Merge pull request #427 from wernsaar/develop

added experimental support for big numa machines

Author: xianyi

Makefile.rule

@@ -95,6 +95,9 @@ NO_WARMUP = 1
 # If you want to disable CPU/Memory affinity on Linux.
 NO_AFFINITY = 1
 
+# if you are compiling for Linux and you have more than 16 numa nodes or more than 256 cpus
+# BIGNUMA = 1
+
 # Don't use AVX kernel on Sandy Bridge. It is compatible with old compilers
 # and OS. However, the performance is low.
 # NO_AVX = 1

Makefile.system

@@ -803,6 +803,10 @@ ifeq ($(USE_OPENMP), 1)
 CCOMMON_OPT	+= -DUSE_OPENMP
 endif
 
+ifeq ($(BIGNUMA), 1)
+CCOMMON_OPT	+= -DBIGNUMA
+endif
+
 endif
 
 ifeq ($(NO_WARMUP), 1)

driver/others/init.c

@@ -85,8 +85,16 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 #include <unistd.h>
 #include <string.h>
 
+#if defined(BIGNUMA)
+// max number of nodes as defined in numa.h
+// max cpus as defined in sched.h
+#define MAX_NODES	128
+#define MAX_CPUS	CPU_SETSIZE
+#else
 #define MAX_NODES	16
 #define MAX_CPUS	256
+#endif
+
 #define NCPUBITS        (8*sizeof(unsigned long))
 #define MAX_BITMASK_LEN (MAX_CPUS/NCPUBITS)
 #define CPUELT(cpu)	((cpu) / NCPUBITS)
@@ -544,16 +552,26 @@ static inline int is_dead(int id) {
 
   return shmctl(id, IPC_STAT, &ds);
 }
+
 static void open_shmem(void) {
 
   int try = 0;
 
   do {
 
+#if defined(BIGNUMA)
+    // raised to 32768, enough for 128 nodes and 1024 cups
+    shmid = shmget(SH_MAGIC, 32768, 0666);
+#else
     shmid = shmget(SH_MAGIC, 4096, 0666);
+#endif
 
     if (shmid == -1) {
+#if defined(BIGNUMA)
+      shmid = shmget(SH_MAGIC, 32768, IPC_CREAT | 0666);
+#else
       shmid = shmget(SH_MAGIC, 4096, IPC_CREAT | 0666);
+#endif
     }
 
     try ++;