Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1454822
| From | Rafael Aquini <aquini@redhat.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | [PATCH] x86/mm: Add barriers and document switch_mm()-vs-flush synchronization follow-up |
| Date | 2016-08-02 17:50 +0200 |
| Message-ID | <s1K79-6pf-41@gated-at.bofh.it> (permalink) |
| Organization | linux.* mail to news gateway |
While backporting 71b3c126e611 ("x86/mm: Add barriers and document switch_mm()-vs-flush synchronization")
we stumbled across a possibly missing barrier at flush_tlb_page().
Following the reasoning presented while introducing the synchronization
barrier at flush_tlb_mm_range(), for the current->active_mm != mm checkpoint:
if (current->active_mm != mm) {
/* Synchronize with switch_mm. */
smp_mb();
goto out;
}
it suggests the same barrier should be introduced for the similar
outcome at flush_tlb_page(). This patch add that mentioned missing
barrier and documents its case.
Suggested-by: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Rafael Aquini <aquini@redhat.com>
---
arch/x86/mm/tlb.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index 4dbe656..3b4addc 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -375,6 +375,12 @@ void flush_tlb_page(struct vm_area_struct *vma, unsigned long start)
/* Synchronize with switch_mm. */
smp_mb();
}
+ } else {
+ /*
+ * current->active_mm != mm
+ * Synchronize with switch_mm.
+ */
+ smp_mb();
}
if (cpumask_any_but(mm_cpumask(mm), smp_processor_id()) < nr_cpu_ids)
--
2.5.5
Back to linux.kernel | Previous | Next — Next in thread | Find similar | Unroll thread
[PATCH] x86/mm: Add barriers and document switch_mm()-vs-flush synchronization follow-up Rafael Aquini <aquini@redhat.com> - 2016-08-02 17:50 +0200
Re: [PATCH] x86/mm: Add barriers and document switch_mm()-vs-flush synchronization follow-up Nadav Amit <nadav.amit@gmail.com> - 2016-08-03 00:30 +0200
Re: [PATCH] x86/mm: Add barriers and document switch_mm()-vs-flush synchronization follow-up Andy Lutomirski <luto@amacapital.net> - 2016-08-03 00:50 +0200
Re: [PATCH] x86/mm: Add barriers and document switch_mm()-vs-flush synchronization follow-up Rafael Aquini <aquini@redhat.com> - 2016-08-03 01:20 +0200
Re: [PATCH] x86/mm: Add barriers and document switch_mm()-vs-flush synchronization follow-up Nadav Amit <nadav.amit@gmail.com> - 2016-08-03 02:50 +0200
csiph-web