diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig
index e1924cc9687b..ff8bb3770c9d 100644
--- a/arch/ia64/configs/sn2_defconfig
+++ b/arch/ia64/configs/sn2_defconfig
@@ -113,7 +113,7 @@ CONFIG_IOSAPIC=y
 CONFIG_IA64_SGI_SN_XP=m
 CONFIG_FORCE_MAX_ZONEORDER=17
 CONFIG_SMP=y
-CONFIG_NR_CPUS=512
+CONFIG_NR_CPUS=1024
 # CONFIG_HOTPLUG_CPU is not set
 CONFIG_SCHED_SMT=y
 CONFIG_PREEMPT=y
diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index 5b7e736f3b49..028a2b95936c 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -249,3 +249,32 @@ time_init (void)
 	 */
 	set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec);
 }
+
+#define SMALLUSECS 100
+
+void
+udelay (unsigned long usecs)
+{
+	unsigned long start;
+	unsigned long cycles;
+	unsigned long smallusecs;
+
+	/*
+	 * Execute the non-preemptible delay loop (because the ITC might
+	 * not be synchronized between CPUS) in relatively short time
+	 * chunks, allowing preemption between the chunks.
+	 */
+	while (usecs > 0) {
+		smallusecs = (usecs > SMALLUSECS) ? SMALLUSECS : usecs;
+		preempt_disable();
+		cycles = smallusecs*local_cpu_data->cyc_per_usec;
+		start = ia64_get_itc();
+
+		while (ia64_get_itc() - start < cycles)
+			cpu_relax();
+
+		preempt_enable();
+		usecs -= smallusecs;
+	}
+}
+EXPORT_SYMBOL(udelay);
diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c
index c6d40446c2c4..b631cf86ed44 100644
--- a/arch/ia64/kernel/uncached.c
+++ b/arch/ia64/kernel/uncached.c
@@ -53,7 +53,7 @@ static void uncached_ipi_visibility(void *data)
 	if ((status != PAL_VISIBILITY_OK) &&
 	    (status != PAL_VISIBILITY_OK_REMOTE_NEEDED))
 		printk(KERN_DEBUG "pal_prefetch_visibility() returns %i on "
-		       "CPU %i\n", status, get_cpu());
+		       "CPU %i\n", status, raw_smp_processor_id());
 }
 
 
@@ -63,7 +63,7 @@ static void uncached_ipi_mc_drain(void *data)
 	status = ia64_pal_mc_drain();
 	if (status)
 		printk(KERN_WARNING "ia64_pal_mc_drain() failed with %i on "
-		       "CPU %i\n", status, get_cpu());
+		       "CPU %i\n", status, raw_smp_processor_id());
 }
 
 
@@ -105,7 +105,7 @@ uncached_get_new_chunk(struct gen_pool *poolp)
 	status = ia64_pal_prefetch_visibility(PAL_VISIBILITY_PHYSICAL);
 
 	dprintk(KERN_INFO "pal_prefetch_visibility() returns %i on cpu %i\n",
-		status, get_cpu());
+		status, raw_smp_processor_id());
 
 	if (!status) {
 		status = smp_call_function(uncached_ipi_visibility, NULL, 0, 1);
diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S
index 30d8564e9603..73af6267d2ef 100644
--- a/arch/ia64/kernel/vmlinux.lds.S
+++ b/arch/ia64/kernel/vmlinux.lds.S
@@ -177,6 +177,9 @@ SECTIONS
 	}
   . = ALIGN(PAGE_SIZE);		/* make sure the gate page doesn't expose kernel data */
 
+  .data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET)
+        { *(.data.read_mostly) }
+
   .data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET)
         { *(.data.cacheline_aligned) }
 
diff --git a/arch/ia64/sn/kernel/sn2/sn2_smp.c b/arch/ia64/sn/kernel/sn2/sn2_smp.c
index 5d54f5f4e926..471bbaa65d1b 100644
--- a/arch/ia64/sn/kernel/sn2/sn2_smp.c
+++ b/arch/ia64/sn/kernel/sn2/sn2_smp.c
@@ -202,7 +202,7 @@ sn2_global_tlb_purge(struct mm_struct *mm, unsigned long start,
 		     unsigned long end, unsigned long nbits)
 {
 	int i, opt, shub1, cnode, mynasid, cpu, lcpu = 0, nasid, flushed = 0;
-	int mymm = (mm == current->active_mm);
+	int mymm = (mm == current->active_mm && current->mm);
 	volatile unsigned long *ptc0, *ptc1;
 	unsigned long itc, itc2, flags, data0 = 0, data1 = 0, rr_value;
 	short nasids[MAX_NUMNODES], nix;
diff --git a/include/asm-ia64/delay.h b/include/asm-ia64/delay.h
index 57182d6f2b9a..bba702076391 100644
--- a/include/asm-ia64/delay.h
+++ b/include/asm-ia64/delay.h
@@ -84,14 +84,6 @@ __delay (unsigned long loops)
 	ia64_delay_loop (loops - 1);
 }
 
-static __inline__ void
-udelay (unsigned long usecs)
-{
-	unsigned long start = ia64_get_itc();
-	unsigned long cycles = usecs*local_cpu_data->cyc_per_usec;
-
-	while (ia64_get_itc() - start < cycles)
-		cpu_relax();
-}
+extern void udelay (unsigned long usecs);
 
 #endif /* _ASM_IA64_DELAY_H */
diff --git a/include/linux/cache.h b/include/linux/cache.h
index f6b5a46c5f82..0b7ecf3af78a 100644
--- a/include/linux/cache.h
+++ b/include/linux/cache.h
@@ -13,7 +13,7 @@
 #define SMP_CACHE_BYTES L1_CACHE_BYTES
 #endif
 
-#if defined(CONFIG_X86) || defined(CONFIG_SPARC64)
+#if defined(CONFIG_X86) || defined(CONFIG_SPARC64) || defined(CONFIG_IA64)
 #define __read_mostly __attribute__((__section__(".data.read_mostly")))
 #else
 #define __read_mostly