounter.exe 1000000 10
1000000 increments over 10 threads
Counter value: 999211

u...@host:~/code/tests> mono SharedCounter.exe 1000000 10
1000000 increments over 10 threads
Counter value: 999219

Please let me know if you need any additional information. Thanks in
advance,

Darryl
-- 
View this message in context: 
http://www.nabble.com/Mono-2.4-on-IA-64-SGI-Altix-4700-tp26123445p26123445.html
Sent from the Mono - Dev mailing list archive at Nabble.com.

_______________________________________________
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list

Reply via email to