Here is a simple patch to use a per cpu cpumask instead of constructing
one on the stack.
Conditioned by NR_CPUS > BITS_PER_LONG as if less than or equal,
cpumask_of_cpu() generates a simple unsigned long. But the macro is
changed to generate an lvalue so a pointer to cpumask_of_cpu can be
provided.
This removes 25552 bytes of stack usage, as well as reduces the code
generated for each usage.
Based on linux-2.6.25-rc5-mm1
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
---
arch/x86/kernel/apic_32.c | 1 +
arch/x86/kernel/apic_64.c | 1 +
arch/x86/kernel/cpu/intel.c | 1 +
arch/x86/kernel/hpet.c | 1 +
arch/x86/kernel/smp_32.c | 1 +
arch/x86/kernel/smp_64.c | 1 +
arch/x86/kernel/smpcommon_32.c | 1 +
arch/x86/mach-generic/bigsmp.c | 1 +
arch/x86/mach-generic/summit.c | 1 +
include/linux/cpumask.h | 10 +++++++---
include/linux/sched.h | 5 +++++
kernel/sched.c | 10 ++++++++++
12 files changed, 31 insertions(+), 3 deletions(-)
--- linux-2.6.25-rc5.orig/arch/x86/kernel/apic_32.c
+++ linux-2.6.25-rc5/arch/x86/kernel/apic_32.c
@@ -28,6 +28,7 @@
#include <linux/acpi_pmtmr.h>
#include <linux/module.h>
#include <linux/dmi.h>
+#include <linux/sched.h>
#include <asm/atomic.h>
#include <asm/smp.h>
--- linux-2.6.25-rc5.orig/arch/x86/kernel/apic_64.c
+++ linux-2.6.25-rc5/arch/x86/kernel/apic_64.c
@@ -27,6 +27,7 @@
#include <linux/clockchips.h>
#include <linux/acpi_pmtmr.h>
#include <linux/module.h>
+#include <linux/sched.h>
#include <asm/atomic.h>
#include <asm/smp.h>
--- linux-2.6.25-rc5.orig/arch/x86/kernel/cpu/intel.c
+++ linux-2.6.25-rc5/arch/x86/kernel/cpu/intel.c
@@ -6,6 +6,7 @@
#include <linux/smp.h>
#include <linux/thread_info.h>
#include <linux/module.h>
+#include <linux/sched.h>
#include <asm/processor.h>
#include <asm/pgtable.h>
--- linux-2.6.25-rc5.orig/arch/x86/kernel/hpet.c
+++ linux-2.6.25-rc5/arch/x86/kernel/hpet.c
@@ -6,6 +6,7 @@
#include <linux/init.h>
#include <linux/sysdev.h>
#include <linux/pm.h>
+#include <linux/sched.h>
#include <asm/fixmap.h>
#include <asm/hpet.h>
--- linux-2.6.25-rc5.orig/arch/x86/kernel/smp_32.c
+++ linux-2.6.25-rc5/arch/x86/kernel/smp_32.c
@@ -19,6 +19,7 @@
#include <linux/interrupt.h>
#include <linux/cpu.h>
#include <linux/module.h>
+#include <linux/sched.h>
#include <asm/mtrr.h>
#include <asm/tlbflush.h>
--- linux-2.6.25-rc5.orig/arch/x86/kernel/smp_64.c
+++ linux-2.6.25-rc5/arch/x86/kernel/smp_64.c
@@ -18,6 +18,7 @@
#include <linux/kernel_stat.h>
#include <linux/mc146818rtc.h>
#include <linux/interrupt.h>
+#include <linux/sched.h>
#include <asm/mtrr.h>
#include <asm/pgalloc.h>
--- linux-2.6.25-rc5.orig/arch/x86/kernel/smpcommon_32.c
+++ linux-2.6.25-rc5/arch/x86/kernel/smpcommon_32.c
@@ -2,6 +2,7 @@
* SMP stuff which is common to all sub-architectures.
*/
#include <linux/module.h>
+#include <linux/sched.h>
#include <asm/smp.h>
DEFINE_PER_CPU(unsigned long, this_cpu_off);
--- linux-2.6.25-rc5.orig/arch/x86/mach-generic/bigsmp.c
+++ linux-2.6.25-rc5/arch/x86/mach-generic/bigsmp.c
@@ -14,6 +14,7 @@
#include <linux/smp.h>
#include <linux/init.h>
#include <linux/dmi.h>
+#include <linux/sched.h>
#include <asm/mach-bigsmp/mach_apic.h>
#include <asm/mach-bigsmp/mach_apicdef.h>
#include <asm/mach-bigsmp/mach_ipi.h>
--- linux-2.6.25-rc5.orig/arch/x86/mach-generic/summit.c
+++ linux-2.6.25-rc5/arch/x86/mach-generic/summit.c
@@ -13,6 +13,7 @@
#include <linux/string.h>
#include <linux/smp.h>
#include <linux/init.h>
+#include <linux/sched.h>
#include <asm/mach-summit/mach_apic.h>
#include <asm/mach-summit/mach_apicdef.h>
#include <asm/mach-summit/mach_ipi.h>
--- linux-2.6.25-rc5.orig/include/linux/cpumask.h
+++ linux-2.6.25-rc5/include/linux/cpumask.h
@@ -226,8 +226,11 @@ int __next_cpu(int n, const cpumask_t *s
#define next_cpu(n, src) ({ (void)(src); 1; })
#endif
+#if NR_CPUS > BITS_PER_LONG
+#define cpumask_of_cpu(cpu) per_cpu(cpu_mask, cpu)
+#else
#define cpumask_of_cpu(cpu) \
-({ \
+(*({ \
typeof(_unused_cpumask_arg_) m; \
if (sizeof(m) == sizeof(unsigned long)) { \
m.bits[0] = 1UL<<(cpu); \
@@ -235,8 +238,9 @@ int __next_cpu(int n, const cpumask_t *s
cpus_clear(m); \
cpu_set((cpu), m); \
} \
- m; \
-})
+ &m; \
+}))
+#endif
#define CPU_MASK_LAST_WORD BITMAP_LAST_WORD_MASK(NR_CPUS)
--- linux-2.6.25-rc5.orig/include/linux/sched.h
+++ linux-2.6.25-rc5/include/linux/sched.h
@@ -2130,6 +2130,11 @@ static inline void migration_init(void)
#define TASK_STATE_TO_CHAR_STR "RSDTtZX"
+#if NR_CPUS > BITS_PER_LONG
+/* for cpumask_of_cpu() */
+DECLARE_PER_CPU(cpumask_t, cpu_mask);
+#endif
+
#endif /* __KERNEL__ */
#endif
--- linux-2.6.25-rc5.orig/kernel/sched.c
+++ linux-2.6.25-rc5/kernel/sched.c
@@ -7198,6 +7198,12 @@ static void init_tg_rt_entry(struct rq *
}
#endif
+#if NR_CPUS > BITS_PER_LONG
+/* for cpumask_of_cpu() */
+DEFINE_PER_CPU(cpumask_t, cpu_mask);
+EXPORT_PER_CPU_SYMBOL(cpu_mask);
+#endif
+
void __init sched_init(void)
{
int i, j;
@@ -7242,6 +7248,10 @@ void __init sched_init(void)
for_each_possible_cpu(i) {
struct rq *rq;
+#if NR_CPUS > BITS_PER_LONG
+ /* This makes cpumask_of_cpu() work */
+ cpu_set(i, per_cpu(cpu_mask, i));
+#endif
rq = cpu_rq(i);
spin_lock_init(&rq->lock);
lockdep_set_class(&rq->lock, &rq->rq_lock_key);
--
--
