Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1305073
| From | Andy Lutomirski <luto@kernel.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | [RFC 10/13] x86/mm: Factor out remote TLB flushing |
| Date | 2016-01-09 00:20 +0100 |
| Message-ID | <qOOu7-5ka-33@gated-at.bofh.it> (permalink) |
| References | <qOOu6-5ka-5@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
There are three call sites that propagate TLB flushes, and they all
do exactly the same thing. Factor the code out into a helper.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
---
arch/x86/mm/tlb.c | 17 +++++++++++------
1 file changed, 11 insertions(+), 6 deletions(-)
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index 8f4cc3dfac32..b208a33571b0 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -154,6 +154,14 @@ void native_flush_tlb_others(const struct cpumask *cpumask,
smp_call_function_many(cpumask, flush_tlb_func, &info, 1);
}
+static void propagate_tlb_flush(unsigned int this_cpu,
+ struct mm_struct *mm, unsigned long start,
+ unsigned long end)
+{
+ if (cpumask_any_but(mm_cpumask(mm), this_cpu) < nr_cpu_ids)
+ flush_tlb_others(mm_cpumask(mm), mm, 0UL, TLB_FLUSH_ALL);
+}
+
void flush_tlb_current_task(void)
{
struct mm_struct *mm = current->mm;
@@ -166,8 +174,7 @@ void flush_tlb_current_task(void)
local_flush_tlb();
trace_tlb_flush(TLB_LOCAL_SHOOTDOWN, TLB_FLUSH_ALL);
- if (cpumask_any_but(mm_cpumask(mm), smp_processor_id()) < nr_cpu_ids)
- flush_tlb_others(mm_cpumask(mm), mm, 0UL, TLB_FLUSH_ALL);
+ propagate_tlb_flush(smp_processor_id(), mm, 0UL, TLB_FLUSH_ALL);
preempt_enable();
}
@@ -231,8 +238,7 @@ out:
start = 0UL;
end = TLB_FLUSH_ALL;
}
- if (cpumask_any_but(mm_cpumask(mm), smp_processor_id()) < nr_cpu_ids)
- flush_tlb_others(mm_cpumask(mm), mm, start, end);
+ propagate_tlb_flush(smp_processor_id(), mm, start, end);
preempt_enable();
}
@@ -257,8 +263,7 @@ void flush_tlb_page(struct vm_area_struct *vma, unsigned long start)
}
}
- if (cpumask_any_but(mm_cpumask(mm), smp_processor_id()) < nr_cpu_ids)
- flush_tlb_others(mm_cpumask(mm), mm, start, 0UL);
+ propagate_tlb_flush(smp_processor_id(), mm, start, 0UL);
preempt_enable();
}
--
2.5.0
Back to linux.kernel | Previous | Next | Find similar | Unroll thread
[RFC 10/13] x86/mm: Factor out remote TLB flushing Andy Lutomirski <luto@kernel.org> - 2016-01-09 00:20 +0100
csiph-web