path: root/arch/s390/kernel/smp.c
diff options
authorMartin Schwidefsky <schwidefsky@de.ibm.com>2014-04-03 13:55:01 +0200
committerMartin Schwidefsky <schwidefsky@de.ibm.com>2014-04-03 14:31:00 +0200
commit1b948d6caec4f28e3524244ca0f77c6ae8ddceef (patch)
treebc7e1d5800f10c39979d3f47872ba7047568f8a4 /arch/s390/kernel/smp.c
parent02a8f3abb708919149cb657a5202f4603f0c38e2 (diff)
s390/mm,tlb: optimize TLB flushing for zEC12
The zEC12 machines introduced the local-clearing control for the IDTE and IPTE instruction. If the control is set only the TLB of the local CPU is cleared of entries, either all entries of a single address space for IDTE, or the entry for a single page-table entry for IPTE. Without the local-clearing control the TLB flush is broadcasted to all CPUs in the configuration, which is expensive. The reset of the bit mask of the CPUs that need flushing after a non-local IDTE is tricky. As TLB entries for an address space remain in the TLB even if the address space is detached a new bit field is required to keep track of attached CPUs vs. CPUs in the need of a flush. After a non-local flush with IDTE the bit-field of attached CPUs is copied to the bit-field of CPUs in need of a flush. The ordering of operations on cpu_attach_mask, attach_count and mm_cpumask(mm) is such that an underindication in mm_cpumask(mm) is prevented but an overindication in mm_cpumask(mm) is possible. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Diffstat (limited to 'arch/s390/kernel/smp.c')
1 files changed, 6 insertions, 0 deletions
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 366d14460c2b..42a501d13a3b 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -236,6 +236,9 @@ static void pcpu_prepare_secondary(struct pcpu *pcpu, int cpu)
struct _lowcore *lc = pcpu->lowcore;
+ cpumask_set_cpu(cpu, &init_mm.context.cpu_attach_mask);
+ cpumask_set_cpu(cpu, mm_cpumask(&init_mm));
lc->cpu_nr = cpu;
lc->percpu_offset = __per_cpu_offset[cpu];
@@ -760,6 +763,9 @@ void __cpu_die(unsigned int cpu)
+ cpumask_clear_cpu(cpu, mm_cpumask(&init_mm));
+ cpumask_clear_cpu(cpu, &init_mm.context.cpu_attach_mask);
void __noreturn cpu_die(void)